instruction stringlengths 34 186 | input stringlengths 2.02k 93.8k | output stringlengths 2 418 | meta_questiontype stringclasses 6 values | meta_inputlicense stringclasses 6 values | meta_pmid stringlengths 8 8 | meta_safetyreportid int64 9.51M 21M | meta_articlepubdate stringlengths 4 10 |
|---|---|---|---|---|---|---|---|
What was the administration route of drug 'LORAZEPAM'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Oral | DrugAdministrationRoute | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the administration route of drug 'TEMOZOLOMIDE'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Oral | DrugAdministrationRoute | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Alanine aminotransferase increased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Aspartate aminotransferase increased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Asthenia'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Blood lactate dehydrogenase increased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Chest pain'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovering | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Condition aggravated'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Depressed mood'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Drug interaction'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Drug-induced liver injury'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Dyspepsia'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Fatigue'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Gamma-glutamyltransferase increased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Glioblastoma multiforme'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Hepatic enzyme increased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Hepatotoxicity'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Monocyte count decreased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Nausea'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'Weight decreased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
What was the outcome of reaction 'White blood cell count decreased'? | Drug repurposing using transcriptome sequencing and virtual drug screening in a patient with glioblastoma.
Background Precision medicine and drug repurposing are attractive strategies, especially for tumors with worse prognosis. Glioblastoma is a highly malignant brain tumor with limited treatment options and short survival times. We identified novel BRAF (47-438del) and PIK3R1 (G376R) mutations in a glioblastoma patient by RNA-sequencing. Methods The protein expression of BRAF and PIK3R1 as well as the lack of EGFR expression as analyzed by immunohistochemistry corroborated RNA-sequencing data. The expression of additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) emphasized the aggressiveness of the tumor. Then, we screened a chemical library of > 1500 FDA-approved drugs and > 25,000 novel compounds in the ZINC database to find established drugs targeting BRAF47-438del and PIK3R1-G376R mutated proteins. Results Several compounds (including anthracyclines) bound with higher affinities than the control drugs (sorafenib and vemurafenib for BRAF and PI-103 and LY-294,002 for PIK3R1). Subsequent cytotoxicity analyses showed that anthracyclines might be suitable drug candidates. Aclarubicin revealed higher cytotoxicity than both sorafenib and vemurafenib, whereas idarubicin and daunorubicin revealed higher cytotoxicity than LY-294,002. Liposomal formulations of anthracyclines may be suitable to cross the blood brain barrier. Conclusions In conclusion, we identified novel small molecules via a drug repurposing approach that could be effectively used for personalized glioblastoma therapy especially for patients carrying BRAF47-438del and PIK3R1-G376R mutations.
Background
Half of the brain tumors represent diffuse gliomas, and the World Health Organization (WHO) provided classification criteria to predict the clinical behavior of these neoplasms [1]. Glioblastoma is classified as grade IV/IV gliomas with specific characteristics, including high cellularity, cellular pleomorphism, nuclear atypia and necrosis [2]. Glioblastoma has a dismal prognosis despite aggressive treatment [2, 3]. Various genetic mutations and aberration have been identified in glioblastoma patients, such as MGMT methylation [3, 4], BRAFG596A [5], BRAFV600E [5–7], PIK3R1G376R, PIK3R1D560Y, PIK3R1N564K mutations [8].
Mutations in proteins critical for cancer biology usually lead to uncontrolled cell growth, resistance to conventional chemotherapy and subsequently, therapy failure. In the past decade, the demand for effective novel agents to combat cancer has drawn the attention to specific molecular alterations in cancer cells as targets for therapy [9]. The concept of precision medicine implies the application of targeted drugs coupled with specific diagnostic assays in order to determine whether patients are likely to benefit from targeted therapy. The shift from classical, non-selective, cytotoxic chemotherapy to molecular targeted cancer drugs resulted in increased tumor response and patients’ survival rates [10–12].
Therapeutic monoclonal antibodies are considered a successful strategy for targeted cancer therapy. Antibodies have, however, several limitations, including targeting only cellular surface epitopes, high immunogenicity and high production costs [13]. Small molecule inhibitors possess advantages compared to antibodies, as they address both intracellular and surface proteins. A showcase example is the BCR-ABL inhibitor imatinib, which resulted in dramatic improvements in the survival of chronic myeloid leukemia patients. The same applies to small molecule inhibitors directed against targets in solid tumors, e.g. the epidermal growth factor receptor (EGFR) kinase inhibitors gefitinib and erlotinib to treat non-small cell lung cancer and the vascular endothelial growth factor receptor (VEGFR) kinase inhibitor sorafenib against renal cancer [14–16]. In the case of EGFR mutations in the kinase domain of the receptor, it happens that erlotinib is no longer therapeutically effective as cancer cells develop resistance to erlotinib. Numerous mutations appear during tumor progression and cause resistance [17, 18]. Therefore, it is essential to find novel inhibitors, which target and inhibit tumor-related mutant proteins.
The sequencing of tumor genomes and transcriptomes is more and more routinely applied in clinical oncology. The concept of precision medicine by mutations identified by sequence analyses may serve as a basis to select treatment options specifically addressing these mutations. Since such mutations would exclusively occur in tumors but not in normal tissues, it is expected that targeted treatments ought to provoke no or only minimal side effects. However, the vast majority of the data acquired cannot be used for therapeutic purposes yet, as we still lack drugs addressing all relevant mutations. Also, tumors frequently consist of heterogeneous subpopulations with mutational profiles different from the main cell population. Resistance to a targeted drug develop if subpopulations without the treatment-specific mutation appear. This may foster the fatal outcome of malignant diseases. Hence, one wishes to have a larger arsenal of drugs at hand to combat otherwise drug-resistant subpopulations of tumors with mutations, for which no targeted drugs are available currently.
One approach to address this latter problem is to develop individualized tumor vaccination to target mutated tumor surface proteins of individual patients. With the advent of advanced vaccination technologies, it is possible to generate individual vaccines to treat each patient according to the tumor’s individual mutational profile [19, 20]. This approach is difficult to achieve for intracellularly located proteins with tumor-specific mutations since antibodies mainly address extracellular rather than intracellular epitopes in living cells (although endocytic antibody internalization may take place). Therefore, therapeutic strategies are required to address intracellular tumor proteins, which constitute a considerable – if not the largest – portion of the tumor proteome.
In an endeavor to device new strategies for precision medicine based on small molecules to attack mutated intracellular proteins, we developed a concept, which integrates transcriptomic data with virtual drug screening of chemical libraries.
A central element in this context may be the concept of drug repurposing. Drug repurposing is a promising strategy and based on the successful usage of already approved drugs, which were initially developed for other indications than cancer [21]. One of the advantages of these drugs is that they already passed clinical safety testing in clinical trials. Considerable investments in terms of money and manpower are being spent to generate clinical evidence of safety and tolerability of a new drug to fulfill the high standards of the drug-approving authorities. Hence, drug repurposing is attractive from economic as well as medical points of view. The identification of thalidomide against severe erythema nodosum leprosum and retinoic acid against acute promyelocytic leukemia are two successful examples of drug repositioning [22, 23]. Plerixafor was initially developed as HIV drug to block viral entry into the cell, but clinical trials failed. However, leukocytosis in peripheral blood CD34 hematopoietic stem cells led to repurposing of plerixafor as stem cell mobilizing drug [24]. Our group has recently reported on drug repurposing of the anti-malarial artesunate for cancer therapy [25], e.g. prostate carcinoma [26] and colorectal cancer [27] as well as for other diseases such as viral infections [28] and schistosomiasis [29]. In addition, we found that the anti-fungal niclosamide exerted cytotoxic activity towards multidrug-resistant leukemia [30].
In the present study, we identified novel mutations in a glioblastoma patient and first screened a chemical library of more than 1500 FDA-approved drugs to find established drugs, which target novel BRAF and PIK3R1 mutations. Furthermore, we screened more than 25,000 novel compounds from the ZINC database. Then, the cytotoxicity of the selected compounds was evaluated. Furthermore, the protein expression of BRAF, PIK3R1 and other cancer biomarker proteins in the brain tumor tissue obtained from the patient was analyzed by immunohistochemistry. Finally, we identified novel small molecules that effectively targeted cells carrying mutant BRAF and PIK3R1, implying their potential for personalized glioblastoma therapy.
Methods
Patient characteristics
The patient characteristics were recently described [31]. In brief, a 65-year old patient had been diagnosed with glioblastoma WHO grade IV on May 26th, 2014. Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016). The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063). Temozolomide-based radiochemotherapy had been applied according to the current standard of care [32]. Treatment led to stable disease until May 2017, when surgery of the progressive tumor became necessary.
Transcriptome sequencing of tumor and normal tissue was performed at the National Center for Tumor Diseases (NCT, Heidelberg, Germany), which confirmed the histological diagnosis of glioblastoma. In total, 65 nucleotide substitutions, three focal and 17 larger copy number changes, as well as MGMT promoter methylation was found.
Magnetic resonance imaging (MRI) and computer tomography (CT)
Tumor development at the glioblastoma patient within 1 year (before, during and after temozolomide treatment) can be observed from the MRI scans depicted in Fig. 1. Tumor shrinkage occurred after temozolomide treatment.Fig. 1 Computed tomography (CT) scans and magnetic resonance images (MRI) before, during and after the temozolomide treatment of the glioblastoma patient
In silico analyses
A library of FDA-approved drugs (> 1,500 compounds) (http://zinc15.docking.org/substances/subsets/fda/) was screened for established drugs binding with high affinity towards wildtype BRAF, BRAF47-438del and PIK3R1G376R. The PyRx software [33] was used for initial virtual drug screening. The 10 top-ranked compounds with the highest affinities were selected for further analyses. As a second step, the ZINC database library [34] was used (> 25,000 compounds from the “all clean subset with pH 7”) to identify novel non-approved investigational compounds. Again, the 10 top-ranked compounds with the highest affinities towards BRAF47-438del and the 10 top-ranked compounds with the highest affinities towards PIK3R1G376R were selected for further analyses. The selected candidate compounds were further subjected to molecular docking with AutoDock 4 [35]. Whole protein surfaces were taken into account for virtual screening and molecular docking. Three independent docking calculations were conducted, with 25,000,000 energy evaluations and 250 runs by using the Lamarckian Genetic Algorithm. The corresponding lowest binding energies (LBE) were obtained from the docking log files (dlg), and mean values ± SD were calculated. For visualization of docking results, AutoDock Tools and Visual Molecular Dynamics (VMD) were used (Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign) (http://www.ks.uiuc.edu/Research/vmd/). Two known BRAF inhibitors (sorafenib [36] and vemurafenib [37]), two known PIK3R1 inhibitors (PI-103 [38] and LY-294,002 [39]) were used as control drugs.
Immunohistochemistry
For immunohistochemical protein detection, we applied the UltraVision polymer detection method (kit from Thermo Fisher Scientific GmbH, Dreieich, Germany). The method was previously described in detail [26]. Briefly, formalin-fixed, and paraffin-embedded tumor tissue was incubated in a humidified chamber for 1 h at room temperature with primary antibodies against BRAF (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), PI3KR1 (1:100, clone U5, Life Technologies GmbH, Darmstadt, Germany), EGFR (1:50, clone H11, Life Technologies GmbH, Darmstadt, Germany), SRC (1:200, clone 1F11, Life Technologies GmbH, Darmstadt, Germany), AKT (1:100, clone 9Q7, Life Technologies GmbH, Darmstadt, Germany), mTOR (1:100, clone F11, Life Technologies Europe BV, Bleiswijk, Netherlands), NF-κB (1:100, polyclonal, Life Technologies GmbH, Darmstadt, Germany), or Ki-67 (ab16667, dilution 1:100, Abcam, Cambridge, UK). Afterwards, Primary Antibody Amplifier Quanto (Thermo Scientific) was applied for 10 min at room temperature, and HRP Polymer Quanto (Thermo Scientific) was applied for another 10 min. Protein expression was visualized by diaminobenzidine (DAB). Quanto chromogen (Thermo Scientific) was mixed with 1 ml DAB Quanto. The tissues were counterstained in hemalaun solution (Merck KGaA, Darmstadt, Germany). Standard histochemical staining was performed with hematoxylin-eosin (HE).
The immunostained slides were scanned by Panoramic Desk (3D Histotech Pannoramic digital slide scanner, Budapest, Ungary) and quantified by panoramic viewer software (NuclearQuant and MembraneQuant, 3D HISTECH) as previously described [26].
NCI cell lines
The panel of cell lines of the Developmental Therapeutics Program (National Cancer Institute, USA) consists of 7 brain tumor cell lines (SF-295, SF-539, SNB-19, SNB-75, SNB-78, U251, XF498) and of cell lines from other tumor origins (leukemia, melanoma as well as carcinoma of the breast, ovary, prostate, lung, colon, and kidney). The origin of the cell lines has been described [40]. Up to 70 cell lines have been tested per drug using a sulforhodamine assay [41]. Some of these drugs identified by our virtual drug screening approach have been tested in this panel of cell lines, and the log10IC50 values have been deposited at the NCI website (https://dtp.cancer.gov/) [42].
Cytotoxicity
Fifty per cent inhibitory concentrations (IC50) values for the selected compounds from virtual screening of FDA approved drugs towards the brain cancer cell line panel of the NCI cell lines were selected from the National Cancer Institute (NCI) database (http://dtp.nci.nih.gov). The cytotoxicity for the NCI cell line panel was assayed by the sulforhodamine B test, as previously described [43].
Results
Genotyping
The copy number variation plot with selected mutations and variations obtained from tumor genome sequencing of a glioblastoma patient is depicted in Fig. 2. Specific mutations have been identified in BRAF (BRAF47-438del, BRAF-TTYH3 fusion), PIK3R1, MAP3K4, MAP3K5, and PTK2. The promoter of MGMT was methylated. Additionally, several genes were completely deleted, i.e. ABCA13, ABCB4, CDK13, EGFR, EIF4H, and HGF).
Fig. 2 Copy number variation plot and the relevant mutations in the glioblastoma patient
Immunohistochemistry
In order to confirm the relevance of the results of RNA-sequencing, we performed immunohistochemistry for selected markers, i.e. BRAF, PIK3R1, and EGFR. In addition, signaling molecules in the EGFR downstream signaling cascade have been investigated, e.g. kinases (AKT, SRC) and transcription factors (mTOR, NF-κB). Furthermore, general tumor features have been investigated by hematoxilin-eosin staining to monitor the typical glioblastoma multiforme morphology and by the immunohistochemical detection of Ki-67 to estimate the proliferation rate of the tumor. As shown in Fig. 3, all tumor markers except EGFR revealed strong immunopositivity. The genotyping data revealed EGFR whole gene deletion and this is confirmed by EGFR staining. The protein expression of BRAF and PIK3R1, as well as the lack of EGFR expression, fit well to the data obtained by RNA-sequencing. The expression of the additional markers (AKT, SRC, mTOR, NF-κB, Ki-67) are clues for the aggressiveness of the tumor.
Fig. 3 Immunohistochemical detection of BRAF, PIK3R1, EGFR, AKT, SRC, mTOR, NF-kB, Bar, 50 µM. HE, hematoxylin-eosin staining; NC, negative control (w/o primary antibody)
In silico analyses
In order to search for novel treatment options based on individual mutational profiles of glioblastoma genomes, we used the mutations of this patient for further bioinformatical analyses. We focused on the BRAF47-438del and PIK3R1G376R mutations because genetic alterations in these two genes are frequently assumed as driver mutations with relevance for tumor development and progression. The other mutations in the MAP3K4, MSP3K5 and PTK2 genes were not further considered.
The BRAF47-438del is a novel mutation found in this patient for the first time. Therefore, we compared this novel mutation with the most common BRAFV600E mutation. The BRAF-TTYH3 fusion protein could not be used for virtual drug screening, as no crystal structure of this protein was available in the Protein Data Bank (https://www.rcsb.org/). Therefore, a reliable homology model of this fusion protein could not be generated on the basis of the single wildtype BRAF and TTYH3 proteins.
Using PyRx and AutoDock 4.2, virtual screening of the library of FDA-approved drugs revealed drugs with high affinities towards BRAF47-438del and PIK3R1G376R (Table 1). The docking poses of top 10 compounds as well as of vemurafenib and sorafenib (control drugs for BRAF), LY-294,002 and PI-103 (control drugs for PIK3R1) are visualized in Fig. 4. With regard to the PyRx screening results, all 10 compounds revealed a stronger affinity to BRAF47-438del than vemurafenib and sorafenib. Concerning PIK3R1G376R, the top 10 compounds revealed stronger interaction than LY-294,002. All top 10 compounds except tubocurarine, metocurine and idarubicin possess higher affinity than PI-103. Concerning the BRAF47-438del AutoDock results, STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol revealed stronger interaction than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also have stronger affinity than vemurafenib. Within the compounds revealed by PIK3R1G376R docking results, all top 10 compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002. The established anti-BRAF drug sorafenib bound with slightly less affinity to BRAF47-438del (-9.39 ± 0.09 vs. -9.57 ± 0.22 kcal/mol) than to the corresponding wildtype protein as observed in molecular docking analyses. The established anti-PIK3R1 inhibitors; LY-294,002 and PI-103 revealed stronger binding to PIK3R1G376R mutant protein than wildtype protein (-6.47 ± < 0.01 vs. -5.19 ± 0.02 for LY-294,002 and − 7.22 ± 0.01 vs -5.76 ± 0.06 kcal/mol for PI-103). Accordingly, it was possible to identify drugs that bind stronger to BRAF47-438del than sorafenib and vemurafenib and drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103, all of those can specifically target mutant proteins.Table 1 Top 10 out of > 1500 FDA-approved established drugs with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
Tubocurarine -11.0 -8.61 ± 0.01 0.48 ± < 0.01 Met1, Ala3, Ser5, Gly11, Ala12, Glu13, Gly15, Leu18, Gly21, Asp22, Glu26
STK396645 -10.9 -13.78 ± 0.45 0.0001 ± < 0.001 Lys288, Ala296, Lys298, Arg299, Lys291, Cys293, Pro294, Lys295, Leu329, Pro330
Celecoxib -10.7 -8.94 ± 0.01 0.280 ± 0.004 Ser340, Arg343, Glu153, Phe156, Met158, Leu161, Thr259, Asn292, Glu338
Aclarubicin -10.6 -8.39 ± 0.81 1.09 ± 0.88 Ser222, Tyr368, Ser75, Phe76, Ala105, Asn108, Asp184, Lys186, Phe203, Gly204, Leu205, Ala206, Thr207, Gln220, Ala370, Val373
Pimozide -10.4 -10.55 ± 0.27 0.02 ± 0.01 Gly223, Lys186, Leu221, Ser222, Gly223, Ser224, Ile225, Met228, Leu262, Arg270, Ile363, Ala365, Ala370, Phe371, Val373, His374
Folidan -10.3 -10.98 ± 0.30 0.009 ± 0.004 Lys288, Ser291, Lys298, Arg299, Asn292, Cys293, Pro294, Lys295, Arg334, Ser335
Acetophenone -10.2 -10.48 ± 0.04 0.02 ± 0.02 Ser340, Arg343, Phe156, Glu157, Met158, Leu161, Met258, Gln261, Leu286, Asn292, Glu338, Pro339, Leu341
LS-194,959 -10.2 -13.34 ± 0.21 0.0002 ± < 0.001 Lys288, Ser291, Lys295, Lys298, Arg299, Cys293, Pro294, Ala296
Danazol -10.2 -9.41 ± < 0.01 0.126 ± 0.0004 His150, Glu153, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Leu341, Asn342
Triamterene -10.1 -7.69 ± 0.01 2.29 ± 0.01 Leu149, His150, Met258, Glu153, Phe156, Glu157, Met158, Leu161, Asn292, Glu338, Leu341
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
LS-194,959 -9.9 -8.07 ± 0.05 1.22 ± 0.10 Arg40, Lys79, Leu80, Ile81, Lys82, Tyr116
STK396645 -9.7 -9.66 ± 0.16 0.08 ± 0.02 Asp37, Lys130, Ile142, Met282, Thr285, Gln286, Lys287, Gly288, Val289, Arg290
Carminomycin -9.3 -7.36 ± 0.51 5.21 ± 4.75 Glu143, Gly146, Leu149, His150, Leu284, Thr285, Val289, Gln291
Albamycin -9.3 -5.97 ± 0.07 42.39 ± 4.96 Lys275, Thr285, Trp35, Ile38, Glu42, Lys46, Val128, Gln132, Asp278, Gln279, Leu281, Met282,
Gliquidone -9.2 -7.38 ± 0.59 5.56 ± 5.67 Trp35, Lys46, Thr50, Ala51, Thr54, Tyr126, Pro127, Val128, Lys130, Gln132, Gln133, Gln279, Met282
Fazadon -9.2 -7.73 ± 0.06 2.15 ± 0.19 Glu143, Gly146, Leu149, Asn153, Leu284, Val289, Lys293
Daunorubicin -9.1 -6.40 ± 0.67 27.83 ± 19.78 Asn153, Gly146, Leu149, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Tubocurarine -9.0 -6.72 ± 0.02 11.76 ± 0.38 Leu149, Glu42, Arg277, Leu281, Leu284, Thr285, Gln291, Lys293
Metocurine -8.9 -7.30 ± 0.01 4.48 ± 0.03 Gly146, Leu149, His150, Asn153, Arg277, Leu281, Leu284, Val289, Gln291
Idarubicin -8.8 -7.32 ± 0.76 7.61 ± 9.51 Trp35, Asp37, Glu42, Lys46, Leu281, Met282, Thr285
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 4 Docking poses of the top 10 ranked FDA-approved established drugs on the BRAF47-438del and PIK3R1G376R mutant proteins
Furthermore, 10 investigational non-approved compounds from the ZINC database were identified for BRAF47-438del and PIK3R1G376R mutated proteins (Table 2). Residues forming hydrogen bonds were labeled in bold. Docking poses are depicted in Fig. 5. With regards to the PyRx results, all compounds revealed stronger interaction than the control compounds. BRAF47-438del AutoDock results pointed out that all compounds interact stronger than sorafenib. ZINC09339473, ZINC22798105, ZINC33068262, ZINC00691692, ZINC08667624, ZINC33068261, ZINC09219428, ZINC31840966 interact stronger than both vemurafenib and sorafenib. PIK3R1G376R AutoDock results revealed that all compounds except ZINC02690584 interact stronger than both PI-103 and LY-294,002 (Table 3).Table 2 Top 10 out of > 25,000 non-approved investigational compounds from the ZINC database with highest binding affinities towards mutant proteins (LBE = lowest binding energy, kcal/mol, pKi = predicted inhibition constant, µM)
LBE
BRAF47-438del PyRx AutoDock pKi Interacting residues
ZINC09339473 -12.3 -11.30 ± 0.28 0.006 ± 0.003 Phe156, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Glu338, Ser340, Asn342, Arg343, Ala344
ZINC10578766 -12.3 -10.16 ± 0.47 0.043 ± 0.033 Leu149, His150, Glu153, Thr154, Met258, Thr259, Gln261, Arg290, Glu338, Ser340, Leu341, Asn342
ZINC22798105 -12.2 -10.32 ± 0.24 0.03 ± 0.01 Ser5, Gly6, Gly9, Ala12, Gly15, Ala17, Leu18, Gly21, Asp22, Glu24, Glu26
ZINC33068262 -11.8 -11.39 ± 0.36 0.005 ± 0.003 Leu149, Met258, Thr259, Gln261, Leu286, Val289, Arg290, Asn292, Glu338, Pro339, Ser340, Arg343, Ala344
ZINC00691692 -11.8 -10.87 ± 0.21 0.014 ± 0.006 Leu161, Met258, Thr259, Gln261, Ser265, Leu286, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Arg343
ZINC08667624 -11.8 -12.35 ± 0.23 0.001 ± < 0.001 Glu153, Phe156, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Glu338, Pro339, Ser340, Leu341, Arg343
ZINC33068261 -11.7 -11.41 ± 0.10 0.004 ± 0.001 Leu149, His150, Glu153, Phe156, Leu161, Met258, Thr259, Gly260, Gln261, Arg290, Glu338, Pro339, Ser340
ZINC09219428 -11.7 -10.19 ± 0.39 0.039 ± 0.027 His150, Glu153, Leu161, Met258, Thr259, Gln261, Arg290, Asn292, Cys293, Glu338, Ser340, Leu341, Asn342, Arg343
ZINC21793973 -11.6 -9.66 ± 0.01 0.083 ± 0.002 Ser222, Gly223, Ile225, Leu226, Met228, Ile233, Leu262, Ile267, Arg270, Ile273, Ile363, Ala370, Phe371, Val373
ZINC31840966 -11.6 -10.81 ± 0.17 0.012 ± 0.004 Leu149, His150, Met158, Leu161, Met258, Thr259, Gln261, Asn292, Cys293, Glu338, Ser340, Leu341, Asn 342, Arg343, Ala344
sorafenib -8.5 -9.39 ± 0.09 0.13 ± 0.02 Leu49, Gly50, Arg51, Arg52, Asp57, Trp58, Ile60, Gln64, Trp84, Met125
vemurafenib -7.8 -10.17 ± 0.16 0.04 ± 0.01 Leu149, His150, Glu153, Phe156, Met258, Thr259, Gly260, Gln261, Arg290, Asn292, Pro339, Ser340, Asn342, Arg343
PIK3R1G376R PyRx AutoDock pKi Interacting residues
ZINC12583338 -10.8 -8.73 ± 0.24 0.42 ± 0.15 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Val289, Gln291, Lys293
ZINC13828412 -10.3 9.27 ± < 0.01 0.16 ± < 0.01 Ile142, Glu143, Gly146, Leu149, His150, Asn153, Leu281, Leu284, Thr285, Val289, Arg290, Gln291
ZINC08854569 -10.1 -8.08 ± 0.20 1.25 ± 0.45 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Gln291
ZINC01801780 -10 -7.36 ± 0.01 4.00 ± 0.01 Glu143, Glu146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291
ZINC02277300 -10 -7.66 ± 0.01 2.41 ± 0.02 Ile142, Glu143, Gly146, Leu281, Leu284, Val289, Gln291, Lys293
ZINC09358971 -9.9 -7.85 ± 0.04 1.77 ± 0.12 Glu143, Gly146, Leu149, Leu284, Thr285, Val289, Gln291, Lys293
ZINC02690584 -9.8 -6.73 ± 0.01 11.59 ± 0.22 Gly146, Leu149, Leu284, Thr285, Val289, Lys292
ZINC09153343 -9.8 -8.04 ± 0.12 1.29 ± 0.27 Ile142, Glu143, Gly146, Lys147, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13730374 -9.8 -9.02 ± 0.01 0.24 ± < 0.01 Ile142, Glu143, Gly146, Leu149, Leu281, Leu284, Thr285, Val289, Gln291, Lys293
ZINC13828408 -9.8 -7.87 ± 0.01 1.69 ± 0.02 Ile142, Glu143, Gly146, Leu149, His150, Leu284, Val289, Gln291, Lys293
PI-103 -9.0 -7.22 ± 0.01 5.10 ± 0.08 Ile142, Gly146, His150, Leu281, Leu284, Val289, Gln291, Lys293
LY-294,002 -8.1 -6.47 ± < 0.01 18.07 ± 0.02 Ile142, Glu143, Gly146, Leu284, Val289, Lys293
Fig. 5 Docking poses of the top 10 ranked non-approved investigational compounds of the ZINC database to the BRAF47-438del and PIK3R1G376R mutant proteins
Table 3 Disease indications and modes of action of FDA-approved drugs identified by virtual drug screening
Drug Disease/application Mode of action Origin Potentially recommendable for therapy
BRAF47-438del:
Tubocurarine Arrow poison (main component of curare) Non-depolarizing muscle relaxant; competitive inhibitor of acetylcholine receptors at the postsynaptic membrane; inhibition of ligand-driven natrium ion channels. Condrodendron tomentosum No
STK396645 No
Celecoxib Degenerative joint diseases (arthrosis), rheumatoid arthritis; Morbus Bechterew (Spondylitis ankylosans) Non-steroidal anti-rheumatic; selective COX2 inhibitor; inhibition of prostaglandin synthesis Synthetic Yes
Aclarubicin Cancer Anthracycline; histone eviction from chromatin upon DNA intercalation Strepomyces galilaeus Yes
Pimozide Chronic schizophrenic psychoses Antipsychotic; postsynaptic inhibition of dopamin receptors and thereby stimulation of presynaptic dopamin release; inhibition of acidic sphingomyelinase Synthetic Yes
Folidan (Calcium folinate, Leucovorin) Thrombocytopenia, neutropenia, anemia; Folic acid source to prevent depletion by folic acid antagonists; counteracts toxicity by folic acid antagonists Synthetic Yes
Acetophenone chemical precursor for the synthesis of flagnaces and pharmaceuticals Hypnotic, anticonvulsant, sedative Ingredient of essential oils (labdanum, castoreum, Stirlingia latifolia) No
LS-194,959 Cancer Inhibitor of CDK2 synthetic No
Danazol Endometriosis Testosterone derivative; inhibitor of gonatropine release (FSH, LH); interruption of telomere erosion Semisynthetic Yes
Triamterene Hypertonia; chronic heart insufficiency Natrium-sparing diuretic; inhibition of aldosteron-dependent epithelial natrium channels in late distal tubuli Synthetic Yes
PIK3R1G376R:
LS-194,959 see above No
STK396645 see above No
Carminomycin Cancer Anthracycline; DNA intercalation; DNA topoisomerase II inhibitor Actinomadura carminata Yes
Novobiocin Bacerial infections Aminocoumarin antibiotic; inhibition of bacterial gyrase, subunit GyrB Streptomyces niveus Yes
Gliquidone Diabetes mellitus Sulyonylurea; inhibition of natrium channels and depolarization of B-cells leading to opening of current-regulated potassium channels. The potassium influx depletes insulin from storage vesicles and increases insulin release into the blood. Synthetic Yes
Fazadon (fazadinium bromide) Anaesthetic in otorhinolaryngological surgery Non-depolarizing muscle relaxant; antagonist of acetylcholine through neuromuscular blockage Synthetic Yes
Daunorubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor; generation of cytotoxic superoxide. And hydroxyl-radicals Streptomyces peucetius and S. coeruleorubidus Yes
Tubocurare see above
Metocurine Convulsive conditions: anesthetic adjuvant Non-depolarizing muscle relaxant; antagonist of acetylcholine by competitive binding to cholinergic receptors at the motor-end plate Synthetic Yes
Idarubicin Acute leukemia Anthracycline; DNA intercalaction; DNA topoisomerase II inhibitor Semisynthetic Yes
Cytotoxicity
Virtual drug screening revealed that anthracyclines, among other drugs, might target the mutated BRAF and PIK3R1 proteins. Since these drugs are not clinically used for the treatment of glioblastoma multiforme, we explored whether these compounds are able to kill glioblastoma cells in vitro with comparable efficacy than cell lines from other tumor origin. For this reason, we screened the NCI database for test results of the compounds by our virtual drug screening approach. The log10IC50 values of several brain tumor cell lines for the corresponding compounds included into this database are shown in Fig. 6. For comparison, the log10IC50 mean values for each of the test compounds of 46–63 other tumor cell lines have been calculated. The following observations were made: (1) the anthracyclines (daunorubicin, idarubicin, aclacinomycin) were the most cytotoxic compounds in this panel of drugs; (2) Pimazole also showed considerable cytotoxicity against tumor cells, although this is a psychoactive drug used for the treatment of chronic schizophrenic psychoses; (3) The cytotoxicity of the compounds tested were not significantly different between glioblastoma cell lines and all other tumor cell lines, indicating that these compounds might be effective in glioblastoma patients, given that appropriate formulations will be chosen to cross the blood-brain barrier. Aclarubicin revealed cytotoxicity (log10IC50) in the range of -7.2 to -7.6, higher than both sorafenib (in the range of -5.6 to -5.8) and vemurafenib (in the range of -5.3 to -5.6) (Fig. 6a). Idarubicin and daunorubicin (in the range of -7.4 to -7.9) revealed higher cytotoxicity than LY-294,002 (in the range of -4.7 to -5.6) (Fig. 6b).Fig. 6 Cytotoxicity of the available top 10 ranked FDA-approved established drugs (a BRAF-47-438del screening, b PIK3R1-G376R screening) and the known inhibitors towards brain cancer cell lines of the NCI (Bethesda, USA) panel. Shown are log10IC50 values (M)
Discussion
Since many years, MRI/CT imaging is routinely used for monitoring treatment success. However, imaging cannot deliver hints, which salvage therapy could be applied if standard therapy fails. A novel concept in precision medicine is to sequence tumor transcriptomes to identify patient-specific mutations, which can be then addressed by targeted therapeutics [44].
In the present investigation, we explored the possibility to use transcriptomic information for virtual drug screening, in order to indicate novel treatment options for individual patients. For this purpose, we used the mutational profile of a glioblastoma patient as an example to prove the feasibility of this approach. We first focused on screening of FDA-approved drugs, since repurposing of drugs initially approved for other diseases than cancer provided interesting treatment alternatives in many cases. The mutation profile obtained by genotyping of the patient’s tumor has several implications for therapy:The BRAF inhibitors vemurafenib and sorafenib may be less efficient to inhibit mutated BRAF proteins than wildtype BRAF.
Promoter-methylation of the MGMT gene indicates that MGMT protein expression may be downregulated. As MGMT represents a resistance factor for temozolomide [45], this drug may be exquisitely efficient to treat this patient. Indeed, the treatment history of the patient demonstrated a very good response to the temozolomide-based radiochemotherapy protocol.
The EGFR deletion indicated that EGFR-directed drugs (such as erlotinib, gefitinib and others) might not be effective.
Small molecules addressing the other mutations are not available yet, which limits the therapeutic options in this patient.
The HGF gene deletion might be relevant for toxic side effects. HGF (hepatocyte growth factor) is known to protect from drug-induced hepatotoxicity [46–48]. In fact, this patient experienced hepatotoxicity by radiochemotherapy with temozolomide and compassionate use of artesunate and Chinese herbs [31]. It can be speculated that the observed hepatotoxicity might be explained at least in part by the HGF deletion.
To validate the RNA-sequencing data, we performed immunohistochemical analyses. The strong immunostaining of BRAF and PI3KR1 indicated that the primary antibodies used for immunohistochemistry detect not only wildtype but also mutated forms of these proteins and that both proteins were expressed at high levels in this glioblastoma case. Accordingly, immuno-positive staining of tissue slices by these antibodies is probably not the suitable indicator for the usage of the compounds newly identified, whereas sequencing of tissue is more appropriate for the detection of the new mutations.
Furthermore, the sequencing results showed that the EGFR gene was deleted, which was confirmed by immunohistochemistry since the tumor was EGFR-negative in immunohistochemical staining. Even if EGFR is not expressed in this tumor, the question arises, whether downstream signaling pathways are still operative, which might then not be linked to EGFR but to other signaling molecules. We observed strong expressions of signaling kinases (SRC, AKT) and transcription factors (mTOR, NF-κB). Hence, these signaling molecules may still be susceptible to specific small molecule inhibitors against SRC, AKT, mTOR, or NF-κB) [49–52].
A limitation of the tumor sequencing approach is that therapeutic antibodies and small molecules are approved only for some of the major tumor-related proteins, but the vast majority of mutations cannot be therapeutically addressed as of yet. A premise to exploit the full potential of precision medicine would be that the therapeutic options keep pace with the diagnostic power and that ideally hundreds of drugs are available to inhibit the numerous tumor-related mutated proteins. An approach to reach this goal may be for instance tumor vaccination based on mutanomic profiling [20]. Comparable approaches are difficult to realize for small molecules, because the clinical approval of novel drugs is laborious and cost-intensive, and it can take years to bring a new drug to the market.
Therefore, it may be feasible to search for drugs, which have been already approved for other diseases and conditions and which also show activity against cancer by inhibiting tumor-related proteins. This approach has been termed drug repurposing and recently led to astonishing treatment successes [53–55].
In the present investigation, we attempted to utilize the wealth of information of RNA-sequencing data obtained from tumor genotyping to identify already FDA-approved drugs that bound with high affinity to mutated proteins in a glioblastoma patient. A database of > 1,500 FDA-approved drugs was first screened by PyRx. The results obtained were then refined by molecular docking with AutoDock 4.2. Furthermore, we used the novel BRAF47-438del mutation and the known PIK3R1G376R mutation as models to prove the feasibility of this approach. It is reasonable to search for drugs that stronger bind to BRAF47-438del than sorafenib and vemurafenib and to search for drugs that stronger bind to PIK3R1G376R than LY-294,002 and PI-103 to identify compounds that can specifically target mutant proteins. The BRAF47-438del docking results pointed out that STK396645, pimozide, folidan, acetophenone, LS-194,959, danazol bound stronger than sorafenib. STK396645, pimozide, folidan, acetophenone, LS-194,959 also had stronger affinities than vemurafenib. Considering the PIK3R1G376R results, all compounds except albamycin, daunorubicin, tubocurarine had higher affinities than PI-103, whereas all compounds except albamycin, daunorubicin revealed stronger binding than LY-294,002.
Virtual drug screening unraveled a panel of drugs that bound with high affinity to BRAF47-438del or PI3KR1G376R. These drugs were from diverse chemical classes and had different pharmacological activities. Strong poisons such as tubocurarine appeared in the screening, which cannot be considered for treatment. Remarkably, anthracyclines were also among the top-ranking drugs, which are known for their strong anticancer activity (daunorubicin, idarubicin, aclarubicin, carminomycin). Further drugs identified by virtual drug screening revealed anti-inflammatory (celecoxib), psycho-active (pimozide, acetophenone), muscle relaxant (fazadon, metocurine), diuretic (triamterene) antidiabetic (gliquidone), antibiotic (novobiocin), anti-endometriotic (danazol) or other pharmacological activities. The wide variety of drugs can be taken as a clue for the potential of repurposing drugs with diverse pharmacological modes of action for cancer therapy. The question arises, which of these drugs may be useful to treat the glioblastoma of the patient presented in our study. The decision may depend not only on the known side effects of these drugs but also on whether they are able to cross the blood-brain barrier (BBB). While this is well known for psychoactive drugs, anthracyclines are known substrates of the ATP-binding cassette (ABC) transporter, P-glycoprotein, which is an important constituent of the BBB. Anthracyclines such as daunorubicin, idarubicin and others are usually not used for the treatment of glioblastoma, since sufficient amounts cannot cross the BBB to exert considerable therapeutic anticancer effects in the brain [56]. This problem may, however, be tackled by novel nanotechnologically engineered anthracycline-releasing devices to overcome the BBB [57–59]. Since several anthracyclines appeared as top-ranked drugs to bind with high affinity to BRAF47-438del and PI3KR1G376R, it is worth reconsidering them for glioblastoma therapy, if appropriate nanotechnological formulations allow BBB-crossing. This problem may, however, be overcome, if appropriate nanotechnological formulations will be applied. There are daunorubicin liposomes preparations available on the market [60], and anthracycline liposomes have been shown to cross the blood-brain barrier [61–63].
Of course, anthracyclines cannot be viewed as mono-specific drugs addressing the two mutations in BRAF and PI3KR1 in an exclusive manner. As classical drugs of natural origin, they can be reconsidered to act in a multi-specific fashion. Anthracyclines are known to intercalate into DNA, to inhibit DNA topoisomerase II, and to generate cytotoxic superoxide- and hydroxyl radicals. Several other on- and off-target effects can be assumed, and binding to mutated BRAF and PI3KR1 proteins may belong to these still unknown modes of action of anthracyclines.
Having the multi-specific nature of many drugs in mind, it is reasonable to investigate their cytotoxic potential in cell lines with diverse mutations. Therefore, we compared the cytotoxicity of the drugs identified by our virtual drug screening approach in the panel of brain tumor cell lines of the Developmental Therapeutics Program of the National Cancer Institute (NCI, USA) and compared their activity with those of cell lines from other tumor origins. These results may deliver additional information, whether these drugs - initially developed for the treatment of other diseases - are suitable for drug repurposing in glioblastoma therapy. The data obtained with the NCI panel of cell lines pointed out aclarubicin, daunorubicin and idarubicin for glioblastoma treatment since they revealed higher cytotoxicity than the known inhibitors (sorafenib, vemurafenib and LY-294,002). Although the standard drug for glioblastoma treatment is temozolomide, anthracyclines also revealed strong cytotoxicity in vitro against brain tumor cell lines.
Pimozide is a psychoactive drug that does cross the BBB. Since this drug also revealed cytotoxicity towards brain tumor cell lines in vitro, its repurposing for clinical glioblastoma treatment may also be considered.
At the time point, when we obtained these results, the glioblastoma was already at a progressive state and the patient decided not to undergo any further drug treatment anymore. Unfortunately, the patient died before we could recommend an individualized treatment with a liposome-based anthracycline. Hence, a final proof of the clinical success of the concept presented in his paper could not be given. Nevertheless, our study represents a preclinical feasibility approach and a method to identify possible treatment candidates in cases that lack therapeutic options, which is a scenario frequently occurring in clinical practice.
We suggest that further investigations using the tumor genomes of patients should be made to predict active drugs and to observe whether treatment of patients with these drugs really leads to clinically significant response rates.
In general, repurposing of already approved drugs may not deliver more drugs for the individual treatment of cancer patients due to the multiplicity of tumor-related mutations. Therefore, we applied a second approach and used virtual drug screening to screen a chemical library of > 25,000 non-approved investigational compounds. We analyzed whether novel compounds can be identified that also bind to the BRAF and PI3KR1 mutations with high affinities. Indeed, most of the selected compounds revealed stronger binding than the known inhibitors. For instance, all of the selected compounds bound stronger to BRAF47-438del mutant protein than sorafenib and selected compounds except ZINC21793973 bound stronger than vemurafenib. The identified substances may be used as a starting point for further drug development to improve glioblastoma therapy in the future.
Despite the attractiveness of the virtual drug screening approach based on tumor sequencing data, there are also some limitations of this approach that have to be discussed:Virtual drug screening of both FDA-approved drug as well as non-approved investigational compounds may result in the identification of a certain rate of false positive candidates [64, 65]. Therefore, results from virtual drug screening should be verified by in vitro or in vivo experiments.
The sequencing of tumor genomes and transcriptomes delivers information on both relevant driver as well as irrelevant passenger mutations regarding tumor development and progression. Hence, careful visual inspection of the sequencing data is advisable to make rational decisions, which mutated proteins are most suitable for virtual drug screening to find suitable candidate drugs with the therapeutic potential to significantly and sustainably improve tumor response to treatment and to prolong the survival time of patients.
Conclusions
Systematic genome-dependent repurposing may reveal putative drug candidates for the further development of precision medicine in cancer and other gene-dependent diseases. A combined concept consisting of multiple techniques, i.e. tumor transcriptomics, diverse virtual drug screening methods, chemical libraries for drug repurposing and in vitro testing may help to make beneficial predictions on small molecules to inhibit distinct mutated proteins and to improve cancer therapy for each individual patient. We term the concept presented here “ReSeqtion” (Repurposing of drugs by genome Sequencing and bioinformatic calculation).
Author contributions
M.E.M.S., O.K. A.Y. and K.M. performed the bioinformatical in silico analyses. M.E.M.S. performed the immunohistochemical staining. F.W. and F.A.G. were the treating physicians. H.J.G. read and revised the manuscript. T.E. designed the concept and wrote the manuscript.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Data availability
The data are available upon reasonable request.
Compliance with ethical standards
Ethical approval
The patient was enrolled in the INTRAGO trial [32], which was registered at clinictrials.gov, no. NCT02104882 (registration date: 03/26/2014) (https://clinicaltrials.gov/ct2/show/NCT02104882). INTRAGO was approved by the local ethics committee (Medical Ethics Commission II of the Faculty of Mannheim, University of Heidelberg 2013-548S-MA) and the Federal Office of Radiation Protection (Z 5-2246/2-2013-063).
Informed consent
Informed written consent had been given by the patient that the data are allowed to be scientifically used and published (dated: January 26, 2016).
Conflict of interest
All authors declare there is no conflict of interest. However, patent application #EP18211231 “A method to determine agents for personalized use” is disclosed.
Abbreviations
AKT AKT serine/threonine kinase 1
BBB blood brain barrier
BRAF B-Raf proto-oncogene, serine/threonine kinase
CD34 cluster of differentiation marker 34
CRISPR/Cas9 Clustered regularly interspaced short palindromic repeats/CRISPR-associated 3
CT computer tomography
EGFR epidermal growth factor receptor
FDA Food and Drug Administration
HGF hepatocyte growth factor
LBE lowest binding energy
MAP3K4/5 Mitogen-activated protein kinase kinase kinase 4/5
MGMT O-6-methylguanine-DNA methyltransferase
MRI magnetic resonance imaging
NCI National Cancer Institute
NF-κB nuclear factor-kappa B
PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1
PTK2 Protein tyrosine kinase 2
RMSD root mean square deviation
SRC SRC proto-oncogene, non-receptor tyrosine kinase
TTYH1 Tweety family member 1
VEGFR vascular endothelial growth factor receptor
VMD Visual Molecular Dynamics
WHO World Health Organization
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33313992 | 15,046,217 | 2021-06 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Aplasia'. | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | CYTARABINE, DAUNORUBICIN HYDROCHLORIDE | DrugsGivenReaction | CC BY | 33314627 | 18,923,057 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'COVID-19 pneumonia'. | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | CYTARABINE, DAUNORUBICIN HYDROCHLORIDE | DrugsGivenReaction | CC BY | 33314627 | 18,923,057 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective for unapproved indication'. | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | MEROPENEM, PIPERACILLIN SODIUM\TAZOBACTAM SODIUM, POSACONAZOLE | DrugsGivenReaction | CC BY | 33314627 | 18,898,999 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Multiple organ dysfunction syndrome'. | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | CYTARABINE, DAUNORUBICIN HYDROCHLORIDE | DrugsGivenReaction | CC BY | 33314627 | 18,923,057 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | MEROPENEM, PIPERACILLIN SODIUM\TAZOBACTAM SODIUM, POSACONAZOLE | DrugsGivenReaction | CC BY | 33314627 | 18,898,999 | 2021-01 |
What was the dosage of drug 'DAUNORUBICIN HYDROCHLORIDE'? | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | UNK (3 + 7) | DrugDosageText | CC BY | 33314627 | 18,923,057 | 2021-01 |
What was the outcome of reaction 'Aplasia'? | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | Fatal | ReactionOutcome | CC BY | 33314627 | 18,923,057 | 2021-01 |
What was the outcome of reaction 'COVID-19 pneumonia'? | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | Fatal | ReactionOutcome | CC BY | 33314627 | 18,923,057 | 2021-01 |
What was the outcome of reaction 'Drug ineffective for unapproved indication'? | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | Fatal | ReactionOutcome | CC BY | 33314627 | 18,898,999 | 2021-01 |
What was the outcome of reaction 'Multiple organ dysfunction syndrome'? | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | Fatal | ReactionOutcome | CC BY | 33314627 | 18,923,057 | 2021-01 |
What was the outcome of reaction 'Off label use'? | Characteristics and mechanisms to control a COVID-19 outbreak on a leukemia and stem cell transplantation unit.
Immunosuppressed patients like patients with leukemia or lymphoma, but also patients after autologous or allogeneic stem cell transplantation are at particular risk for an infection with COVID-19. We describe a COVID-19 outbreak on our leukemia and stem cell transplantation unit (LSCT-Unit) originating from a patient with newly diagnosed acute myeloid leukemia. The patient was treated with intensive induction chemotherapy and we characterize the subsequent outbreak of COVID-19 on a LSCT-Unit. We describe the characteristics of the 36 contacts among the medical team, the results of their PCR and antibody tests and clinical aspects and features of infected employees. Of these 36 close contacts, 9 employees of the LSCT-Unit were infected and were tested positive by PCR and/or antibody-testing. 8/9 of them were symptomatic, 3/9 with severe, 5/9 with mild symptoms, and one person without symptoms. Due to stringent hygiene measures, the outbreak did not lead to infections of other patients despite ongoing clinical work. Moreover, we demonstrate that incubation period and clinical course of a COVID-19 infection in an immunosuppressed patient could be unusual compared to that of immunocompetent patients. Consistent PCR and antibody testing are helpful to understand, control, and prevent outbreaks. For the safety of health-care workers and patients alike, all employees wore FFP2 masks and were trained to adhere to several further safety guidelines. The implementation of rigorous hygiene measures is the key to controlling an outbreak and preventing infections of other patients.
1 INTRODUCTION
In December 2019, the Corona Virus Disease 2019 (COVID‐19) outbreak started in China and rapidly developed into a pandemic threatening the population worldwide. COVID‐19 signifies a great risk especially for the elderly, patients with chronic diseases, and immunosuppressed patients, particularly for patients with hematological malignancies like leukemia or lymphoma and patients after autologous and allogeneic stem cell transplantation. In patients with hematological malignancies, a high mortality of COVID‐19 could be expected but at this time, there are only reports on small cohorts but no systematic clinical studies available.
1
,
2
,
3
,
4
Apart from the individual risk an infection carries for these patients, there is a high risk for all health‐care facilities that members of their staff will be infected by the virus and develop COVID‐19,
5
thereby dangerously reducing the number of health‐care workers able to treat patients sufficiently.
6
There also exists the additional high risk in all health‐care facilities that personnel become virus carriers and develop transmission chains between health‐care workers and patients.
7
In particular, as asymptomatic carriers can infect other people, apparently healthy medical staff have the potential of infecting patients with severe hematological diseases during their treatment. Similarly, patients switching between outpatient and inpatient treatment could be initially asymptomatic carriers potentially turning into the source of an outbreak in the clinical units. However, due to the long incubation time, initial PCR testing is no guarantee to have noninfected patients.
In this report, we describe a COVID‐19 outbreak on our leukemia and stem cell transplantation unit (LSCT‐Unit) originating from a 59‐year‐old female patient who was newly diagnosed with acute myeloid leukemia (AML), which measures we took to control the outbreak among the employees and how we avoided further spreading of the disease.
Our patient developed fever and atypical pneumonia in aplasia. We did a swab for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) detection, although at this point an infection with SARS‐CoV‐2 seemed to be rather unlikely, considering that our AML patient had been in hospital for over 10 days already. Our patient tested positive for COVID‐19. Thus, even patients who have been in hospital longer need surveillance testing for COVID‐19.
8
,
9
Due to the fact, that COVID‐19 was only detected after the patient had spent 2 weeks in hospital, the patient had 36 category I contacts, that is, cumulative face‐to‐face contact for at least 15 minutes, or exposure during aerosol‐forming procedures, or as part of a medical examination.
We will outline the probable path of infection in our clinic, the experiences we made, and the successful management of disease control during ongoing clinical work. The process changed during this time because of varying requirements of the local and national health‐care authorities but also due to temporary lack of assays and/or swab tubes. We describe the cluster, timeline, type and number of contacts, tests and test results and development of symptoms of the nine infected members of staff, and the measures taken to stop the spreading of the disease and to ensure staff and patient safety.
2 METHODS
2.1 Detection and contact assessment
SARS‐CoV‐2 was detected by nasopharyngeal swab of a hematological patient whose clinical course is detailed in the results section. As soon as the disease was detected, all contact persons were identified and data about their SARS‐CoV‐2 positivity assessed by PCR and/or antibody testing in 92% of all category I contact persons. Those members of staff, who had contact with the patient but were off when COVID‐19 was diagnosed, were under orders to stay home for as long as was feasible and thus were not tested at the time. All contact persons, their symptoms as well as the time course of possible disease development were recorded.
2.2 PCR
RT‐PCR for the novel coronavirus was performed according to Corman et al.
10
2.3 Antibody testing
IgG/IgM rapid tests were performed for the detection of antibodies.
11
COVID‐19 IgG/IgM rapid test (Biomerica, Irvine, CA, USA) is a lateral flow chromatographic immunoassay to detect antibodies specific to SARS‐CoV‐2. We used serum samples throughout.
We have verified the rapid tests with a Roche antibody test using the Cobas e411 System (Test Elecsys Anti‐SARS‐CoV‐2, Kit # 09203095190, Roche, Penzberg, Germany).
2.4 Contact person definition
According to the German national health institute Robert‐Koch Institute (RKI)
12
contact persons are persons who had contact to a confirmed case of COVID‐19 within 48 h before onset of symptoms in the index‐case. The end of the infectious period is currently not clear,
13
especially in immunocompromised patients.
Contacts are defined as follows. Category I contacts with close contact (higher risk of infection): People with cumulative face‐to‐face contact for at least 15 min. These include members of the same household, persons with direct contact with secretions or body fluids, in particular with respiratory secretions from a confirmed COVID‐19 case, persons who are exposed to aerosol‐forming procedures, medical personnel with contact to a confirmed COVID‐19 case as part of care or a medical examination (≤2 m), and without the use of protective equipment.
Category II contacts (lower risk of infection). Persons who were in the same room as a confirmed COVID‐19 case, but had cumulative face‐to‐face contact with that case for under 15 min. Medical personnel who were in the same room as the confirmed COVID‐19 case without the use of adequate protective clothing, but who kept a distance of more than 2 m at all times.
Since there is an obligation to report COVID‐19, contact persons of category I were reported to the public health office, there was a close cooperation. The public health office, traced contact persons and also got in touch with the respective persons, thus had a structured overview of all individuals who had tested positive.
2.5 Hygiene regulations
Our LSCT‐Unit offers 16 beds for treatment according to our hygiene standards of care. These include that patients undergoing allogeneic stem cell transplantation are treated in the four HEPA filtered single rooms, but patients receiving an induction therapy for acute leukemias may be treated in double rooms, except when carrying infectious diseases. Prior to the COVID‐19 pandemic, our staff used basic personal protective equipment (PPE) that is, medical/surgical masks, apron‐style polyethylene gowns, non‐sterile gloves, and disinfectants when in contact with patients with proven transmissible infectious disease or when in contact with aplastic patients undergoing allogeneic stem cell transplantation. In cases of fever of unknown origin with no infectious agent detectable, no PPE was worn. Hence, when attending our AML patient, we did not wear PPE before COVID‐19 was detected, as to this time no transmissible infectious agents had been found.
During each shift, a patient is assigned one particular doctor and one nurse. However, due to shift changes, necessary diagnostic procedures, doctors’ visits, visits from the support team, and so on, our COVID‐19 patient had accumulated a considerable number of contacts. Despite this, work on the ward had to be continued. Therefore, detailed instructions were issued in close consultation with the local health authorities on who was quarantined and who was allowed to keep working. In addition to the usual safety measures on an LSCT‐Unit, as soon as the COVID‐19‐infection was detected, all patients on the ward were transferred to individual single rooms, no visitors were allowed, and patients were not allowed to leave the ward anymore. The usual equipment of non‐sterile gloves and disinfectants was used, as well as spunbond polyethylene gowns and FFP2 masks were made available for all staff members for every workday for the 2 weeks after detection of the COVID‐19 infection considered the high‐risk infectious time. Strict social distancing was decreed, a 2 m‐distance rule during breaks was strictly adhered to, contacts were reduced to a minimum, and employees were specifically trained for the situation. They were also not allowed to consume food while on the ward as that would have meant taking their masks off. Visitors to the hospital in general had already been prohibited. Staff who showed any kind of symptoms that could be typical for COVID‐19 at the time stayed home until cessation of symptoms for at least 2 days and the performance of a negative PCR test. As far as possible, contact persons were quarantined and replaced by other personnel. The remaining staff were allowed to continue working under the condition they adhere strictly to hygienic rules, wear masks at all times, and keep constant vigilance as to the development of symptoms.
As soon as antibody testing was possible, all available contacts of category I were tested except for one nurse who now lives in another city, one health‐care assistant who finished her contract, and the art therapist who has gone on a sabbatical (Table 1).
TABLE 1 Contacts and infections during the COVID‐19 outbreak at our clinic. Of the 36 category I contacts of our AML/COVID‐19 patient who had incubated the SARS‐CoV‐2 virus, and was admitted to our clinic, nine staff members were infected. Those contacts are listed in the Table and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team. The nine infected employees are shown in red, all of them were either positive in the PCR or in antibody tests (also in red). Almost all of them showed symptoms, the onset of symptoms is indicated in the symptoms column. We have used the same numbers in Figure 1 as in the table, to be consistent. In the column comments, we described some forms of contact to patient/staff, however more are described throughout the manuscript
# Function in clinic Age (y) m/f Date PCR Result PCR Symptoms Ab‐test date/n.p. Result Ab‐test Comments
1 Doctor 34 f 15.03. Neg No 30.04.20 Neg
2 Doctor 48 m n.p. No 09.04.20 Neg Endoscopist who performed BAL
3 Doctor 35 f n.p. No 08.04.20 Neg
4 Doctor 28 f 15.03. Neg Fever, cough, headache, fatigue, slight dyspnea, and onset of symptoms 22.03. Quarantine for 14 days
23.03
Pos
07.04. Neg 09.04.20
Pos
5 Doctor 44 f 15.03. Neg No
17.03. Neg
6 Doctor 49 m 15.03. Neg No
16.03 Neg
07.04.20 Neg
17.03. Neg
7 Doctor 32 m n.p. No 30.04.20 Neg
8 Doctor 30 f 15.03. Neg Fever, cough, and onset of symptoms 17.03. 27.04.20 Neg Quarantine until cessation of symptoms and PCR negativity
25.03. Neg
9 Doctor 41 f 16.03. Neg No
17.03. Neg 09.04.20 Neg
10 Doctor 37 m 15.03. Neg No 09.04.20 Neg Endoscopist who performed BAL
17.03. Neg
20.03. Neg
11 Nurse 26 f 16.03. Neg No 24.04.20 Neg
12 Nurse 30 f 16.03. Neg No 24.04.20 Neg
13 Nurse 27 f 16.03. Neg Slight sore throat and onset of symptoms 18.03. Flat share with #16, quarantine for 14 days
08.04.20
Pos
14 Nurse 37 f 15.03. Neg Altered taste sensation, and onset of symptoms 22.03.
09.04.20
Pos
15 Nurse 33 f 15.03. Neg No
16.03. Neg
27.04.20 Neg
17.03. Neg
16 Nurse 29 f 16.03.
Pos
Fever, cough, sore throat, dyspnea, altered taste sensation, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
01.04. Neg
09.04.20
Pos
02.04. Neg
17 Nurse 25 f 16.03. Neg No
17.03. Neg 29.04.20 Neg
18 Nurse 26 f 15.03. Neg Cough, sore throat, and onset of symptoms 16.03. Quarantine until cessation of symptoms and PCR negativity
20.03. Neg 24.04.20
Pos
19 Nurse 24 f 15.03. Neg No 07.04.20 Neg
20 Nurse 37 f 17.03. Neg No 24.04.20
Pos
21 Nurse 26 f 16.03. Neg No 07.04.20 Neg
22 Nurse 41 f 16.03. Neg Fever, cough, and onset of symptoms 16.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
17.03. Neg
09.04.20
Pos
18.03. Neg
23 Nurse 24 n.p. n.p.
24 Diagnostic staff (EKG) 55 f 16.03. Neg Slight sore throat, fatigue, slight dyspnea, and onset of symptoms 20.03.
29.04.20
Pos
25 Health‐care assistant 20 f 16.03. Neg No
17.03. Neg 30.04.20 Neg
26 Health‐care assistant 20 f 16.03. Neg No
Neg
17.03. Neg
21.04.20
27 Cleaner 46 f 18.03.
Pos
Fever, cough, chest pain, and onset symptoms 13.03. Quarantine for 14 days until cessation of symptoms and PCR negativity
02.04.
Pos
20.04.20
Pos
08.04. Neg
28 MTRA 35 f n.p. No 29.04.20 Neg
29 MTRA 24 f n.p. No 30.04.20 Neg
30 MTA Endoscopy 44 m n.p. No 04.05.20 Neg
31 MTA Endoscopy 21 m n.p. No 29.04.20 Neg
32 MTLA 58 f 15.03. Neg No 29.04.20 Neg
33 Transplant coordinator 45 f n.p. No 21.04.20 Neg
34 Art therapist 66 f n.p. No n.p.
35 Pastoral care 54 f n.p. No 29.04.20 Neg
36 Psychologist 52 f 16.03. Neg No 27.04.20 Neg Mild sore throat 12.03.
19.03. Neg
John Wiley & Sons, Ltd3 RESULTS
3.1 Clinical course of the patient
The patient was admitted to the clinic because of symptomatic anemia; further laboratory tests showed pancytopenia. Figure 1 shows the timelines of the clinical course, the COVID‐19 infection and infected staff members.
FIGURE 1 Timelines of infection. The upper part of the figure shows the clinical course of the patient 59y (♀) including admission to the hospital (day 0), diagnosis of acute myeloid leukemia (AML, day 1) time‐point of infection, symptoms and COVID‐19 infection (blue boxes). The patient showed first symptoms more than 3 weeks after her last contact to the proposed person‐0 who is the origin of infection. The SARS‐CoV‐2 PCR was positive day 12 after admission to the hospital. From the time‐point 2 days before the onset of symptoms until positive result of PCR the patient had close contact with 36 persons of the staff of the leukemia and stem cell transplantation unit and infected 9 persons of these 36 staff member. Eight of them had symptoms. The lower part of the figure (in red) shows the nine infected persons of the staff, the symptoms, the PCR, and antibody tests performed and the course of infection. PCR testing was not performed in all persons as staff member that were off duty and developed symptoms were advised not to go to the clinic and our mobile services came some weeks later. Importantly, due to our measures the outbreak resulted in no further infections of other patients despite ongoing clinical work
Bone marrow biopsy showed myelodysplastic changes as well as a 20%–30% infiltration by myeloid blasts (Figure S1
). Abdominal ultrasound, chest X‐ray (Figure S2), pulmonary function testing, and echocardiography showed no abnormal findings, thus induction chemotherapy was started (Daunorubicin/Cytarabine 3 + 7). On day 3, the patient developed temperatures up to 39°C, at first without further symptoms apart from trouble swallowing and loss of appetite. Antibiotic treatment with Piperacillin/Tazobactam was initiated. Over the next 2 days oxygen saturation dropped repeatedly below 90% so extra oxygen was supplied via nasal cannula. The fever did not respond to antibiotic treatment, which was switched to Meropenem and Posaconazole was added. The patient continued to display high temperatures up to 39°C. A CT‐scan (Figure S2) showed pneumonia in the lower lobes, predominantly the left side and a bronchoalveolar lavage (BAL, Figure S3) was performed, in which neither any typical respiratory viruses nor bacteria were found. Over the following 2 days, the patient quickly deteriorated with oxygen saturation dropping further so that the patient was transferred to the intensive care unit. She developed a productive cough and still had a persistent high fever. At this point repeated questioning of the relatives revealed that the patient had contact to another person who had since tested positive for COVID‐19 (person‐0, Figure 1). As that contact had taken place more than 2 weeks before the patient's admission into hospital, the patient had not mentioned it. Our patient tested positive for SARS‐CoV‐2 and the following day needed to be intubated and started on mechanical ventilation. Over the next 10 days, the antibiotic regimen was repeatedly adjusted with no improvement of the respiratory situation. Very severe aplasia after chemotherapy persisted (Figure S3). Multi‐organ failure eventually set in and the patient died 10 days after diagnosis of COVID‐19.
3.2 All category I contacts among clinical staff from the LSCT‐Unit (Table)
During the time from initial hospital admission to diagnosis of COVID‐19, the patient had contact (category I) with 36 members of staff. Those contacts were listed (Table) and consisted of 10 medical doctors, 13 nurses, 2 health‐care assistants, 1 cleaning staff, 6 members of the diagnostic department, the transplant coordinator, and 3 members of the support team.
Contacts were listed only, if the contact had taken place within 48 h before the first symptoms of the patient or afterwards. Of these 36 contacts, 9 members of the staff of the LSCT‐Unit (1 MD, 6 nursing staff, 1 diagnostic, and 1 cleaning staff) were infected with COVID‐19, most of them symptomatic, and were tested positive by PCR and/or antibody testing. Due to changes in the management of coronavirus infection by local and central health offices and due to the temporary lack of laboratory capacities and ingredients, testing procedures are not the same for each staff member. Importantly, none of the other patients on the ward were tested positive or showed signs of COVID‐19 until end of June 2020.
13
Four members of staff (Figure 1, #4, #16, #22, and #27) had severe symptoms but no one was hospitalized. The others had relatively mild symptoms. One younger person (#13, flat mate of #16) had almost no notable symptoms except for mild sore throat. One nurse (#20, 37y) had no symptoms at all, although she had cared for the patient for 1 week.
The symptoms were quite diverse; among them were fever, cough but also chest pain, and three staff members experienced dyspnea. Headache and altered taste sensation were also among the symptoms, some of which were only later brought into association with COVID‐19. 4/9 infected had cough, 4/9 sore throat, 3/9 fever, 3/9 dyspnea, 2/9 altered taste sensation, 2/9 fatigue, 1/9 chest pain, and 1/9 headache.
14
,
15
Based on our measures taken, none of the other patients on the ward were tested positive by PCR analysis in the further clinical course or showed signs of COVID‐19 until June 2020.
3.3 Cluster of outbreak on LSCT‐Unit and clinical history of patient
Figure 1 pictures the timeline of the outbreak including information about infected staff members but also the clinical course of the AML patient who succumbed to the disease. 8/9 staff members had clinical symptoms as described earlier. Not all staff members were tested by PCR, because some were at home on time off and not called back in just to be tested (mobile swab teams testing people in their homes were only installed later). However, 25 contact persons were tested, some of them repeatedly, 47 PCRs were performed in total, yielding positive results in 3 members of staff. 9/36 contact person tested positive for antibodies later on. All infected employees are described with relevant contact periods, tests, and symptoms in Figure 1. #2 was tested negative by PCR 10 days after the first of several contacts, then developed symptoms and was tested again on day 19 with a positive result. Interestingly, the nurse #6 developed typical symptoms, tested negative in PCR, and positive in antibody testing later.
In the process of the outbreak on our LSCT‐Unit, we had to alter or adjust test numbers and methods due to changes in regulatory requirements, testing capacities, material availability, and alterations in the recommendations for testing.
4 DISCUSSION
Stem cell transplantation units are especially sensitive care units for particularly vulnerable patients, therefore, in the current climate an understandable fear of a COVID‐19 outbreak is ever‐present. Our patient with newly detected AML incubated the SARS‐CoV‐2 for more than 2 weeks and experienced a late, unexpected and atypical outbreak, thereby exposing a comparatively large number of staff to the potential risk of infection.
16
Due to the attentiveness and sensitivity of the team and the consistent implementation of hygiene measures, the outbreak resulted in no further infections of other patients despite ongoing clinical activity. All infected staff except one showed symptoms often described in COVID‐19 patients,
17
their symptoms resolved and all personnel are back at work.
Of the 36 close contacts who were analyzed for SARS‐CoV‐2 positivity after outbreak discovery, 9 (25%) were tested positive by antibody testing and 8/9 of these staff members turned out to have been symptomatic, albeit some of them in such a mild way that it did not register at the time. PCR testing was not performed in all contact persons as staff members who were off duty or who could be replaced with other personnel were advised not to come to the clinic and stay in quarantine for 14 days. Staff members who became symptomatic while in the clinic were tested immediately and sent home to quarantine. The data collected showed a cluster of SARS‐CoV‐2 infections not all of which showed up in the PCR‐tests. Therefore, these data have to be discussed carefully. In one person, there were no detectable symptoms during the time of observation. In this person, it could be an asymptomatic infection but also false positive testing could be possible, which is a critical aspect if those tests are used as a “safety pass” for employees in the clinical setting in the future. However, in our test series, the antibody testing worked very well, similar to the literature
18
but larger clinical studies are necessary.
Another point clearly demonstrated in our data are that the RT‐PCR of the nasopharyngeal swab on any given day is just a real‐time snapshot and has to be repeated frequently. For example, the MD who had contact to the COVID‐19 patient on several days over a period of more than 1 week, was infected, however was tested negative on day 10 after the first contact. She then developed symptoms and subsequently became positive in the SARS‐CoV‐2 RT‐PCR on day 19 after the first contact and was immediately quarantined. From this it can be concluded that testing for possible infection is essential, but diagnostic measures still need improvement, and a systematic evaluation of the PCR and antibody tests of both patients and employees is required.
Guidelines on testing and testing procedures have changed rapidly in recent weeks here, as in all countries. After the experiences we made with the management of COVID‐19 on our LSCT‐Unit, it has been very advantageous to introduce consistent testing at inpatient admission and to test patients immediately when symptomatic. This was not yet possible to the same extent at the beginning of the outbreak described here. Testing employees was and is particularly important, since our data show that young employees can have few symptoms and while being contagious. We suggest consistently performing PCR but also antibody tests for patients and also employees. Although testing strategies were frequently subject to change, a fairly coherent testing of all category I contacts and all patients on the LSCT‐Unit was eventually achieved.
Notable here is that with a combination of stringent hygienic measures and repeated testing no other patients were infected despite being cared for by the same personnel.
Our cluster of infection showed also interesting data regarding virus transmission. Different statements in current research literature about the transmission rate in families have been reported.
19
,
20
Of two nurses who both cared for the patient and who share a flat, only one became symptomatic quickly and tested positive. The other stayed in quarantine with her despite negativity in PCR. She developed the mildest of symptoms and had a positive antibody test later on. In contrast, some intensive contacts did not result in an infection. Surprisingly the employees who performed the bronchoalveolar lavage (BAL) on our AML/COVID‐19 patient remained free of symptoms and negative in all tests for SARS‐CoV‐2, although they had come into extremely close contact to the supposed infection site for a prolonged period of time.
21
,
22
Another important aspect highlighted in our data are that the clinical course of COVID‐19 in immunosuppressed patients could be different from that of other COVID‐19 cases. The estimated incubation period for COVID‐19 is supposedly no more than 14 days. However, looking back at the development of COVID‐19 in our AML patient and the onset of symptoms, the clinical findings suggest an incubation period of more than 3 weeks as there was a clear contact to a positive person and our patient had no other contacts to potential COVID‐19 sources after that. A possible explanation could be that in hematological patients a COVID‐19 infection might set off slowly, as the impaired immune system is unable to build up an immediate strong response. In acute leukemia, especially after allogeneic stem cell transplantation, immune reactions against immunogenic antigens like viral antigens but also other immunogenic antigenic structures, are weaker and differ in quality compared to those of healthy controls.
23
,
24
,
25
Intensive mouthwashes, medication or other unknown factors could also explain a longer incubation period in leukemia patients. Ultimately, the underlying cause of death of our AML/COVID‐19 patient was pneumonia but also persistent aplasia which prevented her from building up a proper immune response; an interesting question here is whether COVID‐19 was a causative agent in prolonging aplasia. There exist no data about this so far. Therefore, further data from patients with different leukemias and other hematological malignancies but also from patients with other immune‐compromising diseases and conditions are necessary to elucidate this interesting issue.
In conclusion, outbreaks of COVID‐19 in hospital units with immunosuppressed patients, including hematological and transplant units, might become increasingly more common. In leukemia patients, these infections can possibly take a course different from that in immunocompetent healthy individuals. Due to our measures, no further infection among the patients has occurred. Consistent hygiene management and continuously improved testing in the future might help to save patients with severe hematological malignancies.
5 AUTHORSHIP CONTRIBUTION STATEMENT
Jochen Greiner designed the study, analyzed and interpreted the data, and wrote the manuscript. Marlies Götz analyzed the data, searched literature, and reviewed the manuscript. Waltraud Malner‐Wagner interpreted and collected the data. Constanze Wendt reviewed the manuscript. Martin Enders interpreted the data and reviewed the manuscript. Christine Durst interpreted and collected the data. Detlef Michel interpreted the data and reviewed the manuscript. Stephanie von Harsdorf interpreted the data and reviewed the manuscript. Susanne Jung designed the study and figures, analyzed and interpreted the data, and wrote the manuscript.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
ETHICAL APPROVAL
Ethical approval was sought for this study.
Supporting information
Fig S1
Click here for additional data file.
Fig S2
Click here for additional data file.
Fig S3
Click here for additional data file.
ACKNOWLEDGMENTS
The authors thank all members of the staff for the donation of samples.
DATA AVAILABILITY STATEMENT
Not applicable. | Fatal | ReactionOutcome | CC BY | 33314627 | 18,898,999 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Breast cancer metastatic'. | Gastric metastases from invasive lobular carcinoma of the breast: Case report.
Invasive lobular carcinoma is the second most common type of invasive carcinoma of the breast. Although rare, invasive lobular carcinoma can lead to gastric metastases, which may appear several years after the initial diagnosis. The diagnosis is difficult, either because of its rarity or because of overlapping symptoms and imaging findings with primary gastric carcinoma. Immunohistochemistry is the key to diagnosis. We report a case of a 40-year-old woman with a previous history of invasive lobular carcinoma of the breast 2 years before, who presented recurrent and nonspecific gastrointestinal symptoms. Imaging findings revealed linitis plastica and the biopsy showed the presence of signet ring cell neoplasia. After gastrectomy, immunohistochemistry demonstrated diffuse expression of GATA-3 and the presence of estrogen receptors in some neoplastic cells with CK20-, leading to the final diagnosis of gastric metastases from invasive lobular carcinoma of the breast.
Introduction
The 2 most common histologic types of invasive breast cancer are invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC). ILC corresponds to less than 10% of all malignant breast tumors [1,2] and has a higher rate of bilaterality and multifocality than does IDC. It is the most difficult tumor to diagnose on mammography, with architectural distortion being the most frequent alteration. These lesions uncommonly present themselves with microcalcifications. For this reason, at the time of diagnosis, 60% of all patients will have lymph node or distant organ metastases [2]. IDC usually metastasize to local and distant lymph nodes, liver, lungs, brain, and bone [3,4]. In contrast, ILC shows a distinct systemic metastatic pattern and has a higher incidence of metastases to the gastrointestinal tract, gynecological organs, meninges, pleura, peritoneum, and skin [1], [2], [3], [4], [5], [6].
Breast cancer is one of the most common malignancies that metastasizes to the gastrointestinal tract, along with melanoma and lung carcinoma [1]. However, gastric metastases of lobular breast cancer are uncommon, with a reported incidence varying between 2.8% and 27% [2]. Gastric metastases can be diagnosed after a long period of time from primary tumor diagnosis (on average 7 years) [3], [4], [5]. The attributable symptoms are indistinguishable from the symptoms of a primary gastric cancer: Indigestion, dyspepsia, anorexia, pyrosis, nausea, epigastric pain, early satiety, vomiting, bleeding, and weight loss [1], [2], [3],7]. In addition, the correct diagnosis may be difficult because metastatic disease involving the stomach is hard to differentiate from a primary gastric cancer on clinical, imaging, and pathological examinations [1,4,8].
Radiological findings are nonspecific and the most frequent change is asymmetrical or diffuse thickening of the gastric wall. Endoscopic examination may also be very similar to primary gastric carcinomas and the most common pattern is a linitis plastica with diffuse infiltration of the submucosa and muscularis propria in 73%-83% of cases [1,4,9]. On pathologic examination of the gastric biopsies, the presence of signet ring-shaped cells may be interpreted as primary gastric cancer. Ultimately, the definitive diagnosis is based on immunohistochemical analysis and supported by previous clinical history [3].
Case report
A 40-year-old woman came to the emergency service with complaints of recurrent low back pain. A computed tomography scan was performed, showing diffuse lytic bone lesions in the axial skeleton suspicious of metastases. After a physical examination, a hard lump was palpated in the upper right breast, associated with skin retraction. On the mammography, a focal asymmetry was observed in the upper right breast, with an extension of approximately 9.5 cm. On the breast ultrasound, this alteration corresponded to a suspicious hypoechogenic and ill-defined lesion, with posterior acoustic shadowing (Fig. 1). Ipsilateral axillary adenopathy was also present. Ultrasound-guided biopsy was performed and grade 2 invasive lobular carcinoma was diagnosed. Estrogen receptors (ER) were positive (100%), progesterone receptors (PR) were positive (10%) and Her-2 receptors were negative, with no E-cadherin expression (Fig. 2). The final diagnosis was stage IV luminal B right breast carcinoma. After a discussion in a multidisciplinary meeting, the patient started neoadjuvant chemotherapy with desunomabⓇ and placlitaxelⓇ.Fig. 1 On mammography, craniocaudal (A) and mediolateral oblique (B) views show a focal asymmetry in the upper right breast (circle) with an extension of approximately 9.5 cm. On ultrasound (C), this alteration corresponds to a hypoechogenic and ill-defined lesion with posterior acoustic shadowing (arrows). Computed tomography images (D-E) show diffuse lytic bone lesions consistent with metastases in the axial skeleton (arrows).
Fig 1Fig. 2 Histology from the breast biopsy shows discohesive neoplastic cells invading the stroma, individually dispersed or arranged in single-file linear cords (Hematoxylin and eosin (H&E) staining, 100×) (A); immunohistochemical analysis reveals diffuse positivity for estrogen receptors (Immunohistochemistry (IHC) for ER, 100×) (B), and absence of E-Cadherin expression (IHC for E-Cadherin, 100×) (C). Note the retained expression of E-cadherin in a non-neoplastic duct (arrow).
Fig. 2
After finishing 9 cycles of chemotherapy, magnetic resonance imaging was performed, showing a residual small lesion in the upper outer quadrant of the right breast. Conservative surgery followed by radiation therapy to the right breast and the ipsilateral axillary was performed. Due to the presence of bone metastases, the patient was also treated with fulvestrantⓇ and palbociclibⓇ, achieving clinical and imaging stability.
After 2 years of follow-up, the patient showed with complaints of recurrent nausea, epigastric discomfort, early satiety, and weight loss. An abdominal computed tomography depicted diffuse thickening of the gastric wall with a small lumen, in a leather bottle-like appearance. No other significant abdominal findings were visualized, such as abdominal adenopathy, liver metastases, peritoneal implants, or ascites. Upper gastrointestinal endoscopy revealed marked rigidity for the gastric wall with a narrow lumen and a heterogeneous mucosa with thickened folds, which was suspicious of diffuse malignant infiltration of the submucosa (linitis plastica, Fig. 3). Gastric biopsies were compatible with adenocarcinoma with signet ring cells, which was interpreted as probable primary gastric neoplasia. Consequently, the patient underwent a total gastrectomy. The pathological evaluation of the surgical specimen revealed involvement of the entire stomach by a malignant neoplasm, invading the entire wall thickness, and formed by poorly cohesive cells with focal signet ring features. The subsequent immunohistochemical study revealed diffuse expression of GATA-3 and estrogen receptors (about 10% of the cells), in the absence of CK20, and E-cadherin expression (Fig. 4). Thus, immunohistochemical analysis was compatible with gastric metastases of previously diagnosed lobular breast carcinoma.Fig. 3 Axial (A) and sagittal (B) contrast-enhanced computed tomography images show diffuse thickening of the gastric wall (arrow), with a narrow lumen, typical of linitis plastica. Endoscopy shows marked rigidity of the gastric wall, with mucosal integrity but slight gastric fold swelling (C). These characteristics are suspicious for infiltrative invasion of the submucosa.
Fig. 3Fig. 4 Histology of the surgical specimen shows diffuse involvement of the gastric wall by neoplasia, consisting of poorly cohesive cells (H&E staining, 100×) (A); immunohistochemical analysis reveals absence of E-cadherin expression (IHC for E-cadherin, 100×) (B), weak immunoreactivity for estrogen receptors in about 10% of the neoplastic cells (IHC for ER, 100×) (C); diffuse expression of GATA-3, compatible with gastric metastases from a previously diagnosed lobular breast carcinoma (IHC for GATA-3, 100×) (D). Note that the residual E-cadherin positivity in the preserved gastric glands (B).
Fig. 4
Discussion
Invasive lobular carcinoma of the breast usually presents with a distinct metastatic pattern in comparison to other invasive breast carcinomas. This may be explained by the fact that around 90% of ILCs have E-cadherin loss, a molecule responsible for cell-cell adhesion [1,4,[7], [8], [9]]. Consequently, the ILC is formed by noncohesive small cells, with preferential growth at sites of metastases. Frequently, gastric metastases spread to the submucosal layer in a diffuse infiltrative pattern without major involvement of the mucosa, which may accordingly lead to normal endoscopic examinations in up to 50% of cases and misleading false-negative biopsies [1,4,7].
The histological features of metastatic ILC to the stomach consist of infiltration of the gastric tissue by noncohesive small tumor cells with an occasional intracytoplasmic lumen arranged on linear cords between the normal gastric glands [5,7,10]. Therefore, as in the breast, metastatic ILC tends to infiltrate the affected organs in a diffuse process instead of forming a tumor nodule [11]. On imaging studies, the infiltration of the stomach wall can give an appearance of linitis plastica (water-bottle stomach), created by circumferential thickening and stiffness of the gastric wall, with narrowed lumen [11]. Peritoneal and retroperitoneal spread typically appears as tiny nodules that tend to become confluent and may cause “omental caking.” In the genitourinary system, the most frequent findings are bilateral cystic and solid ovarian masses (Krukenberg syndrome) [11].
The differentiation between primary gastric carcinoma and metastases of breast carcinoma is challenging, especially when gastric biopsies contain signet ring-shaped cells on pathologic examination [1,5]. Tumor cells with these features are characteristic of a subtype of primary gastric malignancy: Signet ring cell type gastric carcinoma [5]. However, tumor cells of ILC also have this morphology, making diagnosis a difficult task [1,2,6]. Therefore, the only way to reach the diagnosis is through immunohistochemical study. The immunohistochemical study is essential for the diagnosis of metastases in rare locations. For example, Singh T et al reported an extremely rare case of duodenal metastasization from endometrial carcinoma, which was confirmed through immunohistochemistry [12].
ILCs usually are ER and PR positive, without overexpression or amplification of the human epidermal growth factor receptor 2 (HER-2/neu) and E-cadherin [1,7]. ER and PR can be used as markers; however, they are not always suitable diagnostic markers to confirm tumor has originated. These receptors may be positive in patients with primary gastric carcinoma (ER in 32% and PR in 12% of the cases) [3], and if the primary lesion is negative for ER and PR, these markers are not useful in the diagnosis of breast cancer metastases in the stomach [8]. In addition, it is well known that ER and PR may change in expression at metastatic sites over the course of disease progression, usually resulting in loss or decrease in expression, with discrepancy between primary breast cancer and metastases in 15%-40% of the cases [9].
Other markers have emerged to distinguish between gastric metastases from breast cancer and primary gastric malignancy. While metastatic breast carcinoma is usually positive for cytokeratin 7 (CK7, 90%), gross cystic disease fluid protein 15 (GCDFP-15), and negative for cytokeratin 20 (CK20) [[1], [2], [3],6,9], primary gastric carcinoma is negative for CK7 and GCDFP‐15 and positive for CK20 [6], [7], [8]. Recently, GATA-3 has emerged as a marker of urothelial and breast carcinoma. It has 100% positivity in lobular breast carcinoma and 96% positivity in ductal carcinoma of the breast [3]. In primary gastric carcinoma it is positive in only 5% (in well-differentiated adenocarcinomas, with no reported cases in carcinomas with poorly cohesive cells, such as signet ring carcinomas) [13].
Our patient had a strong diffuse nuclear expression of GATA-3, which together with a previous medical history of ILC of the breast, CK20-, slight ER positivity, and absence of E-cadherin was consistent with the diagnosis of gastric metastases from invasive lobular carcinoma of the breast. In this case, there was a change of the expression of ER in the gastric metastases, as the expression was diffuse in the neoplastic cells in the breast and only discreet in the metastatic cells (not exceeding 10%).
Unfortunately, the definitive diagnosis in this case was only performed after total gastrectomy, due to the fact that the immunohistochemical study was not performed on the previous gastric biopsy and was interpreted as primary gastric carcinoma. Despite the similar clinical, endoscopic, and histological characteristics, the differentiation between primary and metastatic gastric carcinoma is pivotal, because the treatment and prognosis are dissimilar [2,6,9]. The treatment recommendation for gastric metastases of breast cancer is predictably systemic treatment with chemotherapy and hormone therapy [1,2,7]. Surgical intervention should be reserved for palliation or certain cases of solitary resectable gastrointestinal tract metastases [5], [6]. On the other hand, in the case of primary gastric cancer, surgical resection is the primary treatment in the absence of distant metastases [9]. Additionally, some authors state the importance of regular endoscopy in patients with a history of invasive lobular breast cancer. The hypothesis of gastric metastases should always be considered in these patients and an immunohistochemical study carried out for the definitive diagnosis [1].
In conclusion, although gastric metastases from ILCs are rare, this clinical hypothesis should always be considered in patients with gastrointestinal symptoms (such as nausea, epigastric pain, early satiety, vomiting, and weight loss) and endoscopic changes (for example, gastric wall rigidity and heterogeneous mucosa with thickened folds). The final diagnosis may be challenging due to endoscopic limitations (endoscopy can be normal along with falsely negative biopsies) and pathological interpretation (overlapping features with primary gastric carcinoma). In general, immunohistochemical study offers the key to the definitive diagnosis.
Patient consent statement
Unfortunately, the patient in this clinical case died last year, so it was not possible to obtain informed consent.
Acknowledgments
Our acknowledgment to Dr. Irene Gullo and Dr. Isabel Amendoeira from the Pathology Department of Centro Hospitalar e Universitário de São João for contributing to this clinical case with the pathological images.
Declarations of competing interest: None. | DENOSUMAB, FULVESTRANT, PACLITAXEL, PALBOCICLIB | DrugsGivenReaction | CC BY-NC-ND | 33318776 | 19,164,904 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Death'. | Gastric metastases from invasive lobular carcinoma of the breast: Case report.
Invasive lobular carcinoma is the second most common type of invasive carcinoma of the breast. Although rare, invasive lobular carcinoma can lead to gastric metastases, which may appear several years after the initial diagnosis. The diagnosis is difficult, either because of its rarity or because of overlapping symptoms and imaging findings with primary gastric carcinoma. Immunohistochemistry is the key to diagnosis. We report a case of a 40-year-old woman with a previous history of invasive lobular carcinoma of the breast 2 years before, who presented recurrent and nonspecific gastrointestinal symptoms. Imaging findings revealed linitis plastica and the biopsy showed the presence of signet ring cell neoplasia. After gastrectomy, immunohistochemistry demonstrated diffuse expression of GATA-3 and the presence of estrogen receptors in some neoplastic cells with CK20-, leading to the final diagnosis of gastric metastases from invasive lobular carcinoma of the breast.
Introduction
The 2 most common histologic types of invasive breast cancer are invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC). ILC corresponds to less than 10% of all malignant breast tumors [1,2] and has a higher rate of bilaterality and multifocality than does IDC. It is the most difficult tumor to diagnose on mammography, with architectural distortion being the most frequent alteration. These lesions uncommonly present themselves with microcalcifications. For this reason, at the time of diagnosis, 60% of all patients will have lymph node or distant organ metastases [2]. IDC usually metastasize to local and distant lymph nodes, liver, lungs, brain, and bone [3,4]. In contrast, ILC shows a distinct systemic metastatic pattern and has a higher incidence of metastases to the gastrointestinal tract, gynecological organs, meninges, pleura, peritoneum, and skin [1], [2], [3], [4], [5], [6].
Breast cancer is one of the most common malignancies that metastasizes to the gastrointestinal tract, along with melanoma and lung carcinoma [1]. However, gastric metastases of lobular breast cancer are uncommon, with a reported incidence varying between 2.8% and 27% [2]. Gastric metastases can be diagnosed after a long period of time from primary tumor diagnosis (on average 7 years) [3], [4], [5]. The attributable symptoms are indistinguishable from the symptoms of a primary gastric cancer: Indigestion, dyspepsia, anorexia, pyrosis, nausea, epigastric pain, early satiety, vomiting, bleeding, and weight loss [1], [2], [3],7]. In addition, the correct diagnosis may be difficult because metastatic disease involving the stomach is hard to differentiate from a primary gastric cancer on clinical, imaging, and pathological examinations [1,4,8].
Radiological findings are nonspecific and the most frequent change is asymmetrical or diffuse thickening of the gastric wall. Endoscopic examination may also be very similar to primary gastric carcinomas and the most common pattern is a linitis plastica with diffuse infiltration of the submucosa and muscularis propria in 73%-83% of cases [1,4,9]. On pathologic examination of the gastric biopsies, the presence of signet ring-shaped cells may be interpreted as primary gastric cancer. Ultimately, the definitive diagnosis is based on immunohistochemical analysis and supported by previous clinical history [3].
Case report
A 40-year-old woman came to the emergency service with complaints of recurrent low back pain. A computed tomography scan was performed, showing diffuse lytic bone lesions in the axial skeleton suspicious of metastases. After a physical examination, a hard lump was palpated in the upper right breast, associated with skin retraction. On the mammography, a focal asymmetry was observed in the upper right breast, with an extension of approximately 9.5 cm. On the breast ultrasound, this alteration corresponded to a suspicious hypoechogenic and ill-defined lesion, with posterior acoustic shadowing (Fig. 1). Ipsilateral axillary adenopathy was also present. Ultrasound-guided biopsy was performed and grade 2 invasive lobular carcinoma was diagnosed. Estrogen receptors (ER) were positive (100%), progesterone receptors (PR) were positive (10%) and Her-2 receptors were negative, with no E-cadherin expression (Fig. 2). The final diagnosis was stage IV luminal B right breast carcinoma. After a discussion in a multidisciplinary meeting, the patient started neoadjuvant chemotherapy with desunomabⓇ and placlitaxelⓇ.Fig. 1 On mammography, craniocaudal (A) and mediolateral oblique (B) views show a focal asymmetry in the upper right breast (circle) with an extension of approximately 9.5 cm. On ultrasound (C), this alteration corresponds to a hypoechogenic and ill-defined lesion with posterior acoustic shadowing (arrows). Computed tomography images (D-E) show diffuse lytic bone lesions consistent with metastases in the axial skeleton (arrows).
Fig 1Fig. 2 Histology from the breast biopsy shows discohesive neoplastic cells invading the stroma, individually dispersed or arranged in single-file linear cords (Hematoxylin and eosin (H&E) staining, 100×) (A); immunohistochemical analysis reveals diffuse positivity for estrogen receptors (Immunohistochemistry (IHC) for ER, 100×) (B), and absence of E-Cadherin expression (IHC for E-Cadherin, 100×) (C). Note the retained expression of E-cadherin in a non-neoplastic duct (arrow).
Fig. 2
After finishing 9 cycles of chemotherapy, magnetic resonance imaging was performed, showing a residual small lesion in the upper outer quadrant of the right breast. Conservative surgery followed by radiation therapy to the right breast and the ipsilateral axillary was performed. Due to the presence of bone metastases, the patient was also treated with fulvestrantⓇ and palbociclibⓇ, achieving clinical and imaging stability.
After 2 years of follow-up, the patient showed with complaints of recurrent nausea, epigastric discomfort, early satiety, and weight loss. An abdominal computed tomography depicted diffuse thickening of the gastric wall with a small lumen, in a leather bottle-like appearance. No other significant abdominal findings were visualized, such as abdominal adenopathy, liver metastases, peritoneal implants, or ascites. Upper gastrointestinal endoscopy revealed marked rigidity for the gastric wall with a narrow lumen and a heterogeneous mucosa with thickened folds, which was suspicious of diffuse malignant infiltration of the submucosa (linitis plastica, Fig. 3). Gastric biopsies were compatible with adenocarcinoma with signet ring cells, which was interpreted as probable primary gastric neoplasia. Consequently, the patient underwent a total gastrectomy. The pathological evaluation of the surgical specimen revealed involvement of the entire stomach by a malignant neoplasm, invading the entire wall thickness, and formed by poorly cohesive cells with focal signet ring features. The subsequent immunohistochemical study revealed diffuse expression of GATA-3 and estrogen receptors (about 10% of the cells), in the absence of CK20, and E-cadherin expression (Fig. 4). Thus, immunohistochemical analysis was compatible with gastric metastases of previously diagnosed lobular breast carcinoma.Fig. 3 Axial (A) and sagittal (B) contrast-enhanced computed tomography images show diffuse thickening of the gastric wall (arrow), with a narrow lumen, typical of linitis plastica. Endoscopy shows marked rigidity of the gastric wall, with mucosal integrity but slight gastric fold swelling (C). These characteristics are suspicious for infiltrative invasion of the submucosa.
Fig. 3Fig. 4 Histology of the surgical specimen shows diffuse involvement of the gastric wall by neoplasia, consisting of poorly cohesive cells (H&E staining, 100×) (A); immunohistochemical analysis reveals absence of E-cadherin expression (IHC for E-cadherin, 100×) (B), weak immunoreactivity for estrogen receptors in about 10% of the neoplastic cells (IHC for ER, 100×) (C); diffuse expression of GATA-3, compatible with gastric metastases from a previously diagnosed lobular breast carcinoma (IHC for GATA-3, 100×) (D). Note that the residual E-cadherin positivity in the preserved gastric glands (B).
Fig. 4
Discussion
Invasive lobular carcinoma of the breast usually presents with a distinct metastatic pattern in comparison to other invasive breast carcinomas. This may be explained by the fact that around 90% of ILCs have E-cadherin loss, a molecule responsible for cell-cell adhesion [1,4,[7], [8], [9]]. Consequently, the ILC is formed by noncohesive small cells, with preferential growth at sites of metastases. Frequently, gastric metastases spread to the submucosal layer in a diffuse infiltrative pattern without major involvement of the mucosa, which may accordingly lead to normal endoscopic examinations in up to 50% of cases and misleading false-negative biopsies [1,4,7].
The histological features of metastatic ILC to the stomach consist of infiltration of the gastric tissue by noncohesive small tumor cells with an occasional intracytoplasmic lumen arranged on linear cords between the normal gastric glands [5,7,10]. Therefore, as in the breast, metastatic ILC tends to infiltrate the affected organs in a diffuse process instead of forming a tumor nodule [11]. On imaging studies, the infiltration of the stomach wall can give an appearance of linitis plastica (water-bottle stomach), created by circumferential thickening and stiffness of the gastric wall, with narrowed lumen [11]. Peritoneal and retroperitoneal spread typically appears as tiny nodules that tend to become confluent and may cause “omental caking.” In the genitourinary system, the most frequent findings are bilateral cystic and solid ovarian masses (Krukenberg syndrome) [11].
The differentiation between primary gastric carcinoma and metastases of breast carcinoma is challenging, especially when gastric biopsies contain signet ring-shaped cells on pathologic examination [1,5]. Tumor cells with these features are characteristic of a subtype of primary gastric malignancy: Signet ring cell type gastric carcinoma [5]. However, tumor cells of ILC also have this morphology, making diagnosis a difficult task [1,2,6]. Therefore, the only way to reach the diagnosis is through immunohistochemical study. The immunohistochemical study is essential for the diagnosis of metastases in rare locations. For example, Singh T et al reported an extremely rare case of duodenal metastasization from endometrial carcinoma, which was confirmed through immunohistochemistry [12].
ILCs usually are ER and PR positive, without overexpression or amplification of the human epidermal growth factor receptor 2 (HER-2/neu) and E-cadherin [1,7]. ER and PR can be used as markers; however, they are not always suitable diagnostic markers to confirm tumor has originated. These receptors may be positive in patients with primary gastric carcinoma (ER in 32% and PR in 12% of the cases) [3], and if the primary lesion is negative for ER and PR, these markers are not useful in the diagnosis of breast cancer metastases in the stomach [8]. In addition, it is well known that ER and PR may change in expression at metastatic sites over the course of disease progression, usually resulting in loss or decrease in expression, with discrepancy between primary breast cancer and metastases in 15%-40% of the cases [9].
Other markers have emerged to distinguish between gastric metastases from breast cancer and primary gastric malignancy. While metastatic breast carcinoma is usually positive for cytokeratin 7 (CK7, 90%), gross cystic disease fluid protein 15 (GCDFP-15), and negative for cytokeratin 20 (CK20) [[1], [2], [3],6,9], primary gastric carcinoma is negative for CK7 and GCDFP‐15 and positive for CK20 [6], [7], [8]. Recently, GATA-3 has emerged as a marker of urothelial and breast carcinoma. It has 100% positivity in lobular breast carcinoma and 96% positivity in ductal carcinoma of the breast [3]. In primary gastric carcinoma it is positive in only 5% (in well-differentiated adenocarcinomas, with no reported cases in carcinomas with poorly cohesive cells, such as signet ring carcinomas) [13].
Our patient had a strong diffuse nuclear expression of GATA-3, which together with a previous medical history of ILC of the breast, CK20-, slight ER positivity, and absence of E-cadherin was consistent with the diagnosis of gastric metastases from invasive lobular carcinoma of the breast. In this case, there was a change of the expression of ER in the gastric metastases, as the expression was diffuse in the neoplastic cells in the breast and only discreet in the metastatic cells (not exceeding 10%).
Unfortunately, the definitive diagnosis in this case was only performed after total gastrectomy, due to the fact that the immunohistochemical study was not performed on the previous gastric biopsy and was interpreted as primary gastric carcinoma. Despite the similar clinical, endoscopic, and histological characteristics, the differentiation between primary and metastatic gastric carcinoma is pivotal, because the treatment and prognosis are dissimilar [2,6,9]. The treatment recommendation for gastric metastases of breast cancer is predictably systemic treatment with chemotherapy and hormone therapy [1,2,7]. Surgical intervention should be reserved for palliation or certain cases of solitary resectable gastrointestinal tract metastases [5], [6]. On the other hand, in the case of primary gastric cancer, surgical resection is the primary treatment in the absence of distant metastases [9]. Additionally, some authors state the importance of regular endoscopy in patients with a history of invasive lobular breast cancer. The hypothesis of gastric metastases should always be considered in these patients and an immunohistochemical study carried out for the definitive diagnosis [1].
In conclusion, although gastric metastases from ILCs are rare, this clinical hypothesis should always be considered in patients with gastrointestinal symptoms (such as nausea, epigastric pain, early satiety, vomiting, and weight loss) and endoscopic changes (for example, gastric wall rigidity and heterogeneous mucosa with thickened folds). The final diagnosis may be challenging due to endoscopic limitations (endoscopy can be normal along with falsely negative biopsies) and pathological interpretation (overlapping features with primary gastric carcinoma). In general, immunohistochemical study offers the key to the definitive diagnosis.
Patient consent statement
Unfortunately, the patient in this clinical case died last year, so it was not possible to obtain informed consent.
Acknowledgments
Our acknowledgment to Dr. Irene Gullo and Dr. Isabel Amendoeira from the Pathology Department of Centro Hospitalar e Universitário de São João for contributing to this clinical case with the pathological images.
Declarations of competing interest: None. | DENOSUMAB, FULVESTRANT, PACLITAXEL, PALBOCICLIB | DrugsGivenReaction | CC BY-NC-ND | 33318776 | 19,164,904 | 2021-02 |
What was the outcome of reaction 'Death'? | Gastric metastases from invasive lobular carcinoma of the breast: Case report.
Invasive lobular carcinoma is the second most common type of invasive carcinoma of the breast. Although rare, invasive lobular carcinoma can lead to gastric metastases, which may appear several years after the initial diagnosis. The diagnosis is difficult, either because of its rarity or because of overlapping symptoms and imaging findings with primary gastric carcinoma. Immunohistochemistry is the key to diagnosis. We report a case of a 40-year-old woman with a previous history of invasive lobular carcinoma of the breast 2 years before, who presented recurrent and nonspecific gastrointestinal symptoms. Imaging findings revealed linitis plastica and the biopsy showed the presence of signet ring cell neoplasia. After gastrectomy, immunohistochemistry demonstrated diffuse expression of GATA-3 and the presence of estrogen receptors in some neoplastic cells with CK20-, leading to the final diagnosis of gastric metastases from invasive lobular carcinoma of the breast.
Introduction
The 2 most common histologic types of invasive breast cancer are invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC). ILC corresponds to less than 10% of all malignant breast tumors [1,2] and has a higher rate of bilaterality and multifocality than does IDC. It is the most difficult tumor to diagnose on mammography, with architectural distortion being the most frequent alteration. These lesions uncommonly present themselves with microcalcifications. For this reason, at the time of diagnosis, 60% of all patients will have lymph node or distant organ metastases [2]. IDC usually metastasize to local and distant lymph nodes, liver, lungs, brain, and bone [3,4]. In contrast, ILC shows a distinct systemic metastatic pattern and has a higher incidence of metastases to the gastrointestinal tract, gynecological organs, meninges, pleura, peritoneum, and skin [1], [2], [3], [4], [5], [6].
Breast cancer is one of the most common malignancies that metastasizes to the gastrointestinal tract, along with melanoma and lung carcinoma [1]. However, gastric metastases of lobular breast cancer are uncommon, with a reported incidence varying between 2.8% and 27% [2]. Gastric metastases can be diagnosed after a long period of time from primary tumor diagnosis (on average 7 years) [3], [4], [5]. The attributable symptoms are indistinguishable from the symptoms of a primary gastric cancer: Indigestion, dyspepsia, anorexia, pyrosis, nausea, epigastric pain, early satiety, vomiting, bleeding, and weight loss [1], [2], [3],7]. In addition, the correct diagnosis may be difficult because metastatic disease involving the stomach is hard to differentiate from a primary gastric cancer on clinical, imaging, and pathological examinations [1,4,8].
Radiological findings are nonspecific and the most frequent change is asymmetrical or diffuse thickening of the gastric wall. Endoscopic examination may also be very similar to primary gastric carcinomas and the most common pattern is a linitis plastica with diffuse infiltration of the submucosa and muscularis propria in 73%-83% of cases [1,4,9]. On pathologic examination of the gastric biopsies, the presence of signet ring-shaped cells may be interpreted as primary gastric cancer. Ultimately, the definitive diagnosis is based on immunohistochemical analysis and supported by previous clinical history [3].
Case report
A 40-year-old woman came to the emergency service with complaints of recurrent low back pain. A computed tomography scan was performed, showing diffuse lytic bone lesions in the axial skeleton suspicious of metastases. After a physical examination, a hard lump was palpated in the upper right breast, associated with skin retraction. On the mammography, a focal asymmetry was observed in the upper right breast, with an extension of approximately 9.5 cm. On the breast ultrasound, this alteration corresponded to a suspicious hypoechogenic and ill-defined lesion, with posterior acoustic shadowing (Fig. 1). Ipsilateral axillary adenopathy was also present. Ultrasound-guided biopsy was performed and grade 2 invasive lobular carcinoma was diagnosed. Estrogen receptors (ER) were positive (100%), progesterone receptors (PR) were positive (10%) and Her-2 receptors were negative, with no E-cadherin expression (Fig. 2). The final diagnosis was stage IV luminal B right breast carcinoma. After a discussion in a multidisciplinary meeting, the patient started neoadjuvant chemotherapy with desunomabⓇ and placlitaxelⓇ.Fig. 1 On mammography, craniocaudal (A) and mediolateral oblique (B) views show a focal asymmetry in the upper right breast (circle) with an extension of approximately 9.5 cm. On ultrasound (C), this alteration corresponds to a hypoechogenic and ill-defined lesion with posterior acoustic shadowing (arrows). Computed tomography images (D-E) show diffuse lytic bone lesions consistent with metastases in the axial skeleton (arrows).
Fig 1Fig. 2 Histology from the breast biopsy shows discohesive neoplastic cells invading the stroma, individually dispersed or arranged in single-file linear cords (Hematoxylin and eosin (H&E) staining, 100×) (A); immunohistochemical analysis reveals diffuse positivity for estrogen receptors (Immunohistochemistry (IHC) for ER, 100×) (B), and absence of E-Cadherin expression (IHC for E-Cadherin, 100×) (C). Note the retained expression of E-cadherin in a non-neoplastic duct (arrow).
Fig. 2
After finishing 9 cycles of chemotherapy, magnetic resonance imaging was performed, showing a residual small lesion in the upper outer quadrant of the right breast. Conservative surgery followed by radiation therapy to the right breast and the ipsilateral axillary was performed. Due to the presence of bone metastases, the patient was also treated with fulvestrantⓇ and palbociclibⓇ, achieving clinical and imaging stability.
After 2 years of follow-up, the patient showed with complaints of recurrent nausea, epigastric discomfort, early satiety, and weight loss. An abdominal computed tomography depicted diffuse thickening of the gastric wall with a small lumen, in a leather bottle-like appearance. No other significant abdominal findings were visualized, such as abdominal adenopathy, liver metastases, peritoneal implants, or ascites. Upper gastrointestinal endoscopy revealed marked rigidity for the gastric wall with a narrow lumen and a heterogeneous mucosa with thickened folds, which was suspicious of diffuse malignant infiltration of the submucosa (linitis plastica, Fig. 3). Gastric biopsies were compatible with adenocarcinoma with signet ring cells, which was interpreted as probable primary gastric neoplasia. Consequently, the patient underwent a total gastrectomy. The pathological evaluation of the surgical specimen revealed involvement of the entire stomach by a malignant neoplasm, invading the entire wall thickness, and formed by poorly cohesive cells with focal signet ring features. The subsequent immunohistochemical study revealed diffuse expression of GATA-3 and estrogen receptors (about 10% of the cells), in the absence of CK20, and E-cadherin expression (Fig. 4). Thus, immunohistochemical analysis was compatible with gastric metastases of previously diagnosed lobular breast carcinoma.Fig. 3 Axial (A) and sagittal (B) contrast-enhanced computed tomography images show diffuse thickening of the gastric wall (arrow), with a narrow lumen, typical of linitis plastica. Endoscopy shows marked rigidity of the gastric wall, with mucosal integrity but slight gastric fold swelling (C). These characteristics are suspicious for infiltrative invasion of the submucosa.
Fig. 3Fig. 4 Histology of the surgical specimen shows diffuse involvement of the gastric wall by neoplasia, consisting of poorly cohesive cells (H&E staining, 100×) (A); immunohistochemical analysis reveals absence of E-cadherin expression (IHC for E-cadherin, 100×) (B), weak immunoreactivity for estrogen receptors in about 10% of the neoplastic cells (IHC for ER, 100×) (C); diffuse expression of GATA-3, compatible with gastric metastases from a previously diagnosed lobular breast carcinoma (IHC for GATA-3, 100×) (D). Note that the residual E-cadherin positivity in the preserved gastric glands (B).
Fig. 4
Discussion
Invasive lobular carcinoma of the breast usually presents with a distinct metastatic pattern in comparison to other invasive breast carcinomas. This may be explained by the fact that around 90% of ILCs have E-cadherin loss, a molecule responsible for cell-cell adhesion [1,4,[7], [8], [9]]. Consequently, the ILC is formed by noncohesive small cells, with preferential growth at sites of metastases. Frequently, gastric metastases spread to the submucosal layer in a diffuse infiltrative pattern without major involvement of the mucosa, which may accordingly lead to normal endoscopic examinations in up to 50% of cases and misleading false-negative biopsies [1,4,7].
The histological features of metastatic ILC to the stomach consist of infiltration of the gastric tissue by noncohesive small tumor cells with an occasional intracytoplasmic lumen arranged on linear cords between the normal gastric glands [5,7,10]. Therefore, as in the breast, metastatic ILC tends to infiltrate the affected organs in a diffuse process instead of forming a tumor nodule [11]. On imaging studies, the infiltration of the stomach wall can give an appearance of linitis plastica (water-bottle stomach), created by circumferential thickening and stiffness of the gastric wall, with narrowed lumen [11]. Peritoneal and retroperitoneal spread typically appears as tiny nodules that tend to become confluent and may cause “omental caking.” In the genitourinary system, the most frequent findings are bilateral cystic and solid ovarian masses (Krukenberg syndrome) [11].
The differentiation between primary gastric carcinoma and metastases of breast carcinoma is challenging, especially when gastric biopsies contain signet ring-shaped cells on pathologic examination [1,5]. Tumor cells with these features are characteristic of a subtype of primary gastric malignancy: Signet ring cell type gastric carcinoma [5]. However, tumor cells of ILC also have this morphology, making diagnosis a difficult task [1,2,6]. Therefore, the only way to reach the diagnosis is through immunohistochemical study. The immunohistochemical study is essential for the diagnosis of metastases in rare locations. For example, Singh T et al reported an extremely rare case of duodenal metastasization from endometrial carcinoma, which was confirmed through immunohistochemistry [12].
ILCs usually are ER and PR positive, without overexpression or amplification of the human epidermal growth factor receptor 2 (HER-2/neu) and E-cadherin [1,7]. ER and PR can be used as markers; however, they are not always suitable diagnostic markers to confirm tumor has originated. These receptors may be positive in patients with primary gastric carcinoma (ER in 32% and PR in 12% of the cases) [3], and if the primary lesion is negative for ER and PR, these markers are not useful in the diagnosis of breast cancer metastases in the stomach [8]. In addition, it is well known that ER and PR may change in expression at metastatic sites over the course of disease progression, usually resulting in loss or decrease in expression, with discrepancy between primary breast cancer and metastases in 15%-40% of the cases [9].
Other markers have emerged to distinguish between gastric metastases from breast cancer and primary gastric malignancy. While metastatic breast carcinoma is usually positive for cytokeratin 7 (CK7, 90%), gross cystic disease fluid protein 15 (GCDFP-15), and negative for cytokeratin 20 (CK20) [[1], [2], [3],6,9], primary gastric carcinoma is negative for CK7 and GCDFP‐15 and positive for CK20 [6], [7], [8]. Recently, GATA-3 has emerged as a marker of urothelial and breast carcinoma. It has 100% positivity in lobular breast carcinoma and 96% positivity in ductal carcinoma of the breast [3]. In primary gastric carcinoma it is positive in only 5% (in well-differentiated adenocarcinomas, with no reported cases in carcinomas with poorly cohesive cells, such as signet ring carcinomas) [13].
Our patient had a strong diffuse nuclear expression of GATA-3, which together with a previous medical history of ILC of the breast, CK20-, slight ER positivity, and absence of E-cadherin was consistent with the diagnosis of gastric metastases from invasive lobular carcinoma of the breast. In this case, there was a change of the expression of ER in the gastric metastases, as the expression was diffuse in the neoplastic cells in the breast and only discreet in the metastatic cells (not exceeding 10%).
Unfortunately, the definitive diagnosis in this case was only performed after total gastrectomy, due to the fact that the immunohistochemical study was not performed on the previous gastric biopsy and was interpreted as primary gastric carcinoma. Despite the similar clinical, endoscopic, and histological characteristics, the differentiation between primary and metastatic gastric carcinoma is pivotal, because the treatment and prognosis are dissimilar [2,6,9]. The treatment recommendation for gastric metastases of breast cancer is predictably systemic treatment with chemotherapy and hormone therapy [1,2,7]. Surgical intervention should be reserved for palliation or certain cases of solitary resectable gastrointestinal tract metastases [5], [6]. On the other hand, in the case of primary gastric cancer, surgical resection is the primary treatment in the absence of distant metastases [9]. Additionally, some authors state the importance of regular endoscopy in patients with a history of invasive lobular breast cancer. The hypothesis of gastric metastases should always be considered in these patients and an immunohistochemical study carried out for the definitive diagnosis [1].
In conclusion, although gastric metastases from ILCs are rare, this clinical hypothesis should always be considered in patients with gastrointestinal symptoms (such as nausea, epigastric pain, early satiety, vomiting, and weight loss) and endoscopic changes (for example, gastric wall rigidity and heterogeneous mucosa with thickened folds). The final diagnosis may be challenging due to endoscopic limitations (endoscopy can be normal along with falsely negative biopsies) and pathological interpretation (overlapping features with primary gastric carcinoma). In general, immunohistochemical study offers the key to the definitive diagnosis.
Patient consent statement
Unfortunately, the patient in this clinical case died last year, so it was not possible to obtain informed consent.
Acknowledgments
Our acknowledgment to Dr. Irene Gullo and Dr. Isabel Amendoeira from the Pathology Department of Centro Hospitalar e Universitário de São João for contributing to this clinical case with the pathological images.
Declarations of competing interest: None. | Fatal | ReactionOutcome | CC BY-NC-ND | 33318776 | 19,164,904 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Poisoning deliberate'. | Patterns and outcomes of admissions to the medical acute care unit of a tertiary teaching hospital in South Africa.
A Medical Acute Care Unit (MACU) was established at Chris Hani Baragwanath Academic Hospital (CHBAH) to provide comprehensive medical specialist care to the patients presenting with acute medical emergencies. Improved healthcare delivery systems at the MACU may result in shorter hospital stays, better outcomes, and less mortality.
The study's objective was to describe the demographics, diagnoses, disease patterns, and outcomes, including patient's mortality, admitted to the MACU at CHBAH.
Records of 200 patients admitted, between March 2015 to August 2015, to the MACU at CHBAH were reviewed. Patient demographics, diagnosis at admission, duration of stay, and outcomes were documented. Patients transferred to the medical ward, the Intensive Care Unit (ICU), or discharge. The leading causes of mortality were documented.
Of the 200 patients, 59% were females. The patients' mean age was 46 (17.2) years, and the mean duration of stay at the MACU was 1.45 (1.25) days. Non-communicable diseases accounted for 76% of admissions. The most frequently diagnosed conditions included: diabetic ketoacidosis acidosis (DKA) and hyperosmolar non-ketotic (HONK) (17.5%), non-accidental self-poisoning (16%), hypertensive emergencies (9.5%), decompensated cardiac failure (8%) and ischemic heart disease (7%). Infectious diseases comprised 14% of the diagnoses, of which cases of pneumonia were the most common (5%). Most patients (77.5%) were transferred to medical wards, 12% to ICU, while 10% demised at the MACU. The leading causes of death included sepsis (25%), DKA/HONK (20%), non-accidental self-poisoning (10%), and cardiac failure (10%).
Non-communicable diseases, particularly diabetic emergencies, were the leading causes of admission to the MACU at CHBAH. During the study period, high rates of case improvement, patient discharge, shorter hospital stay, and less mortality were observed. The leading cause of mortality was sepsis related.
African relevance
• This study was conducted in the largest hospital in Africa, the Chris Hani Baragwanath Academic Hospital.
• Most patients in this study were resident in the surrounding urban township, Soweto.
• This cross-sectional study highlights the acute medical conditions that patients have and their outcomes
Introduction
As a result, hospitals looked for structural reforms to improve the quality of care. The admission process of the Emergency centre (EC) for acutely ill medical patients to Internal Medicine needed to be improved [7]. This led to the introduction of a Medical Acute Care Unit (MACU), and “Acute Medicine” emerged as a branch of Internal Medicine in the developed world [7]. Acute Medicine is a subspecialty of Internal Medicine focused on the immediate and early specialist management of acute medical patients presenting to hospitals as emergencies [7]. The MACU is a dedicated ward where this takes place [8].
This model of health care has been widely implemented in the United Kingdom (U.K.) [7], Australia [9], and New Zealand [10], resulting in reports of good outcomes in terms of patients care and service delivery [11,12]. Currently, Acute Medicine is not a formally recognised specialty of Internal Medicine in South Africa. The MACU health care model is a new concept that we have adapted at Chris Hani Baragwanath Academic Hospital (CHBAH), a tertiary institution located in Soweto, South Africa (S.A.).
Since the MACU was established at the CHBAH, there have not been studies regarding disease and mortality patterns. It is essential to understand acute medical admissions causes to develop or amend preventive and therapeutic protocols for specific diseases. This information is also essential for health care planners as it identifies areas of priority for ongoing service development.
Objectives
This study aims to describe the pattern of diseases and outcomes, including mortality, in acutely ill medical patients admitted to the MACU at the CHBAH.
Methods
Study setting
Chris Hani Baragwanath Academic Hospital is a tertiary referral hospital in Soweto, South Africa. It provides medical care to an indigent population of 3.6 million in all specialties. The Department of Internal Medicine has 500 beds. It is the hospital's busiest department with admits 36,000 admissions annually, with an average of 100 patients per day. Patients are referred from the EC, secondary hospitals, and clinics. Patients are assessed first by EC doctors and then referred to the medical registrar allocated to the MACU. The MACU is a 16-bed facility located close to the EC and radiology. It is a specifically equipped ward where haemodynamic monitoring and specific therapeutic services, excluding mechanical ventilation, can be provided. It is staffed by general medical registrars, nurses, and allied health professionals and supervised by a specialist physician. Patients with acute reversible illnesses with predicted favourable outcomes are accepted to the MACU. The general medicine specialist on duty for the day regularly reviews the patients and initiates the post-admission rounds at the MACU. The resuscitation and subsequent observations to monitor response to the therapy given are ensured. Any predicted adverse outcomes are documented and acted upon immediately. Once the acute illness is resolved, patients can be discharged home or transferred to the medical wards. Patients requiring mechanical ventilation or invasive haemodynamic monitoring are referred to the Intensive Care Unit or coronary care unit depending on the acute illness.
Study population
We included a convenient sample of 200 patients 18 years and older, with any form of medical emergency admitted to the MACU between March 2015 to August 2015. This period is not limited to one season.
Study design
A retrospective review of the admission register of the MACU was performed. Demographic data, including gender and age, initial diagnosis, and outcomes, were recorded. In addition to the MACU register, patients' hospital files containing clinical details, duration of stay, and mortality were reviewed.
The initial diagnosis was assigned to systemic subgroups according to the organ system affected: cardiovascular, respiratory, renal, neurology, endocrine, non-accidental poisoning, and others. The initial diagnosis was further subdivided into specific diagnostic categories to assess the pattern of diseases.
The outcome was defined as the patient's discharge endpoint, i.e., directly home, transfer to the medical wards, ICU/ High Care, or death. The leading causes of mortality were documented.
Statistical analysis
The statistical package, STATA®, version 12, was used for the data analysis. For descriptive data, means with standard deviations and medians with inter-quartile ranges were used. Demographic characteristics were expressed as frequencies and percentages. Analytical data were expressed using the Chi-square test. Variables having a two-tailed p < 0.05 were considered significant.
Ethics permission
The study was approved by the Human Research Ethics Committee of the University of the Witwatersrand (certificate no: M159953).
Results
In the study cohort, there was a predominance of females, and the mean age of the patients was 46 (17.2) years. Patients in the 46–60 age group were the most frequently admitted, constituting a third of all admissions (Table 1).Table 1 Demographics of patients admitted to the MACU at Chris Hani Baragwanath Academic Hospital, South Africa (n = 200).
Table 1Characteristic n (%)
Gender
Male 82.0 (41)
Female 118 (59)
Age groups in years
18–30 41 (20.5)
31–45 52 (26.0)
46–60 65 (32.5)
61–75 34 (17.0)
>75 8.0 (4.0)
Ethnicity
African 182 (91)
Asian 8.0 (4.0)
White 6.0 (3.0)
Mixed ancestry 4.0 (2.0)
The central organ systems affected in the study group included: cardiac (24.5%), endocrine (19.5%), and non-accidental self-poisoning (18.5%) (Table 2).Table 2 Reasons for admission by organ system affected, mean age and gender distribution of the study population at MACU, Chris Hani Baragwanath Academic Hospital, South Africa (n = 200).
Table 2Affected organ system Mean age in years Male Female n (%)
Cardiac 50(1.21) 22 27 49 (24.5%)
Endocrine 48(1.11) 23 16 39 (19.5%)
Non-accidental self-poisoning 28(1.21) 13 24 37(18.5%)
Others 46(1.41) 11 20 31(15.5%)
Respiratory 43(1.31) 7.0 15 22 (11.0%)
Neurology 60(1.41) 7.0 8.0 15 (7.5%)
Renal 55(1.21) 2.0 5.0 7.0 (3.5%)
Non-accidental self-poisoning occurred more commonly in younger patients with a mean age of 28(1.21) years. Non-accidental self-poisoning and respiratory system disorders affected mainly females in the study population.
The most common diagnoses of the patients on admission to the MACU included diabetic ketoacidosis/hyperosmolar non-ketotic (17.5%), non-accidental self-poisoning with organophosphate and other agents (16%), hypertensive emergencies (9.5%), decompensated cardiac failure (8%), and ischemic heart disease (7%). Infectious diseases (14%) such as pneumonia, malaria, gastroenteritis, tuberculosis, and meningitis were noted. (Table 3).Table 3 Frequency of the Diagnoses of patients admitted to the MACU at Chris Hani Baragwanath Academic Hospital, South Africa (n = 200).
Table 3Diagnosis Frequency n (%)
Diabetic ketoacidosis/Hyperosmolar non-ketotic 35 (17.5)
Hypertensive emergency 19 (9.5)
Non-accidental self-poisoning with organophosphates 17 (8.5)
Decompensated cardiac failure 16 (8.0)
Non-accidental self-poisoning with other toxic agents 15 (7.5)
Myocardial infarction 14 (7.0)
Cerebrovascular accident 13 (6.5)
Pneumonia 10 (5.0)
Exacerbation of asthma 6.0 (3.0)
Exacerbation of Chronic obstructive pulmonary disease 6.0 (3.0)
Gastroenteritis 6.0 (3.0)
Malaria 6.0 (3.0)
Non-accidental self-poisoning with paracetamol 5.0 (2.5)
Septic shock 5.0 (2.5)
Pulmonary embolism 4.0 (2.0)
Disseminated Tuberculosis 4.0 (2.0)
Acute renal failure 4.0 (2.0)
Chronic renal failure 3.0 (1.5)
Meningitis 2.0 (1.0)
Epilepsy 2.0 (1.0)
Hypoglycaemia 2.0 (1.0)
Thyroid storm 2.0 (1.0)
Alcohol intoxication 1.0 (0.5)
Systemic lupus erythematosus 1.0 (0.5)
Pyelonephritis 1.0 (0.5)
Thrombotic thrombocytopenic purpura 1.0 (0.5)
The duration of stay of the study population at the MACU was short, with 22.5% of patients stayed for less than one day (Fig. 1).Fig. 1 Duration of stay of patients at MACU, Chris Hani Baragwanath Academic Hospital, South Africa (n = 200).
PE = Pulmonary Embolus, HONK = Hyperosmolar non-ketotic.
Fig. 1
The mean duration of stay at the MACU was 1.45(1.25) days, which differed in the different age groups. It was longer, 1.90 (1.44) days in the younger patients 18–30 years old. The shortest mean duration of stay, 1.00 (1.69), was noted in older patients >75 years old. However, this difference in duration of stay was not significant (p-value 0.07). The duration of stay did not differ significantly among males versus females. There was no significant relationship between duration of stay and the organ system affected or diagnosis.
The outcomes of admissions to the MACU were favourable in most patients (77.5%), showed recovery, and transfer to the general medical wards. A few patients (12%) required invasive haemodynamic monitoring and were subsequently transferred to the ICU, and (1%) were discharged home. A proportion of 10% of the patients admitted to the MACU, demised.
Of 20 patients who demised in MACU, the leading causes of death were sepsis-related (25%), diabetic ketoacidosis/hyperosmolar non-ketotic (20%), non-accidental self-poisoning with organophosphates and other toxic agents (15%), cardiac failure (10%), and hypertension (5%) (Fig. 2).Fig. 2 Causes of death of study patients who demised in MACU presented as percentages (n−20).
Fig. 2
Discussion
To the best of our knowledge, this was the first study describing the patterns of diseases in acute medical admissions to the MACU in S.A.
In this study, most patients were females (59%), in keeping with demographics seen in MACUs from the developed world. [13,14] The predominant age group of all the patients admitted at the MACU was 40–60 years (32.5%), also reported elsewhere [13,14]. Most of the patients in this sample were of African ethnicity (91%).
In the present study, 76% of admissions at the MACU were due to non-communicable diseases such as diabetic ketoacidosis/ hyperosmolar non-ketotic, hypertensive emergency, non-accidental self-poisoning, cardiac failure, ischemic heart disease, and cerebrovascular accident. The most commonly encountered disorders were within the scope of cardiology, endocrinology, non-accidental self-poisoning, and neurology, which is like data reported by other medical acute units in the developed world [7,13,14,15] (Table 4).Table 4 Patterns of diseases at the MACU, Chris Hani Baragwanath Academic Hospital, South Africa, and other international units.
Table 4South Africa
Present study United Kingdom
[7] United Kingdom
[13] United Kingdom
[14] Ireland
[15]
DKA/HONK Nonspecific chest pain Cellulitis Chest pain Heart failure
Self-poisoning Pneumonia Psychiatric Falls Atrial fibrillation
Hypertensive emergency Urinary tract infection Endocrine Pneumonia Diabetes
Cardiac failure COPD CVA COPD Hyponatremia
Ischemic heart disease Acute bronchitis Alcohol excess Gastrointestinal bleeding COPD
CVA Cardiac dysrhythmias Self-poisoning Diarrhoea and vomiting Anaemia
Pneumonia Coronary artery disease Collapse Urinary tract infection Altered mental status
Exacerbation of asthma Skin and soft tissue infection Headache/Migraine CVA Pneumonia
Exacerbation of COPD Epilepsy Urinary tract infection Self-poisoning Neoplasia
Malaria Cerebrovascular disease Gastritis CVA Acute myocardial infarction
CHBAH-Chris Hani Baragwanath Academic Hospital CVA = Cerebrovascular accident, COPD = Chronic obstructive airway disease, DKA = Diabetic ketoacidosis, HONK=Hyperosmolar non-ketotic, MACU = Medical Acute Care Unit.
There is a shortage of data on the pattern of diseases at the MACUs in a developing country like S.A.; therefore, a local comparison was not possible. However, the findings of the current study could be explained by several reasons: there is a rising prevalence of the non-communicable diseases of urbanisation that were previously unknown in rural S.A. [16], such as diabetes [17] and cardiovascular diseases. [18] Chris Hani Baragwanath Academic Hospital serves Soweto's population, where risk factors for these diseases [19], such as obesity and smoking [20,21,22], are highly prevalent predisposing the individuals towards non-communicable diseases. Communicable diseases like HIV/AIDS and tuberculosis were the causes of epidemics in S.A. [23]. However, the reduced frequency of these disorders observed in this study might reflect effective case management with specific therapies. Widespread use of highly active antiretroviral therapy (HAART) in S.A. since 2005/6 resulted in increased survival of patients with HIV/AIDS with an accompanying rise in non-communicable disease co-morbidities in this subgroup [24]. Interestingly, metabolic syndrome, altered glucose metabolism, dyslipidaemia, and lipodystrophy are seen frequently in patients with HIV/AIDS [26,26]. The use of some antiretroviral drugs in these patients, such as zidovudine, didanosine, and protease inhibitors, can predispose them to an increased risk of diabetes [27]. However, data from the present study did not include information on the HIV status of patients. Also, patients with advanced HIV/AIDS or disseminated tuberculosis with poor prognosis may not meet the criteria for admission in the MACU and are admitted directly to the medical wards at CHBAH. For the same reasons, infectious diseases such as pneumonia, gastroenteritis, malaria, and meningitis were found in small numbers (12%), possibly because they also admitted directly to the internal medicine wards.
Non-accidental self-poisoning was noted as a frequent reason for admission and mortality in the present study, especially in young African females, as previously reported in S.A. [28]. Most of these cases were individuals who attempted suicide [29]. The types of toxic agents used include organophosphates, paracetamol, cocaine, and other substances [30]. This could be explained based on the high prevalence of psychosocial stresses, such as untreated mental illness [31], substance abuse [32], family circumstances, and poverty [33].
During the study period, outcomes of admissions to the MACU were favourable in most cases. Most patients improved and were discharged to the medical wards (77.5%). The improved quality of care in the MACU healthcare model may partly explain this result.
The duration of stay at the MACU was short, 1.45 (1.2) days. Similarly, a small duration of stay was reported elsewhere [13,14]. It is possible that most uncomplicated non-communicable diseases and acute communicable diseases may be treated within a shorter time period. The short duration of stay might have a positive benefit on local government health finances.
The mortality rate was lower in the MACU than the general medical wards at 10% and 13%, respectively. The differences in mortality reflect more intensive care in MACU and a different spectrum of illnesses in MACU as compared to the general medical wards. The most frequently reported causes of death (sepsis, DKA, self-poisoning, cardiac failure, and hypertension) may also be attributed to the high prevalence of these disorders and the increased percentage of older individuals in the present research. The high mortality associated with diabetes also raises concerns about whether suboptimal care is offered to diabetics at a community health clinic level [34,35] or whether these patients may delay in presenting to healthcare facilities.
Sepsis remains a problem in the South African context [36]. High mortality due to sepsis in this study may indicate loopholes in the management and failure to institute the time-sensitive resuscitation process, which is vital to the control of sepsis.
Our results on mortality patterns were like reported elsewhere in the developed world [15]. However, due to the lack of data from MACUs in South Africa, our results do not have local comparisons.
The current study has several limitations. One of them is poor record-keeping, as is described in retrospective record reviews. We tried to overcome this through a precise search and retrieval of the data available. We excluded patients with incomplete data. There is a possibility of diagnostic errors due to a lack of diagnostic standards available for the study. Unfortunately, the percentage of patients that were discharged home after being discharged from MACU is not known. A further weakness is that only patients admitted to the MACU were included in the study, and the data does not consider acute medical patients that required direct admission to the general ward or ICU. Nonetheless, the data represents disease patterns, not the actual number of patients with acute conditions.
Furthermore, this study was conducted over a short duration. The conduction of similar studies over a more extended period would offer more robust evidence for these findings. Considering the study's limitations, the community's actual disease pattern may not be accurately reflected. However, this study provides a valuable foundation for further studies on acute admission patterns at Chris Hani Baragwanath Academic Hospital despite these limitations.
Conclusion
Non-communicable diseases, particularly diabetic emergencies, were the leading causes of admission to the MACU at CHBAH. During the study period, outcomes of admissions to the MACU were favourable in most cases. High rates of case improvement, patient discharge, shorter hospital stay, and less mortality were observed. The leading causes of mortality were sepsis-related, diabetes, and non-accidental self-poisoning.
Dissemination of results
The results of this study were presented at the academic meeting at the Chris Hani Baragwanath Academic Hospital.
Author's contribution
Authors contributed as follow to the conception or design of the work; the acquisition, analysis, or interpretation of data for the work; and drafting the work or revising it critically for important intellectual content: UK contributed 40%; CM 30%; and NG 30%. All authors approved the version to be published and agreed to be accountable for all aspects of the work.
Declaration of competing interest
The authors declared no conflicts of interest. | ACETAMINOPHEN, CODEINE | DrugsGivenReaction | CC BY-NC-ND | 33318914 | 19,208,356 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Respiratory failure'. | Post-transplant patients with COVID-19 associated acute respiratory distress syndrome, a role for Tociluzumab: A case series.
COVID-19 is the disease caused by SARS-CoV-2 that portends both a relatively high mortality rate as well as high rate of intensive care admission amongst all age groups; however effective therapy remains poorly characterized. Post-transplant patients are especially high risk and underrepresented in the literature. In these patients, cytokine release may play a significant role in the development of acute respiratory distress syndrome, raising the hypothesis that interleukin-6 inhibitors such as tocilizumab may be of benefit. Here, we describe two high-risk post-transplant patients who were treated with single-dose tocilizumab after intubation for moderate acute respiratory distress syndrome secondary to confirmed COVID-19 infection. Both patients recovered rapidly and were successfully extubated and discharged from the hospital without need for supplemental oxygen shortly thereafter, and their clinical improvement correlated with response in interleukin-6 levels. Tocilizumab appears to hold promise for critically ill COVID-19 patients who require mechanical ventilation when given shortly after intubation.
1 Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel beta coronavirus which originated in Wuhan, China in 2019. This virus causes coronavirus disease of 2019 (COVID-19). SARS-CoV-2 has a high virulence, no vaccine, no antiviral treatment, and relatively high mortality of 1.8–3.4%. Disease severity of COVID-19 necessitating admission to intensive care units is estimated at 5–11.5% amongst all age groups [1]. The United States is at particularly high risk for SARS-CoV-2-related mortality due to high prevalence of risk factors for poor outcomes, which include age greater than 65, cardiovascular disease, cerebrovascular disease, hypertension, diabetes mellitus, history of tobacco use, and chronic obstructive pulmonary disease [[2], [3], [4]]. Transplant patients are likely at higher risk given immunosuppression and underrepresented in current literature.
Several treatments are being actively investigated including hydroxychloroquine and azithromycin combination therapy, lopinavir/ritonavir, and remdesivir. These agents have shown antiviral activity against SARS-CoV-2 in vitro [[5], [6], [7]] as well as promise in humans [8,9], though this data is limited. Utility of these agents is likely limited to early disease since they theoretically function as antiviral; late complications of COVID-19 include acute respiratory distress syndrome (ARDS), which may be a result of cytokine storm. Cytokine storm is triggered by release of interferon gamma by T-cells which leads to release of tumor necrosis factor alpha, IL-6, and IL-10 by macrophages which can go on to cause significant lung injury and ARDS, as well as distributive shock and multisystem organ failure [10]. Studies have already shown that lopinavir/ritonavir did not have significant benefit when used late in disease course for varying severities of COVID-19, including ARDS [11,12].
Mortality of late stage ARDS secondary to COVID-19 is high, with studies showing mortality from 48% up to 90% once intubated and placed on mechanical ventilation, which is significantly higher than mortality associated with intubation for other viral pneumonias which is around 22% [13,14]. The reason for this higher mortality could be related to cytokine storm as critically ill COVID-19 patients have cytokine profiles resembling macrophage activation syndrome and secondary hemophagocytic lymphohistiocytosis with significant elevations in IL-1B, IL-2, IL-6, IL-17, IL-8 and tumor necrosis factor [15], with similar serologic markers such as elevated ferritin, elevated liver enzymes, and coagulopathies [16]. Presence of cytokine storm is further supported by significant proportion of COVID-19 patients who are intubated and requiring vasopressors for distributive shock despite no evidence of bacterial superinfection [17]. Thus, an effective therapy is in dire need.
IL-6 inhibitors such as tocilizumab and siltuximab/sarilumab have shown benefit in treatment of cytokine storm in chimeric antigen receptor T-cell patients. These agents have also shown promise for treatment of ARDS in severely ill COVID-19 patients via suppression of cytokine storm [18]. We present two critically ill post-transplant patients, each requiring intubation and mechanical ventilation for severe COVID-19 disease who were treated with single-dose tocilizumab and experienced rapid subsequent improvement at a large tertiary referral center in Gainesville, Florida, United States.
2 Case 1
Case 1 is a 50-year-old male with past medical history of coronary artery disease, nonischemic cardiomyopathy, type 2 diabetes mellitus, essential hypertension, and prior stroke who recently underwent kidney and heart transplant in February 2020. His post-transplant course was complicated with upper respiratory symptoms about one month after transplant, and his nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. During that admission, he did not require supplemental oxygen, and was discharged home to complete a total five-day course of hydroxychloroquine. Three days after discharge, he presented to the emergency department for severe dyspnea. While being evaluated in the emergency department, he experienced rapid and profound hypoxic respiratory decompensation requiring intubation and mechanical ventilation. Initial lab work was significant for IL-6 level of 45 pg/mL, ferritin of 648 ng/mL, LDH of 426 U/L, D-Dimer of 4.84 UG/mL, high-sensitivity CRP of 74.9 mg/L. These elevated markers of inflammation and cell death were consistent with potential cytokine storm. Admission chest radiograph showed significant burden of bilateral airspace opacities (Fig. 1). He was given 400 mg of tocilizumab on his first hospital day, approximately 5 hours after intubation. He was also started on broad spectrum antibiotics (vancomycin and cefepime), azithromycin 500 mg daily, and hydroxychloroquine 200 mg twice a day. His transplant immunosuppression with tacrolimus was continued, he was started on stress dose hydrocortisone 50mg every 6 hours and his mycophenolate was held per International Society for Heart and Lung Transplantation guidelines for moderate-severe COVID-19 ARDS [19]. Due to progressively worsening hypoxia and ARDS (pAO2:FiO2 ratio of 117), he was started on inhaled epoprostenol and subsequently underwent brief neuromuscular blockade with cisatracurium for 4 h on his first hospital night. He continued to improve clinically in the subsequent days with supportive care and lung-protective ventilation per ARDSNet protocol [20], associated with rapid improvement in the airspace opacities seen on admission chest radiograph on repeat on his fourth hospital day (Fig. 2). He did not require proning or vasopressive support during his hospitalization. He was weaned off inhaled epoprostenol on his fifth hospital day, extubated on his seventh hospital day, and was discharged home without supplemental oxygen requirement on his eleventh hospital day. His IL-6 level peaked at 303 pg/mL and decreased to 21 pg/mL on the date of discharge.Fig. 1 Chest radiograph of Case 1 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 1
Fig. 2 Chest radiograph of Case 1 on hospital day 8 post-extubation which shows improvement in multifocal airspace opacities.
Fig. 2
3 Case 2
Case 2 is a 67-year-old male with past medical history of chronic hepatitis B complicated by hepatocellular carcinoma status post orthotopic liver transplant in September 2016, with repeat orthotopic Roux-en-Y liver transplant with pyloric exclusion in November of 2016, history of cryptococcus pneumonia with documented clearance after 28 days of fluconazole in 2018, hypertension, and type 2 diabetes mellitus who presented to the emergency department for one week of fever and dyspnea, and he was admitted to the medical intensive care unit due to acute hypoxic respiratory failure requiring intubation and mechanical ventilation. Admission chest radiograph showed multifocal airspace opacities (Fig. 3), and computerized tomography (CT) of the chest on day of admission showed multifocal mixed ground glass opacities and dense consolidations throughout bilateral lungs (Fig. 4). Initial arterial blood gas showed a pAO2:FiO2 ratio of 116, consistent with moderate ARDS. His nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. His initial lab work was notable for IL-6 level of 31 pg/mL, ferritin of 320.4 ng/mL, LDH of 321 U/L, D-Dimer of 3.25 UG/mL, high-sensitivity CRP of 244 mg/L, which suggested a high level of inflammation and cell death consistent with potential cytokine storm. He was started on broad spectrum antibiotics (vancomycin and piperacillin/tazobactam), azithromycin 500 mg daily, and hydroxychloroquine 400 mg twice a day for one day followed by 200 mg twice a day. His home oral hydrocortisone for underlying adrenal insufficiency was continued and he did not require stress dose steroids. He was given single 400 mg dose of tocilizumab on his first hospital day, approximately 5 h after intubation, after which transplant hepatology recommended holding home mycophenolate mofetil and tacrolimus given potent immunosuppressive effects of tocilizumab. He was extubated on his second hospital day, and chest radiograph afterwards showed improvement in the multifocal airspace disease (Fig. 5). Tacrolimus was started seven days after administration of tocilizumab. He did not require neuromuscular blockade, vasopressive support, or proning during his hospitalization. He continued to improve and was discharged home on his eighth hospital day without supplemental oxygen requirement. His IL-6 level peaked at 318 pg/mL and decreased to 78 pg/mL on the date of discharge. His mycophenolate was restarted two weeks after administration of tocilizumab.Fig. 3 Chest radiograph of Case 2 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 3
Fig. 4 Axial CT chest on day of admission with multifocal mixed ground glass and dense consolidations throughout bilateral lungs.
Fig. 4
Fig. 5 Chest radiograph of Case 2 on hospital day 4 with slight improvement in multifocal airspace opacities.
Fig. 5
4 Discussion
We have presented two successful cases of single-dose tocilizumab use in critically ill post-transplant patients with moderate COVID-19 ARDS.
Although both patients had rapid improvement with tocilizumab, there are several unique factors. Notably, both patients were already immunosuppressed prior to hospitalization due to their post-transplant status, theoretically causing their respective immune systems to be less able to mount a strong cytokine release in the setting of COVID-19. This could perhaps explain how our cases differ from the results of the study by Lou et al. in Wuhan, China which showed that single-dose tocilizumab often failed to improve disease activity in those who were critically ill, who often needed multiple doses of tocilizumab due to persistently elevated IL-6 levels even when given in conjunction with high-dose glucocorticoids [18]. Both of our patients had a peak IL-6 level over 300 pg/mL during their hospitalization which subsequently trended downward prior to discharge. This particular population of post-transplant patients with severe COVID-19 is underrepresented in current literature and provides a unique perspective in use of novel therapies.
Of note, both patients were also treated with hydroxychloroquine and azithromycin since treatment took part prior to recent studies that revealed no benefit to COVID-19 patients when treated with hydroxychloroquine [21] with or without azithromycin [22]. In fact, some studies found an increase in adverse events for group treated with hydroxychloroquine [21]. Despite confounding factors and limitations, the rapid improvement seen in both patients is both promising and reassuring given that COVID-19 patients who require intubation typically require prolonged mechanical ventilation for average of 10 days [4]. Both patients also have several comorbidities, notably history of solid organ transplant on immunosuppression, and advanced age which puts them in higher risk for extended mechanical ventilation duration and mortality from COVID-19. The therapeutic window for IL-6 inhibitors such as tocilizumab is likely quite narrow. These agents must be given just before or immediately at onset of ARDS and cytokine storm, which tends to occur at day 7–10 of illness, and only in the minority of patients who experience rapid deterioration requiring admission to intensive care for mechanical ventilation (approximately 5–10% of all patients) [1]. Once ARDS has fully evolved, inhibition of immune system would likely do little to help remove proteinaceous fluid in alveoli, thus long mechanical ventilation and admission in the intensive care setting is to be expected.
Moving forward, further randomized controlled trials are necessary to support the findings we report. Given the robust response, we recommend consideration of tocilizumab, especially if other IL-6 inhibitors/blockers are not available, to critically ill patients with confirmed COVID-19 who require intubation for hypoxic respiratory failure with evidence of bilateral ground glass opacities on chest radiograph or CT, as aggressive empiric early treatment will likely have most impact on preventing ARDS and thus decreasing mortality.
Funding sources
This research did not receive any grant support.
Author contributions
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented.
Declaration of competing interest
The authors have no conflicts of interest to disclose.
Acknowledgements
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented. | FLUCONAZOLE, HYDROCORTISONE, MYCOPHENOLATE MOFETIL, TACROLIMUS | DrugsGivenReaction | CC BY | 33318918 | 18,853,137 | 2021 |
What was the administration route of drug 'HYDROCORTISONE'? | Post-transplant patients with COVID-19 associated acute respiratory distress syndrome, a role for Tociluzumab: A case series.
COVID-19 is the disease caused by SARS-CoV-2 that portends both a relatively high mortality rate as well as high rate of intensive care admission amongst all age groups; however effective therapy remains poorly characterized. Post-transplant patients are especially high risk and underrepresented in the literature. In these patients, cytokine release may play a significant role in the development of acute respiratory distress syndrome, raising the hypothesis that interleukin-6 inhibitors such as tocilizumab may be of benefit. Here, we describe two high-risk post-transplant patients who were treated with single-dose tocilizumab after intubation for moderate acute respiratory distress syndrome secondary to confirmed COVID-19 infection. Both patients recovered rapidly and were successfully extubated and discharged from the hospital without need for supplemental oxygen shortly thereafter, and their clinical improvement correlated with response in interleukin-6 levels. Tocilizumab appears to hold promise for critically ill COVID-19 patients who require mechanical ventilation when given shortly after intubation.
1 Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel beta coronavirus which originated in Wuhan, China in 2019. This virus causes coronavirus disease of 2019 (COVID-19). SARS-CoV-2 has a high virulence, no vaccine, no antiviral treatment, and relatively high mortality of 1.8–3.4%. Disease severity of COVID-19 necessitating admission to intensive care units is estimated at 5–11.5% amongst all age groups [1]. The United States is at particularly high risk for SARS-CoV-2-related mortality due to high prevalence of risk factors for poor outcomes, which include age greater than 65, cardiovascular disease, cerebrovascular disease, hypertension, diabetes mellitus, history of tobacco use, and chronic obstructive pulmonary disease [[2], [3], [4]]. Transplant patients are likely at higher risk given immunosuppression and underrepresented in current literature.
Several treatments are being actively investigated including hydroxychloroquine and azithromycin combination therapy, lopinavir/ritonavir, and remdesivir. These agents have shown antiviral activity against SARS-CoV-2 in vitro [[5], [6], [7]] as well as promise in humans [8,9], though this data is limited. Utility of these agents is likely limited to early disease since they theoretically function as antiviral; late complications of COVID-19 include acute respiratory distress syndrome (ARDS), which may be a result of cytokine storm. Cytokine storm is triggered by release of interferon gamma by T-cells which leads to release of tumor necrosis factor alpha, IL-6, and IL-10 by macrophages which can go on to cause significant lung injury and ARDS, as well as distributive shock and multisystem organ failure [10]. Studies have already shown that lopinavir/ritonavir did not have significant benefit when used late in disease course for varying severities of COVID-19, including ARDS [11,12].
Mortality of late stage ARDS secondary to COVID-19 is high, with studies showing mortality from 48% up to 90% once intubated and placed on mechanical ventilation, which is significantly higher than mortality associated with intubation for other viral pneumonias which is around 22% [13,14]. The reason for this higher mortality could be related to cytokine storm as critically ill COVID-19 patients have cytokine profiles resembling macrophage activation syndrome and secondary hemophagocytic lymphohistiocytosis with significant elevations in IL-1B, IL-2, IL-6, IL-17, IL-8 and tumor necrosis factor [15], with similar serologic markers such as elevated ferritin, elevated liver enzymes, and coagulopathies [16]. Presence of cytokine storm is further supported by significant proportion of COVID-19 patients who are intubated and requiring vasopressors for distributive shock despite no evidence of bacterial superinfection [17]. Thus, an effective therapy is in dire need.
IL-6 inhibitors such as tocilizumab and siltuximab/sarilumab have shown benefit in treatment of cytokine storm in chimeric antigen receptor T-cell patients. These agents have also shown promise for treatment of ARDS in severely ill COVID-19 patients via suppression of cytokine storm [18]. We present two critically ill post-transplant patients, each requiring intubation and mechanical ventilation for severe COVID-19 disease who were treated with single-dose tocilizumab and experienced rapid subsequent improvement at a large tertiary referral center in Gainesville, Florida, United States.
2 Case 1
Case 1 is a 50-year-old male with past medical history of coronary artery disease, nonischemic cardiomyopathy, type 2 diabetes mellitus, essential hypertension, and prior stroke who recently underwent kidney and heart transplant in February 2020. His post-transplant course was complicated with upper respiratory symptoms about one month after transplant, and his nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. During that admission, he did not require supplemental oxygen, and was discharged home to complete a total five-day course of hydroxychloroquine. Three days after discharge, he presented to the emergency department for severe dyspnea. While being evaluated in the emergency department, he experienced rapid and profound hypoxic respiratory decompensation requiring intubation and mechanical ventilation. Initial lab work was significant for IL-6 level of 45 pg/mL, ferritin of 648 ng/mL, LDH of 426 U/L, D-Dimer of 4.84 UG/mL, high-sensitivity CRP of 74.9 mg/L. These elevated markers of inflammation and cell death were consistent with potential cytokine storm. Admission chest radiograph showed significant burden of bilateral airspace opacities (Fig. 1). He was given 400 mg of tocilizumab on his first hospital day, approximately 5 hours after intubation. He was also started on broad spectrum antibiotics (vancomycin and cefepime), azithromycin 500 mg daily, and hydroxychloroquine 200 mg twice a day. His transplant immunosuppression with tacrolimus was continued, he was started on stress dose hydrocortisone 50mg every 6 hours and his mycophenolate was held per International Society for Heart and Lung Transplantation guidelines for moderate-severe COVID-19 ARDS [19]. Due to progressively worsening hypoxia and ARDS (pAO2:FiO2 ratio of 117), he was started on inhaled epoprostenol and subsequently underwent brief neuromuscular blockade with cisatracurium for 4 h on his first hospital night. He continued to improve clinically in the subsequent days with supportive care and lung-protective ventilation per ARDSNet protocol [20], associated with rapid improvement in the airspace opacities seen on admission chest radiograph on repeat on his fourth hospital day (Fig. 2). He did not require proning or vasopressive support during his hospitalization. He was weaned off inhaled epoprostenol on his fifth hospital day, extubated on his seventh hospital day, and was discharged home without supplemental oxygen requirement on his eleventh hospital day. His IL-6 level peaked at 303 pg/mL and decreased to 21 pg/mL on the date of discharge.Fig. 1 Chest radiograph of Case 1 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 1
Fig. 2 Chest radiograph of Case 1 on hospital day 8 post-extubation which shows improvement in multifocal airspace opacities.
Fig. 2
3 Case 2
Case 2 is a 67-year-old male with past medical history of chronic hepatitis B complicated by hepatocellular carcinoma status post orthotopic liver transplant in September 2016, with repeat orthotopic Roux-en-Y liver transplant with pyloric exclusion in November of 2016, history of cryptococcus pneumonia with documented clearance after 28 days of fluconazole in 2018, hypertension, and type 2 diabetes mellitus who presented to the emergency department for one week of fever and dyspnea, and he was admitted to the medical intensive care unit due to acute hypoxic respiratory failure requiring intubation and mechanical ventilation. Admission chest radiograph showed multifocal airspace opacities (Fig. 3), and computerized tomography (CT) of the chest on day of admission showed multifocal mixed ground glass opacities and dense consolidations throughout bilateral lungs (Fig. 4). Initial arterial blood gas showed a pAO2:FiO2 ratio of 116, consistent with moderate ARDS. His nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. His initial lab work was notable for IL-6 level of 31 pg/mL, ferritin of 320.4 ng/mL, LDH of 321 U/L, D-Dimer of 3.25 UG/mL, high-sensitivity CRP of 244 mg/L, which suggested a high level of inflammation and cell death consistent with potential cytokine storm. He was started on broad spectrum antibiotics (vancomycin and piperacillin/tazobactam), azithromycin 500 mg daily, and hydroxychloroquine 400 mg twice a day for one day followed by 200 mg twice a day. His home oral hydrocortisone for underlying adrenal insufficiency was continued and he did not require stress dose steroids. He was given single 400 mg dose of tocilizumab on his first hospital day, approximately 5 h after intubation, after which transplant hepatology recommended holding home mycophenolate mofetil and tacrolimus given potent immunosuppressive effects of tocilizumab. He was extubated on his second hospital day, and chest radiograph afterwards showed improvement in the multifocal airspace disease (Fig. 5). Tacrolimus was started seven days after administration of tocilizumab. He did not require neuromuscular blockade, vasopressive support, or proning during his hospitalization. He continued to improve and was discharged home on his eighth hospital day without supplemental oxygen requirement. His IL-6 level peaked at 318 pg/mL and decreased to 78 pg/mL on the date of discharge. His mycophenolate was restarted two weeks after administration of tocilizumab.Fig. 3 Chest radiograph of Case 2 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 3
Fig. 4 Axial CT chest on day of admission with multifocal mixed ground glass and dense consolidations throughout bilateral lungs.
Fig. 4
Fig. 5 Chest radiograph of Case 2 on hospital day 4 with slight improvement in multifocal airspace opacities.
Fig. 5
4 Discussion
We have presented two successful cases of single-dose tocilizumab use in critically ill post-transplant patients with moderate COVID-19 ARDS.
Although both patients had rapid improvement with tocilizumab, there are several unique factors. Notably, both patients were already immunosuppressed prior to hospitalization due to their post-transplant status, theoretically causing their respective immune systems to be less able to mount a strong cytokine release in the setting of COVID-19. This could perhaps explain how our cases differ from the results of the study by Lou et al. in Wuhan, China which showed that single-dose tocilizumab often failed to improve disease activity in those who were critically ill, who often needed multiple doses of tocilizumab due to persistently elevated IL-6 levels even when given in conjunction with high-dose glucocorticoids [18]. Both of our patients had a peak IL-6 level over 300 pg/mL during their hospitalization which subsequently trended downward prior to discharge. This particular population of post-transplant patients with severe COVID-19 is underrepresented in current literature and provides a unique perspective in use of novel therapies.
Of note, both patients were also treated with hydroxychloroquine and azithromycin since treatment took part prior to recent studies that revealed no benefit to COVID-19 patients when treated with hydroxychloroquine [21] with or without azithromycin [22]. In fact, some studies found an increase in adverse events for group treated with hydroxychloroquine [21]. Despite confounding factors and limitations, the rapid improvement seen in both patients is both promising and reassuring given that COVID-19 patients who require intubation typically require prolonged mechanical ventilation for average of 10 days [4]. Both patients also have several comorbidities, notably history of solid organ transplant on immunosuppression, and advanced age which puts them in higher risk for extended mechanical ventilation duration and mortality from COVID-19. The therapeutic window for IL-6 inhibitors such as tocilizumab is likely quite narrow. These agents must be given just before or immediately at onset of ARDS and cytokine storm, which tends to occur at day 7–10 of illness, and only in the minority of patients who experience rapid deterioration requiring admission to intensive care for mechanical ventilation (approximately 5–10% of all patients) [1]. Once ARDS has fully evolved, inhibition of immune system would likely do little to help remove proteinaceous fluid in alveoli, thus long mechanical ventilation and admission in the intensive care setting is to be expected.
Moving forward, further randomized controlled trials are necessary to support the findings we report. Given the robust response, we recommend consideration of tocilizumab, especially if other IL-6 inhibitors/blockers are not available, to critically ill patients with confirmed COVID-19 who require intubation for hypoxic respiratory failure with evidence of bilateral ground glass opacities on chest radiograph or CT, as aggressive empiric early treatment will likely have most impact on preventing ARDS and thus decreasing mortality.
Funding sources
This research did not receive any grant support.
Author contributions
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented.
Declaration of competing interest
The authors have no conflicts of interest to disclose.
Acknowledgements
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented. | Oral | DrugAdministrationRoute | CC BY | 33318918 | 18,853,137 | 2021 |
What was the outcome of reaction 'Acute respiratory distress syndrome'? | Post-transplant patients with COVID-19 associated acute respiratory distress syndrome, a role for Tociluzumab: A case series.
COVID-19 is the disease caused by SARS-CoV-2 that portends both a relatively high mortality rate as well as high rate of intensive care admission amongst all age groups; however effective therapy remains poorly characterized. Post-transplant patients are especially high risk and underrepresented in the literature. In these patients, cytokine release may play a significant role in the development of acute respiratory distress syndrome, raising the hypothesis that interleukin-6 inhibitors such as tocilizumab may be of benefit. Here, we describe two high-risk post-transplant patients who were treated with single-dose tocilizumab after intubation for moderate acute respiratory distress syndrome secondary to confirmed COVID-19 infection. Both patients recovered rapidly and were successfully extubated and discharged from the hospital without need for supplemental oxygen shortly thereafter, and their clinical improvement correlated with response in interleukin-6 levels. Tocilizumab appears to hold promise for critically ill COVID-19 patients who require mechanical ventilation when given shortly after intubation.
1 Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel beta coronavirus which originated in Wuhan, China in 2019. This virus causes coronavirus disease of 2019 (COVID-19). SARS-CoV-2 has a high virulence, no vaccine, no antiviral treatment, and relatively high mortality of 1.8–3.4%. Disease severity of COVID-19 necessitating admission to intensive care units is estimated at 5–11.5% amongst all age groups [1]. The United States is at particularly high risk for SARS-CoV-2-related mortality due to high prevalence of risk factors for poor outcomes, which include age greater than 65, cardiovascular disease, cerebrovascular disease, hypertension, diabetes mellitus, history of tobacco use, and chronic obstructive pulmonary disease [[2], [3], [4]]. Transplant patients are likely at higher risk given immunosuppression and underrepresented in current literature.
Several treatments are being actively investigated including hydroxychloroquine and azithromycin combination therapy, lopinavir/ritonavir, and remdesivir. These agents have shown antiviral activity against SARS-CoV-2 in vitro [[5], [6], [7]] as well as promise in humans [8,9], though this data is limited. Utility of these agents is likely limited to early disease since they theoretically function as antiviral; late complications of COVID-19 include acute respiratory distress syndrome (ARDS), which may be a result of cytokine storm. Cytokine storm is triggered by release of interferon gamma by T-cells which leads to release of tumor necrosis factor alpha, IL-6, and IL-10 by macrophages which can go on to cause significant lung injury and ARDS, as well as distributive shock and multisystem organ failure [10]. Studies have already shown that lopinavir/ritonavir did not have significant benefit when used late in disease course for varying severities of COVID-19, including ARDS [11,12].
Mortality of late stage ARDS secondary to COVID-19 is high, with studies showing mortality from 48% up to 90% once intubated and placed on mechanical ventilation, which is significantly higher than mortality associated with intubation for other viral pneumonias which is around 22% [13,14]. The reason for this higher mortality could be related to cytokine storm as critically ill COVID-19 patients have cytokine profiles resembling macrophage activation syndrome and secondary hemophagocytic lymphohistiocytosis with significant elevations in IL-1B, IL-2, IL-6, IL-17, IL-8 and tumor necrosis factor [15], with similar serologic markers such as elevated ferritin, elevated liver enzymes, and coagulopathies [16]. Presence of cytokine storm is further supported by significant proportion of COVID-19 patients who are intubated and requiring vasopressors for distributive shock despite no evidence of bacterial superinfection [17]. Thus, an effective therapy is in dire need.
IL-6 inhibitors such as tocilizumab and siltuximab/sarilumab have shown benefit in treatment of cytokine storm in chimeric antigen receptor T-cell patients. These agents have also shown promise for treatment of ARDS in severely ill COVID-19 patients via suppression of cytokine storm [18]. We present two critically ill post-transplant patients, each requiring intubation and mechanical ventilation for severe COVID-19 disease who were treated with single-dose tocilizumab and experienced rapid subsequent improvement at a large tertiary referral center in Gainesville, Florida, United States.
2 Case 1
Case 1 is a 50-year-old male with past medical history of coronary artery disease, nonischemic cardiomyopathy, type 2 diabetes mellitus, essential hypertension, and prior stroke who recently underwent kidney and heart transplant in February 2020. His post-transplant course was complicated with upper respiratory symptoms about one month after transplant, and his nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. During that admission, he did not require supplemental oxygen, and was discharged home to complete a total five-day course of hydroxychloroquine. Three days after discharge, he presented to the emergency department for severe dyspnea. While being evaluated in the emergency department, he experienced rapid and profound hypoxic respiratory decompensation requiring intubation and mechanical ventilation. Initial lab work was significant for IL-6 level of 45 pg/mL, ferritin of 648 ng/mL, LDH of 426 U/L, D-Dimer of 4.84 UG/mL, high-sensitivity CRP of 74.9 mg/L. These elevated markers of inflammation and cell death were consistent with potential cytokine storm. Admission chest radiograph showed significant burden of bilateral airspace opacities (Fig. 1). He was given 400 mg of tocilizumab on his first hospital day, approximately 5 hours after intubation. He was also started on broad spectrum antibiotics (vancomycin and cefepime), azithromycin 500 mg daily, and hydroxychloroquine 200 mg twice a day. His transplant immunosuppression with tacrolimus was continued, he was started on stress dose hydrocortisone 50mg every 6 hours and his mycophenolate was held per International Society for Heart and Lung Transplantation guidelines for moderate-severe COVID-19 ARDS [19]. Due to progressively worsening hypoxia and ARDS (pAO2:FiO2 ratio of 117), he was started on inhaled epoprostenol and subsequently underwent brief neuromuscular blockade with cisatracurium for 4 h on his first hospital night. He continued to improve clinically in the subsequent days with supportive care and lung-protective ventilation per ARDSNet protocol [20], associated with rapid improvement in the airspace opacities seen on admission chest radiograph on repeat on his fourth hospital day (Fig. 2). He did not require proning or vasopressive support during his hospitalization. He was weaned off inhaled epoprostenol on his fifth hospital day, extubated on his seventh hospital day, and was discharged home without supplemental oxygen requirement on his eleventh hospital day. His IL-6 level peaked at 303 pg/mL and decreased to 21 pg/mL on the date of discharge.Fig. 1 Chest radiograph of Case 1 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 1
Fig. 2 Chest radiograph of Case 1 on hospital day 8 post-extubation which shows improvement in multifocal airspace opacities.
Fig. 2
3 Case 2
Case 2 is a 67-year-old male with past medical history of chronic hepatitis B complicated by hepatocellular carcinoma status post orthotopic liver transplant in September 2016, with repeat orthotopic Roux-en-Y liver transplant with pyloric exclusion in November of 2016, history of cryptococcus pneumonia with documented clearance after 28 days of fluconazole in 2018, hypertension, and type 2 diabetes mellitus who presented to the emergency department for one week of fever and dyspnea, and he was admitted to the medical intensive care unit due to acute hypoxic respiratory failure requiring intubation and mechanical ventilation. Admission chest radiograph showed multifocal airspace opacities (Fig. 3), and computerized tomography (CT) of the chest on day of admission showed multifocal mixed ground glass opacities and dense consolidations throughout bilateral lungs (Fig. 4). Initial arterial blood gas showed a pAO2:FiO2 ratio of 116, consistent with moderate ARDS. His nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. His initial lab work was notable for IL-6 level of 31 pg/mL, ferritin of 320.4 ng/mL, LDH of 321 U/L, D-Dimer of 3.25 UG/mL, high-sensitivity CRP of 244 mg/L, which suggested a high level of inflammation and cell death consistent with potential cytokine storm. He was started on broad spectrum antibiotics (vancomycin and piperacillin/tazobactam), azithromycin 500 mg daily, and hydroxychloroquine 400 mg twice a day for one day followed by 200 mg twice a day. His home oral hydrocortisone for underlying adrenal insufficiency was continued and he did not require stress dose steroids. He was given single 400 mg dose of tocilizumab on his first hospital day, approximately 5 h after intubation, after which transplant hepatology recommended holding home mycophenolate mofetil and tacrolimus given potent immunosuppressive effects of tocilizumab. He was extubated on his second hospital day, and chest radiograph afterwards showed improvement in the multifocal airspace disease (Fig. 5). Tacrolimus was started seven days after administration of tocilizumab. He did not require neuromuscular blockade, vasopressive support, or proning during his hospitalization. He continued to improve and was discharged home on his eighth hospital day without supplemental oxygen requirement. His IL-6 level peaked at 318 pg/mL and decreased to 78 pg/mL on the date of discharge. His mycophenolate was restarted two weeks after administration of tocilizumab.Fig. 3 Chest radiograph of Case 2 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 3
Fig. 4 Axial CT chest on day of admission with multifocal mixed ground glass and dense consolidations throughout bilateral lungs.
Fig. 4
Fig. 5 Chest radiograph of Case 2 on hospital day 4 with slight improvement in multifocal airspace opacities.
Fig. 5
4 Discussion
We have presented two successful cases of single-dose tocilizumab use in critically ill post-transplant patients with moderate COVID-19 ARDS.
Although both patients had rapid improvement with tocilizumab, there are several unique factors. Notably, both patients were already immunosuppressed prior to hospitalization due to their post-transplant status, theoretically causing their respective immune systems to be less able to mount a strong cytokine release in the setting of COVID-19. This could perhaps explain how our cases differ from the results of the study by Lou et al. in Wuhan, China which showed that single-dose tocilizumab often failed to improve disease activity in those who were critically ill, who often needed multiple doses of tocilizumab due to persistently elevated IL-6 levels even when given in conjunction with high-dose glucocorticoids [18]. Both of our patients had a peak IL-6 level over 300 pg/mL during their hospitalization which subsequently trended downward prior to discharge. This particular population of post-transplant patients with severe COVID-19 is underrepresented in current literature and provides a unique perspective in use of novel therapies.
Of note, both patients were also treated with hydroxychloroquine and azithromycin since treatment took part prior to recent studies that revealed no benefit to COVID-19 patients when treated with hydroxychloroquine [21] with or without azithromycin [22]. In fact, some studies found an increase in adverse events for group treated with hydroxychloroquine [21]. Despite confounding factors and limitations, the rapid improvement seen in both patients is both promising and reassuring given that COVID-19 patients who require intubation typically require prolonged mechanical ventilation for average of 10 days [4]. Both patients also have several comorbidities, notably history of solid organ transplant on immunosuppression, and advanced age which puts them in higher risk for extended mechanical ventilation duration and mortality from COVID-19. The therapeutic window for IL-6 inhibitors such as tocilizumab is likely quite narrow. These agents must be given just before or immediately at onset of ARDS and cytokine storm, which tends to occur at day 7–10 of illness, and only in the minority of patients who experience rapid deterioration requiring admission to intensive care for mechanical ventilation (approximately 5–10% of all patients) [1]. Once ARDS has fully evolved, inhibition of immune system would likely do little to help remove proteinaceous fluid in alveoli, thus long mechanical ventilation and admission in the intensive care setting is to be expected.
Moving forward, further randomized controlled trials are necessary to support the findings we report. Given the robust response, we recommend consideration of tocilizumab, especially if other IL-6 inhibitors/blockers are not available, to critically ill patients with confirmed COVID-19 who require intubation for hypoxic respiratory failure with evidence of bilateral ground glass opacities on chest radiograph or CT, as aggressive empiric early treatment will likely have most impact on preventing ARDS and thus decreasing mortality.
Funding sources
This research did not receive any grant support.
Author contributions
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented.
Declaration of competing interest
The authors have no conflicts of interest to disclose.
Acknowledgements
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented. | Recovered | ReactionOutcome | CC BY | 33318918 | 18,853,137 | 2021 |
What was the outcome of reaction 'COVID-19'? | Post-transplant patients with COVID-19 associated acute respiratory distress syndrome, a role for Tociluzumab: A case series.
COVID-19 is the disease caused by SARS-CoV-2 that portends both a relatively high mortality rate as well as high rate of intensive care admission amongst all age groups; however effective therapy remains poorly characterized. Post-transplant patients are especially high risk and underrepresented in the literature. In these patients, cytokine release may play a significant role in the development of acute respiratory distress syndrome, raising the hypothesis that interleukin-6 inhibitors such as tocilizumab may be of benefit. Here, we describe two high-risk post-transplant patients who were treated with single-dose tocilizumab after intubation for moderate acute respiratory distress syndrome secondary to confirmed COVID-19 infection. Both patients recovered rapidly and were successfully extubated and discharged from the hospital without need for supplemental oxygen shortly thereafter, and their clinical improvement correlated with response in interleukin-6 levels. Tocilizumab appears to hold promise for critically ill COVID-19 patients who require mechanical ventilation when given shortly after intubation.
1 Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel beta coronavirus which originated in Wuhan, China in 2019. This virus causes coronavirus disease of 2019 (COVID-19). SARS-CoV-2 has a high virulence, no vaccine, no antiviral treatment, and relatively high mortality of 1.8–3.4%. Disease severity of COVID-19 necessitating admission to intensive care units is estimated at 5–11.5% amongst all age groups [1]. The United States is at particularly high risk for SARS-CoV-2-related mortality due to high prevalence of risk factors for poor outcomes, which include age greater than 65, cardiovascular disease, cerebrovascular disease, hypertension, diabetes mellitus, history of tobacco use, and chronic obstructive pulmonary disease [[2], [3], [4]]. Transplant patients are likely at higher risk given immunosuppression and underrepresented in current literature.
Several treatments are being actively investigated including hydroxychloroquine and azithromycin combination therapy, lopinavir/ritonavir, and remdesivir. These agents have shown antiviral activity against SARS-CoV-2 in vitro [[5], [6], [7]] as well as promise in humans [8,9], though this data is limited. Utility of these agents is likely limited to early disease since they theoretically function as antiviral; late complications of COVID-19 include acute respiratory distress syndrome (ARDS), which may be a result of cytokine storm. Cytokine storm is triggered by release of interferon gamma by T-cells which leads to release of tumor necrosis factor alpha, IL-6, and IL-10 by macrophages which can go on to cause significant lung injury and ARDS, as well as distributive shock and multisystem organ failure [10]. Studies have already shown that lopinavir/ritonavir did not have significant benefit when used late in disease course for varying severities of COVID-19, including ARDS [11,12].
Mortality of late stage ARDS secondary to COVID-19 is high, with studies showing mortality from 48% up to 90% once intubated and placed on mechanical ventilation, which is significantly higher than mortality associated with intubation for other viral pneumonias which is around 22% [13,14]. The reason for this higher mortality could be related to cytokine storm as critically ill COVID-19 patients have cytokine profiles resembling macrophage activation syndrome and secondary hemophagocytic lymphohistiocytosis with significant elevations in IL-1B, IL-2, IL-6, IL-17, IL-8 and tumor necrosis factor [15], with similar serologic markers such as elevated ferritin, elevated liver enzymes, and coagulopathies [16]. Presence of cytokine storm is further supported by significant proportion of COVID-19 patients who are intubated and requiring vasopressors for distributive shock despite no evidence of bacterial superinfection [17]. Thus, an effective therapy is in dire need.
IL-6 inhibitors such as tocilizumab and siltuximab/sarilumab have shown benefit in treatment of cytokine storm in chimeric antigen receptor T-cell patients. These agents have also shown promise for treatment of ARDS in severely ill COVID-19 patients via suppression of cytokine storm [18]. We present two critically ill post-transplant patients, each requiring intubation and mechanical ventilation for severe COVID-19 disease who were treated with single-dose tocilizumab and experienced rapid subsequent improvement at a large tertiary referral center in Gainesville, Florida, United States.
2 Case 1
Case 1 is a 50-year-old male with past medical history of coronary artery disease, nonischemic cardiomyopathy, type 2 diabetes mellitus, essential hypertension, and prior stroke who recently underwent kidney and heart transplant in February 2020. His post-transplant course was complicated with upper respiratory symptoms about one month after transplant, and his nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. During that admission, he did not require supplemental oxygen, and was discharged home to complete a total five-day course of hydroxychloroquine. Three days after discharge, he presented to the emergency department for severe dyspnea. While being evaluated in the emergency department, he experienced rapid and profound hypoxic respiratory decompensation requiring intubation and mechanical ventilation. Initial lab work was significant for IL-6 level of 45 pg/mL, ferritin of 648 ng/mL, LDH of 426 U/L, D-Dimer of 4.84 UG/mL, high-sensitivity CRP of 74.9 mg/L. These elevated markers of inflammation and cell death were consistent with potential cytokine storm. Admission chest radiograph showed significant burden of bilateral airspace opacities (Fig. 1). He was given 400 mg of tocilizumab on his first hospital day, approximately 5 hours after intubation. He was also started on broad spectrum antibiotics (vancomycin and cefepime), azithromycin 500 mg daily, and hydroxychloroquine 200 mg twice a day. His transplant immunosuppression with tacrolimus was continued, he was started on stress dose hydrocortisone 50mg every 6 hours and his mycophenolate was held per International Society for Heart and Lung Transplantation guidelines for moderate-severe COVID-19 ARDS [19]. Due to progressively worsening hypoxia and ARDS (pAO2:FiO2 ratio of 117), he was started on inhaled epoprostenol and subsequently underwent brief neuromuscular blockade with cisatracurium for 4 h on his first hospital night. He continued to improve clinically in the subsequent days with supportive care and lung-protective ventilation per ARDSNet protocol [20], associated with rapid improvement in the airspace opacities seen on admission chest radiograph on repeat on his fourth hospital day (Fig. 2). He did not require proning or vasopressive support during his hospitalization. He was weaned off inhaled epoprostenol on his fifth hospital day, extubated on his seventh hospital day, and was discharged home without supplemental oxygen requirement on his eleventh hospital day. His IL-6 level peaked at 303 pg/mL and decreased to 21 pg/mL on the date of discharge.Fig. 1 Chest radiograph of Case 1 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 1
Fig. 2 Chest radiograph of Case 1 on hospital day 8 post-extubation which shows improvement in multifocal airspace opacities.
Fig. 2
3 Case 2
Case 2 is a 67-year-old male with past medical history of chronic hepatitis B complicated by hepatocellular carcinoma status post orthotopic liver transplant in September 2016, with repeat orthotopic Roux-en-Y liver transplant with pyloric exclusion in November of 2016, history of cryptococcus pneumonia with documented clearance after 28 days of fluconazole in 2018, hypertension, and type 2 diabetes mellitus who presented to the emergency department for one week of fever and dyspnea, and he was admitted to the medical intensive care unit due to acute hypoxic respiratory failure requiring intubation and mechanical ventilation. Admission chest radiograph showed multifocal airspace opacities (Fig. 3), and computerized tomography (CT) of the chest on day of admission showed multifocal mixed ground glass opacities and dense consolidations throughout bilateral lungs (Fig. 4). Initial arterial blood gas showed a pAO2:FiO2 ratio of 116, consistent with moderate ARDS. His nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. His initial lab work was notable for IL-6 level of 31 pg/mL, ferritin of 320.4 ng/mL, LDH of 321 U/L, D-Dimer of 3.25 UG/mL, high-sensitivity CRP of 244 mg/L, which suggested a high level of inflammation and cell death consistent with potential cytokine storm. He was started on broad spectrum antibiotics (vancomycin and piperacillin/tazobactam), azithromycin 500 mg daily, and hydroxychloroquine 400 mg twice a day for one day followed by 200 mg twice a day. His home oral hydrocortisone for underlying adrenal insufficiency was continued and he did not require stress dose steroids. He was given single 400 mg dose of tocilizumab on his first hospital day, approximately 5 h after intubation, after which transplant hepatology recommended holding home mycophenolate mofetil and tacrolimus given potent immunosuppressive effects of tocilizumab. He was extubated on his second hospital day, and chest radiograph afterwards showed improvement in the multifocal airspace disease (Fig. 5). Tacrolimus was started seven days after administration of tocilizumab. He did not require neuromuscular blockade, vasopressive support, or proning during his hospitalization. He continued to improve and was discharged home on his eighth hospital day without supplemental oxygen requirement. His IL-6 level peaked at 318 pg/mL and decreased to 78 pg/mL on the date of discharge. His mycophenolate was restarted two weeks after administration of tocilizumab.Fig. 3 Chest radiograph of Case 2 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 3
Fig. 4 Axial CT chest on day of admission with multifocal mixed ground glass and dense consolidations throughout bilateral lungs.
Fig. 4
Fig. 5 Chest radiograph of Case 2 on hospital day 4 with slight improvement in multifocal airspace opacities.
Fig. 5
4 Discussion
We have presented two successful cases of single-dose tocilizumab use in critically ill post-transplant patients with moderate COVID-19 ARDS.
Although both patients had rapid improvement with tocilizumab, there are several unique factors. Notably, both patients were already immunosuppressed prior to hospitalization due to their post-transplant status, theoretically causing their respective immune systems to be less able to mount a strong cytokine release in the setting of COVID-19. This could perhaps explain how our cases differ from the results of the study by Lou et al. in Wuhan, China which showed that single-dose tocilizumab often failed to improve disease activity in those who were critically ill, who often needed multiple doses of tocilizumab due to persistently elevated IL-6 levels even when given in conjunction with high-dose glucocorticoids [18]. Both of our patients had a peak IL-6 level over 300 pg/mL during their hospitalization which subsequently trended downward prior to discharge. This particular population of post-transplant patients with severe COVID-19 is underrepresented in current literature and provides a unique perspective in use of novel therapies.
Of note, both patients were also treated with hydroxychloroquine and azithromycin since treatment took part prior to recent studies that revealed no benefit to COVID-19 patients when treated with hydroxychloroquine [21] with or without azithromycin [22]. In fact, some studies found an increase in adverse events for group treated with hydroxychloroquine [21]. Despite confounding factors and limitations, the rapid improvement seen in both patients is both promising and reassuring given that COVID-19 patients who require intubation typically require prolonged mechanical ventilation for average of 10 days [4]. Both patients also have several comorbidities, notably history of solid organ transplant on immunosuppression, and advanced age which puts them in higher risk for extended mechanical ventilation duration and mortality from COVID-19. The therapeutic window for IL-6 inhibitors such as tocilizumab is likely quite narrow. These agents must be given just before or immediately at onset of ARDS and cytokine storm, which tends to occur at day 7–10 of illness, and only in the minority of patients who experience rapid deterioration requiring admission to intensive care for mechanical ventilation (approximately 5–10% of all patients) [1]. Once ARDS has fully evolved, inhibition of immune system would likely do little to help remove proteinaceous fluid in alveoli, thus long mechanical ventilation and admission in the intensive care setting is to be expected.
Moving forward, further randomized controlled trials are necessary to support the findings we report. Given the robust response, we recommend consideration of tocilizumab, especially if other IL-6 inhibitors/blockers are not available, to critically ill patients with confirmed COVID-19 who require intubation for hypoxic respiratory failure with evidence of bilateral ground glass opacities on chest radiograph or CT, as aggressive empiric early treatment will likely have most impact on preventing ARDS and thus decreasing mortality.
Funding sources
This research did not receive any grant support.
Author contributions
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented.
Declaration of competing interest
The authors have no conflicts of interest to disclose.
Acknowledgements
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented. | Recovered | ReactionOutcome | CC BY | 33318918 | 18,853,137 | 2021 |
What was the outcome of reaction 'Respiratory failure'? | Post-transplant patients with COVID-19 associated acute respiratory distress syndrome, a role for Tociluzumab: A case series.
COVID-19 is the disease caused by SARS-CoV-2 that portends both a relatively high mortality rate as well as high rate of intensive care admission amongst all age groups; however effective therapy remains poorly characterized. Post-transplant patients are especially high risk and underrepresented in the literature. In these patients, cytokine release may play a significant role in the development of acute respiratory distress syndrome, raising the hypothesis that interleukin-6 inhibitors such as tocilizumab may be of benefit. Here, we describe two high-risk post-transplant patients who were treated with single-dose tocilizumab after intubation for moderate acute respiratory distress syndrome secondary to confirmed COVID-19 infection. Both patients recovered rapidly and were successfully extubated and discharged from the hospital without need for supplemental oxygen shortly thereafter, and their clinical improvement correlated with response in interleukin-6 levels. Tocilizumab appears to hold promise for critically ill COVID-19 patients who require mechanical ventilation when given shortly after intubation.
1 Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel beta coronavirus which originated in Wuhan, China in 2019. This virus causes coronavirus disease of 2019 (COVID-19). SARS-CoV-2 has a high virulence, no vaccine, no antiviral treatment, and relatively high mortality of 1.8–3.4%. Disease severity of COVID-19 necessitating admission to intensive care units is estimated at 5–11.5% amongst all age groups [1]. The United States is at particularly high risk for SARS-CoV-2-related mortality due to high prevalence of risk factors for poor outcomes, which include age greater than 65, cardiovascular disease, cerebrovascular disease, hypertension, diabetes mellitus, history of tobacco use, and chronic obstructive pulmonary disease [[2], [3], [4]]. Transplant patients are likely at higher risk given immunosuppression and underrepresented in current literature.
Several treatments are being actively investigated including hydroxychloroquine and azithromycin combination therapy, lopinavir/ritonavir, and remdesivir. These agents have shown antiviral activity against SARS-CoV-2 in vitro [[5], [6], [7]] as well as promise in humans [8,9], though this data is limited. Utility of these agents is likely limited to early disease since they theoretically function as antiviral; late complications of COVID-19 include acute respiratory distress syndrome (ARDS), which may be a result of cytokine storm. Cytokine storm is triggered by release of interferon gamma by T-cells which leads to release of tumor necrosis factor alpha, IL-6, and IL-10 by macrophages which can go on to cause significant lung injury and ARDS, as well as distributive shock and multisystem organ failure [10]. Studies have already shown that lopinavir/ritonavir did not have significant benefit when used late in disease course for varying severities of COVID-19, including ARDS [11,12].
Mortality of late stage ARDS secondary to COVID-19 is high, with studies showing mortality from 48% up to 90% once intubated and placed on mechanical ventilation, which is significantly higher than mortality associated with intubation for other viral pneumonias which is around 22% [13,14]. The reason for this higher mortality could be related to cytokine storm as critically ill COVID-19 patients have cytokine profiles resembling macrophage activation syndrome and secondary hemophagocytic lymphohistiocytosis with significant elevations in IL-1B, IL-2, IL-6, IL-17, IL-8 and tumor necrosis factor [15], with similar serologic markers such as elevated ferritin, elevated liver enzymes, and coagulopathies [16]. Presence of cytokine storm is further supported by significant proportion of COVID-19 patients who are intubated and requiring vasopressors for distributive shock despite no evidence of bacterial superinfection [17]. Thus, an effective therapy is in dire need.
IL-6 inhibitors such as tocilizumab and siltuximab/sarilumab have shown benefit in treatment of cytokine storm in chimeric antigen receptor T-cell patients. These agents have also shown promise for treatment of ARDS in severely ill COVID-19 patients via suppression of cytokine storm [18]. We present two critically ill post-transplant patients, each requiring intubation and mechanical ventilation for severe COVID-19 disease who were treated with single-dose tocilizumab and experienced rapid subsequent improvement at a large tertiary referral center in Gainesville, Florida, United States.
2 Case 1
Case 1 is a 50-year-old male with past medical history of coronary artery disease, nonischemic cardiomyopathy, type 2 diabetes mellitus, essential hypertension, and prior stroke who recently underwent kidney and heart transplant in February 2020. His post-transplant course was complicated with upper respiratory symptoms about one month after transplant, and his nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. During that admission, he did not require supplemental oxygen, and was discharged home to complete a total five-day course of hydroxychloroquine. Three days after discharge, he presented to the emergency department for severe dyspnea. While being evaluated in the emergency department, he experienced rapid and profound hypoxic respiratory decompensation requiring intubation and mechanical ventilation. Initial lab work was significant for IL-6 level of 45 pg/mL, ferritin of 648 ng/mL, LDH of 426 U/L, D-Dimer of 4.84 UG/mL, high-sensitivity CRP of 74.9 mg/L. These elevated markers of inflammation and cell death were consistent with potential cytokine storm. Admission chest radiograph showed significant burden of bilateral airspace opacities (Fig. 1). He was given 400 mg of tocilizumab on his first hospital day, approximately 5 hours after intubation. He was also started on broad spectrum antibiotics (vancomycin and cefepime), azithromycin 500 mg daily, and hydroxychloroquine 200 mg twice a day. His transplant immunosuppression with tacrolimus was continued, he was started on stress dose hydrocortisone 50mg every 6 hours and his mycophenolate was held per International Society for Heart and Lung Transplantation guidelines for moderate-severe COVID-19 ARDS [19]. Due to progressively worsening hypoxia and ARDS (pAO2:FiO2 ratio of 117), he was started on inhaled epoprostenol and subsequently underwent brief neuromuscular blockade with cisatracurium for 4 h on his first hospital night. He continued to improve clinically in the subsequent days with supportive care and lung-protective ventilation per ARDSNet protocol [20], associated with rapid improvement in the airspace opacities seen on admission chest radiograph on repeat on his fourth hospital day (Fig. 2). He did not require proning or vasopressive support during his hospitalization. He was weaned off inhaled epoprostenol on his fifth hospital day, extubated on his seventh hospital day, and was discharged home without supplemental oxygen requirement on his eleventh hospital day. His IL-6 level peaked at 303 pg/mL and decreased to 21 pg/mL on the date of discharge.Fig. 1 Chest radiograph of Case 1 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 1
Fig. 2 Chest radiograph of Case 1 on hospital day 8 post-extubation which shows improvement in multifocal airspace opacities.
Fig. 2
3 Case 2
Case 2 is a 67-year-old male with past medical history of chronic hepatitis B complicated by hepatocellular carcinoma status post orthotopic liver transplant in September 2016, with repeat orthotopic Roux-en-Y liver transplant with pyloric exclusion in November of 2016, history of cryptococcus pneumonia with documented clearance after 28 days of fluconazole in 2018, hypertension, and type 2 diabetes mellitus who presented to the emergency department for one week of fever and dyspnea, and he was admitted to the medical intensive care unit due to acute hypoxic respiratory failure requiring intubation and mechanical ventilation. Admission chest radiograph showed multifocal airspace opacities (Fig. 3), and computerized tomography (CT) of the chest on day of admission showed multifocal mixed ground glass opacities and dense consolidations throughout bilateral lungs (Fig. 4). Initial arterial blood gas showed a pAO2:FiO2 ratio of 116, consistent with moderate ARDS. His nasopharyngeal nucleic acid amplification test for SARS-CoV-2 returned positive. His initial lab work was notable for IL-6 level of 31 pg/mL, ferritin of 320.4 ng/mL, LDH of 321 U/L, D-Dimer of 3.25 UG/mL, high-sensitivity CRP of 244 mg/L, which suggested a high level of inflammation and cell death consistent with potential cytokine storm. He was started on broad spectrum antibiotics (vancomycin and piperacillin/tazobactam), azithromycin 500 mg daily, and hydroxychloroquine 400 mg twice a day for one day followed by 200 mg twice a day. His home oral hydrocortisone for underlying adrenal insufficiency was continued and he did not require stress dose steroids. He was given single 400 mg dose of tocilizumab on his first hospital day, approximately 5 h after intubation, after which transplant hepatology recommended holding home mycophenolate mofetil and tacrolimus given potent immunosuppressive effects of tocilizumab. He was extubated on his second hospital day, and chest radiograph afterwards showed improvement in the multifocal airspace disease (Fig. 5). Tacrolimus was started seven days after administration of tocilizumab. He did not require neuromuscular blockade, vasopressive support, or proning during his hospitalization. He continued to improve and was discharged home on his eighth hospital day without supplemental oxygen requirement. His IL-6 level peaked at 318 pg/mL and decreased to 78 pg/mL on the date of discharge. His mycophenolate was restarted two weeks after administration of tocilizumab.Fig. 3 Chest radiograph of Case 2 on day of admission which shows bilateral multifocal airspace opacities.
Fig. 3
Fig. 4 Axial CT chest on day of admission with multifocal mixed ground glass and dense consolidations throughout bilateral lungs.
Fig. 4
Fig. 5 Chest radiograph of Case 2 on hospital day 4 with slight improvement in multifocal airspace opacities.
Fig. 5
4 Discussion
We have presented two successful cases of single-dose tocilizumab use in critically ill post-transplant patients with moderate COVID-19 ARDS.
Although both patients had rapid improvement with tocilizumab, there are several unique factors. Notably, both patients were already immunosuppressed prior to hospitalization due to their post-transplant status, theoretically causing their respective immune systems to be less able to mount a strong cytokine release in the setting of COVID-19. This could perhaps explain how our cases differ from the results of the study by Lou et al. in Wuhan, China which showed that single-dose tocilizumab often failed to improve disease activity in those who were critically ill, who often needed multiple doses of tocilizumab due to persistently elevated IL-6 levels even when given in conjunction with high-dose glucocorticoids [18]. Both of our patients had a peak IL-6 level over 300 pg/mL during their hospitalization which subsequently trended downward prior to discharge. This particular population of post-transplant patients with severe COVID-19 is underrepresented in current literature and provides a unique perspective in use of novel therapies.
Of note, both patients were also treated with hydroxychloroquine and azithromycin since treatment took part prior to recent studies that revealed no benefit to COVID-19 patients when treated with hydroxychloroquine [21] with or without azithromycin [22]. In fact, some studies found an increase in adverse events for group treated with hydroxychloroquine [21]. Despite confounding factors and limitations, the rapid improvement seen in both patients is both promising and reassuring given that COVID-19 patients who require intubation typically require prolonged mechanical ventilation for average of 10 days [4]. Both patients also have several comorbidities, notably history of solid organ transplant on immunosuppression, and advanced age which puts them in higher risk for extended mechanical ventilation duration and mortality from COVID-19. The therapeutic window for IL-6 inhibitors such as tocilizumab is likely quite narrow. These agents must be given just before or immediately at onset of ARDS and cytokine storm, which tends to occur at day 7–10 of illness, and only in the minority of patients who experience rapid deterioration requiring admission to intensive care for mechanical ventilation (approximately 5–10% of all patients) [1]. Once ARDS has fully evolved, inhibition of immune system would likely do little to help remove proteinaceous fluid in alveoli, thus long mechanical ventilation and admission in the intensive care setting is to be expected.
Moving forward, further randomized controlled trials are necessary to support the findings we report. Given the robust response, we recommend consideration of tocilizumab, especially if other IL-6 inhibitors/blockers are not available, to critically ill patients with confirmed COVID-19 who require intubation for hypoxic respiratory failure with evidence of bilateral ground glass opacities on chest radiograph or CT, as aggressive empiric early treatment will likely have most impact on preventing ARDS and thus decreasing mortality.
Funding sources
This research did not receive any grant support.
Author contributions
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented.
Declaration of competing interest
The authors have no conflicts of interest to disclose.
Acknowledgements
ML, FLV, and PBM had full access to both cases in this study and drafted the original manuscript, and JL provided critical revision. ML, FLV, PBM, and JL were involved in the conception of the project and the approval of the final manuscript, and they will take responsibility for the integrity and accuracy of the data presented. | Recovered | ReactionOutcome | CC BY | 33318918 | 18,853,137 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Haemoglobin decreased'. | Large pelvic hematoma following UroLift procedure causing renal failure requiring dialysis.
The prostatic urethral lift procedure is a minimally invasive treatment option for lower urinary tract symptoms due to benign prostatic hyperplasia, with reported benefit of less adverse effects than traditional treatments. While complications are usually minimal, our patient developed a large pelvic hematoma and the first case of organ failure after prostatic urethral lift. He required temporary dialysis during his extended postoperative admission, and his chronic kidney disease permanently progressed from stage III to stage IV. This case highlights the need for research into the safest preoperative and operative approach for prostatic urethral lift procedures in patients with comorbidities.
Introduction
Compared to traditional modalities such as transurethral resection of the prostate (TURP), the UroLift® prostatic urethral lift procedure effectively treats voiding symptoms from benign prostatic hyperplasia (BPH) with less bleeding and shorter duration postoperative incontinence.1,2 The most frequently described UroLift® complications are self-limiting dysuria and hematuria.1 Two recent reports describe pelvic hematoma after UroLift®: one underwent conservative treatment, the other fulguration and tying off of small vessel under general anesthesia.3 We describe a third large pelvic hematoma, with blood loss contributing to acute renal failure requiring dialysis. After extensive literature search, we believe this is the first incidence of single organ failure after UroLift®.
Case presentation
An 83-year-old male with prostate cancer treated with radiation in 2017 presented to urology clinic with voiding complaints. Medical history included chronic kidney disease (CKD) stage III [baseline creatinine (Cr) 1.92 mg/dL, glomerular filtration rate (GFR) 34] and atrial fibrillation on warfarin. He endorsed urinary frequency and nocturia every 2 h, stream intermittency, and incomplete emptying. International Prostate Symptom Score was 17 (moderate voiding symptoms). Prostate measured 25 g on transrectal ultrasound. Cystoscopy revealed visually obstructing prostate with non-protruding median lobe, no urethral strictures. Uroflow demonstrated: low flow rate, small voided volume, post void residual 97 mL. Maximal medical therapy was unsuccessful. One week before surgery, he was bridged from warfarin to enoxaparin, with final enoxaparin dose given the morning of procedure.
UroLift® was performed under general anesthesia, employing five prostatic implants: two in each lateral lobe, one pinning the median lobe from left to right per recommended technique (Fig. 1, Fig. 2).4 At completion, an anterior channel was visible. No complications occurred intraoperatively; blood loss was minimal. Foley catheter placed for removal in recovery, to which he was transferred in stable condition.Fig. 1 Visualization of intraurethral portions of UroLift® prostatic urethral lift devices.
Fig. 1Fig. 2 Visualization of capsular portions of UroLift® prostatic urethral lift devices.
Fig. 2
Upon catheter removal, he experienced syncope. He was monitored 2 h on bed rest while receiving 2 L intravenous normal saline. He then experienced another syncopal episode upon standing. Systolic blood pressure (BP) was 80 mmHg from preoperative 120 mmHg. Suprapubic mass palpated. He was transferred to a local Emergency Department (ED) for evaluation.
ED evaluation revealed BP 63/33 and hemoglobin (Hb) 8.7 g/dL, ~2 points below preoperative values. Abdomen was moderately distended. After resuscitation, BP improved temporarily. Hb dropped further to 7.3 g/dL with normal internal normalized ratio. Despite Cr 1.99 mg/dL (GFR 34), CT torso with intravenous contrast was performed to rule out significant vascular injury, demonstrating 15cm hematoma in the space of Retzius (Fig. 3), no active extravasation. He was transfused two units packed red blood cells (PRBCs) and admitted to surgical intensive care unit (SICU) for monitoring. Foley catheter placed for strict fluid balance; hematuria noted.Fig. 3 Large pelvic hematoma compressing bladder several hours after UroLift® procedure.
Fig. 3
By postoperative day (POD) 2, six more units PRBCs were given, stabilizing Hgb to 7.5 g/dL. Systolic BP remained 80–100 mmHg with no vasopressors. Creatinine continued to rise to peak of 6.71 mg/dL (GFR 8). He became hyperkalemic and oliguric; hemodialysis was initiated POD3. Final 2 units PRBCs were required that week for Hb target >8.0, bringing total transfusions to 10. Renal function slowly improved; last dialysis was POD8, the day he left SICU for floor care. Catheter removed POD9, revealing new onset urge incontinence. Discharge on POD12 was to rehabilitation due to deconditioning, then to home POD18. Discharge laboratory values included new baseline Cr of 2.64 mg/dL (GFR 23). He had progressed to CKD stage IV, due to prolonged hypotension from acute blood loss and contrast loading during hypovolemic state. In the 90 days postoperatively, he had three readmissions for recurrent fever and failure to thrive, twice with urinary tract infection. Six months postoperatively, he was diagnosed with bilateral lower extremity deep vein thrombosis and aortic stenosis causing systolic heart failure.
In urologic follow up, cystoscopy at two months demonstrated edema of bladder with external compression, no visible clips. Imaging at eight months revealed no pelvic hematoma. One year postoperatively, urinary frequency was worse compared to preop, with urge incontinence requiring pads. Comparative improvements included better flow and complete emptying.
Discussion
The UroLift® procedure effectively treats BPH.2 Improvements in voiding parameters are reported with only mild dysuria and hematuria, less risk of blood loss, and shorter duration transient urge incontinence than traditional modalities such as TURP.1 This remains true even when used for obstructive median lobes.4 To date, there have been no cases of bleeding requiring massive blood transfusion.2 One probability estimate of such an event was nearly zero.2 Two case reports describe pelvic hematoma after UroLift®; one required conservative management, the other return to operating room.3 In contrast, our patient required ten units PRBCs and developed acute renal failure requiring temporary dialysis, with progression from CKD III to IV.
Contributors to renal dysfunction included hypotension from acute blood loss anemia and use of intravenous contrast in a CKD patient. While contrast-induced nephropathy is not an independent risk factor for dialysis, acute and chronic renal failure may raise risk.5 As first line treatment for pelvic hematoma is conservative, an alternate approach would have been to perform serial imaging, monitoring need for delayed intervention.
When reviewing studies of morbidities after surgery, patient selection should be discussed. At this time, there have been no studies of UroLift® in patients with CKD nor in previously radiated prostates. There are likewise no studies examining Urolift® complications in anticoagulated patients. Current guidance focuses on emphasizing complications to higher-risk patients.
Conclusion
The overall incidence of bleeding after UroLift® remains relatively low, with blood transfusion described as “not a risk due to the nature of the UroLift® procedure.”2 Our case and select others indicate this is no longer true with UroLift® rising in usage. While there was likely no further preoperative workup to predict our patient's complication, knowledge of added risks in patient selection provided from case studies may be helpful in treatment or technique choice. Research is necessary to determine the morbidity risk for Urolift® patients on anticoagulation, and with comorbidities such as CKD and radiation, as these are prevalent in urologic populations. In this way, the safest preoperative and operative approach can be designed.
Consent
Patient information de-identified prior to submission of case report.
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
The authors certify that they have no affiliations with or involvement in any organization of entity with any financial interest or non-financial interest in the subject matter discussed in this manuscript. | WARFARIN SODIUM | DrugsGivenReaction | CC BY-NC-ND | 33318939 | 18,771,341 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pelvic haematoma'. | Large pelvic hematoma following UroLift procedure causing renal failure requiring dialysis.
The prostatic urethral lift procedure is a minimally invasive treatment option for lower urinary tract symptoms due to benign prostatic hyperplasia, with reported benefit of less adverse effects than traditional treatments. While complications are usually minimal, our patient developed a large pelvic hematoma and the first case of organ failure after prostatic urethral lift. He required temporary dialysis during his extended postoperative admission, and his chronic kidney disease permanently progressed from stage III to stage IV. This case highlights the need for research into the safest preoperative and operative approach for prostatic urethral lift procedures in patients with comorbidities.
Introduction
Compared to traditional modalities such as transurethral resection of the prostate (TURP), the UroLift® prostatic urethral lift procedure effectively treats voiding symptoms from benign prostatic hyperplasia (BPH) with less bleeding and shorter duration postoperative incontinence.1,2 The most frequently described UroLift® complications are self-limiting dysuria and hematuria.1 Two recent reports describe pelvic hematoma after UroLift®: one underwent conservative treatment, the other fulguration and tying off of small vessel under general anesthesia.3 We describe a third large pelvic hematoma, with blood loss contributing to acute renal failure requiring dialysis. After extensive literature search, we believe this is the first incidence of single organ failure after UroLift®.
Case presentation
An 83-year-old male with prostate cancer treated with radiation in 2017 presented to urology clinic with voiding complaints. Medical history included chronic kidney disease (CKD) stage III [baseline creatinine (Cr) 1.92 mg/dL, glomerular filtration rate (GFR) 34] and atrial fibrillation on warfarin. He endorsed urinary frequency and nocturia every 2 h, stream intermittency, and incomplete emptying. International Prostate Symptom Score was 17 (moderate voiding symptoms). Prostate measured 25 g on transrectal ultrasound. Cystoscopy revealed visually obstructing prostate with non-protruding median lobe, no urethral strictures. Uroflow demonstrated: low flow rate, small voided volume, post void residual 97 mL. Maximal medical therapy was unsuccessful. One week before surgery, he was bridged from warfarin to enoxaparin, with final enoxaparin dose given the morning of procedure.
UroLift® was performed under general anesthesia, employing five prostatic implants: two in each lateral lobe, one pinning the median lobe from left to right per recommended technique (Fig. 1, Fig. 2).4 At completion, an anterior channel was visible. No complications occurred intraoperatively; blood loss was minimal. Foley catheter placed for removal in recovery, to which he was transferred in stable condition.Fig. 1 Visualization of intraurethral portions of UroLift® prostatic urethral lift devices.
Fig. 1Fig. 2 Visualization of capsular portions of UroLift® prostatic urethral lift devices.
Fig. 2
Upon catheter removal, he experienced syncope. He was monitored 2 h on bed rest while receiving 2 L intravenous normal saline. He then experienced another syncopal episode upon standing. Systolic blood pressure (BP) was 80 mmHg from preoperative 120 mmHg. Suprapubic mass palpated. He was transferred to a local Emergency Department (ED) for evaluation.
ED evaluation revealed BP 63/33 and hemoglobin (Hb) 8.7 g/dL, ~2 points below preoperative values. Abdomen was moderately distended. After resuscitation, BP improved temporarily. Hb dropped further to 7.3 g/dL with normal internal normalized ratio. Despite Cr 1.99 mg/dL (GFR 34), CT torso with intravenous contrast was performed to rule out significant vascular injury, demonstrating 15cm hematoma in the space of Retzius (Fig. 3), no active extravasation. He was transfused two units packed red blood cells (PRBCs) and admitted to surgical intensive care unit (SICU) for monitoring. Foley catheter placed for strict fluid balance; hematuria noted.Fig. 3 Large pelvic hematoma compressing bladder several hours after UroLift® procedure.
Fig. 3
By postoperative day (POD) 2, six more units PRBCs were given, stabilizing Hgb to 7.5 g/dL. Systolic BP remained 80–100 mmHg with no vasopressors. Creatinine continued to rise to peak of 6.71 mg/dL (GFR 8). He became hyperkalemic and oliguric; hemodialysis was initiated POD3. Final 2 units PRBCs were required that week for Hb target >8.0, bringing total transfusions to 10. Renal function slowly improved; last dialysis was POD8, the day he left SICU for floor care. Catheter removed POD9, revealing new onset urge incontinence. Discharge on POD12 was to rehabilitation due to deconditioning, then to home POD18. Discharge laboratory values included new baseline Cr of 2.64 mg/dL (GFR 23). He had progressed to CKD stage IV, due to prolonged hypotension from acute blood loss and contrast loading during hypovolemic state. In the 90 days postoperatively, he had three readmissions for recurrent fever and failure to thrive, twice with urinary tract infection. Six months postoperatively, he was diagnosed with bilateral lower extremity deep vein thrombosis and aortic stenosis causing systolic heart failure.
In urologic follow up, cystoscopy at two months demonstrated edema of bladder with external compression, no visible clips. Imaging at eight months revealed no pelvic hematoma. One year postoperatively, urinary frequency was worse compared to preop, with urge incontinence requiring pads. Comparative improvements included better flow and complete emptying.
Discussion
The UroLift® procedure effectively treats BPH.2 Improvements in voiding parameters are reported with only mild dysuria and hematuria, less risk of blood loss, and shorter duration transient urge incontinence than traditional modalities such as TURP.1 This remains true even when used for obstructive median lobes.4 To date, there have been no cases of bleeding requiring massive blood transfusion.2 One probability estimate of such an event was nearly zero.2 Two case reports describe pelvic hematoma after UroLift®; one required conservative management, the other return to operating room.3 In contrast, our patient required ten units PRBCs and developed acute renal failure requiring temporary dialysis, with progression from CKD III to IV.
Contributors to renal dysfunction included hypotension from acute blood loss anemia and use of intravenous contrast in a CKD patient. While contrast-induced nephropathy is not an independent risk factor for dialysis, acute and chronic renal failure may raise risk.5 As first line treatment for pelvic hematoma is conservative, an alternate approach would have been to perform serial imaging, monitoring need for delayed intervention.
When reviewing studies of morbidities after surgery, patient selection should be discussed. At this time, there have been no studies of UroLift® in patients with CKD nor in previously radiated prostates. There are likewise no studies examining Urolift® complications in anticoagulated patients. Current guidance focuses on emphasizing complications to higher-risk patients.
Conclusion
The overall incidence of bleeding after UroLift® remains relatively low, with blood transfusion described as “not a risk due to the nature of the UroLift® procedure.”2 Our case and select others indicate this is no longer true with UroLift® rising in usage. While there was likely no further preoperative workup to predict our patient's complication, knowledge of added risks in patient selection provided from case studies may be helpful in treatment or technique choice. Research is necessary to determine the morbidity risk for Urolift® patients on anticoagulation, and with comorbidities such as CKD and radiation, as these are prevalent in urologic populations. In this way, the safest preoperative and operative approach can be designed.
Consent
Patient information de-identified prior to submission of case report.
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
The authors certify that they have no affiliations with or involvement in any organization of entity with any financial interest or non-financial interest in the subject matter discussed in this manuscript. | WARFARIN SODIUM | DrugsGivenReaction | CC BY-NC-ND | 33318939 | 18,771,341 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Renal cell carcinoma stage III'. | Double partial nephrectomy in allograft transplanted kidney.
A 61-year-old female presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior. As the 10 year survival of living transplant recipients increases, malignancy presentations will continue to rise. Nephron sparing surgery in renal allografts is sparse due to difficult operative dissection and complicated hila vascular control. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as a viable option to safely perform partial nephrectomy in a transplanted kidney.
Introduction
Transplant recipients are twice as likely to develop a malignancy compared to the rest of the population. However, the risk of renal cell carcinoma (RCC) in this population is only mildly elevated at 4.6% compared with 3% in the general population. Only 10% of these tumours occur in the allograft. End stage kidney disease and chronic immunosuppression is known to pre-dispose native kidneys to malignancy. However, the aetiology of malignancy in allografts is poorly understood with no consensus on treatment.1
Historically, radical nephrectomy (RN) was considered first line treatment due to the potential for rapidly progressive RCC in an immunosuppressed patient. This placed patients back on dialysis with associated mortality and morbidity. Nephron sparing surgical (NSS) approaches are considered risky due to the hostile operative environment. However, sparse case report data indicate technical feasibility.2 We present a novel technique to perform off clamp NSS in a transplant recipient presenting with two tumours in an allograft kidney.
Case report
A 61-year-old women presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior into the right iliac fossa. This was uncomplicated and her serum creatinine post-transplant ranged between 95 and 110μmol/L. Maintenance immunosuppressive therapy included tacrolimus, mycophenolate and low dose prednisolone (3mg daily). Her background included steroid induced insulin-dependent diabetes mellitus and osteopenia. She was independent with all activities of daily living.
A computed tomography (CT) scan identified two enhancing renal lesions which were partly exophytic with some cystic components. The larger 3cm lesion was located in the anterior mid-pole region and the smaller 1.5cm lesion was located in the anterior upper pole region of the transplanted kidney (Fig. 1). No metastatic spread was found on staging. An open dual partial nephrectomy of the allograft kidney was planned.Fig. 1 Pre-operative planning CT scan of the abdomen and pelvis with intravenous contrast. A – sagittal view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). B – sagittal view showing cystic components of the allograft lesion (blue arrow). C – axial view showing an enhancing anterior 1.5cm upper pole solid allograft lesion (green arrow). D – axial view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 1
The patient was positioned supine and a 10cm incision was made through the previous Gibson incision. Dissection to the renal parenchyma was performed exposing the anterior and upper pole of the kidney. Intraoperative ultrasonography was used to identify both tumours (Fig. 2). Dissection of the hilum was not performed due to adhesions, short hilar vessels and to limit trauma to the transplanted vessels. Dual partial nephrectomy using cold cut and diathermy achieved excision of both tumours. A curved bowel clamp was utilised to apply local graded manual pressure directly underneath the tumour site during the resection of the tumour and subsequent renorrhaphy closure to minimise haemorrhage (Fig. 3A/B). A 2-0 V-loc suture was used for renorrhaphy closure. There was approximately 500mL of blood loss. The post-operative course was uncomplicated and the patient was discharged day 6 with a serum creatinine of 111μmol/L. The histopathology revealed Grade 3 clear cell RCC with negative margins at both sites and at one month follow up the patient remained well (Fig. 3C/D). The maintenance immunosuppressants were not altered before or after surgery.Fig. 2 Dissection to the allograft kidney. A – superior portion of kidney with peritoneal reflection (1), bulging upper tumour (2) and healthy renal parenchyma (3) B – inferior portion of kidney with lower pole tumour (4) and healthy renal parenchyma (3).
Fig. 2Fig. 3 Nephron sparing resection of allograft tumours. A – novel use of soft bowel clamp (1) to the upper pole tumour. B – The same is shown in for the lower pole tumour. C – upper pole allograft tumour resected. D – lower pole allograft tumour resected.
Fig. 3
Discussion
Transplant recipients are living longer than ever before. According to the 2018 Australian and New Zealand Scientific Registry of Transplant Patients, the 10 year survival for living transplant recipients has increased from 84% to 88% compared to the previous decade. This will likely see increasing presentations of malignancy within the allograft. Although the aetiology for RCC in allografts is not completely understood, theories exist within the literature. A recent study from Germany examined 1655 transplant patients for approximately 12 years.3 Twenty-six cases of RCC after transplantation were diagnosed. Post-transplant RCC was significantly associated with longer durations of pre-transplant haemodialysis (p = 0.007), post-transplant immunosuppression with cyclosporine (p = 0.029) and/or mycophenolate (p = 0.020) and with post-transplant prednisolone (p = 0.042). Cyclosporine A and mycophenolate usage were also independent risk factors for RCC development. The patient in our study had been on mycophenolate and prednisolone for over 10 years.
The literature on nephron sparing surgery in renal allografts is sparse and mostly limited to patients with T1a allograft masses.2 Traditionally all other masses are managed with full RN of the allograft kidney. The European Organisation for Research and Treatment of Cancer (EORTC) randomised trial 30904 is the only multi-centre international randomised control trial which compared NSS and RN in 541 non-transplant patients with small (≤5 cm) singular renal tumours.4 NSS was shown to reduce post-operative moderate renal dysfunction (eGFR <60) by 21% (95% CI, 13.8–28.3) when compared with RN with no difference in cancer-specific survival or oncological clearance. Re-operation rates in NSS was slightly higher compared to RN (4.4% compared with 2.4%) with bleeding accounting for 71% of surgical take-back. Although this study was completed in native kidney disease, results suggest NSS is a viable alternative.
Partial nephrectomy of the renal allograft is a challenging procedure due to adhesions. Severe adhesions make kidney mobilisation, hilum control and parenchymal resection difficult. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as an alternative to hilum vascular control. As shown in Fig. 3A/B close proximal clamping of the parenchyma surrounding the tumour during resection provides tamponade pressure and allows safe resection of the tumour until haemostatic sutures and/or cautery can be attended. This eliminates the need to dissect out the renal hilum of the allograft and provides zero global ischemia to maximise nephron sparing.
Immunosuppressive adjustment reduces allograft malignancy without compromising rejection. The CONVERT trial randomised 830 renal allograft recipients 6–120 months post-transplant to remain on calcineurin inhibitiors (CNI) or convert CNI to mammalian target of rapamycin (mTOR) inhibitor.5 Conversion to an mTOR inhibitior was associated with a lower incidence of all malignancy (12% compared with 21% p=<0.001) with no significant difference in biopsy confirmed acute rejection. The patient in our case would therefore benefit from conversion of CNI to mTOR inhibitor.
Conclusion
Partial nephrectomy is a safe and viable option to consider in the management of small renal masses in the transplanted kidney. The use of manual atraumatic graded bowel clamp pressure around the resected tumour is an alternative to hilum vascular control and can facilitate safe effective resection.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.
Declaration of competing interest
Nil.
Acknowledgements
Nil. | MYCOPHENOLIC ACID, PREDNISOLONE, TACROLIMUS | DrugsGivenReaction | CC BY-NC-ND | 33318944 | 19,597,832 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Renal cell carcinoma'. | Double partial nephrectomy in allograft transplanted kidney.
A 61-year-old female presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior. As the 10 year survival of living transplant recipients increases, malignancy presentations will continue to rise. Nephron sparing surgery in renal allografts is sparse due to difficult operative dissection and complicated hila vascular control. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as a viable option to safely perform partial nephrectomy in a transplanted kidney.
Introduction
Transplant recipients are twice as likely to develop a malignancy compared to the rest of the population. However, the risk of renal cell carcinoma (RCC) in this population is only mildly elevated at 4.6% compared with 3% in the general population. Only 10% of these tumours occur in the allograft. End stage kidney disease and chronic immunosuppression is known to pre-dispose native kidneys to malignancy. However, the aetiology of malignancy in allografts is poorly understood with no consensus on treatment.1
Historically, radical nephrectomy (RN) was considered first line treatment due to the potential for rapidly progressive RCC in an immunosuppressed patient. This placed patients back on dialysis with associated mortality and morbidity. Nephron sparing surgical (NSS) approaches are considered risky due to the hostile operative environment. However, sparse case report data indicate technical feasibility.2 We present a novel technique to perform off clamp NSS in a transplant recipient presenting with two tumours in an allograft kidney.
Case report
A 61-year-old women presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior into the right iliac fossa. This was uncomplicated and her serum creatinine post-transplant ranged between 95 and 110μmol/L. Maintenance immunosuppressive therapy included tacrolimus, mycophenolate and low dose prednisolone (3mg daily). Her background included steroid induced insulin-dependent diabetes mellitus and osteopenia. She was independent with all activities of daily living.
A computed tomography (CT) scan identified two enhancing renal lesions which were partly exophytic with some cystic components. The larger 3cm lesion was located in the anterior mid-pole region and the smaller 1.5cm lesion was located in the anterior upper pole region of the transplanted kidney (Fig. 1). No metastatic spread was found on staging. An open dual partial nephrectomy of the allograft kidney was planned.Fig. 1 Pre-operative planning CT scan of the abdomen and pelvis with intravenous contrast. A – sagittal view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). B – sagittal view showing cystic components of the allograft lesion (blue arrow). C – axial view showing an enhancing anterior 1.5cm upper pole solid allograft lesion (green arrow). D – axial view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 1
The patient was positioned supine and a 10cm incision was made through the previous Gibson incision. Dissection to the renal parenchyma was performed exposing the anterior and upper pole of the kidney. Intraoperative ultrasonography was used to identify both tumours (Fig. 2). Dissection of the hilum was not performed due to adhesions, short hilar vessels and to limit trauma to the transplanted vessels. Dual partial nephrectomy using cold cut and diathermy achieved excision of both tumours. A curved bowel clamp was utilised to apply local graded manual pressure directly underneath the tumour site during the resection of the tumour and subsequent renorrhaphy closure to minimise haemorrhage (Fig. 3A/B). A 2-0 V-loc suture was used for renorrhaphy closure. There was approximately 500mL of blood loss. The post-operative course was uncomplicated and the patient was discharged day 6 with a serum creatinine of 111μmol/L. The histopathology revealed Grade 3 clear cell RCC with negative margins at both sites and at one month follow up the patient remained well (Fig. 3C/D). The maintenance immunosuppressants were not altered before or after surgery.Fig. 2 Dissection to the allograft kidney. A – superior portion of kidney with peritoneal reflection (1), bulging upper tumour (2) and healthy renal parenchyma (3) B – inferior portion of kidney with lower pole tumour (4) and healthy renal parenchyma (3).
Fig. 2Fig. 3 Nephron sparing resection of allograft tumours. A – novel use of soft bowel clamp (1) to the upper pole tumour. B – The same is shown in for the lower pole tumour. C – upper pole allograft tumour resected. D – lower pole allograft tumour resected.
Fig. 3
Discussion
Transplant recipients are living longer than ever before. According to the 2018 Australian and New Zealand Scientific Registry of Transplant Patients, the 10 year survival for living transplant recipients has increased from 84% to 88% compared to the previous decade. This will likely see increasing presentations of malignancy within the allograft. Although the aetiology for RCC in allografts is not completely understood, theories exist within the literature. A recent study from Germany examined 1655 transplant patients for approximately 12 years.3 Twenty-six cases of RCC after transplantation were diagnosed. Post-transplant RCC was significantly associated with longer durations of pre-transplant haemodialysis (p = 0.007), post-transplant immunosuppression with cyclosporine (p = 0.029) and/or mycophenolate (p = 0.020) and with post-transplant prednisolone (p = 0.042). Cyclosporine A and mycophenolate usage were also independent risk factors for RCC development. The patient in our study had been on mycophenolate and prednisolone for over 10 years.
The literature on nephron sparing surgery in renal allografts is sparse and mostly limited to patients with T1a allograft masses.2 Traditionally all other masses are managed with full RN of the allograft kidney. The European Organisation for Research and Treatment of Cancer (EORTC) randomised trial 30904 is the only multi-centre international randomised control trial which compared NSS and RN in 541 non-transplant patients with small (≤5 cm) singular renal tumours.4 NSS was shown to reduce post-operative moderate renal dysfunction (eGFR <60) by 21% (95% CI, 13.8–28.3) when compared with RN with no difference in cancer-specific survival or oncological clearance. Re-operation rates in NSS was slightly higher compared to RN (4.4% compared with 2.4%) with bleeding accounting for 71% of surgical take-back. Although this study was completed in native kidney disease, results suggest NSS is a viable alternative.
Partial nephrectomy of the renal allograft is a challenging procedure due to adhesions. Severe adhesions make kidney mobilisation, hilum control and parenchymal resection difficult. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as an alternative to hilum vascular control. As shown in Fig. 3A/B close proximal clamping of the parenchyma surrounding the tumour during resection provides tamponade pressure and allows safe resection of the tumour until haemostatic sutures and/or cautery can be attended. This eliminates the need to dissect out the renal hilum of the allograft and provides zero global ischemia to maximise nephron sparing.
Immunosuppressive adjustment reduces allograft malignancy without compromising rejection. The CONVERT trial randomised 830 renal allograft recipients 6–120 months post-transplant to remain on calcineurin inhibitiors (CNI) or convert CNI to mammalian target of rapamycin (mTOR) inhibitor.5 Conversion to an mTOR inhibitior was associated with a lower incidence of all malignancy (12% compared with 21% p=<0.001) with no significant difference in biopsy confirmed acute rejection. The patient in our case would therefore benefit from conversion of CNI to mTOR inhibitor.
Conclusion
Partial nephrectomy is a safe and viable option to consider in the management of small renal masses in the transplanted kidney. The use of manual atraumatic graded bowel clamp pressure around the resected tumour is an alternative to hilum vascular control and can facilitate safe effective resection.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.
Declaration of competing interest
Nil.
Acknowledgements
Nil. | MYCOPHENOLATE MOFETIL, PREDNISOLONE, TACROLIMUS | DrugsGivenReaction | CC BY-NC-ND | 33318944 | 19,656,620 | 2021-03 |
What was the dosage of drug 'MYCOPHENOLATE MOFETIL'? | Double partial nephrectomy in allograft transplanted kidney.
A 61-year-old female presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior. As the 10 year survival of living transplant recipients increases, malignancy presentations will continue to rise. Nephron sparing surgery in renal allografts is sparse due to difficult operative dissection and complicated hila vascular control. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as a viable option to safely perform partial nephrectomy in a transplanted kidney.
Introduction
Transplant recipients are twice as likely to develop a malignancy compared to the rest of the population. However, the risk of renal cell carcinoma (RCC) in this population is only mildly elevated at 4.6% compared with 3% in the general population. Only 10% of these tumours occur in the allograft. End stage kidney disease and chronic immunosuppression is known to pre-dispose native kidneys to malignancy. However, the aetiology of malignancy in allografts is poorly understood with no consensus on treatment.1
Historically, radical nephrectomy (RN) was considered first line treatment due to the potential for rapidly progressive RCC in an immunosuppressed patient. This placed patients back on dialysis with associated mortality and morbidity. Nephron sparing surgical (NSS) approaches are considered risky due to the hostile operative environment. However, sparse case report data indicate technical feasibility.2 We present a novel technique to perform off clamp NSS in a transplant recipient presenting with two tumours in an allograft kidney.
Case report
A 61-year-old women presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior into the right iliac fossa. This was uncomplicated and her serum creatinine post-transplant ranged between 95 and 110μmol/L. Maintenance immunosuppressive therapy included tacrolimus, mycophenolate and low dose prednisolone (3mg daily). Her background included steroid induced insulin-dependent diabetes mellitus and osteopenia. She was independent with all activities of daily living.
A computed tomography (CT) scan identified two enhancing renal lesions which were partly exophytic with some cystic components. The larger 3cm lesion was located in the anterior mid-pole region and the smaller 1.5cm lesion was located in the anterior upper pole region of the transplanted kidney (Fig. 1). No metastatic spread was found on staging. An open dual partial nephrectomy of the allograft kidney was planned.Fig. 1 Pre-operative planning CT scan of the abdomen and pelvis with intravenous contrast. A – sagittal view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). B – sagittal view showing cystic components of the allograft lesion (blue arrow). C – axial view showing an enhancing anterior 1.5cm upper pole solid allograft lesion (green arrow). D – axial view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 1
The patient was positioned supine and a 10cm incision was made through the previous Gibson incision. Dissection to the renal parenchyma was performed exposing the anterior and upper pole of the kidney. Intraoperative ultrasonography was used to identify both tumours (Fig. 2). Dissection of the hilum was not performed due to adhesions, short hilar vessels and to limit trauma to the transplanted vessels. Dual partial nephrectomy using cold cut and diathermy achieved excision of both tumours. A curved bowel clamp was utilised to apply local graded manual pressure directly underneath the tumour site during the resection of the tumour and subsequent renorrhaphy closure to minimise haemorrhage (Fig. 3A/B). A 2-0 V-loc suture was used for renorrhaphy closure. There was approximately 500mL of blood loss. The post-operative course was uncomplicated and the patient was discharged day 6 with a serum creatinine of 111μmol/L. The histopathology revealed Grade 3 clear cell RCC with negative margins at both sites and at one month follow up the patient remained well (Fig. 3C/D). The maintenance immunosuppressants were not altered before or after surgery.Fig. 2 Dissection to the allograft kidney. A – superior portion of kidney with peritoneal reflection (1), bulging upper tumour (2) and healthy renal parenchyma (3) B – inferior portion of kidney with lower pole tumour (4) and healthy renal parenchyma (3).
Fig. 2Fig. 3 Nephron sparing resection of allograft tumours. A – novel use of soft bowel clamp (1) to the upper pole tumour. B – The same is shown in for the lower pole tumour. C – upper pole allograft tumour resected. D – lower pole allograft tumour resected.
Fig. 3
Discussion
Transplant recipients are living longer than ever before. According to the 2018 Australian and New Zealand Scientific Registry of Transplant Patients, the 10 year survival for living transplant recipients has increased from 84% to 88% compared to the previous decade. This will likely see increasing presentations of malignancy within the allograft. Although the aetiology for RCC in allografts is not completely understood, theories exist within the literature. A recent study from Germany examined 1655 transplant patients for approximately 12 years.3 Twenty-six cases of RCC after transplantation were diagnosed. Post-transplant RCC was significantly associated with longer durations of pre-transplant haemodialysis (p = 0.007), post-transplant immunosuppression with cyclosporine (p = 0.029) and/or mycophenolate (p = 0.020) and with post-transplant prednisolone (p = 0.042). Cyclosporine A and mycophenolate usage were also independent risk factors for RCC development. The patient in our study had been on mycophenolate and prednisolone for over 10 years.
The literature on nephron sparing surgery in renal allografts is sparse and mostly limited to patients with T1a allograft masses.2 Traditionally all other masses are managed with full RN of the allograft kidney. The European Organisation for Research and Treatment of Cancer (EORTC) randomised trial 30904 is the only multi-centre international randomised control trial which compared NSS and RN in 541 non-transplant patients with small (≤5 cm) singular renal tumours.4 NSS was shown to reduce post-operative moderate renal dysfunction (eGFR <60) by 21% (95% CI, 13.8–28.3) when compared with RN with no difference in cancer-specific survival or oncological clearance. Re-operation rates in NSS was slightly higher compared to RN (4.4% compared with 2.4%) with bleeding accounting for 71% of surgical take-back. Although this study was completed in native kidney disease, results suggest NSS is a viable alternative.
Partial nephrectomy of the renal allograft is a challenging procedure due to adhesions. Severe adhesions make kidney mobilisation, hilum control and parenchymal resection difficult. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as an alternative to hilum vascular control. As shown in Fig. 3A/B close proximal clamping of the parenchyma surrounding the tumour during resection provides tamponade pressure and allows safe resection of the tumour until haemostatic sutures and/or cautery can be attended. This eliminates the need to dissect out the renal hilum of the allograft and provides zero global ischemia to maximise nephron sparing.
Immunosuppressive adjustment reduces allograft malignancy without compromising rejection. The CONVERT trial randomised 830 renal allograft recipients 6–120 months post-transplant to remain on calcineurin inhibitiors (CNI) or convert CNI to mammalian target of rapamycin (mTOR) inhibitor.5 Conversion to an mTOR inhibitior was associated with a lower incidence of all malignancy (12% compared with 21% p=<0.001) with no significant difference in biopsy confirmed acute rejection. The patient in our case would therefore benefit from conversion of CNI to mTOR inhibitor.
Conclusion
Partial nephrectomy is a safe and viable option to consider in the management of small renal masses in the transplanted kidney. The use of manual atraumatic graded bowel clamp pressure around the resected tumour is an alternative to hilum vascular control and can facilitate safe effective resection.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.
Declaration of competing interest
Nil.
Acknowledgements
Nil. | FOR OVER 10 YEARS | DrugDosageText | CC BY-NC-ND | 33318944 | 19,656,620 | 2021-03 |
What was the dosage of drug 'TACROLIMUS'? | Double partial nephrectomy in allograft transplanted kidney.
A 61-year-old female presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior. As the 10 year survival of living transplant recipients increases, malignancy presentations will continue to rise. Nephron sparing surgery in renal allografts is sparse due to difficult operative dissection and complicated hila vascular control. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as a viable option to safely perform partial nephrectomy in a transplanted kidney.
Introduction
Transplant recipients are twice as likely to develop a malignancy compared to the rest of the population. However, the risk of renal cell carcinoma (RCC) in this population is only mildly elevated at 4.6% compared with 3% in the general population. Only 10% of these tumours occur in the allograft. End stage kidney disease and chronic immunosuppression is known to pre-dispose native kidneys to malignancy. However, the aetiology of malignancy in allografts is poorly understood with no consensus on treatment.1
Historically, radical nephrectomy (RN) was considered first line treatment due to the potential for rapidly progressive RCC in an immunosuppressed patient. This placed patients back on dialysis with associated mortality and morbidity. Nephron sparing surgical (NSS) approaches are considered risky due to the hostile operative environment. However, sparse case report data indicate technical feasibility.2 We present a novel technique to perform off clamp NSS in a transplant recipient presenting with two tumours in an allograft kidney.
Case report
A 61-year-old women presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior into the right iliac fossa. This was uncomplicated and her serum creatinine post-transplant ranged between 95 and 110μmol/L. Maintenance immunosuppressive therapy included tacrolimus, mycophenolate and low dose prednisolone (3mg daily). Her background included steroid induced insulin-dependent diabetes mellitus and osteopenia. She was independent with all activities of daily living.
A computed tomography (CT) scan identified two enhancing renal lesions which were partly exophytic with some cystic components. The larger 3cm lesion was located in the anterior mid-pole region and the smaller 1.5cm lesion was located in the anterior upper pole region of the transplanted kidney (Fig. 1). No metastatic spread was found on staging. An open dual partial nephrectomy of the allograft kidney was planned.Fig. 1 Pre-operative planning CT scan of the abdomen and pelvis with intravenous contrast. A – sagittal view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). B – sagittal view showing cystic components of the allograft lesion (blue arrow). C – axial view showing an enhancing anterior 1.5cm upper pole solid allograft lesion (green arrow). D – axial view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 1
The patient was positioned supine and a 10cm incision was made through the previous Gibson incision. Dissection to the renal parenchyma was performed exposing the anterior and upper pole of the kidney. Intraoperative ultrasonography was used to identify both tumours (Fig. 2). Dissection of the hilum was not performed due to adhesions, short hilar vessels and to limit trauma to the transplanted vessels. Dual partial nephrectomy using cold cut and diathermy achieved excision of both tumours. A curved bowel clamp was utilised to apply local graded manual pressure directly underneath the tumour site during the resection of the tumour and subsequent renorrhaphy closure to minimise haemorrhage (Fig. 3A/B). A 2-0 V-loc suture was used for renorrhaphy closure. There was approximately 500mL of blood loss. The post-operative course was uncomplicated and the patient was discharged day 6 with a serum creatinine of 111μmol/L. The histopathology revealed Grade 3 clear cell RCC with negative margins at both sites and at one month follow up the patient remained well (Fig. 3C/D). The maintenance immunosuppressants were not altered before or after surgery.Fig. 2 Dissection to the allograft kidney. A – superior portion of kidney with peritoneal reflection (1), bulging upper tumour (2) and healthy renal parenchyma (3) B – inferior portion of kidney with lower pole tumour (4) and healthy renal parenchyma (3).
Fig. 2Fig. 3 Nephron sparing resection of allograft tumours. A – novel use of soft bowel clamp (1) to the upper pole tumour. B – The same is shown in for the lower pole tumour. C – upper pole allograft tumour resected. D – lower pole allograft tumour resected.
Fig. 3
Discussion
Transplant recipients are living longer than ever before. According to the 2018 Australian and New Zealand Scientific Registry of Transplant Patients, the 10 year survival for living transplant recipients has increased from 84% to 88% compared to the previous decade. This will likely see increasing presentations of malignancy within the allograft. Although the aetiology for RCC in allografts is not completely understood, theories exist within the literature. A recent study from Germany examined 1655 transplant patients for approximately 12 years.3 Twenty-six cases of RCC after transplantation were diagnosed. Post-transplant RCC was significantly associated with longer durations of pre-transplant haemodialysis (p = 0.007), post-transplant immunosuppression with cyclosporine (p = 0.029) and/or mycophenolate (p = 0.020) and with post-transplant prednisolone (p = 0.042). Cyclosporine A and mycophenolate usage were also independent risk factors for RCC development. The patient in our study had been on mycophenolate and prednisolone for over 10 years.
The literature on nephron sparing surgery in renal allografts is sparse and mostly limited to patients with T1a allograft masses.2 Traditionally all other masses are managed with full RN of the allograft kidney. The European Organisation for Research and Treatment of Cancer (EORTC) randomised trial 30904 is the only multi-centre international randomised control trial which compared NSS and RN in 541 non-transplant patients with small (≤5 cm) singular renal tumours.4 NSS was shown to reduce post-operative moderate renal dysfunction (eGFR <60) by 21% (95% CI, 13.8–28.3) when compared with RN with no difference in cancer-specific survival or oncological clearance. Re-operation rates in NSS was slightly higher compared to RN (4.4% compared with 2.4%) with bleeding accounting for 71% of surgical take-back. Although this study was completed in native kidney disease, results suggest NSS is a viable alternative.
Partial nephrectomy of the renal allograft is a challenging procedure due to adhesions. Severe adhesions make kidney mobilisation, hilum control and parenchymal resection difficult. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as an alternative to hilum vascular control. As shown in Fig. 3A/B close proximal clamping of the parenchyma surrounding the tumour during resection provides tamponade pressure and allows safe resection of the tumour until haemostatic sutures and/or cautery can be attended. This eliminates the need to dissect out the renal hilum of the allograft and provides zero global ischemia to maximise nephron sparing.
Immunosuppressive adjustment reduces allograft malignancy without compromising rejection. The CONVERT trial randomised 830 renal allograft recipients 6–120 months post-transplant to remain on calcineurin inhibitiors (CNI) or convert CNI to mammalian target of rapamycin (mTOR) inhibitor.5 Conversion to an mTOR inhibitior was associated with a lower incidence of all malignancy (12% compared with 21% p=<0.001) with no significant difference in biopsy confirmed acute rejection. The patient in our case would therefore benefit from conversion of CNI to mTOR inhibitor.
Conclusion
Partial nephrectomy is a safe and viable option to consider in the management of small renal masses in the transplanted kidney. The use of manual atraumatic graded bowel clamp pressure around the resected tumour is an alternative to hilum vascular control and can facilitate safe effective resection.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.
Declaration of competing interest
Nil.
Acknowledgements
Nil. | FOR OVER 10 YEARS | DrugDosageText | CC BY-NC-ND | 33318944 | 19,656,620 | 2021-03 |
What was the outcome of reaction 'Renal cell carcinoma stage III'? | Double partial nephrectomy in allograft transplanted kidney.
A 61-year-old female presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior. As the 10 year survival of living transplant recipients increases, malignancy presentations will continue to rise. Nephron sparing surgery in renal allografts is sparse due to difficult operative dissection and complicated hila vascular control. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as a viable option to safely perform partial nephrectomy in a transplanted kidney.
Introduction
Transplant recipients are twice as likely to develop a malignancy compared to the rest of the population. However, the risk of renal cell carcinoma (RCC) in this population is only mildly elevated at 4.6% compared with 3% in the general population. Only 10% of these tumours occur in the allograft. End stage kidney disease and chronic immunosuppression is known to pre-dispose native kidneys to malignancy. However, the aetiology of malignancy in allografts is poorly understood with no consensus on treatment.1
Historically, radical nephrectomy (RN) was considered first line treatment due to the potential for rapidly progressive RCC in an immunosuppressed patient. This placed patients back on dialysis with associated mortality and morbidity. Nephron sparing surgical (NSS) approaches are considered risky due to the hostile operative environment. However, sparse case report data indicate technical feasibility.2 We present a novel technique to perform off clamp NSS in a transplant recipient presenting with two tumours in an allograft kidney.
Case report
A 61-year-old women presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior into the right iliac fossa. This was uncomplicated and her serum creatinine post-transplant ranged between 95 and 110μmol/L. Maintenance immunosuppressive therapy included tacrolimus, mycophenolate and low dose prednisolone (3mg daily). Her background included steroid induced insulin-dependent diabetes mellitus and osteopenia. She was independent with all activities of daily living.
A computed tomography (CT) scan identified two enhancing renal lesions which were partly exophytic with some cystic components. The larger 3cm lesion was located in the anterior mid-pole region and the smaller 1.5cm lesion was located in the anterior upper pole region of the transplanted kidney (Fig. 1). No metastatic spread was found on staging. An open dual partial nephrectomy of the allograft kidney was planned.Fig. 1 Pre-operative planning CT scan of the abdomen and pelvis with intravenous contrast. A – sagittal view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). B – sagittal view showing cystic components of the allograft lesion (blue arrow). C – axial view showing an enhancing anterior 1.5cm upper pole solid allograft lesion (green arrow). D – axial view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 1
The patient was positioned supine and a 10cm incision was made through the previous Gibson incision. Dissection to the renal parenchyma was performed exposing the anterior and upper pole of the kidney. Intraoperative ultrasonography was used to identify both tumours (Fig. 2). Dissection of the hilum was not performed due to adhesions, short hilar vessels and to limit trauma to the transplanted vessels. Dual partial nephrectomy using cold cut and diathermy achieved excision of both tumours. A curved bowel clamp was utilised to apply local graded manual pressure directly underneath the tumour site during the resection of the tumour and subsequent renorrhaphy closure to minimise haemorrhage (Fig. 3A/B). A 2-0 V-loc suture was used for renorrhaphy closure. There was approximately 500mL of blood loss. The post-operative course was uncomplicated and the patient was discharged day 6 with a serum creatinine of 111μmol/L. The histopathology revealed Grade 3 clear cell RCC with negative margins at both sites and at one month follow up the patient remained well (Fig. 3C/D). The maintenance immunosuppressants were not altered before or after surgery.Fig. 2 Dissection to the allograft kidney. A – superior portion of kidney with peritoneal reflection (1), bulging upper tumour (2) and healthy renal parenchyma (3) B – inferior portion of kidney with lower pole tumour (4) and healthy renal parenchyma (3).
Fig. 2Fig. 3 Nephron sparing resection of allograft tumours. A – novel use of soft bowel clamp (1) to the upper pole tumour. B – The same is shown in for the lower pole tumour. C – upper pole allograft tumour resected. D – lower pole allograft tumour resected.
Fig. 3
Discussion
Transplant recipients are living longer than ever before. According to the 2018 Australian and New Zealand Scientific Registry of Transplant Patients, the 10 year survival for living transplant recipients has increased from 84% to 88% compared to the previous decade. This will likely see increasing presentations of malignancy within the allograft. Although the aetiology for RCC in allografts is not completely understood, theories exist within the literature. A recent study from Germany examined 1655 transplant patients for approximately 12 years.3 Twenty-six cases of RCC after transplantation were diagnosed. Post-transplant RCC was significantly associated with longer durations of pre-transplant haemodialysis (p = 0.007), post-transplant immunosuppression with cyclosporine (p = 0.029) and/or mycophenolate (p = 0.020) and with post-transplant prednisolone (p = 0.042). Cyclosporine A and mycophenolate usage were also independent risk factors for RCC development. The patient in our study had been on mycophenolate and prednisolone for over 10 years.
The literature on nephron sparing surgery in renal allografts is sparse and mostly limited to patients with T1a allograft masses.2 Traditionally all other masses are managed with full RN of the allograft kidney. The European Organisation for Research and Treatment of Cancer (EORTC) randomised trial 30904 is the only multi-centre international randomised control trial which compared NSS and RN in 541 non-transplant patients with small (≤5 cm) singular renal tumours.4 NSS was shown to reduce post-operative moderate renal dysfunction (eGFR <60) by 21% (95% CI, 13.8–28.3) when compared with RN with no difference in cancer-specific survival or oncological clearance. Re-operation rates in NSS was slightly higher compared to RN (4.4% compared with 2.4%) with bleeding accounting for 71% of surgical take-back. Although this study was completed in native kidney disease, results suggest NSS is a viable alternative.
Partial nephrectomy of the renal allograft is a challenging procedure due to adhesions. Severe adhesions make kidney mobilisation, hilum control and parenchymal resection difficult. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as an alternative to hilum vascular control. As shown in Fig. 3A/B close proximal clamping of the parenchyma surrounding the tumour during resection provides tamponade pressure and allows safe resection of the tumour until haemostatic sutures and/or cautery can be attended. This eliminates the need to dissect out the renal hilum of the allograft and provides zero global ischemia to maximise nephron sparing.
Immunosuppressive adjustment reduces allograft malignancy without compromising rejection. The CONVERT trial randomised 830 renal allograft recipients 6–120 months post-transplant to remain on calcineurin inhibitiors (CNI) or convert CNI to mammalian target of rapamycin (mTOR) inhibitor.5 Conversion to an mTOR inhibitior was associated with a lower incidence of all malignancy (12% compared with 21% p=<0.001) with no significant difference in biopsy confirmed acute rejection. The patient in our case would therefore benefit from conversion of CNI to mTOR inhibitor.
Conclusion
Partial nephrectomy is a safe and viable option to consider in the management of small renal masses in the transplanted kidney. The use of manual atraumatic graded bowel clamp pressure around the resected tumour is an alternative to hilum vascular control and can facilitate safe effective resection.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.
Declaration of competing interest
Nil.
Acknowledgements
Nil. | Recovering | ReactionOutcome | CC BY-NC-ND | 33318944 | 19,597,832 | 2021-03 |
What was the outcome of reaction 'Renal cell carcinoma'? | Double partial nephrectomy in allograft transplanted kidney.
A 61-year-old female presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior. As the 10 year survival of living transplant recipients increases, malignancy presentations will continue to rise. Nephron sparing surgery in renal allografts is sparse due to difficult operative dissection and complicated hila vascular control. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as a viable option to safely perform partial nephrectomy in a transplanted kidney.
Introduction
Transplant recipients are twice as likely to develop a malignancy compared to the rest of the population. However, the risk of renal cell carcinoma (RCC) in this population is only mildly elevated at 4.6% compared with 3% in the general population. Only 10% of these tumours occur in the allograft. End stage kidney disease and chronic immunosuppression is known to pre-dispose native kidneys to malignancy. However, the aetiology of malignancy in allografts is poorly understood with no consensus on treatment.1
Historically, radical nephrectomy (RN) was considered first line treatment due to the potential for rapidly progressive RCC in an immunosuppressed patient. This placed patients back on dialysis with associated mortality and morbidity. Nephron sparing surgical (NSS) approaches are considered risky due to the hostile operative environment. However, sparse case report data indicate technical feasibility.2 We present a novel technique to perform off clamp NSS in a transplant recipient presenting with two tumours in an allograft kidney.
Case report
A 61-year-old women presented with an incidental anterior mid pole renal mass on ultrasound. She had previously undergone live directed donor renal transplantation 13 years prior into the right iliac fossa. This was uncomplicated and her serum creatinine post-transplant ranged between 95 and 110μmol/L. Maintenance immunosuppressive therapy included tacrolimus, mycophenolate and low dose prednisolone (3mg daily). Her background included steroid induced insulin-dependent diabetes mellitus and osteopenia. She was independent with all activities of daily living.
A computed tomography (CT) scan identified two enhancing renal lesions which were partly exophytic with some cystic components. The larger 3cm lesion was located in the anterior mid-pole region and the smaller 1.5cm lesion was located in the anterior upper pole region of the transplanted kidney (Fig. 1). No metastatic spread was found on staging. An open dual partial nephrectomy of the allograft kidney was planned.Fig. 1 Pre-operative planning CT scan of the abdomen and pelvis with intravenous contrast. A – sagittal view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). B – sagittal view showing cystic components of the allograft lesion (blue arrow). C – axial view showing an enhancing anterior 1.5cm upper pole solid allograft lesion (green arrow). D – axial view showing an enhancing, complex anterior mid-pole 3cm allograft lesion (green arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 1
The patient was positioned supine and a 10cm incision was made through the previous Gibson incision. Dissection to the renal parenchyma was performed exposing the anterior and upper pole of the kidney. Intraoperative ultrasonography was used to identify both tumours (Fig. 2). Dissection of the hilum was not performed due to adhesions, short hilar vessels and to limit trauma to the transplanted vessels. Dual partial nephrectomy using cold cut and diathermy achieved excision of both tumours. A curved bowel clamp was utilised to apply local graded manual pressure directly underneath the tumour site during the resection of the tumour and subsequent renorrhaphy closure to minimise haemorrhage (Fig. 3A/B). A 2-0 V-loc suture was used for renorrhaphy closure. There was approximately 500mL of blood loss. The post-operative course was uncomplicated and the patient was discharged day 6 with a serum creatinine of 111μmol/L. The histopathology revealed Grade 3 clear cell RCC with negative margins at both sites and at one month follow up the patient remained well (Fig. 3C/D). The maintenance immunosuppressants were not altered before or after surgery.Fig. 2 Dissection to the allograft kidney. A – superior portion of kidney with peritoneal reflection (1), bulging upper tumour (2) and healthy renal parenchyma (3) B – inferior portion of kidney with lower pole tumour (4) and healthy renal parenchyma (3).
Fig. 2Fig. 3 Nephron sparing resection of allograft tumours. A – novel use of soft bowel clamp (1) to the upper pole tumour. B – The same is shown in for the lower pole tumour. C – upper pole allograft tumour resected. D – lower pole allograft tumour resected.
Fig. 3
Discussion
Transplant recipients are living longer than ever before. According to the 2018 Australian and New Zealand Scientific Registry of Transplant Patients, the 10 year survival for living transplant recipients has increased from 84% to 88% compared to the previous decade. This will likely see increasing presentations of malignancy within the allograft. Although the aetiology for RCC in allografts is not completely understood, theories exist within the literature. A recent study from Germany examined 1655 transplant patients for approximately 12 years.3 Twenty-six cases of RCC after transplantation were diagnosed. Post-transplant RCC was significantly associated with longer durations of pre-transplant haemodialysis (p = 0.007), post-transplant immunosuppression with cyclosporine (p = 0.029) and/or mycophenolate (p = 0.020) and with post-transplant prednisolone (p = 0.042). Cyclosporine A and mycophenolate usage were also independent risk factors for RCC development. The patient in our study had been on mycophenolate and prednisolone for over 10 years.
The literature on nephron sparing surgery in renal allografts is sparse and mostly limited to patients with T1a allograft masses.2 Traditionally all other masses are managed with full RN of the allograft kidney. The European Organisation for Research and Treatment of Cancer (EORTC) randomised trial 30904 is the only multi-centre international randomised control trial which compared NSS and RN in 541 non-transplant patients with small (≤5 cm) singular renal tumours.4 NSS was shown to reduce post-operative moderate renal dysfunction (eGFR <60) by 21% (95% CI, 13.8–28.3) when compared with RN with no difference in cancer-specific survival or oncological clearance. Re-operation rates in NSS was slightly higher compared to RN (4.4% compared with 2.4%) with bleeding accounting for 71% of surgical take-back. Although this study was completed in native kidney disease, results suggest NSS is a viable alternative.
Partial nephrectomy of the renal allograft is a challenging procedure due to adhesions. Severe adhesions make kidney mobilisation, hilum control and parenchymal resection difficult. We present the use of manual atraumatic graded bowel clamp pressure around the resected tumour as an alternative to hilum vascular control. As shown in Fig. 3A/B close proximal clamping of the parenchyma surrounding the tumour during resection provides tamponade pressure and allows safe resection of the tumour until haemostatic sutures and/or cautery can be attended. This eliminates the need to dissect out the renal hilum of the allograft and provides zero global ischemia to maximise nephron sparing.
Immunosuppressive adjustment reduces allograft malignancy without compromising rejection. The CONVERT trial randomised 830 renal allograft recipients 6–120 months post-transplant to remain on calcineurin inhibitiors (CNI) or convert CNI to mammalian target of rapamycin (mTOR) inhibitor.5 Conversion to an mTOR inhibitior was associated with a lower incidence of all malignancy (12% compared with 21% p=<0.001) with no significant difference in biopsy confirmed acute rejection. The patient in our case would therefore benefit from conversion of CNI to mTOR inhibitor.
Conclusion
Partial nephrectomy is a safe and viable option to consider in the management of small renal masses in the transplanted kidney. The use of manual atraumatic graded bowel clamp pressure around the resected tumour is an alternative to hilum vascular control and can facilitate safe effective resection.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for profit sectors.
Declaration of competing interest
Nil.
Acknowledgements
Nil. | Recovering | ReactionOutcome | CC BY-NC-ND | 33318944 | 19,656,620 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Necrotising herpetic retinopathy'. | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | CYCLOSPORINE, GANCICLOVIR, PREDNISONE | DrugsGivenReaction | CC BY-NC-ND | 33323907 | 18,867,747 | 2021-07-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Necrotising retinitis'. | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | CYCLOSPORINE, PREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33323907 | 20,477,398 | 2021-07-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Product use issue'. | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | CYCLOSPORINE, GANCICLOVIR, PREDNISONE | DrugsGivenReaction | CC BY-NC-ND | 33323907 | 18,867,747 | 2021-07-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Retinopathy'. | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | CYCLOSPORINE, GANCICLOVIR, PREDNISONE | DrugsGivenReaction | CC BY-NC-ND | 33323907 | 20,499,297 | 2021-07-01 |
What was the outcome of reaction 'Cytomegalovirus chorioretinitis'? | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | Recovered | ReactionOutcome | CC BY-NC-ND | 33323907 | 20,477,398 | 2021-07-01 |
What was the outcome of reaction 'Glaucoma'? | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | Recovered | ReactionOutcome | CC BY-NC-ND | 33323907 | 18,867,747 | 2021-07-01 |
What was the outcome of reaction 'Necrotising herpetic retinopathy'? | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | Recovered | ReactionOutcome | CC BY-NC-ND | 33323907 | 18,867,747 | 2021-07-01 |
What was the outcome of reaction 'Necrotising retinitis'? | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | Recovered | ReactionOutcome | CC BY-NC-ND | 33323907 | 20,477,398 | 2021-07-01 |
What was the outcome of reaction 'Retinopathy'? | NEOVASCULAR COMPLICATIONS FROM CYTOMEGALOVIRUS NECROTIZING RETINOPATHY IN PATIENTS AFTER HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION.
OBJECTIVE
To report the incidence and clinical features of neovascular complications from cytomegalovirus (CMV) necrotizing retinopathy in patients after haploidentical hematopoietic stem cell transplantation.
METHODS
Thirty-nine patients (58 eyes) of CMV necrotizing retinopathy after haploidentical hematopoietic stem cell transplantation in our institute between January 2018 and June 2020 were retrospectively reviewed, and cases that developed neovascular complications during follow-up were identified and described.
RESULTS
Two (2 eyes) cases that developed neovascular glaucoma from CMV necrotizing retinopathy were identified. Both of them manifested as granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which were consistent with chronic retinal necrosis. Insidious progression of isolated immune-mediated occlusive vasculitis that could only be observed on fundus fluorescein angiography without active retinitis or intraocular inflammation was recognized to be the cause in one of two cases.
CONCLUSIONS
Neovascular glaucoma developed in 5.1%/cases and 3.4%/eyes complicated by CMV chronic retinal necrosis and vasculitis in patients after haploidentical hematopoietic stem cell transplantation, which warrants the needs for long-term follow-up. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with a minimal immune deviation and may only be found on fundus fluorescein angiography, which emphasizes the importance of fundus fluorescein angiography on a regular basis during follow-up.
Haploidentical hematopoietic stem cell transplantation (HHSCT) expanded the selection range of donors and makes it easier to obtain donor lymphocytes in subsequent adoptive immunotherapy.1,2 But a T-cell repletion and depletion approach around HHSCT and subsequent immunomodulatory therapy increases the risk of cytomegalovirus (CMV) disease after HHSCT.3 With a growing number of patients receiving HHSCT worldwide,2 CMV necrotizing retinopathy is becoming more and more common in ophthalmic clinic.
Chronic retinal necrosis (CRN), which was first reported by Schneider et al in 2013,4 is caused by CMV infection and mainly affects patients with limited immune dysfunction, such as aging and diabetes. Its clinical characteristics include slowly progressive granular retinitis, occlusive panretinal vasculitis, and varying degrees of intraocular inflammation, which resemble those of acute retinal necrosis except for the slow progression and a more limited extent of the retinitis.5 Several cases with neovascular complications secondary to CRN had been reported.4,6,7 Occlusive vasculitis and large area of nonperfusion on the retina that already existed at initial presentation was the main cause. There were only two CRN cases reported after Schneider et al, and both of them developed neovascular glaucoma (NVG) during follow-up.6,7 To the best of our knowledge, no cases of neovascular complications/NVG have been reported in patients with CMV CRN after HHSCT. Besides, there has been no detailed report on CRN and its neovascular complication in China.
Our study was to report the incidence and clinical features of cases developing neovascular complications/NVG from CMV CRN after HHSCT. Particularly, we noted that isolated insidious immune-mediated CMV retinal vasculitis without the progression of retinitis or evidence of intraocular inflammation could result in enlarging the nonperfusion area and finally causes NVG.
Patients and Methods
We performed a retrospective study of all CMV necrotizing retinopathy after HHSCT in the Peking University People's Hospital between January 2018 and June 2020. Cytomegalovirus necrotizing retinopathy diagnosis was established by a recent history of HHSCT, presence of suggestive clinical and fundus imaging features, positive CMV-DNA load but no other human herpes virus DNA in aqueous, and exclusion of other possible etiologies that are clinically similar to CMV necrotizing retinopathy, such as syphilis, tuberculosis, and toxoplasmosis. Neovascular complications were defined by the following criteria: 1) neovascularization of iris and/or angle with/without anterior synechiae; 2) neovascularization of retina on fundus examination or fundus fluorescein angiography (FFA); 3) large nonperfusion area shown by FFA that corresponded to neovascularization; and 4) no other causes could be attributed to, such as retinal vein occlusion and diabetic retinopathy. Increased intraocular pressure (IOP) together with neovascular complications is defined as NVG. We collected information on clinical features, multimodal fundus images, treatments, and outcomes.
This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Peking University People's hospital under grant No. 2018PHB196-01. Written informed consent was obtained from each patient before enrollment.
Results
A total of 39 cases (58 eyes) of CMV necrotizing retinopathy after HHSCT were identified, and all of them were HIV-seronegative. Among, two cases (2/39, 5.1% cases; 2/58, 3.4% eyes) fulfilled the criteria for neovascular complications and both of them developed NVG during follow-up. Both patients manifested as peripheral granular retinitis, occlusive panretinal vasculitis, and certain degrees of intraocular inflammation, which were consistent with CRN described by Schneider et al.4 Aqueous cells were found in one patient, and vitritis existed in both patients. Both patients showed negative whole blood CMV-DNA (＜1 × 103 IU/mL)8 when CMV necrotizing retinopathy was diagnosed.
Considering the potential effect of myelosuppression9 and delaying recovery of CMV-specific T-cell responses of ganciclovir,10,11 borderline and progressively subtherapeutic vitreous concentrations,12 and suboptimal effect for macula and/or optic disk–threatening disease when ganciclovir was given intravenously,13 together with negative results of whole blood CMV-DNA, after discussion with hematologists and the two patients, intravitreal ganciclovir injection combined with a dose reduction of immunomodulatory drugs for chronic graft-versus-host disease without systemic ganciclovir were prescribed for both of them.14,15 Intravitreal injections were given as loading doses twice per week, followed by maintenance dosing once a week. Aqueous CMV-DNA was monitored by quantitative nucleic acid amplification testing.16 Retinitis regressed, lesions healed, and aqueous CMV-DNA decreased to negative (<1 × 103 IU/mL)8 after series of injections in both cases.
Neovascular glaucoma developed within weeks in one patient and 9 months later in the other. Although antivascular endothelial growth factor drug intravitreous injection, panretinal photocoagulation, and antiglaucoma surgery when necessary were able to control IOP, the outcome of visual acuity was poor.
Insidious progression of isolated occlusive vasculitis that could only be observed on FFA without active retinitis or intraocular inflammation was observed in the second patient, which was rather different from the first.
Case Presentations
Case 1
A 29-year-old man was referred for evaluation of progressive vision loss in the left eye over the past 20 days. He was in this status after HHSCT + 165 days because of acute lymphocytic leukemia. On presentation, his immunosuppressive medications included prednisone 5 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The engraftment status was well, with neutrophil count 2.83 × 109/L and platelet 88 × 109/L. The total T-lymphocyte count was 1.850 × 109/L. The patient was treated with a 2-week course of oral ganciclovir after a single-positive whole blood CMV titer (1.32 × 103 IU/mL) 1 month ago. No further evidence of CMV-DNAemia was noted despite regular surveillance.
The visual acuity was 20/20 in the right eye and 20/50 in the left eye. Intraocular pressure was 14 mmHg and 28 mmHg, respectively. Biomicroscopy revealed 1+ aqueous and 2+ vitreous cell in the left eye. Fundus ophthalmoscopy revealed a 2-o'clock hour patch of granular retinitis in the temporal periphery and a linear lesion with one end pointing to the optic disk lying in the nasal equator, as well as a few retinal hemorrhages along retinal vessels (Figure 1, A and B). Cytomegalovirus necrotizing retinopathy was suspected and aqueous sample from the left eye was obtained. Quantitative nucleic acid amplification testing for CMV in aqueous was 3.99 × 105 IU/mL, whereas whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Fig. 1. Patient 1. A and B. Fundus photograph at presentation. A 2-o'clock hour patch of granular retinitis in the temporal periphery and a fusiform lesion with one end pointing to the optic disk lying in the equator, as well as a few retinal hemorrhages along retinal vessels in the left eye. The right eye was unremarkable. C and D. Fundus photograph after 6 times of intravitreous injection of ganciclovir and aqueous cytomegalovirus DNA was negative by then. Intraocular inflammation and granular lesion seemed regressed but superficial hemorrhage along retinal vessel exaggerated in the left eye. The right eye was unremarkable. E and F. Corresponding fundus fluorescein angiography of (C) and (D) confirmed the presence of 360° retinal nonperfusion. The right eye was unremarkable.
Ganciclovir intravitreal injection was administered as well as reducing the dose of cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After 6 times of injections, aqueous CMV-DNA decreased to negative, but the visual acuity dropped to hand motion and IOP increased to 45 mmHg. Although intraocular inflammation and granular lesion seemed regressed, superficial hemorrhage along retinal vessels exaggerated (Figure 1, C and D). Neovascularization was observed in the iris and an angle with 360° anterior synechia. Fluorescein angiography confirmed the presence of 360° retinal nonperfusion (Figure 1, E and F), proposing the diagnosis of NVG. Antivascular endothelial growth factor drug intravitreal injection, Ahmed valve implantation, and panretinal photocoagulation were given. Final visual acuity, 13 months after presentation, was light perception in the left eye. There was no involvement of the right eye and no recurrent retinitis in the left eye during the follow-up period.
Case 2
A 34-year-old woman reported a 2-week history of progressive vision loss in both eyes. The medical history was notable for HHSCT status + 145 days because of acute lymphocytic leukemia. She was taking prednisone 10 mg daily and cyclosporine 50 mg twice daily for chronic graft-versus-host disease. The patient was treated with oral ganciclovir for 3 weeks because of a 2-week course CMV-DNAemia with peak whole blood CMV-DNA 2.33 × 104 IU/mL 2 months ago. No further evidence of CMV-DNAemia was noted despite regular surveillance. The neutrophil count was 1.91 × 109/L, platelet 76 × 109/L, and total T-lymphocyte count was 1.630 × 109/L the day before she presented.
At initial evaluation, the visual acuity was 20/30 and 20/50 in the right and left eye, respectively, and IOP was 16 mmHg and 15 mmHg in the right and left eye, respectively. There was minimal anterior chamber inflammation but 2+ vitreous cells in both eyes. Fundus ophthalmoscopy showed fan-shaped granular retinitis that started from the right fovea and extended to the inferotemporal in the right eye, and a 2-o'clock hour patch of granular retinitis in the temporal periphery of the left eye. Bilateral CMV necrotizing retinopathy was suspected, and aqueous CMV-DNA was 4.56 × 105 IU/mL and 3.43 × 105 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was negative (<1 × 103 IU/mL). Additional workup including fluorescent treponemal antibody absorption, HIV serologies, serum toxoplasma IgG and IgM, T-SPOT.TB, and chest X-ray were all negative.
Intravitreal injection of ganciclovir was prescribed for both eyes, combined with reducing the dose of prednisolone to 5 mg daily and cyclosporine to 50 mg daily. Quantitative nucleic acid amplification testing for CMV in the aqueous was performed during each time of injection. After five injections, aqueous CMV-DNA was negative (<1 × 103 IU/mL) and granular retinitis regressed in both eyes (Figure 2, A and C). Fluorescence fundus angiography revealed a small patch of nonperfusion area in the temporal periphery in both eyes (Figure 2, B and D). The patient was discharged and followed regularly. Two months later, the patient reported another course of vision decreasing in both eyes. The visual acuity was 20/30 and 20/60 for the right and left eye, respectively. Fundus examination indicated recurrence of CMV necrotizing retinopathy from the border of the former scar, and aqueous CMV-DNA was 1.32 × 104 IU/mL and 2.14 × 104 IU/mL for the right and left eye, respectively. Whole blood CMV-DNA was still negative. After another four times of ganciclovir intravitreal injection, aqueous CMV-DNA turned negative again. Fundus examination indicated an enlarged scar (Figure 2, E and G) as well as a nonperfusion area in the temporal periphery on FFA (Figure 2, F and H) in both eyes. Over the next 8 months, the patient was followed up using fundus ophthalmoscopy and fundus photographs. No signs of recurrence were observed in either eye except growing vascular sheathing in the left side, and IOP was always within the normal range. But when she showed up again 9 months later, the visual acuity of the left eye dropped to 20/200 with IOP increased to 35 mmHg. The visual acuity and IOP in the right eye were 20/25 and 18 mmHg, respectively. Neovascularization was observed in the left angle without anterior synechiae. No cells were found in the anterior chamber or the vitreous. Fundus ophthalmoscopy revealed vessel sheathing all around the left fundus. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescence angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connect to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina. Neovascular glaucoma was diagnosed, and antivascular endothelial growth factor and panretinal photocoagulation was given. One week later, IOP decreased to 19 mmHg and angle neovascularization regressed, but the visual acuity remained at 20/200. The patient was followed for another 6 months, and no disease progression was further observed.
Fig. 2. Patient 2. A–D. Fundus photograph and corresponding FFA after the first episode of CMV necrotizing retinopathy. Granular lesions on both sides regressed leaving a small patch of nonperfusion area in the temporal periphery in the left eye. E–H. Fundus photograph and corresponding FFA after the secondary episode of CMV chronic necrotizing retinopathy. Bilateral scar enlarged as well as the nonperfusion area in the temporal periphery on FFA. I–L. Nine months later without intraocular CMV reactivation, wide-spread vessel sheathing was found all around the left fundus besides the scar in the temporal periphery. No additional findings could be found in the right eye compared with 9 months ago. Fundus fluorescein angiography indicated a small patch of nonperfusion area in the temporal periphery that did not connected to the retinal scar in the right eye and 360° retinal nonperfusion in the left eye, including the nasal retina.
Discussion
It is estimated that approximately 5,000 allo-HSCT procedures are performed in China annually,17 and the proportion of HHSCT was 30.8%.1 The cumulative incidence of CMV necrotizing retinopathy was reported to be 2.3% 1 year after HHSCT.18 In this study, intraocular neovascular complications further developed in 5.1%/cases and 3.4%/eyes in these patients. Thus, the incidence of intraocular neovascular complications was about 1.81/person-year within the first year after HHSCT in China. Although rare, considering the growing number of patients receiving HHSCT in China1 and the poor visual outcome for patients who developed such complications during follow-up, attentions should be drawn and early detection and intervention are important.
Ever since the first case of neovascular event complicating CMV necrotizing retinopathy reported by Saran et al in 1996,19 a dozen of similar cases could be found in the literature at present.6,7,20–22 Neovascular complications of HIV-related CMV retinitis are rare, but the incidence was recognized to be higher in non-HIV patients.21,22 All the cases reported share similar clinical manifestation including granular peripheral retinitis, panretinal occlusive vasculitis, and some degree of intraocular inflammation, which fit the criteria of acute retinal necrosis defined by the American Uveitis Society.5 Schneider et al4 proposed that CRN caused by CMV is related to the immune status of the host. In limited immuno-compromised patients, the manifestation of CMV retinitis may be a spectrum of mixture of acute retinal necrosis and CMV retinitis, from acute retinal necrosis-like end in patients with lesser degrees of immune dysfunction to classic CMV retinitis-like end in whom more serious immune compromise exists. In their case series, patients with CRN were reported to have symptoms for weeks to months and were observed to have little progression even without antiviral management.
In Schneider's report, granular retinitis, occlusive panretinal vasculitis, and intraocular inflammation occurred at the same time. No progression or reactivation of retinitis was noted after antiviral treatment, but no further FFA findings were described during follow-up. Four of the five patients developed neovascular complications, and vessel occlusion and extensive retinal nonperfusion seem to have already existed at the initial presentation in all four cases.4 A similar clinical picture was seen in our first case and in cases reported by Matsuoka et al in 20177 and Cho et al in 2018.6 Our second case was different in that, at the first episode, the appearance and enlargement of nonperfusion area was accompanied by active retinitis and was relatively small. After the second episode of CMV reactivation, intraocular inflammation and retinitis seemed calm all the time on fundus biomicroscopy and fundus photographs, but actually, the nonperfusion area continued growing during the 8 months' follow-up and could only be observed on FFA, which emphasizes the importance of FFA on a regular basis to monitor patients with CMV necrotizing retinopathy. Limited immune dysfunction, the continuous replication of CMV in vascular endothelial cells and CMV-specific T-cell-mediated endothelial cell damage may be the possible mechanism for this phenomenon.4,6,20,22 In addition to the spectrum proposed by Schneider et al,4 insidious immune-mediated CMV vasculitis may be an isolated manifestation for CMV necrotizing retinopathy when immune deviation was even lesser than patients with CRN.
The discrepancy of incidence of neovascular complications between Schneider's report and ours (80% vs. 5.1%) was most probably because of patient selection, as we included all patients with CMV necrotizing retinopathy irrespective of their clinical appearance, but Schneider et al only included patients manifested as CRN. Patients with CRN had minimal symptoms in the early stage of the disease and showed up only when the retinal nonperfusion area grew large enough to cause significant visual impairment. But for patients with classic CMV retinitis (fulminant/edema type), floaters and vision loss caused by vitritis and macular involvement prompted them to see doctors immediately when disease occurred and then retinal vessel occlusion could be stopped with proper management. Neovascular complications were more common in patients with CRN when cases of CMV necrotizing retinopathy were retrospectively reviewed.
It is interesting that both patients developed unilateral neovascular complications, especially the enlargement of nonperfusion area only happened to the left eye but not the right eye of the second patient. The exact reason for this was unknown. Ocular immune-privilege effect and local CMV-specific T-cell immune deviation may be a feasible explanation22–25 but further investigations were needed.
Intravenous ganciclovir is considered the first-line treatment for CMV disease after HSCT26 and solid organ transplantation27 and was recommended for CMV retinitis in the 2017 European Conference on Infections in Leukemia guideline.26 We had concerns about using it in our patients due to its potential effect of myelosuppression9 and delaying the recovery of CMV-specific T-cell responses,10,11 combined with its relatively low permeability into the vitreous12 and the negative results of whole blood CMV-DNA testing at presentation. After discussion with hematologists and the two patients, local ganciclovir injection combined with dose reduction of immunomodulatory drugs without systemic ganciclovir were prescribed. Retinitis was sufficiently controlled, which was confirmed by negative aqueous CMV-DNA in the end in both cases. Considering the possible mechanism of isolated CMV vasculitis, systemic ganciclovir may produce a better outcome and reduce the incidence of neovascular complications, but more observations and evidences are needed. And its benefits must be balanced with its potential side effects. Besides, considering the high risk of NVG and its poor visual outcome, prophylactic panretinal photocoagulation should be performed right at once when extensive retinal nonperfusion was found on FFA.
In conclusion, although rare, NVG developed in 5.1%/cases and 3.4%/eyes of patients with CMV necrotizing retinopathy after HHSCT. Immune-mediated CMV vasculitis could be an isolated manifestation in patients with minimal immune deviation and could only be found on FFA, which warrants the needs for long-term follow-up and FFA on a regular basis.
Supported by the National Natural Science Foundation of China provided financial support in the form of National Natural Science Foundation of China under Grant No. 81800847, the 10th “Academic Star” of Peking University People's Hospital under Grant No. RS2018-05, and Technological Innovation and Cultivation Fund for Medical youth of Peking University under Grant No. BMU2020PYB014.
None of the authors has any financial/conflicting interests to disclose.
Z. Long and J. Hou contributed equally to this study. | Recovering | ReactionOutcome | CC BY-NC-ND | 33323907 | 20,499,297 | 2021-07-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pancreatitis necrotising'. | Metastatic Donor-derived Malignancies Following Simultaneous Pancreas-kidney Transplant: Three Case Reports and Management Strategies.
Stopping immunosuppression in a transplant patient with donor-derived malignancy offers the theoretical benefit that reconstitution of the patient's immune system will allow "rejection" of the malignancy, as the malignancy also originates from allogeneic tissue. However, this option exists with the caveat that the patient's allograft(s) will likely be rejected too. In simultaneous pancreas-kidney (SPK) recipients, the normal continued functioning and possible absence of malignancy in either the unaffected kidney or pancreas further complicate this decision.
The charts of 3 patients with donor-derived metastatic malignancies after SPK were retrospectively reviewed in detail. We provide treatment and management recommendations based on successful outcomes and a review of the existing literature.
Consistent with a broad review of the literature, in all 3 cases, complete immunosuppression cessation, removal of both grafts, and in 1 case treatment with an immune checkpoint inhibitor to augment the immune response was successful. One patient is doing well 1 year after successfully undergoing kidney retransplantation, while a second patient is active on the waitlist for SPK retransplantation after no evidence of metastatic disease for 2 years.
The successful management of metastatic donor-derived malignancies requires allograft removal, immunosuppression cessation, and adjuvant therapy that includes occasional use of checkpoint inhibitors to augment the immune response.
INTRODUCTION
Simultaneous pancreas-kidney (SPK) recipients require aggressive immunosuppression to temporize the alloimmune response and the recurrence of autoimmunity. SPK programs generally use lymphocyte-depleting induction regimens, followed by a relatively significant burden of maintenance immunosuppression compared with regimens typically used after solitary kidney transplant.1 Consequently, immunosuppression modification when recipients develop a new malignancy is challenging. For donor-derived malignancies, the option of reducing or stopping immunosuppression to “reject” the tumor presents a unique therapeutic option with the caveat that the allograft will likely be rejected.2 This approach’s theoretical basis is that the host immune system’s reconstitution will trigger an alloimmune response against the malignancy, as the malignancy originates from donor tissue. However, an algorithm for stopping immunosuppression, allograft removal, and implementing medical measures such as chemotherapy is not well defined. Decision-making can be challenging in the setting of a well-functioning kidney or pancreas allograft, which may be uninvolved in the metastatic cancer, as allograft removal needs to be strongly considered to allow immunosuppression withdrawal. Given the paucity of literature on this subject and accompanying decisions, the purpose of this study is to present 3 cases of donor-derived metastatic malignancies after SPK and provide management recommendations based on successful outcomes and a review of the existing literature.
MATERIALS AND METHODS
Three charts of SPK recipients with donor-derived malignancy were retrospectively reviewed. Our institution does not require IRB review for clinical case study reports on up to 3 clinical experiences identified during clinical care. For privacy, all Health Insurance Portability and Accountability Act identifiers have been removed. All 3 patients underwent donor pancreas implantation into the right iliac vessels (systemic endocrine drainage) with enteric exocrine drainage (donor duodenum to recipient ileum) and donor kidney implantation into the left iliac vessels with a ureteroneocystostomy.
CASE SUMMARIES
Case No. 1: Donor-transmitted Pancreatic Adenocarcinoma Detected 6 Months Post SPK
Clinical History
At the time of her SPK, patient 1 was a 42-year-old woman with a history of end-stage renal disease (ESRD) secondary to type 1 diabetes (DM1) with a calculated panel reactive antibodies (cPRA) of 91. She underwent thymoglobulin induction and transitioned to a maintenance immunosuppression of tacrolimus (trough goal 5–15 µg/L), everolimus (trough goal 5–7 µg/L), mycophenolate 540 mg BID, and prednisone 5 mg daily. In the 6 months following transplant, she was seen >10 times at multiple hospitals for recurrent abdominal pain. She received steroids and thymoglobulin for possible rejection. Two months post SPK, a computerized tomography (CT) abdomen-pelvis showed stranding around the pancreas allograft suggestive of possible pancreatitis but with normal lipase. At 6-month posttransplant, it became known that the patient’s donor had transmitted an adenocarcinoma to 3 other recipients who received organs from the same donor. At that time, a positron emission tomography (PET) scan revealed diffuse nodal uptake within the left supraclavicular, mediastinal, mesenteric, and right external iliac nodal regions.
Management and Outcome
Three days after the PET scan, given the concern for donor-derived malignancy, the patient underwent removal of both grafts, followed by immunosuppression cessation. Based on the concern for metastatic disease and the confirmed transmission of aggressive malignancy to the other recipients of organs from the same donor, no preoperative biopsy was attempted as the transplants were going to be removed regardless of any biopsy findings. The PET-avid lymph nodes were also not biopsied at the time of surgery. Explant pathology confirmed widespread adenocarcinoma in the pancreas most consistent with a pancreatic primary with extensive lymphovascular invasion and periovarian and fallopian tube involvement. Tumor genotyping identified mutations that would potentially allow serial monitoring via cell-free DNA (cf-DNA) testing. By 6 months following the allograft removals, the previously seen PET-avid lesions had resolved. The patient continued to have no evidence of cancer for the following year and a half, by both cross-sectional imaging and cf-DNA testing. When being considered for repeat kidney transplant, she was noted to have several new donor-specific antibodies (DSA). Two years after her graft removals, given the substantial time period without evidence of malignancy, the patient underwent a second kidney transplant and restarted immunosuppression. The patient was also listed for a pancreas transplant but received a kidney-only offer, which was accepted given lower immunosuppression requirements and concern over an extended delay with waiting for a compatible pancreas as she had a cPRA of 100% at the time. Six months later, she underwent treatment for an antibody-mediated rejection episode, and 11 months after her second kidney transplant had a PET scan that found no evidence of cancer. The patient is now 3½ years from her original SPK, 3 years from her graft removals, and 1 year from her kidney retransplant. She has been considered for pancreas transplant but is not currently thought to be a candidate due to her recent rejection treatment and a continued cPRA of 100%.
Case No. 2: Donor-derived Pancreatic Adenocarcinoma Detected ≥10 Years Post SPK
Clinical History
At the time of her SPK, patient 2 was a 28-year-old woman with a history of DM1 and ESRD since age 12 with a cPRA of 0%. She underwent thymoglobulin induction and was transitioned to a maintenance regimen of tacrolimus (trough goal 5–15 µg/L), mycophenolate 540 mg BID, and everolimus (trough goal 2–3 µg/L). She had an uneventful course with excellent function of both grafts for 10 years, when she developed chronic abdominal pain. Initial imaging and endoscopies did not identify an etiology for the pain. Twelve years after her SPK, the patient was treated with thymoglobulin for suspected rejection due to an elevated lipase, but no biopsy was done at that time. The lipase returned to normal, but imaging showed fullness in the donor duodenum wall. A subsequent fine-needle aspiration showed adenocarcinoma (Figure 1). The patient underwent PET-CT scan, which showed mediastinal, left supraclavicular, and retroperitoneal lymphadenopathy in addition to marked hypermetabolism of the transplanted pancreas’s head and duodenal cuff.
FIGURE 1. Images of the lesion associated with the allograft pancreas in the right lower quadrant in case 2. A, Coronal CT scan. B, Axial CT scan. C, Successful ultrasound-guided fine-needle aspiration of the lesion in the transplanted duodenum associated with the allograft pancreas. CT, computed tomography.
Management and Outcome
The patient had her tacrolimus trough goal lowered to 3–5 µg/L, mycophenolate lowered to 180 mg BID, and everolimus continued with troughs of 2–3 µg/L. The plan was made to discontinue immunosuppression entirely if the cancer was confirmed to be donor-derived. Short-tandem repeat-based identity mapping was performed on the tumor, which confirmed donor-derived malignancy. After this result, the patient underwent removal of both grafts so that immunosuppression could be discontinued to treat the metastatic donor-derived pancreatic adenocarcinoma. Final pathology confirmed a 6.5 -cm poorly differentiated adenocarcinoma involving the pancreatic head with invasion into the duodenum, ampulla, and peripancreatic soft tissue and metastatic adenocarcinoma in 5 of 16 lymph nodes. Three months after her graft removals, repeat PET scan showed no hypermetabolic lesions to suggest malignancy, and her CA 19-9 decreased from elevated at 85 before her graft removals to within the normal range. PET scans in the 11 months after pancreatectomy and nephrectomy continued to display no evidence of cancer, and her CA 19-9 has remained normal. Her cPRA is currently 99%, and she was also noted to have a new HLA class 1 DSA. The patient is now 14 years from her original SPK, 2 years from her graft pancreatectomy and nephrectomy, and is now listed for another SPK. The oncology service will continue to monitor her on the waitlist with PET scans and serum CA-19-9.
Case No. 3: Donor-derived Renal Cell Carcinoma Detected 13 Years Post SPK
Clinical History
At the time of her SPK, patient 3 was a 33-year-old woman with a history of DM1 and ESRD with a cPRA of 24%. She underwent thymoglobulin induction and transitioned to an immunosuppression regimen of tacrolimus (trough goal 5–15 µg/L), mycophenolate 360 mg BID, and prednisone 5 mg daily. She had excellent graft function and an uncomplicated course for 7 years. She developed biopsy-proven chronic allograft nephropathy 7 years after transplant and returned to dialysis but remained insulin independent. Thirteen years after her SPK, a CT scan performed to evaluate nondescript abdominal pain showed a new 3 -cm mass in the transplant kidney. Ultrasound surveillance imaging done 6 months later described this lesion as a vascular lesion in the transplant kidney’s superior pole, which had grown to 4.7 cm. Previously anuric, over the next month, the patient developed hematuria with increasing abdominal pain. A PET/CT scan done revealed a hypermetabolic lesion in the transplant kidney without evidence of metastatic disease.
Management and Outcome
Given the concern for donor-derived malignancy in the transplanted kidney, 2 days after the PET scan, the patient underwent transplant nephrectomy. Final pathology revealed a 5.8 -cm, stage T1b renal cell carcinoma (RCC) of unclassified subtype. Given this diagnosis, the patient’s immunosuppression regimen was reduced to a tacrolimus trough goal of 3–5 µg/L but otherwise maintained to preserve allograft pancreas function. Over the subsequent 6 months, surveillance CT scans showed no evidence of metastatic disease. However, 11 months following the transplant nephrectomy, a CT scan identified a 7-mm soft tissue nodule in the left lower quadrant deemed to be low likelihood for recurrence given the RCC’s low stage. Serial imaging identified the lesion’s slow growth to 1.6 cm. She was referred to oncology, who elected to treat this lesion as local recurrence with stereotactic radiation therapy. The lesion showed slight interval growth over the subsequent 9 months without metastatic spread despite radiation therapy.
Two years after transplant nephrectomy, she was found to have new retroperitoneal, pulmonary, and hepatic nodules suspicious for metastatic disease. She was still insulin independent. Over the course of several multidisciplinary discussions, immunosuppression cessation was recommended, as was tyrosine kinase inhibition. She deferred these recommendations due to the dramatic improvement in quality of life her pancreas transplant had provided and challenges with blindness and insulin administration. Two months later, she was admitted with severe pain and was found to have a malignant pleural effusion and hepatic metastases. She then agreed to discontinue immunosuppression and initiate the checkpoint inhibitor nivolumab. She was still resistant to graft pancreatectomy, opting to wait to see if she might remain insulin independent. Four weeks after nivolumab initiation, she developed severe right lower quadrant pain with fever. CT scan showed dramatic improvement in hepatic and retroperitoneal metastatic disease burden, but her transplant pancreas was necrotic appearing. The patient underwent immediate transplant pancreatectomy, which was challenging due to profound local inflammation. A bovine pericardial patch was required to reconstruct the iliac vein. On the first day after pancreatectomy, the iliac vein thrombosed, necessitating thrombectomy and placement of a bare-metal stent across the narrowing. She was subsequently discharged with continued CT scan surveillance showing resolution of her retroperitoneal lymphadenopathy and decrease in the size of her liver metastases (Figure 2). The patient is now 16 years from her original SPK, 2½ years from her transplant nephrectomy, and 3 months from her transplant pancreatectomy. For her donor-derived RCC, she will continue on monthly nivolumab for 1 year with regular follow-up with oncology.
FIGURE 2. The left-sided panels show the metastatic liver lesions in case 3 before explantation of the allograft pancreas and cycle 2 of nivolumab. Multiple liver lesions can be seen measuring 2.2, 0.6, and 2.3 cm in the left-top, left-middle, and left-bottom panels, respectively. The right-sided panels show the liver after explanation of the allograft pancreas and multiple cycles of nivolumab. The liver lesion in the top panels has reduced from 2.2 to 1.0 cm in size, while the liver lesion in bottom panels has reduced from 2.3 to 0.85 cm in size. The liver lesion in the middle panel was resolved and was not seen on subsequent scans. Lesions are all indicated by white arrows.
DISCUSSION
This series reports the successful initial treatment of 3 SPK recipients with donor-derived metastatic malignancies. The literature on the treatment of metastatic donor-derived malignancy is limited overall and particularly limited in SPK recipients. In the larger kidney transplant literature, several systematic reviews have investigated the management of donor transmitted malignancy. In 2013, Xiao et al identified 104 donor-transmitted cases and showed that 67% of patients underwent graft nephrectomy and withdrawal of immunosuppression, representing the most common approach pursued. The use of adjuvant chemotherapy, radiotherapy, and immunotherapy was highly variable ranging from 0% to 80% use depending on tumor tissue of origin.3 In 2020, Eccher et al presented a similar analysis of 234 recipients with cancer of donor origin and noted that metastatic disease marked the most significant predictor of death even in this population. Because of the option of return to dialysis, most kidney recipients were treated maximally with immunosuppression cessation, graft removal regardless of tumor tissue of origin, grade, or subtype. However, adjuvant medical treatments were pursued and individualized based on tumor tissue of origin, grade, and subtype.4 For instance, 2 recent case reports have described the successful treatment of metastatic melanoma in a kidney-only recipients through immunosuppression cessation, allograft explantation, and based on the relative success in general metastatic melanoma, adjuvant immune checkpoint therapy.5,6
The literature in pancreas transplants alone (PTA) is limited to single case reports. The first donor-transmitted malignancy in a PTA was reported by Perosa et al in 2010. The malignancy was limited to the pancreas and treated successfully with graft pancreatectomy and immunosuppression cessation.7 Nagaraju et al reported a case of a soft tissue sarcoma arising in a pancreas allograft, which was not tested for donor origin but nevertheless successfully treated with graft pancreatectomy and immunosuppression cessation.8
Focusing on SPK patients, Roza et al reported the first donor-derived malignancy in a transplanted pancreas in an SPK recipient in 2001. However, this patient died of malignancy after transplant pancreatectomy, immunosuppression cessation, and 2 chemotherapy courses.9 In 2020, Meier et al reported the successful treatment of a widely metastatic BK virus-associated renal carcinoma in an SPK patient with graft nephrectomy and IL-2 immunotherapy. In their case, rejection of the pancreas resulted in a spontaneous rupture of a pseudoaneurysm of the pancreas arterial anastomosis requiring emergent surgery.10 To our knowledge, there is no other literature on the successful treatment of metastatic donor-derived malignancy in SPK recipients.
Our series adds to this literature by describing the successful initial treatment of 3 SPK patients with donor-derived metastatic malignancy. Patient 1 had donor-transmitted malignancy, meaning the malignancy was present in the donor at the time of donation, while patients 2 and 3 had malignancy that likely originated from donor tissue years after transplantation. Although the timing of the development of these donor-derived malignancies was disparate, in all 3 cases, the malignancy expanded under the surveillance of recipient immune systems that were suppressed to prevent rejection. The common strategy used in all 3 cases forms the basis for the algorithm outlined in Figure 3. The strategies and proposed algorithm are based on our experience and a comprehensive literature review, which demonstrate that successful treatment of metastatic malignancies in transplant recipients is dependent on early allograft removal once donor-derived malignancy is identified, which allows immune reconstitution via immunosuppression cessation. These steps also permit the additional option of treatment with a checkpoint inhibitor, which can cause vigorous rejection if a transplanted organ remains in place.
FIGURE 3. Algorithm outlining management strategy of metastatic donor-derived malignancy in SPK patients. aMalignancy not squamous, skin or lymphoma and <24 mo from transplant, or current/previous lesion noted in transplanted organ. bTesting options include FISH, HLA typing, and nucleic acid–based testing. cRecommendation to remove remaining graft arguably stronger for a remaining pancreas than a remaining kidney, due to a higher risk of rejection and complications with the pancreas in particular. dExperimental directions for assessing response to therapy may include cPRA retesting as a potential biomarker for “rejection” of donor antigens. Cell-free DNA may experimentally be used to comment on tumor burden and response. eAdjuvant chemo- or targeted-therapy may also be indicated based on the malignancy’s tissue of origin, subtype, and grade. Immunotherapy may be preferred on a theoretical basis. However, most adjuvant recommendations are based on the general oncology literature as there are no standard of care recommendations for adjuvant treatment of donor-derived malignancy. fRemoval of remaining grafts before initiation of immunotherapy due to the high risk of rejection and complications with immunotherapy. SPK, simultaneous pancreas-kidney; PET-CT, positron emission tomography-computerized tomography scan; NED, no evidence of disease.
In the cases above, the methods for distinguishing donor from recipient tumor origin included analysis for differences in microsatellites, that is, short-tandem repeats or checking for a panel of specific gene mutations. These genomic techniques represent just a few options in a larger arsenal of techniques that can be applied to distinguish between donor and recipient tissue. These approaches also include HLA-typing or using fluorescence in situ hybridization for karyotyping.11-15 Using genomic techniques allowed us to apply cell-free-DNA testing as an adjunct to support our determination of no tumor recurrence in case 1. Cell free-DNA testing for malignancy has been described as potentially useful in determining cancer recurrence. However, it remains an evolving, experimental diagnostic tool.16
Focusing on the management approach to each case, case 1 was clearly a tumor that was transmitted with the donor pancreas, as the heart and liver recipients had metastatic pancreatic adenocarcinoma with similar histopathology. In contrast to SPK cases, stopping immunosuppression or removing the transplants in the cases of the heart and liver recipients would have been fatal. As a result, the heart and liver recipient ultimately died from metastatic pancreatic adenocarcinoma. In our case, the kidney and pancreas could be immediately removed due to the options for dialysis and insulin, respectively. At the time of allograft pancreatectomy and nephrectomy (6 mo after transplant), the tumor had metastasized to the adjacent fallopian tube. Although there was no evidence of tumor in the explanted kidney, a PET scan at the time of pancreatectomy and nephrectomy revealed diffuse nodal uptake within the left supraclavicular, mediastinal, mesenteric, and right external iliac regions. Despite the tumor’s locally aggressive nature with lymphovascular invasion, stopping all immunosuppression resulted in normalization of the PET scan within 6 months and no evidence of metastatic disease. In this case, simply restoring the immune response allowed “rejection” of the donor-derived tumor. It is difficult to determine how much of the tumor control is related to restoring immune-mediated tumor surveillance by stopping immunosuppression versus alloimmune rejection of tumor-bearing donor HLA. The fact that the alloimmune response likely provided a substantial contribution to eliminating the tumor is reflected by the dramatic increase in anti-HLA antibodies to a cPRA of 100%. Despite her high panel reactive antibodies, she received a 0-mismatched kidney transplant approximately 2 years after her explant and continues to do well with no evidence of tumor recurrence.
The malignancy in case 2 occurred in the transplanted pancreas of an SPK recipient 10 years after transplant. Following the confirmation of the tumor’s donor origin, both the pancreas and kidney were explanted, and immunosuppression was stopped. Like case 1, removal of both organs allowed complete immunosuppression cessation and resulted in resolution of the systemic lymphadenopathy and hypermetabolism at the site of the pancreas transplant observed with sequential PET scans. The decision to remove the normally functioning kidney transplant with no evidence of disease was difficult, but since the tumor was donor-derived, we were concerned about occult disease. We also wanted to be prepared to use checkpoint inhibitors to augment the immune response if immunosuppression withdrawal was insufficient to clear the tumor cells. In case 2, like case 1, immunosuppression withdrawal was also associated with an increase of donor specific HLA antibody and a cPRA of nearly 100%, suggesting that the alloimmune response contributed to the control of the tumor. There has been no evidence of recurrence 2 years following the explants, and the patient is active for both kidney and pancreas retransplantation.
Case 3 is different than the others in that the donor-derived malignancy occurred in a nonfunctioning kidney allograft 13 years following SPK, and the pancreas transplant was functioning normally. At the time of nephrectomy, there was no evidence of metastatic disease, so immunosuppression was lowered but not stopped based on the pancreas’s ongoing excellent function. Transplant pancreatectomy would have allowed immunosuppression cessation and would have been in line with the literature in single organ transplants where graft removal and immunosuppression cessation are the standard approaches. However, the treatment of SPK patients remains more complex than the treatment of kidney or PTA transplants as providers must weigh the risks of removing a normally functioning transplant that may be tumor-free. Thus, this decision involved a risk-benefit assessment by the transplant team and a conversation with the patient regarding preferences. The recipient ultimately did not want to abandon immunosuppression and felt having a functioning pancreas’s benefits outweighed the disease recurrence risks. When metastatic disease became apparent (pulmonary effusion, liver lesions), immunosuppression was stopped. She continued to have normal pancreas function but developed severe pain at the local recurrence site and required a chest tube for drainage of the malignant effusion. At this point, checkpoint inhibitors were initiated, but due to her frail condition and normal function (at the time), the pancreas allograft was not removed. However, within weeks of starting checkpoint inhibitors, her increasing insulin requirements and severe pain over the pancreas allograft prompted emergent transplant pancreatectomy. The checkpoint inhibitor therapy’s potency was remarkable, and the aggressive rejection induced was almost immediate. In retrospect, pancreatectomy should have been performed before checkpoint inhibitor initiation, as removal of the markedly inflamed and vascular allograft was challenging. The complications of higher blood loss and postexplant deep venous thrombosis might have been avoided. Therefore, removing the pancreas before the massive inflammatory response checkpoint inhibitors can induce may be safest. Despite the negative aspects of the rejection induced by checkpoint inhibitor therapy, the aggressive immune response was also associated with her metastatic disease’s rapid improvement. Within 3 months of immunotherapy initiation, the pulmonary, liver, and lymph lesions have improved substantially based on imaging studies.
We elected to reactivate the patients for either kidney alone or SPK in cases 1 and 2 following 2 years of being cancer free. Although the decision to proceed with transplantation for all potential recipients with a history of treated cancers is dependent on disease-free survival estimates for each malignancy, these data are not available for the scenarios described here. We are optimistic that the immune memory for the donor HLA will be able to maintain adequate control of the original donor-derived tumor, but rigorous follow-up will be necessary to ensure that immunosuppression reinitiation has not compromised immune-mediated control of the tumor. The ongoing presence of DSA would suggest ongoing antitumor activity and could be monitored for future study purposes.
In summary, the finding of metastatic donor-derived malignancy following SPK should prompt immediate removal of the allograft with the primary lesion, immunosuppression cessation, and strong consideration for removal of the second allograft too. If the natural immune surveillance associated with stopping immunosuppression fails to control the metastatic disease, checkpoint inhibition can augment the natural immune response and successfully control aggressive metastatic disease. This algorithm is only possible for kidney or SPK recipients since these patients have alternative medical therapies following allograft removal, unlike heart, lung, or liver recipients. This strategy is consistent with a broad literature review, demonstrating that successful management of metastatic donor-derived malignancies requires allograft removal, immunosuppression cessation, and adjuvant therapy dictated by tumor tissue of origin that can include occasional immunotherapy use to augment the immune response.
ACKNOWLEDGMENTS
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Published online 8 December, 2020.
D.A. and P.G.S. did conception and design. D.A., S.W., and G.W. did acquisition of data. D.A., E.A.C., T.F., G.R.R., R.H., and P.G.S. did analysis and interpretation of data. D.A. and P.G.S. drafted the article. S.W., H.J.B., E.A.C., T.F., G.R.R., R.H., and P.G.S. critically revised the article.
H.J.B. was funded by the National Institutes of Health Grant Number T32AI125222.
The authors declare no conflicts of interest. | NIVOLUMAB | DrugsGivenReaction | CC BY-NC-ND | 33324741 | 20,787,052 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Paraneoplastic Sensory Polyneuropathy Related to Anti-PD-L1-including Anticancer Treatment in a Patient with Lung Cancer.
Paraneoplastic neurological syndromes (PNS), such as sensory polyneuropathy, are rare, and serum neuronal antibodies that are used for diagnosing this syndrome are occasionally positive. Similarly, neurological immune-related adverse events due to immune checkpoint inhibitors (ICIs) are also rare. However, their etiologies and the relationship between them remain unclear. We herein report a patient with suspected small cell lung cancer who showed sensory polyneuropathy after treatment with atezolizumab in combination with cytotoxic agents (carboplatin and etoposide) and was doubly positive for serum anti-Hu and anti-SOX-1 antibodies. Treatment with ICI and cytotoxic agents may sometimes lead to the development of PNS.
Introduction
Paraneoplastic neurological syndromes (PNS) are defined as remote effects of cancer unrelated to direct effects of tumors or metastases and are exceedingly rare, affecting less than 1/10,000 cancer patients (1). Neuronal anti-Hu and anti-SOX-1 antibodies are sometimes detected in patients with PNS in the serum or cerebrospinal fluid, and the clinical signs are different from the main symptoms of cancer.
In general, neurological symptoms tend to progress subacutely and cause severe physical dysfunction. In about 80% of patients, the onset of neurological symptoms and the detection of antibodies precedes the detection of cancer by months to years; therefore, early antibody detection may be useful for predicting PNS.
Immune checkpoint inhibitors (ICIs) have widely been used in the treatment of multiple types of cancer. The main side effects of ICIs are immunological; thyroid function disorders are common, and organ-related side effects are also prevalent (2). Among these side effects, neurological immune-related adverse events (irAEs) due to ICIs are rare; only 0.93% of patients with ICI treatment experience neurological side effects, such as neuropathy, noninfective meningitis, encephalitis, and neuromuscular disorders (3).
Among neuronal autoantibodies related to PNS, anti-Hu antibody recognizes the nucleus of all central nervous system neurons and causes limbic encephalitis (LE), subacute cerebellar degeneration, and polyneuropathy. PNS with anti-Hu antibody was commonly seen in 78%, of small-cell lung cancer (SCLC) patients; 73% showed signs and symptoms of multifocal involvement of the nervous system, while 74% had sensory neuronopathy (2). In addition to anti-Hu antibody, anti-SOX-1 antibody recognizes the nuclear protein of glial cells and causes axonal neuropathy, demyelinating neuropathy, sensory and motor disorders (4). Five of 55 (9.1%) anti-SOX-1 antibody-positive patients reportedly had coexisting anti-Hu antibodies in SCLC patients (5).
Sensory polyneuropathy has clinically been recognized due to the advent of ICIs (3); however, neurological adverse events (nAEs) of ICIs, such as sensory polyneuropathy, are considered PNS and are still rare in patients with lung cancer. We herein report the first SCLC patient with sensory polyneuropathy who tested positive for anti-neuronal antibodies after initial treatment with anti programmed death-ligand 1 (PD-L1) antibody combined with cytotoxic agents.
Case Report
A 70-year-old current-smoking (80 pack-years) Japanese man with a history of hypertension and nonrecurrent colon cancer was admitted to our hospital. Two months earlier, he had experienced dizziness and been diagnosed with brain stem hemorrhaging due to cavernous malformation (hemangioma). In addition, multiple nodules suspected of being brain metastases had been seen on enhanced magnetic resonance imaging (MRI). He has been treated with concentrated glycerin and fructose at a previous hospital, and his symptoms had completely disappeared.
Upon admission to our hospital, a physical examination revealed the following: height, 171.5 cm; body weight, 74.1 kg; body temperature, 35.9℃; heart rate, 103 bpm; blood pressure, 107/71 mmHg; and oxygen saturation, 95% (room air, at rest); no neurological abnormalities were observed. Laboratory results (Table 1) showed a high titer of pro-gastrin-releasing peptide (pro-GRP; 541.5 pg/mL) and a low titer of carcinoembryonic antigen (CEA; 2.5 ng/mL) and cytokeratin fragment 21-1 (CYFRA21-1; 1.7 ng/mL). Chest radiography and high-resolution computed tomography showed two nodules in the right lower lobe with hilar lymphadenopathies. Sputum samples for a cytological examination were not available. Invasive pathological examinations, including bronchoscopy, could not be carried out because the patient was at a high risk of recurrent intracranial hemorrhaging due to cavernous hemangioma. The serum pro-GRP titer has a sensitivity of 86.4%, specificity of 96.4%, positive predictive value of 96.7%, and negative predictive value of 84.4% for discriminating SCLC from non-SCLC (cut-off 77.8 pg/mL) (6), and a titer of over 329.3 pg/mL pro-GRP also suggests extensive SCLC rather than limited disease (6). The high serum level of pro-GRP (541.5 pg/mL) in this patient was thus considered a marker of SCLC. Consequently, he was clinically diagnosed with stage IV SCLC based on the radiological and serological findings.
Table 1. Laboratory Data on Admission.
<Blood cell counts> T-bil 0.8 mg/dL ANA (-)
WBC 5,100 /μL AST 28 IU/L Anti-SS-A (-)
Neutrophils 72.2 % ALT 37 IU/L Anti-SS-B (-)
Lymphocytes 21.1 % LDH 224 IU/L vit.B1 41.9 ng/mL
Eosinophils 0.6 % γ-GTP 49 IU/L vit.B12 539 pg/mL
Monocytes 5.5 % BUN 11.3 mg/dL CEA 2.5 ng/mL
Basophils 0.6 % Cre 0.88 mg/dL CYFRA21-1 1.7 ng/mL
RBC 5.36×106 /μL Na 141 mEq/L Pro-GRP 541.5 pg/mL
Hb 17.1 g/dL K 4.1 mEq/L <Cerebrospinal fluid>
Ht 49.1 % Cl 107 mEq/L Cell 1 /μL
Platelets 18.3×104 /μL Glucose 156 mg/dL Mono 1 /μL
<Blood chemistry> HbA1c 6.3 % Protein 146 mg/dL
TP 7.9 g/dL CRP 0.02 mg/dL Glucose 57 mg/dL
Alb 4.6 g/dL RF <0.1 IU/mL Cytology No malignancy
WBC: white blood cell, RBC: red blood cell, Hb: hemoglobin, Ht: hematocrit, TP: total protein, Alb: albumin, T-bil: total bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, γ-GTP: gamma-glutamyl transferase, BUN: blood urea nitrogen, Cre: creatinine, CRP: c-reactive protein, RF: rheumatoid factor, ANA: anti nuclear antibody, Anti-SS-A: anti Sjögren syndrome-A antibody, Anti-SS-B: anti Sjögren syndrome-B antibody, vit.B1: vitamin B1, vit.B12: vitamin B12, CEA: carcinoembryonic antigen, CYFRA: cytokeratin fragment, Pro-GRP: pro-gastrin-releasing peptide
As the first-line treatment, combination therapy with carboplatin [area under the concentration-time curve (AUC) 5], etoposide (100 mg/m2), and atezolizumab (1,200 mg) was administered (Figure). Three weeks after the chemotherapy, whole-brain irradiation (3×10 Gy) was performed. Four weeks after chemotherapy, tactile and pain disturbances consistent with L2-3 dermatome, loss of bilateral patellar tendon reflexes, and loss of bilateral Achilles tendon reflexes were revealed. No obvious weakness was observed in the manual muscle strength test, Babinski reflexes were negative, and lower limb Barre's sign was positive. Coordination was poor in the finger-nose-finger test and knee-heel test, and gait was oscillating. Lumbar puncture results (Table 1) showed protein cell dissociation and no malignant cells. Enhanced brain MRI showed no evidence of recurrent brain stem hemorrhaging, and the size of all metastatic brain tumors had decreased. Lumbar spinal contrast-enhanced MRI showed no abnormal contrast enhancement of peripheral nerves, and nerve root compression consistent with the symptoms was observed. Serum anti-Hu and anti-SOX-1 antibody tests were both positive, and tests for onconeural antibodies (including anti-CV2, anti-Yo, anti-Ri, anti-amphiphysin, anti-paraneoplastic antigen MA2, anti-recoverin, anti-titin, anti-ZIC4, anti-GAD65 and anti-Tr) were all negative. Nerve conduction testing revealed that amplitude of sensory nerves was not evoked, and no apparent abnormalities in motor nerves were observed (Table 2).
Figure. Clinical course of the patient. IVIg: intravenous immunoglobulin therapy, Pro-GRP: pro-gastrin-releasing peptide
Table 2. The Nerve Conduction Study.
MCS Distal latency
(msec) CMAP amplitude
(mV) MCV
(m/s) SNAP
(μV) SCV
(m/s)
Rt. Median 3.4 18.9 50.9 18.1 50.9
Rt. Ulnar 3 15.4 46.8 12.5 47
Rt. Peroneal 5.4 13.8 42 - -
Rt. Tibial 6.1 18.1 38.3 - -
Rt. Sural 3.8 - - not evoked not evoked
MCS: motor conduction study, SCS: sensory conduction study, CAMP: compound muscle action potential, SNAP: sensory nerve action potential, MCV: motor conduction velocity, SCV: sensory conduction velocity, Rt: right
Based on the neurological findings and results of the nerve conduction test, the patient was diagnosed with sensory polyneuropathy. After systemic chemotherapy with atezolizumab and radiotherapy to the brain, the chest and brain tumors had shrunk, and the elevated serum level of pro-GRP had drastically decreased from 541.5 to 43.9 pg/mL. The clinical timeline of the onset of neurological symptoms was indicative of chemotherapy, particularly atezolizumab-induced. Intravenous immunoglobulin (IVIg) therapy was administered after the first round of systemic chemotherapy, but his neurological symptoms did not improve.
Discussion
Our patient with suspected SCLC who experienced an nAE of sensory polyneuropathy tested positive for anti-neuronal antibodies after initial treatment with atezolizumab in combination with carboplatin and etoposide.
Anti-acetylcholine receptor antibodies are generally pathogenic (e.g., in myasthenia gravis), but not all lung cancer patients positive for anti-Hu antibodies show neurological symptoms; indeed, among 196 SCLC patients with anti-Hu antibodies, 31 patients (16%) had no neurological symptoms (7). In relation to the treatment with ICIs and PNS with anti-Hu antibody, nivolumab (anti-PD-1 antibody)-induced sensory neuropathy (8), nivolumab-induced LE (9), sintilimab-induced LE, and enteric neuropathy (10) have been reported.
Regarding the etiology of PNS, the decreased expression of multiple Treg-related genes involved in immune regulation in SCLC patients might cause impaired immune tolerance, tissue damage due to autoimmune mechanisms, and PNS (11). PD-L1-coated beads can induce Tregs in vitro, and PD-L1 increases Foxp3 expression and enhances the immunosuppressive ability of Tregs (12), suggesting that anti-PD-L1 antibody treatment may suppress Treg infiltration into tumors. Hence, the present patient was positive for anti-neuronal antibodies, but the neurological symptoms manifested after chemotherapy, suggesting that the PNS was evoked by an ICI, such as atezolizumab, and cytotoxic agents.
In the management of PNS, underlying disease treatments, such as anticancer therapy, are prioritized, which may partly improve neurological symptoms; however, only 10-20% of patients achieve improvement in their neurological symptoms (13). For patients in whom neurological symptoms persist, plasma exchange, systemic corticosteroids, immunosuppressants, or IVIg therapy are considered but usually prove to be ineffective (14). Thus far, there have been no available data concerning randomized controlled trials for the treatment of PNS; the available data have only been collected from case series, case reports, or expert opinions (class IV evidence) regarding the effect of immunomodulation (IVIg, steroid treatment, plasma exchange, or chemotherapy) on paraneoplastic neuropathy (15). In our case study, although the patient was treated with a high-dose immunoglobulin, no neurological improvement was noted. In addition, the relationship between the presence of anti-neurological antibodies and the prognosis of neurological symptoms remains controversial (16,17).
Several limitations associated with the present study warrant mention. First, even though anticancer treatment was performed, there was no pathological evidence of lung cancer. There might have been other differential diagnoses, such as non-SCLC, carcinoid, large-cell neuroendocrine carcinoma, and infectious diseases. Second, as this patient had already presented with neurological symptoms at a previous hospital due to hemorrhaging of cavernous malformation, the discrimination of neurological symptoms of irAEs should have been carefully deliberated. However, the intracranial hemorrhagic symptoms had completely disappeared following treatment with concentrated glycerin and fructose before systemic chemotherapy, and sensory disturbances due to irAEs appeared four weeks after chemotherapy; the disturbances were consistent with those of PNS (Figure). Therefore, we believe that the etiology of the two neurological symptoms are clearly distinguishable. Third, the patient was treated with atezolizumab in combination with carboplatin and etoposide; therefore, the specific agent causing PNS was clinically unclear. However, both ICIs and cytotoxic agents have been reported to cause PNS, suggesting that combination therapy with ICIs and cytotoxic agents may increase the incidence of PNS.
Conclusion
In conclusion, we herein report an anti-Hu and anti-SOX-1 antibody dual-positive patient with suspected SCLC induced by combination treatment with carboplatin, etoposide, and atezolizumab. This irAE is often intractable, so physicians should be aware of this side effect, especially when treating patients with anti-neuronal antibodies using the combination of an ICI and cytotoxic agents.
The authors state that they have no Conflict of Interest (COI). | ATEZOLIZUMAB, CARBOPLATIN, ETOPOSIDE, IMMUNE GLOBULIN NOS | DrugsGivenReaction | CC BY-NC-ND | 33328400 | 18,842,917 | 2021-05-15 |
What is the weight of the patient? | Paraneoplastic Sensory Polyneuropathy Related to Anti-PD-L1-including Anticancer Treatment in a Patient with Lung Cancer.
Paraneoplastic neurological syndromes (PNS), such as sensory polyneuropathy, are rare, and serum neuronal antibodies that are used for diagnosing this syndrome are occasionally positive. Similarly, neurological immune-related adverse events due to immune checkpoint inhibitors (ICIs) are also rare. However, their etiologies and the relationship between them remain unclear. We herein report a patient with suspected small cell lung cancer who showed sensory polyneuropathy after treatment with atezolizumab in combination with cytotoxic agents (carboplatin and etoposide) and was doubly positive for serum anti-Hu and anti-SOX-1 antibodies. Treatment with ICI and cytotoxic agents may sometimes lead to the development of PNS.
Introduction
Paraneoplastic neurological syndromes (PNS) are defined as remote effects of cancer unrelated to direct effects of tumors or metastases and are exceedingly rare, affecting less than 1/10,000 cancer patients (1). Neuronal anti-Hu and anti-SOX-1 antibodies are sometimes detected in patients with PNS in the serum or cerebrospinal fluid, and the clinical signs are different from the main symptoms of cancer.
In general, neurological symptoms tend to progress subacutely and cause severe physical dysfunction. In about 80% of patients, the onset of neurological symptoms and the detection of antibodies precedes the detection of cancer by months to years; therefore, early antibody detection may be useful for predicting PNS.
Immune checkpoint inhibitors (ICIs) have widely been used in the treatment of multiple types of cancer. The main side effects of ICIs are immunological; thyroid function disorders are common, and organ-related side effects are also prevalent (2). Among these side effects, neurological immune-related adverse events (irAEs) due to ICIs are rare; only 0.93% of patients with ICI treatment experience neurological side effects, such as neuropathy, noninfective meningitis, encephalitis, and neuromuscular disorders (3).
Among neuronal autoantibodies related to PNS, anti-Hu antibody recognizes the nucleus of all central nervous system neurons and causes limbic encephalitis (LE), subacute cerebellar degeneration, and polyneuropathy. PNS with anti-Hu antibody was commonly seen in 78%, of small-cell lung cancer (SCLC) patients; 73% showed signs and symptoms of multifocal involvement of the nervous system, while 74% had sensory neuronopathy (2). In addition to anti-Hu antibody, anti-SOX-1 antibody recognizes the nuclear protein of glial cells and causes axonal neuropathy, demyelinating neuropathy, sensory and motor disorders (4). Five of 55 (9.1%) anti-SOX-1 antibody-positive patients reportedly had coexisting anti-Hu antibodies in SCLC patients (5).
Sensory polyneuropathy has clinically been recognized due to the advent of ICIs (3); however, neurological adverse events (nAEs) of ICIs, such as sensory polyneuropathy, are considered PNS and are still rare in patients with lung cancer. We herein report the first SCLC patient with sensory polyneuropathy who tested positive for anti-neuronal antibodies after initial treatment with anti programmed death-ligand 1 (PD-L1) antibody combined with cytotoxic agents.
Case Report
A 70-year-old current-smoking (80 pack-years) Japanese man with a history of hypertension and nonrecurrent colon cancer was admitted to our hospital. Two months earlier, he had experienced dizziness and been diagnosed with brain stem hemorrhaging due to cavernous malformation (hemangioma). In addition, multiple nodules suspected of being brain metastases had been seen on enhanced magnetic resonance imaging (MRI). He has been treated with concentrated glycerin and fructose at a previous hospital, and his symptoms had completely disappeared.
Upon admission to our hospital, a physical examination revealed the following: height, 171.5 cm; body weight, 74.1 kg; body temperature, 35.9℃; heart rate, 103 bpm; blood pressure, 107/71 mmHg; and oxygen saturation, 95% (room air, at rest); no neurological abnormalities were observed. Laboratory results (Table 1) showed a high titer of pro-gastrin-releasing peptide (pro-GRP; 541.5 pg/mL) and a low titer of carcinoembryonic antigen (CEA; 2.5 ng/mL) and cytokeratin fragment 21-1 (CYFRA21-1; 1.7 ng/mL). Chest radiography and high-resolution computed tomography showed two nodules in the right lower lobe with hilar lymphadenopathies. Sputum samples for a cytological examination were not available. Invasive pathological examinations, including bronchoscopy, could not be carried out because the patient was at a high risk of recurrent intracranial hemorrhaging due to cavernous hemangioma. The serum pro-GRP titer has a sensitivity of 86.4%, specificity of 96.4%, positive predictive value of 96.7%, and negative predictive value of 84.4% for discriminating SCLC from non-SCLC (cut-off 77.8 pg/mL) (6), and a titer of over 329.3 pg/mL pro-GRP also suggests extensive SCLC rather than limited disease (6). The high serum level of pro-GRP (541.5 pg/mL) in this patient was thus considered a marker of SCLC. Consequently, he was clinically diagnosed with stage IV SCLC based on the radiological and serological findings.
Table 1. Laboratory Data on Admission.
<Blood cell counts> T-bil 0.8 mg/dL ANA (-)
WBC 5,100 /μL AST 28 IU/L Anti-SS-A (-)
Neutrophils 72.2 % ALT 37 IU/L Anti-SS-B (-)
Lymphocytes 21.1 % LDH 224 IU/L vit.B1 41.9 ng/mL
Eosinophils 0.6 % γ-GTP 49 IU/L vit.B12 539 pg/mL
Monocytes 5.5 % BUN 11.3 mg/dL CEA 2.5 ng/mL
Basophils 0.6 % Cre 0.88 mg/dL CYFRA21-1 1.7 ng/mL
RBC 5.36×106 /μL Na 141 mEq/L Pro-GRP 541.5 pg/mL
Hb 17.1 g/dL K 4.1 mEq/L <Cerebrospinal fluid>
Ht 49.1 % Cl 107 mEq/L Cell 1 /μL
Platelets 18.3×104 /μL Glucose 156 mg/dL Mono 1 /μL
<Blood chemistry> HbA1c 6.3 % Protein 146 mg/dL
TP 7.9 g/dL CRP 0.02 mg/dL Glucose 57 mg/dL
Alb 4.6 g/dL RF <0.1 IU/mL Cytology No malignancy
WBC: white blood cell, RBC: red blood cell, Hb: hemoglobin, Ht: hematocrit, TP: total protein, Alb: albumin, T-bil: total bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, γ-GTP: gamma-glutamyl transferase, BUN: blood urea nitrogen, Cre: creatinine, CRP: c-reactive protein, RF: rheumatoid factor, ANA: anti nuclear antibody, Anti-SS-A: anti Sjögren syndrome-A antibody, Anti-SS-B: anti Sjögren syndrome-B antibody, vit.B1: vitamin B1, vit.B12: vitamin B12, CEA: carcinoembryonic antigen, CYFRA: cytokeratin fragment, Pro-GRP: pro-gastrin-releasing peptide
As the first-line treatment, combination therapy with carboplatin [area under the concentration-time curve (AUC) 5], etoposide (100 mg/m2), and atezolizumab (1,200 mg) was administered (Figure). Three weeks after the chemotherapy, whole-brain irradiation (3×10 Gy) was performed. Four weeks after chemotherapy, tactile and pain disturbances consistent with L2-3 dermatome, loss of bilateral patellar tendon reflexes, and loss of bilateral Achilles tendon reflexes were revealed. No obvious weakness was observed in the manual muscle strength test, Babinski reflexes were negative, and lower limb Barre's sign was positive. Coordination was poor in the finger-nose-finger test and knee-heel test, and gait was oscillating. Lumbar puncture results (Table 1) showed protein cell dissociation and no malignant cells. Enhanced brain MRI showed no evidence of recurrent brain stem hemorrhaging, and the size of all metastatic brain tumors had decreased. Lumbar spinal contrast-enhanced MRI showed no abnormal contrast enhancement of peripheral nerves, and nerve root compression consistent with the symptoms was observed. Serum anti-Hu and anti-SOX-1 antibody tests were both positive, and tests for onconeural antibodies (including anti-CV2, anti-Yo, anti-Ri, anti-amphiphysin, anti-paraneoplastic antigen MA2, anti-recoverin, anti-titin, anti-ZIC4, anti-GAD65 and anti-Tr) were all negative. Nerve conduction testing revealed that amplitude of sensory nerves was not evoked, and no apparent abnormalities in motor nerves were observed (Table 2).
Figure. Clinical course of the patient. IVIg: intravenous immunoglobulin therapy, Pro-GRP: pro-gastrin-releasing peptide
Table 2. The Nerve Conduction Study.
MCS Distal latency
(msec) CMAP amplitude
(mV) MCV
(m/s) SNAP
(μV) SCV
(m/s)
Rt. Median 3.4 18.9 50.9 18.1 50.9
Rt. Ulnar 3 15.4 46.8 12.5 47
Rt. Peroneal 5.4 13.8 42 - -
Rt. Tibial 6.1 18.1 38.3 - -
Rt. Sural 3.8 - - not evoked not evoked
MCS: motor conduction study, SCS: sensory conduction study, CAMP: compound muscle action potential, SNAP: sensory nerve action potential, MCV: motor conduction velocity, SCV: sensory conduction velocity, Rt: right
Based on the neurological findings and results of the nerve conduction test, the patient was diagnosed with sensory polyneuropathy. After systemic chemotherapy with atezolizumab and radiotherapy to the brain, the chest and brain tumors had shrunk, and the elevated serum level of pro-GRP had drastically decreased from 541.5 to 43.9 pg/mL. The clinical timeline of the onset of neurological symptoms was indicative of chemotherapy, particularly atezolizumab-induced. Intravenous immunoglobulin (IVIg) therapy was administered after the first round of systemic chemotherapy, but his neurological symptoms did not improve.
Discussion
Our patient with suspected SCLC who experienced an nAE of sensory polyneuropathy tested positive for anti-neuronal antibodies after initial treatment with atezolizumab in combination with carboplatin and etoposide.
Anti-acetylcholine receptor antibodies are generally pathogenic (e.g., in myasthenia gravis), but not all lung cancer patients positive for anti-Hu antibodies show neurological symptoms; indeed, among 196 SCLC patients with anti-Hu antibodies, 31 patients (16%) had no neurological symptoms (7). In relation to the treatment with ICIs and PNS with anti-Hu antibody, nivolumab (anti-PD-1 antibody)-induced sensory neuropathy (8), nivolumab-induced LE (9), sintilimab-induced LE, and enteric neuropathy (10) have been reported.
Regarding the etiology of PNS, the decreased expression of multiple Treg-related genes involved in immune regulation in SCLC patients might cause impaired immune tolerance, tissue damage due to autoimmune mechanisms, and PNS (11). PD-L1-coated beads can induce Tregs in vitro, and PD-L1 increases Foxp3 expression and enhances the immunosuppressive ability of Tregs (12), suggesting that anti-PD-L1 antibody treatment may suppress Treg infiltration into tumors. Hence, the present patient was positive for anti-neuronal antibodies, but the neurological symptoms manifested after chemotherapy, suggesting that the PNS was evoked by an ICI, such as atezolizumab, and cytotoxic agents.
In the management of PNS, underlying disease treatments, such as anticancer therapy, are prioritized, which may partly improve neurological symptoms; however, only 10-20% of patients achieve improvement in their neurological symptoms (13). For patients in whom neurological symptoms persist, plasma exchange, systemic corticosteroids, immunosuppressants, or IVIg therapy are considered but usually prove to be ineffective (14). Thus far, there have been no available data concerning randomized controlled trials for the treatment of PNS; the available data have only been collected from case series, case reports, or expert opinions (class IV evidence) regarding the effect of immunomodulation (IVIg, steroid treatment, plasma exchange, or chemotherapy) on paraneoplastic neuropathy (15). In our case study, although the patient was treated with a high-dose immunoglobulin, no neurological improvement was noted. In addition, the relationship between the presence of anti-neurological antibodies and the prognosis of neurological symptoms remains controversial (16,17).
Several limitations associated with the present study warrant mention. First, even though anticancer treatment was performed, there was no pathological evidence of lung cancer. There might have been other differential diagnoses, such as non-SCLC, carcinoid, large-cell neuroendocrine carcinoma, and infectious diseases. Second, as this patient had already presented with neurological symptoms at a previous hospital due to hemorrhaging of cavernous malformation, the discrimination of neurological symptoms of irAEs should have been carefully deliberated. However, the intracranial hemorrhagic symptoms had completely disappeared following treatment with concentrated glycerin and fructose before systemic chemotherapy, and sensory disturbances due to irAEs appeared four weeks after chemotherapy; the disturbances were consistent with those of PNS (Figure). Therefore, we believe that the etiology of the two neurological symptoms are clearly distinguishable. Third, the patient was treated with atezolizumab in combination with carboplatin and etoposide; therefore, the specific agent causing PNS was clinically unclear. However, both ICIs and cytotoxic agents have been reported to cause PNS, suggesting that combination therapy with ICIs and cytotoxic agents may increase the incidence of PNS.
Conclusion
In conclusion, we herein report an anti-Hu and anti-SOX-1 antibody dual-positive patient with suspected SCLC induced by combination treatment with carboplatin, etoposide, and atezolizumab. This irAE is often intractable, so physicians should be aware of this side effect, especially when treating patients with anti-neuronal antibodies using the combination of an ICI and cytotoxic agents.
The authors state that they have no Conflict of Interest (COI). | 74.1 kg. | Weight | CC BY-NC-ND | 33328400 | 18,842,917 | 2021-05-15 |
What was the administration route of drug 'IMMUNE GLOBULIN NOS'? | Paraneoplastic Sensory Polyneuropathy Related to Anti-PD-L1-including Anticancer Treatment in a Patient with Lung Cancer.
Paraneoplastic neurological syndromes (PNS), such as sensory polyneuropathy, are rare, and serum neuronal antibodies that are used for diagnosing this syndrome are occasionally positive. Similarly, neurological immune-related adverse events due to immune checkpoint inhibitors (ICIs) are also rare. However, their etiologies and the relationship between them remain unclear. We herein report a patient with suspected small cell lung cancer who showed sensory polyneuropathy after treatment with atezolizumab in combination with cytotoxic agents (carboplatin and etoposide) and was doubly positive for serum anti-Hu and anti-SOX-1 antibodies. Treatment with ICI and cytotoxic agents may sometimes lead to the development of PNS.
Introduction
Paraneoplastic neurological syndromes (PNS) are defined as remote effects of cancer unrelated to direct effects of tumors or metastases and are exceedingly rare, affecting less than 1/10,000 cancer patients (1). Neuronal anti-Hu and anti-SOX-1 antibodies are sometimes detected in patients with PNS in the serum or cerebrospinal fluid, and the clinical signs are different from the main symptoms of cancer.
In general, neurological symptoms tend to progress subacutely and cause severe physical dysfunction. In about 80% of patients, the onset of neurological symptoms and the detection of antibodies precedes the detection of cancer by months to years; therefore, early antibody detection may be useful for predicting PNS.
Immune checkpoint inhibitors (ICIs) have widely been used in the treatment of multiple types of cancer. The main side effects of ICIs are immunological; thyroid function disorders are common, and organ-related side effects are also prevalent (2). Among these side effects, neurological immune-related adverse events (irAEs) due to ICIs are rare; only 0.93% of patients with ICI treatment experience neurological side effects, such as neuropathy, noninfective meningitis, encephalitis, and neuromuscular disorders (3).
Among neuronal autoantibodies related to PNS, anti-Hu antibody recognizes the nucleus of all central nervous system neurons and causes limbic encephalitis (LE), subacute cerebellar degeneration, and polyneuropathy. PNS with anti-Hu antibody was commonly seen in 78%, of small-cell lung cancer (SCLC) patients; 73% showed signs and symptoms of multifocal involvement of the nervous system, while 74% had sensory neuronopathy (2). In addition to anti-Hu antibody, anti-SOX-1 antibody recognizes the nuclear protein of glial cells and causes axonal neuropathy, demyelinating neuropathy, sensory and motor disorders (4). Five of 55 (9.1%) anti-SOX-1 antibody-positive patients reportedly had coexisting anti-Hu antibodies in SCLC patients (5).
Sensory polyneuropathy has clinically been recognized due to the advent of ICIs (3); however, neurological adverse events (nAEs) of ICIs, such as sensory polyneuropathy, are considered PNS and are still rare in patients with lung cancer. We herein report the first SCLC patient with sensory polyneuropathy who tested positive for anti-neuronal antibodies after initial treatment with anti programmed death-ligand 1 (PD-L1) antibody combined with cytotoxic agents.
Case Report
A 70-year-old current-smoking (80 pack-years) Japanese man with a history of hypertension and nonrecurrent colon cancer was admitted to our hospital. Two months earlier, he had experienced dizziness and been diagnosed with brain stem hemorrhaging due to cavernous malformation (hemangioma). In addition, multiple nodules suspected of being brain metastases had been seen on enhanced magnetic resonance imaging (MRI). He has been treated with concentrated glycerin and fructose at a previous hospital, and his symptoms had completely disappeared.
Upon admission to our hospital, a physical examination revealed the following: height, 171.5 cm; body weight, 74.1 kg; body temperature, 35.9℃; heart rate, 103 bpm; blood pressure, 107/71 mmHg; and oxygen saturation, 95% (room air, at rest); no neurological abnormalities were observed. Laboratory results (Table 1) showed a high titer of pro-gastrin-releasing peptide (pro-GRP; 541.5 pg/mL) and a low titer of carcinoembryonic antigen (CEA; 2.5 ng/mL) and cytokeratin fragment 21-1 (CYFRA21-1; 1.7 ng/mL). Chest radiography and high-resolution computed tomography showed two nodules in the right lower lobe with hilar lymphadenopathies. Sputum samples for a cytological examination were not available. Invasive pathological examinations, including bronchoscopy, could not be carried out because the patient was at a high risk of recurrent intracranial hemorrhaging due to cavernous hemangioma. The serum pro-GRP titer has a sensitivity of 86.4%, specificity of 96.4%, positive predictive value of 96.7%, and negative predictive value of 84.4% for discriminating SCLC from non-SCLC (cut-off 77.8 pg/mL) (6), and a titer of over 329.3 pg/mL pro-GRP also suggests extensive SCLC rather than limited disease (6). The high serum level of pro-GRP (541.5 pg/mL) in this patient was thus considered a marker of SCLC. Consequently, he was clinically diagnosed with stage IV SCLC based on the radiological and serological findings.
Table 1. Laboratory Data on Admission.
<Blood cell counts> T-bil 0.8 mg/dL ANA (-)
WBC 5,100 /μL AST 28 IU/L Anti-SS-A (-)
Neutrophils 72.2 % ALT 37 IU/L Anti-SS-B (-)
Lymphocytes 21.1 % LDH 224 IU/L vit.B1 41.9 ng/mL
Eosinophils 0.6 % γ-GTP 49 IU/L vit.B12 539 pg/mL
Monocytes 5.5 % BUN 11.3 mg/dL CEA 2.5 ng/mL
Basophils 0.6 % Cre 0.88 mg/dL CYFRA21-1 1.7 ng/mL
RBC 5.36×106 /μL Na 141 mEq/L Pro-GRP 541.5 pg/mL
Hb 17.1 g/dL K 4.1 mEq/L <Cerebrospinal fluid>
Ht 49.1 % Cl 107 mEq/L Cell 1 /μL
Platelets 18.3×104 /μL Glucose 156 mg/dL Mono 1 /μL
<Blood chemistry> HbA1c 6.3 % Protein 146 mg/dL
TP 7.9 g/dL CRP 0.02 mg/dL Glucose 57 mg/dL
Alb 4.6 g/dL RF <0.1 IU/mL Cytology No malignancy
WBC: white blood cell, RBC: red blood cell, Hb: hemoglobin, Ht: hematocrit, TP: total protein, Alb: albumin, T-bil: total bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, γ-GTP: gamma-glutamyl transferase, BUN: blood urea nitrogen, Cre: creatinine, CRP: c-reactive protein, RF: rheumatoid factor, ANA: anti nuclear antibody, Anti-SS-A: anti Sjögren syndrome-A antibody, Anti-SS-B: anti Sjögren syndrome-B antibody, vit.B1: vitamin B1, vit.B12: vitamin B12, CEA: carcinoembryonic antigen, CYFRA: cytokeratin fragment, Pro-GRP: pro-gastrin-releasing peptide
As the first-line treatment, combination therapy with carboplatin [area under the concentration-time curve (AUC) 5], etoposide (100 mg/m2), and atezolizumab (1,200 mg) was administered (Figure). Three weeks after the chemotherapy, whole-brain irradiation (3×10 Gy) was performed. Four weeks after chemotherapy, tactile and pain disturbances consistent with L2-3 dermatome, loss of bilateral patellar tendon reflexes, and loss of bilateral Achilles tendon reflexes were revealed. No obvious weakness was observed in the manual muscle strength test, Babinski reflexes were negative, and lower limb Barre's sign was positive. Coordination was poor in the finger-nose-finger test and knee-heel test, and gait was oscillating. Lumbar puncture results (Table 1) showed protein cell dissociation and no malignant cells. Enhanced brain MRI showed no evidence of recurrent brain stem hemorrhaging, and the size of all metastatic brain tumors had decreased. Lumbar spinal contrast-enhanced MRI showed no abnormal contrast enhancement of peripheral nerves, and nerve root compression consistent with the symptoms was observed. Serum anti-Hu and anti-SOX-1 antibody tests were both positive, and tests for onconeural antibodies (including anti-CV2, anti-Yo, anti-Ri, anti-amphiphysin, anti-paraneoplastic antigen MA2, anti-recoverin, anti-titin, anti-ZIC4, anti-GAD65 and anti-Tr) were all negative. Nerve conduction testing revealed that amplitude of sensory nerves was not evoked, and no apparent abnormalities in motor nerves were observed (Table 2).
Figure. Clinical course of the patient. IVIg: intravenous immunoglobulin therapy, Pro-GRP: pro-gastrin-releasing peptide
Table 2. The Nerve Conduction Study.
MCS Distal latency
(msec) CMAP amplitude
(mV) MCV
(m/s) SNAP
(μV) SCV
(m/s)
Rt. Median 3.4 18.9 50.9 18.1 50.9
Rt. Ulnar 3 15.4 46.8 12.5 47
Rt. Peroneal 5.4 13.8 42 - -
Rt. Tibial 6.1 18.1 38.3 - -
Rt. Sural 3.8 - - not evoked not evoked
MCS: motor conduction study, SCS: sensory conduction study, CAMP: compound muscle action potential, SNAP: sensory nerve action potential, MCV: motor conduction velocity, SCV: sensory conduction velocity, Rt: right
Based on the neurological findings and results of the nerve conduction test, the patient was diagnosed with sensory polyneuropathy. After systemic chemotherapy with atezolizumab and radiotherapy to the brain, the chest and brain tumors had shrunk, and the elevated serum level of pro-GRP had drastically decreased from 541.5 to 43.9 pg/mL. The clinical timeline of the onset of neurological symptoms was indicative of chemotherapy, particularly atezolizumab-induced. Intravenous immunoglobulin (IVIg) therapy was administered after the first round of systemic chemotherapy, but his neurological symptoms did not improve.
Discussion
Our patient with suspected SCLC who experienced an nAE of sensory polyneuropathy tested positive for anti-neuronal antibodies after initial treatment with atezolizumab in combination with carboplatin and etoposide.
Anti-acetylcholine receptor antibodies are generally pathogenic (e.g., in myasthenia gravis), but not all lung cancer patients positive for anti-Hu antibodies show neurological symptoms; indeed, among 196 SCLC patients with anti-Hu antibodies, 31 patients (16%) had no neurological symptoms (7). In relation to the treatment with ICIs and PNS with anti-Hu antibody, nivolumab (anti-PD-1 antibody)-induced sensory neuropathy (8), nivolumab-induced LE (9), sintilimab-induced LE, and enteric neuropathy (10) have been reported.
Regarding the etiology of PNS, the decreased expression of multiple Treg-related genes involved in immune regulation in SCLC patients might cause impaired immune tolerance, tissue damage due to autoimmune mechanisms, and PNS (11). PD-L1-coated beads can induce Tregs in vitro, and PD-L1 increases Foxp3 expression and enhances the immunosuppressive ability of Tregs (12), suggesting that anti-PD-L1 antibody treatment may suppress Treg infiltration into tumors. Hence, the present patient was positive for anti-neuronal antibodies, but the neurological symptoms manifested after chemotherapy, suggesting that the PNS was evoked by an ICI, such as atezolizumab, and cytotoxic agents.
In the management of PNS, underlying disease treatments, such as anticancer therapy, are prioritized, which may partly improve neurological symptoms; however, only 10-20% of patients achieve improvement in their neurological symptoms (13). For patients in whom neurological symptoms persist, plasma exchange, systemic corticosteroids, immunosuppressants, or IVIg therapy are considered but usually prove to be ineffective (14). Thus far, there have been no available data concerning randomized controlled trials for the treatment of PNS; the available data have only been collected from case series, case reports, or expert opinions (class IV evidence) regarding the effect of immunomodulation (IVIg, steroid treatment, plasma exchange, or chemotherapy) on paraneoplastic neuropathy (15). In our case study, although the patient was treated with a high-dose immunoglobulin, no neurological improvement was noted. In addition, the relationship between the presence of anti-neurological antibodies and the prognosis of neurological symptoms remains controversial (16,17).
Several limitations associated with the present study warrant mention. First, even though anticancer treatment was performed, there was no pathological evidence of lung cancer. There might have been other differential diagnoses, such as non-SCLC, carcinoid, large-cell neuroendocrine carcinoma, and infectious diseases. Second, as this patient had already presented with neurological symptoms at a previous hospital due to hemorrhaging of cavernous malformation, the discrimination of neurological symptoms of irAEs should have been carefully deliberated. However, the intracranial hemorrhagic symptoms had completely disappeared following treatment with concentrated glycerin and fructose before systemic chemotherapy, and sensory disturbances due to irAEs appeared four weeks after chemotherapy; the disturbances were consistent with those of PNS (Figure). Therefore, we believe that the etiology of the two neurological symptoms are clearly distinguishable. Third, the patient was treated with atezolizumab in combination with carboplatin and etoposide; therefore, the specific agent causing PNS was clinically unclear. However, both ICIs and cytotoxic agents have been reported to cause PNS, suggesting that combination therapy with ICIs and cytotoxic agents may increase the incidence of PNS.
Conclusion
In conclusion, we herein report an anti-Hu and anti-SOX-1 antibody dual-positive patient with suspected SCLC induced by combination treatment with carboplatin, etoposide, and atezolizumab. This irAE is often intractable, so physicians should be aware of this side effect, especially when treating patients with anti-neuronal antibodies using the combination of an ICI and cytotoxic agents.
The authors state that they have no Conflict of Interest (COI). | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY-NC-ND | 33328400 | 18,842,917 | 2021-05-15 |
What was the outcome of reaction 'Peripheral sensory neuropathy'? | Paraneoplastic Sensory Polyneuropathy Related to Anti-PD-L1-including Anticancer Treatment in a Patient with Lung Cancer.
Paraneoplastic neurological syndromes (PNS), such as sensory polyneuropathy, are rare, and serum neuronal antibodies that are used for diagnosing this syndrome are occasionally positive. Similarly, neurological immune-related adverse events due to immune checkpoint inhibitors (ICIs) are also rare. However, their etiologies and the relationship between them remain unclear. We herein report a patient with suspected small cell lung cancer who showed sensory polyneuropathy after treatment with atezolizumab in combination with cytotoxic agents (carboplatin and etoposide) and was doubly positive for serum anti-Hu and anti-SOX-1 antibodies. Treatment with ICI and cytotoxic agents may sometimes lead to the development of PNS.
Introduction
Paraneoplastic neurological syndromes (PNS) are defined as remote effects of cancer unrelated to direct effects of tumors or metastases and are exceedingly rare, affecting less than 1/10,000 cancer patients (1). Neuronal anti-Hu and anti-SOX-1 antibodies are sometimes detected in patients with PNS in the serum or cerebrospinal fluid, and the clinical signs are different from the main symptoms of cancer.
In general, neurological symptoms tend to progress subacutely and cause severe physical dysfunction. In about 80% of patients, the onset of neurological symptoms and the detection of antibodies precedes the detection of cancer by months to years; therefore, early antibody detection may be useful for predicting PNS.
Immune checkpoint inhibitors (ICIs) have widely been used in the treatment of multiple types of cancer. The main side effects of ICIs are immunological; thyroid function disorders are common, and organ-related side effects are also prevalent (2). Among these side effects, neurological immune-related adverse events (irAEs) due to ICIs are rare; only 0.93% of patients with ICI treatment experience neurological side effects, such as neuropathy, noninfective meningitis, encephalitis, and neuromuscular disorders (3).
Among neuronal autoantibodies related to PNS, anti-Hu antibody recognizes the nucleus of all central nervous system neurons and causes limbic encephalitis (LE), subacute cerebellar degeneration, and polyneuropathy. PNS with anti-Hu antibody was commonly seen in 78%, of small-cell lung cancer (SCLC) patients; 73% showed signs and symptoms of multifocal involvement of the nervous system, while 74% had sensory neuronopathy (2). In addition to anti-Hu antibody, anti-SOX-1 antibody recognizes the nuclear protein of glial cells and causes axonal neuropathy, demyelinating neuropathy, sensory and motor disorders (4). Five of 55 (9.1%) anti-SOX-1 antibody-positive patients reportedly had coexisting anti-Hu antibodies in SCLC patients (5).
Sensory polyneuropathy has clinically been recognized due to the advent of ICIs (3); however, neurological adverse events (nAEs) of ICIs, such as sensory polyneuropathy, are considered PNS and are still rare in patients with lung cancer. We herein report the first SCLC patient with sensory polyneuropathy who tested positive for anti-neuronal antibodies after initial treatment with anti programmed death-ligand 1 (PD-L1) antibody combined with cytotoxic agents.
Case Report
A 70-year-old current-smoking (80 pack-years) Japanese man with a history of hypertension and nonrecurrent colon cancer was admitted to our hospital. Two months earlier, he had experienced dizziness and been diagnosed with brain stem hemorrhaging due to cavernous malformation (hemangioma). In addition, multiple nodules suspected of being brain metastases had been seen on enhanced magnetic resonance imaging (MRI). He has been treated with concentrated glycerin and fructose at a previous hospital, and his symptoms had completely disappeared.
Upon admission to our hospital, a physical examination revealed the following: height, 171.5 cm; body weight, 74.1 kg; body temperature, 35.9℃; heart rate, 103 bpm; blood pressure, 107/71 mmHg; and oxygen saturation, 95% (room air, at rest); no neurological abnormalities were observed. Laboratory results (Table 1) showed a high titer of pro-gastrin-releasing peptide (pro-GRP; 541.5 pg/mL) and a low titer of carcinoembryonic antigen (CEA; 2.5 ng/mL) and cytokeratin fragment 21-1 (CYFRA21-1; 1.7 ng/mL). Chest radiography and high-resolution computed tomography showed two nodules in the right lower lobe with hilar lymphadenopathies. Sputum samples for a cytological examination were not available. Invasive pathological examinations, including bronchoscopy, could not be carried out because the patient was at a high risk of recurrent intracranial hemorrhaging due to cavernous hemangioma. The serum pro-GRP titer has a sensitivity of 86.4%, specificity of 96.4%, positive predictive value of 96.7%, and negative predictive value of 84.4% for discriminating SCLC from non-SCLC (cut-off 77.8 pg/mL) (6), and a titer of over 329.3 pg/mL pro-GRP also suggests extensive SCLC rather than limited disease (6). The high serum level of pro-GRP (541.5 pg/mL) in this patient was thus considered a marker of SCLC. Consequently, he was clinically diagnosed with stage IV SCLC based on the radiological and serological findings.
Table 1. Laboratory Data on Admission.
<Blood cell counts> T-bil 0.8 mg/dL ANA (-)
WBC 5,100 /μL AST 28 IU/L Anti-SS-A (-)
Neutrophils 72.2 % ALT 37 IU/L Anti-SS-B (-)
Lymphocytes 21.1 % LDH 224 IU/L vit.B1 41.9 ng/mL
Eosinophils 0.6 % γ-GTP 49 IU/L vit.B12 539 pg/mL
Monocytes 5.5 % BUN 11.3 mg/dL CEA 2.5 ng/mL
Basophils 0.6 % Cre 0.88 mg/dL CYFRA21-1 1.7 ng/mL
RBC 5.36×106 /μL Na 141 mEq/L Pro-GRP 541.5 pg/mL
Hb 17.1 g/dL K 4.1 mEq/L <Cerebrospinal fluid>
Ht 49.1 % Cl 107 mEq/L Cell 1 /μL
Platelets 18.3×104 /μL Glucose 156 mg/dL Mono 1 /μL
<Blood chemistry> HbA1c 6.3 % Protein 146 mg/dL
TP 7.9 g/dL CRP 0.02 mg/dL Glucose 57 mg/dL
Alb 4.6 g/dL RF <0.1 IU/mL Cytology No malignancy
WBC: white blood cell, RBC: red blood cell, Hb: hemoglobin, Ht: hematocrit, TP: total protein, Alb: albumin, T-bil: total bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, γ-GTP: gamma-glutamyl transferase, BUN: blood urea nitrogen, Cre: creatinine, CRP: c-reactive protein, RF: rheumatoid factor, ANA: anti nuclear antibody, Anti-SS-A: anti Sjögren syndrome-A antibody, Anti-SS-B: anti Sjögren syndrome-B antibody, vit.B1: vitamin B1, vit.B12: vitamin B12, CEA: carcinoembryonic antigen, CYFRA: cytokeratin fragment, Pro-GRP: pro-gastrin-releasing peptide
As the first-line treatment, combination therapy with carboplatin [area under the concentration-time curve (AUC) 5], etoposide (100 mg/m2), and atezolizumab (1,200 mg) was administered (Figure). Three weeks after the chemotherapy, whole-brain irradiation (3×10 Gy) was performed. Four weeks after chemotherapy, tactile and pain disturbances consistent with L2-3 dermatome, loss of bilateral patellar tendon reflexes, and loss of bilateral Achilles tendon reflexes were revealed. No obvious weakness was observed in the manual muscle strength test, Babinski reflexes were negative, and lower limb Barre's sign was positive. Coordination was poor in the finger-nose-finger test and knee-heel test, and gait was oscillating. Lumbar puncture results (Table 1) showed protein cell dissociation and no malignant cells. Enhanced brain MRI showed no evidence of recurrent brain stem hemorrhaging, and the size of all metastatic brain tumors had decreased. Lumbar spinal contrast-enhanced MRI showed no abnormal contrast enhancement of peripheral nerves, and nerve root compression consistent with the symptoms was observed. Serum anti-Hu and anti-SOX-1 antibody tests were both positive, and tests for onconeural antibodies (including anti-CV2, anti-Yo, anti-Ri, anti-amphiphysin, anti-paraneoplastic antigen MA2, anti-recoverin, anti-titin, anti-ZIC4, anti-GAD65 and anti-Tr) were all negative. Nerve conduction testing revealed that amplitude of sensory nerves was not evoked, and no apparent abnormalities in motor nerves were observed (Table 2).
Figure. Clinical course of the patient. IVIg: intravenous immunoglobulin therapy, Pro-GRP: pro-gastrin-releasing peptide
Table 2. The Nerve Conduction Study.
MCS Distal latency
(msec) CMAP amplitude
(mV) MCV
(m/s) SNAP
(μV) SCV
(m/s)
Rt. Median 3.4 18.9 50.9 18.1 50.9
Rt. Ulnar 3 15.4 46.8 12.5 47
Rt. Peroneal 5.4 13.8 42 - -
Rt. Tibial 6.1 18.1 38.3 - -
Rt. Sural 3.8 - - not evoked not evoked
MCS: motor conduction study, SCS: sensory conduction study, CAMP: compound muscle action potential, SNAP: sensory nerve action potential, MCV: motor conduction velocity, SCV: sensory conduction velocity, Rt: right
Based on the neurological findings and results of the nerve conduction test, the patient was diagnosed with sensory polyneuropathy. After systemic chemotherapy with atezolizumab and radiotherapy to the brain, the chest and brain tumors had shrunk, and the elevated serum level of pro-GRP had drastically decreased from 541.5 to 43.9 pg/mL. The clinical timeline of the onset of neurological symptoms was indicative of chemotherapy, particularly atezolizumab-induced. Intravenous immunoglobulin (IVIg) therapy was administered after the first round of systemic chemotherapy, but his neurological symptoms did not improve.
Discussion
Our patient with suspected SCLC who experienced an nAE of sensory polyneuropathy tested positive for anti-neuronal antibodies after initial treatment with atezolizumab in combination with carboplatin and etoposide.
Anti-acetylcholine receptor antibodies are generally pathogenic (e.g., in myasthenia gravis), but not all lung cancer patients positive for anti-Hu antibodies show neurological symptoms; indeed, among 196 SCLC patients with anti-Hu antibodies, 31 patients (16%) had no neurological symptoms (7). In relation to the treatment with ICIs and PNS with anti-Hu antibody, nivolumab (anti-PD-1 antibody)-induced sensory neuropathy (8), nivolumab-induced LE (9), sintilimab-induced LE, and enteric neuropathy (10) have been reported.
Regarding the etiology of PNS, the decreased expression of multiple Treg-related genes involved in immune regulation in SCLC patients might cause impaired immune tolerance, tissue damage due to autoimmune mechanisms, and PNS (11). PD-L1-coated beads can induce Tregs in vitro, and PD-L1 increases Foxp3 expression and enhances the immunosuppressive ability of Tregs (12), suggesting that anti-PD-L1 antibody treatment may suppress Treg infiltration into tumors. Hence, the present patient was positive for anti-neuronal antibodies, but the neurological symptoms manifested after chemotherapy, suggesting that the PNS was evoked by an ICI, such as atezolizumab, and cytotoxic agents.
In the management of PNS, underlying disease treatments, such as anticancer therapy, are prioritized, which may partly improve neurological symptoms; however, only 10-20% of patients achieve improvement in their neurological symptoms (13). For patients in whom neurological symptoms persist, plasma exchange, systemic corticosteroids, immunosuppressants, or IVIg therapy are considered but usually prove to be ineffective (14). Thus far, there have been no available data concerning randomized controlled trials for the treatment of PNS; the available data have only been collected from case series, case reports, or expert opinions (class IV evidence) regarding the effect of immunomodulation (IVIg, steroid treatment, plasma exchange, or chemotherapy) on paraneoplastic neuropathy (15). In our case study, although the patient was treated with a high-dose immunoglobulin, no neurological improvement was noted. In addition, the relationship between the presence of anti-neurological antibodies and the prognosis of neurological symptoms remains controversial (16,17).
Several limitations associated with the present study warrant mention. First, even though anticancer treatment was performed, there was no pathological evidence of lung cancer. There might have been other differential diagnoses, such as non-SCLC, carcinoid, large-cell neuroendocrine carcinoma, and infectious diseases. Second, as this patient had already presented with neurological symptoms at a previous hospital due to hemorrhaging of cavernous malformation, the discrimination of neurological symptoms of irAEs should have been carefully deliberated. However, the intracranial hemorrhagic symptoms had completely disappeared following treatment with concentrated glycerin and fructose before systemic chemotherapy, and sensory disturbances due to irAEs appeared four weeks after chemotherapy; the disturbances were consistent with those of PNS (Figure). Therefore, we believe that the etiology of the two neurological symptoms are clearly distinguishable. Third, the patient was treated with atezolizumab in combination with carboplatin and etoposide; therefore, the specific agent causing PNS was clinically unclear. However, both ICIs and cytotoxic agents have been reported to cause PNS, suggesting that combination therapy with ICIs and cytotoxic agents may increase the incidence of PNS.
Conclusion
In conclusion, we herein report an anti-Hu and anti-SOX-1 antibody dual-positive patient with suspected SCLC induced by combination treatment with carboplatin, etoposide, and atezolizumab. This irAE is often intractable, so physicians should be aware of this side effect, especially when treating patients with anti-neuronal antibodies using the combination of an ICI and cytotoxic agents.
The authors state that they have no Conflict of Interest (COI). | Not recovered | ReactionOutcome | CC BY-NC-ND | 33328400 | 18,842,917 | 2021-05-15 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Organising pneumonia'. | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | AMIODARONE, APIXABAN, ATORVASTATIN, BISOPROLOL, FUROSEMIDE, PANTOPRAZOLE | DrugsGivenReaction | CC BY | 33331164 | 18,799,489 | 2021-09 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pleural effusion'. | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | AMIODARONE, APIXABAN, ATORVASTATIN, BISOPROLOL, FUROSEMIDE, PANTOPRAZOLE | DrugsGivenReaction | CC BY | 33331164 | 18,799,489 | 2021-09 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pulmonary toxicity'. | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | AMIODARONE, APIXABAN, ATORVASTATIN, BISOPROLOL, FUROSEMIDE, PANTOPRAZOLE | DrugsGivenReaction | CC BY | 33331164 | 18,806,921 | 2021-09 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Respiratory failure'. | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | AMIODARONE, APIXABAN, ATORVASTATIN, BISOPROLOL, FUROSEMIDE, PANTOPRAZOLE | DrugsGivenReaction | CC BY | 33331164 | 18,799,489 | 2021-09 |
What is the weight of the patient? | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | 85 kg. | Weight | CC BY | 33331164 | 18,806,921 | 2021-09 |
What was the outcome of reaction 'Pleural effusion'? | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | Recovered | ReactionOutcome | CC BY | 33331164 | 18,799,489 | 2021-09 |
What was the outcome of reaction 'Pulmonary toxicity'? | A nontrivial differential diagnosis in COVID-19 pandemic: a case report and literary review of amiodarone-induced interstitial pneumonia.
Amiodarone is a drug commonly used to treat and prevent cardiac arrhythmias, but it is often associated with several adverse effects, the most serious of which is pulmonary toxicity. A 79-year-old man presented with respiratory failure due to interstitial pneumonia during the COVID-19 pandemic. The viral etiology was nevertheless excluded by repeated nasopharyngeal swabs and serological tests and the final diagnosis was amiodarone-induced organizing pneumonia. The clinical and computed tomography findings improved after amiodarone interruption and steroid therapy. Even during a pandemic, differential diagnosis should always be considered and pulmonary toxicity has to be taken into account in any patient taking amiodarone and who has new respiratory symptoms.
Amiodarone is a bi-iodinated benzofuran derivative class III antiarrhythmic agent (according to Vaughan–Williams classification) [1] used to treat and prevent several cardiac arrhythmias, both supraventricular and ventricular. Amiodarone and its main metabolite mono-N-des-etil-amiodarone have a long half-life (55–60 days) and high lipid solubility, thus accumulating largely in adipose tissue and highly perfused organs, such as liver, lungs and spleen [2–5]. Amiodarone is a very common use drug, but it is frequently associated with several adverse effects, including bradycardia or atrioventricular (AV) blocks, hypothyroidism or hyperthyroidism, blue–grey skin discoloration and photosensitivity, elevated liver enzymes (ALT or AST higher than two-times normal values), corneal microdeposits, anorexia and nausea. Opthalmological evaluation, a yearly ECG and semi-annually thyroid and liver profiles are therefore useful in follow-up. However, the most serious adverse effect is amiodarone pulmonary toxicity (APT) [6], a potentially limiting factor for its use, frequently misdiagnosed, which ranges from acute/subacute interstitial pneumonias, organizing pneumonia (OP), acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage, pulmonary nodules/masses and pleural effusion. An accurate differential diagnosis is therefore mandatory. The incidence of APT is 4–17% [7] and risk factors include dosage and duration of therapy (even if a real‘threshold’ does not exist), increased patient age (threefold for every 10 years in patients over 60 years), male sex, preexisting lung disease, underling pathologies, oxygen administration and invasive or surgical procedures, primarily thoracic ones [8–12]. angiotensin converting enzyme inhibitors-inhibitors and angiotensin receptor blockers seem to be associated with a lower incidence of APT: they increase isoform 2 of ACE expression and activity, which degrades Angiotensin II to Ang1–7, hence diminishing Angiotensin II receptor 1-mediated deleterious effects of enhancing amiodarone-induced apoptosis of alveolar epithelial cell, that in turn plays a central role in the development of acute lung injury [13–15].
Case presentation
We present the case of a 79-year-old man suffering from chronic HF with reduced ejection fraction in postischemic dilated cardiomyopathy, previously implanted with implantable cardioverter-defibrillator in secondary prevention, affected by paroxysmal atrial fibrillation and ascending aortic aneurysm (55 mm), with nonrelevant previous pulmonary history, never smoker, without occupational exposure. Dyspnea, dry cough and signs of respiratory failure without fever appeared at the end of February 2020 and he was hospitalized at the beginning of March 2020.
The patient’s home therapy was pantoprazole 40 mg daily, atorvastatin 20 mg daily, amiodarone 200 mg daily, bisoprolole 3.75 mg, furosemide 25 mg twice a day and apixaban 2.5 mg twice a day (eGFR 38 ml/min) at admission. The initial laboratory examination revealed a normal white blood cells (WBC) count (6.74 × 109/l) with a normal neutrophilic and lymphocyte ratio and increased creatinine value (2.16 mg/dl). A first chest high-resolution computed tomography (HRCT) scan (Figure 1A & B) documented vast areas of bilateral parenchymal consolidation and ground glass opacities (GGO) in the upper lung lobes (Figure 1A), with prevalent perihilar distribution in the lower lobes with air bronchiologram (Figure 1B). These findings were compatible with interstitial pneumonia, in particular OP. CT also showed enlargement of mediastinal lymph nodes (paratracheal and precarenal ones) and pleural effusion, mostly on the left.
Figure 1. High-resolution computed tomography of a 79-year-old man with amiodarone induced organizing pneumonia.
Extended multifocal parenchymal thickening at the (A) apical and (B) lower lobes, bilaterally, with vast ground glass areas and pseudonodular parenchymal consolidations. Progressive resolution of organizing pneumonia after 2 months (C & D) and after 3 months of steroid therapy and drug interruption (E & F), with persisting ‘ground glass’ areas associated with fibrotic-cicatricial manifestations, such as retractions of costal pleural sheets, mostly in lower lobes.
The differential diagnosis was challenging, and it included: coronavirus disease 2019 (COVID-19) pneumonia; cardiogenic pulmonary oedema; viral, bacterial and autoimmune pneumonia; APT. In the high suspicion of COVID-19-related pneumonia, two nasopharyngeal swabs for SARS-CoV-2 were performed (at admission and 48 h later) which resulted negative. The occurrence of two consecutive false negative results was considered highly unlikely and, moreover, serological tests for SARS-CoV-2 1 month after discharge were also negative for both IgM and IgG, confirming the exclusion of COVID etiology. There were no clinical and instrumental signs of acute heart failure (HF). Peripheral edema or ascites were absent. NT-proBNP plasmatic concentration was not elevated in comparison with patient’s baseline value. Transthoracic echocardiogram confirmed postischemic dilated cardiomyopathy with a reduced ejection fraction (32%), unchanged from the previous control. Moreover, HRCT findings were not typical of HF (see ‘discussion’). Therefore, a cardiogenic pulmonary edema was excluded. To rule out other causes of interstitial pneumonia associated with respiratory failure, a large number of laboratory tests were performed, such as plasma level of beta-D-glucan, anti-ENA SSB/La, SSA/Ro, Sm, RNP antibodies, viral serologies and bacterial research in sputum culture. They all resulted negative.
HRCT findings and the exclusion of alternative diagnosis therefore raised the suspicion of amiodarone induced OP. Amiodarone was in fact started 8 months prior to hospital admission with intravenous load, followed by oral administration of 200 mg three-times a day, gradually deescalated to a dosage of 200 mg daily after 8 weeks. Amiodarone was therefore immediately suspended and steroid therapy (prednisone 40 mg/day) was started, with clinical improvement. The CT scans at follow-up in May (Figure 1C & D) and June (Figure 1E & F) showed an absorption stage with a partial resolution of OP characterized by progressive reduction of the parenchymal consolidations of the upper lobes, with persisting ‘ground glass’ areas, and with slight signs of retraction on the pleural sheets and bronchovascular structures. Pleural effusion was absent bilaterally. Signs and symptoms of respiratory insufficiency further improved.
Discussion
Amiodarone is a drug largely used by cardiologists for its efficacy in preventing and treating supraventricular and ventricular arrhythmias. Nevertheless, it is associated with a variety of side effects, including pulmonary toxicity. There are two different categories of pulmonary involvement following amiodarone assumption: asymptomatic lipid pneumonia and APT. In turn APT can be caused by two possible mechanisms: a direct cytotoxic effect or an immuno-mediated mechanism, supported by immunologic markers in the blood stream and lungs of patients and CD8+ lymphocytosis in bronchoalveolar lavage (BAL), with imbalance between T-helper type I and II subpopulations and cytokines [16,17]. APT is less common than thyroid, eye and skin toxicity, but it is the most dangerous one because it may occur as a subacute/chronic onsetting alveolar or interstitial pneumonia with vary degrees of fibrosis, as well as an acute respiratory distress with severe hypoxemia [18]. High cumulative dose and duration of therapy exceeding 2 months, together with pre-existing lung disease, are important risk factors of APT. It affects about 6% of patients receiving a daily dose of 400 mg (or more) over 2 or more months, with a mortality rate of 10–20% [19]. It is characterized by insidious onset of non-productive cough and/or progressive dyspnea on exertion, usually within 6–12 months from starting amiodarone, but it can occur at any time after the treatment is initiated [20]. Low-grade fever or pleuritic chest pain are rarely present. The worst manifestation of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and a typical panel of ARDS. In 5–7% of patients amiodarone pneumonitis is followed by amiodaron-induced pulmonary fibrosis, irreversible and with a poor prognosis. Alveolar hemorrhage and hemoptysis are possible, but unusual [21]. On laboratory data, leukocytosis is often present, rarely due to eosinophilia [22], there could be also a nonspecific elevation of lactic dehydrogenase or serum IL-6, a mucin like glycoprotein expressed on type II pneumocytes and bronchiolar cells. Pulmonary function tests usually show a restrictive syndrome with decreased forced vital and total lung capacities and a reduction in diffusing capacity of the lungs for carbon monoxide more than 15–20% [23]. Pulmonary imaging is essential for the diagnosis and it is characterized by the presence on HRCT of extensive and severe bilateral patchy GGO with honeycombing, localized or diffuse, mono or bilateral, parenchymal (interstitial or alveolar) infiltrates, high attenuation consolidations, also called ‘amiodaronoma’, especially in the right upper lobe [24]. High attenuation, associated with the iodinated properties of the drug, may also appear in the liver and spleen and evidences suggest that lesions >70 Hounsfield Units (HU) may be related to amiodaron toxicity [25]. On microscopic inspection on BAL or transbronchial biopsy, a characteristic finding is the presence of lipid-laden foamy macrophages in alveolar spaces, even if not specific because they are also present in nontoxic patients receiving amiodarone, usually associated to hyperplasia of type II pneumocytes and widening of alveolar septae with a cellular inflammatory infiltrate and varying degrees of interstitial fibrosis [26]. Open lung biopsy should be avoided because APT may worsen after thoracic surgery. Exclusion diagnosis include lung viral or bacterial infection, HF, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma. The disease usually responds to drug discontinuation and corticosteroid administration within a period of 1–6 months. Recurrences are more frequent with steroid tapering in patients with excess adipose tissue [27].
APT is therefore prevalently an exclusion diagnosis, so that a number of alternative conditions have always to be taken into account before considering it.
COVID-19 pneumonia
In the COVID-19 era, the presence of dyspnea and respiratory impairment in patients with interstitial involvement on chest imaging makes it mandatory to suspect a COVID-19 pneumonia, that represents the most common manifestation of the disease. The most diffuse clinical manifestations are fever (>38°C in most cases), dyspnea, dry cough or expectoration with or without rhinorrhea, hypo-anosmia and/or ageusia, fatigue, headache, diarrhea and myalgia up to more severe conditions such as ARDS and respiratory failure, which sometimes require advanced respiratory assistance [28–30]. COVID-19 first affects the terminal bronchioles and surrounding parenchyma, and then develops into infiltration of pulmonary lobules and lastly diffuse alveolar damage [31]. Evidences also suggest a predisposition to thrombotic and thromboembolic disease in these patients [32,33], that were excluded in our case. Lab values in most of COVID-19 patients show normal or low WBC count, elevated neutrophil ratio, serum C-reactive protein, procalcitonin and lactic dehydrogenase and decreased lymphocyte ratio and lymphocyte count. The standard diagnosis of COVID-19 infection requires the identification of viral RNA by the real-time reverse transcriptase PCR essay of respiratory secretions obtained by nasopharyngeal and/or oropharyngeal swab, BAL or tracheal aspirate, with a sensitivity of 32–71% [34–36]. In case of negative result and if persisting high clinical suspicion of COVID-19, it is advised to perform chest CT, that has high sensitivity (75–94%) despite a reduced specificity [37] and shows GGO with bilateral (most) peripheral involvement in multiple lobes progressing to crazy paving pattern, fine reticular opacity and vascular thickening inside the lesions [30,38]. These radiological findings usually present with bilateral and multilobar distribution and a predominant involvement of subpleural/peripheral and posterior lung parenchyma [39], particularly in the lower lobes [40]. Several days after the onset of disease, in most patients linear consolidations and areas of GGO surrounded by peripheral consolidation (reverse halo sign) appear, suggesting OP. Uncommon HCRT features are multifocal nodular appearance with irregular margins, enlargement of mediastinal lymph nodes, pleural effusion and bronchial wall thickening, related to severe disease [38,41]. The most serious pathological pattern of the pulmonary damage caused by SARS-CoV-2 is a condition of acute lung injury, with a wide spectrum of histological pattern ranging from diffuse alveolar damage with hyaline membrane formation to OP [42–44].
Alternative diagnosis
Cardiogenic pulmonary edema is a very common cause of diffuse GGO on HRCT. Typical HRCT features in these patients are the enlargement of the pulmonary veins and smooth thickening of the interlobular septa and peribronchovascular bundles, that were absent in our patient. Besides, HF usually presents a central predominance with sparing of the peripheral portions of the lungs [45], that instead were involved in the presented case. At last, in cardiogenic pulmonary edema the lung lesions can be significantly improved after effective anti-HF treatment. Table 1 summarizes clinical features, radiological findings and laboratory characteristics of APT, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis. Viral, bacterial and autoimmune pneumonias can be usually easily ruled out with laboratory tests including WBC count, specific antibodies search, beta-D-glucan, viral serologies (and in some cases search for the viral genome in blood samples) and bacterial search in sputum culture.
Table 1. Clinical features, radiological findings and laboratory characteristics of amiodarone pulmonary toxicity, COVID-19 pneumonia and cardiogenic pulmonary edema to guide differential diagnosis.
Amiodarone pulmonary toxicity COVID-19 pneumonia Cardiogenic pulmonary edema
Clinical features Dyspnea
Anorexia
Dry cough
Hypoxemia Fever
Cough
Anosmia and ageusia
Shortness of breath
Diarrhea and myalgia Orthopnea
Extreme shortness of breath
Wheezing or gasping for breath
Wheezing
Swelling in lower extremities
Radiological findings GGO
Peripheral involvement (mainly)
Progressive fibrosis GGO
Bilateral peripheral involvement
Fine reticular opacity and vascular thickening inside the lesions
Progressive acute lung injury (hyaline membrane formation up to OP) GGO
Central predominance with sparing of the peripheral portions of the lungs
Peribronchovascular bundles
Pleural effusion
Laboratory/instrumental characteristics Leukocytosis (rarely due to eosinophilia)
Restrictive pattern on spirometry Lymphopenia
Elevated inflammatory markers (CRP; IL-6)
Elevated LDH and D-dimer NT-proBNP elevation
EF reduction on echocardiogram
COVID-19: Coronavirus disease 2019; CRP: C reactive protein; EF: Ejection fraction; GGO: Ground glass opacities; LDH: Lactic dehydrogenase.
Conclusion
Pulmonary toxicity is relatively frequent and occurs in 2–18% of patients receiving amiodarone, usually during long-term therapy with high cumulative doses, and can lead up to lung fibrosis and fatal respiratory failure [6]. The most common CT findings include septal thickening and interstitial pneumonia which can result in OP. The differential diagnosis of APT is mandatory, but can however be challenging, especially in COVID-19 era when interstitial pneumonias are easily attributed to SARS-CoV-2 infection. A temporal relationship of amiodarone intake for months or years could therefore be a key point in the differential diagnosis, as well as the negativity of swabs and serology for viral infection. Moreover, clinical and radiological presentation of interstitial pneumonia can also be similar to those of HF: pulmonary edema causes shortness of breath and CT ground-glass opacities and thickening of interlobular septum, but with prevalent central distribution and higher expansion of small pulmonary veins. When amiodarone therapy is begun it is mandatory to perform a basal and yearly chest x-ray and pulmonary function tests, including a diffusing capacity of the lungs for carbon monoxide. In this context, it is important to investigate when symptoms began or any recent changes in therapy and APT should always be suspected in any patient taking amiodarone who has new or worsening symptoms and/or new infiltrates on chest x-ray or CT scan.
Summary points
The presence of dyspnea, other respiratory symptoms and ground glass opacities on chest high-resolution computed tomography requires a challenging differential diagnosis, especially in COVID-19 era.
The main adverse effects associated to amiodarone involve the lungs, thyroid, eye, liver and skin.
There are two kind of pulmonary involvement due to amiodarone assumption: an asymptomatic lipoid pneumonia and amiodarone pulmonary toxicity (APT), with an incidence that ranges from 4 to 17% and a mortality rate of 10–20%.
Risk factors for APT include dosage and duration of therapy, patient age, male sex, pre-existing lung disease, oxygen administration and invasive or surgical procedures.
The worst complication of APT is a rapidly progressing diffuse pneumonitis with acute respiratory failure and acute respiratory distress syndrome, followed in 5–7% of patients by pulmonary fibrosis, only partially reversible and with a poor prognosis.
On high-resolution computed tomography, APT is characterized by severe bilateral patchy ground glass opacities with honeycombing, localized or diffuse, mono or bilateral, parenchymal infiltrates, high attenuation consolidations, especially in the right upper lobe.
APT is a diagnosis of exclusion, after considering viral or bacterial pneumonias, cardiogenic pulmonary edema, exogenous lipid pneumonia, bronchoalveolar carcinoma and lymphoma.
Clinical status APT-patients usually improves after drug discontinuation and corticosteroid administration within a period of 1–6 months.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Informed consent disclosure
The authors state that they have obtained verbal and written informed consent from the patient/patients for the inclusion of their medical and treatment history within this case report. | Recovered | ReactionOutcome | CC BY | 33331164 | 18,806,921 | 2021-09 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective'. | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | AMPHOTERICIN B, EPINEPHRINE, HYDROXYCHLOROQUINE, LEVOFLOXACIN, LINEZOLID, LOPINAVIR\RITONAVIR, MEROPENEM, PIPERACILLIN SODIUM\TAZOBACTAM SODIUM | DrugsGivenReaction | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the administration route of drug 'AMPHOTERICIN B'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'EPINEPHRINE'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'HYDROXYCHLOROQUINE'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'LEVOFLOXACIN'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'LINEZOLID'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'LOPINAVIR\RITONAVIR'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'MEROPENEM'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
What was the dosage of drug 'PIPERACILLIN SODIUM\TAZOBACTAM SODIUM'? | A challenging complication following SARS-CoV-2 infection: a case of pulmonary mucormycosis.
Severe acute respiratory syndrome coronavirus 2 infection might induce a significant and sustained lymphopenia, increasing the risk of developing opportunistic infections. Mucormycosis is a rare but severe invasive fungal infection, mainly described in immunocompromised patients. The first case of a patient diagnosed with coronavirus disease (COVID-19) who developed a pulmonary mucormycosis with extensive cavitary lesions is here reported. This case highlights how this new coronavirus might impair the immune response, exposing patients to higher risk of developing opportunistic infections and leading to worse outcomes.
pmcIntroduction
Mucormycosis is a rare but severe invasive fungal infection occurring mainly in immunocompromised patients, especially in individuals diagnosed with uncontrolled diabetes mellitus or haematological malignancies, as well as in previously healthy subjects with open wounds contaminated by Mucorales [1–3]. Patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might develop coronavirus disease (COVID-19), which can be associated to significant and sustained lymphopenia compromising the immune system, especially in the most severe cases [4–6]. Some authors described that a significant decrease in lymphocyte count and an increase of neutrophil count together with an inflammatory storm, occur more frequently in patients who developed severe COVID-19 and co-infections [4]. This report describes the first case of a patient with SARS-CoV-2 infection who developed a cavitary pulmonary mucormycosis.
Case report
A 66-year-old male patient was admitted to ICU at the University Hospital, in Sassari, Italy, on March 26, 2020, with a diagnosis of SARS-CoV2 infection. Due to a rapid and progressive deterioration of oxygenation, the patient was intubated after a short period of non-invasive respiratory support. He had a history of arterial hypertension treated with ACE-inhibitors and had recently been diagnosed with urinary tract infection. The beginning of COVID-19 symptoms reportedly started one week before admission. A therapy with hydroxychloroquine and lopinavir-ritonavir was administered for the first 10 days. At ICU admission, the patient was deeply sedated, underwent protective mechanical ventilation, according to the new evidence described for such pulmonary damage phenotype, to avoid ventilator-induced lung injury [7, 8] (tidal volume = 6–7 ml kg−1 *PBW, positive end expiratory pressure (PEEP) = 12 cmH2O; PaO2/FiO2 = 262); he also required circulatory support with vasopressor (norepinephrine = 0.2 mcg kg−1 min). In addition, the patient had multiple organ dysfunction syndrome with sequential organ failure assessment (SOFA) [9] = 14. A few days after admission, respiratory parameters progressively worsened: a lower oxygenation (PaO2/FiO2 = 174), increase of radiological infiltrates and a parenchymal thickening of the left lower lobe were observed. Table 1 describes the main clinical and laboratory data, which include crucial biomarkers reported at ICU admission and during the whole hospitalization. The patient did not present with fever, the white blood count (WBC) and neutrophils increased while lymphocytes progressively decreased (lymphocytes [Nadir] = 400 μl−1; neutrophil/lymphocyte [N/L] ratio = 19.2). As C-reactive protein CRP (= 14.3 mg dl−1) and procalcitonin PCT (= 6.91 ng ml−1) values were elevated he was administered empirical antibiotic therapy (piperacillin-tazobactam 18 gr IV infusion, 24 h a day and levofloxacin 700 mg a day). As a result of a reduced kidney function, a renal replacement therapy (RRT) was required for several days. However, after 2 weeks of empirical therapy, neither were pathogens isolated on microbiological work-up of samples, nor an improvement of oxygenation (PaO2/FiO2 = 130) was observed. Inflammatory biomarkers showed higher values (CRP = 25 mg dl−1; PCT = 13.1 ng ml−1) and a further worsening of pulmonary infiltrates with an increase of parenchymal thickening of the whole left lung was observed by imaging techniques. Empirical therapy was replaced by meropenem 1 gr IV infusion every 24 h (dose-adjusted for renal dysfunction) and linezolid 600 mg every 12 h and a bronchial aspirate (BAS) was repeated to confirm SARS-CoV-2 and detect any additional co-infections. Two weeks after ICU admission a surgical tracheostomy was performed at bedside for prolonged mechanical ventilation. [10] A first BAS negative for SARS-CoV-2 was obtained, while the second BAS tested for co-infections showed rapidly growing cotton-candy like colonies on sabouraud dextrose agar (SDA) at 30 °C. Microscopic examination with lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia containing sporangiospores. The mould was identified as Rhizopus spp. (Fig. 1) based on the phenotype features, however no susceptibility test could be performed at our laboratory. A cranial and thoracic computed tomography (CT) scan was performed to search for specific lesions. Non-encephalic lesions were found, opacification of the left maxillary sinus and thickening with sclerosis of sinus walls were observed and the thoracic scans were suggestive of buried cavitary lesions in the lingula of the left lung upper lobe (Fig. 2). Treatment with liposomal Amphotericin B, 5 mg kg−1 IV was commenced, according to guidelines and after discussing with an infectious disease consultant [11, 12]. A second and a third BAS were consecutively positive for Rhizopus spp. A biopsy of the left maxillary sinus was performed to find the source of mould infection, but samples were positive only for Candida glabrata. A transbronchial biopsy was excluded because of severe hypoxia and high risk of airway bleeding. A probable pulmonary mucormycosis was then hypothesized. After 16 days of antifungal treatment, a slow improvement of gas exchange was noticed. Since the bronchoalveolar lavage (BAL) was still positive for Rhizopus spp., a surgical evaluation was requested, and thoracic CT scan was repeated. The scan revealed a rupture of the cavities previously observed in the pleural space and bilateral pleural effusion was observed. Therefore, a thoracentesis was performed at bedside. Samples of pleural effusion were tested for microbiological and histopathological examination, but neither moulds nor other fungi were isolated. At that stage, surgery was not performed, being regarded as high-risk intervention. Finally, 40 days after ICU admission and 20 days after the beginning of liposomal Amphotericin B treatment, even though Rhizopus spp growth was still observed in BAL samples, the patient clinically improved and recovered from lymphopenia (lymphocyte = 1,800 μl−1, N/L = 3.5). Since persistency of the positive culture on BAL for Rhizopus spp was still occurring, a surgical consultation was planned to eradicate the necrotic lesions from the left lung. At the same time, the antifungal treatment was shifted to Isavuconazole and the liposomal Amphotericin B treatment was suspended [13]. However, since following thoracentesis the oxygenation considerably improved (PaO2/FiO2 > 300), surgery was postponed and also the surgeon evaluated the patient as being at extremely high risk for surgery. Sedation was then suspended, and the patient began a ventilatory weaning process. Furthermore, an improvement of other organ functions was observed, such as in the kidneys, as the urine volume started to increase gradually. The following week, a new clinical deterioration was observed, showing fever, an increase of PCT, severe haemodynamic instability and worsening of kidney and liver functions, probably due to a bacterial co-infection. Although an antifungal treatment with Isavuconazole was maintained together with a prompt empirical antibiotic therapy, the patient died at day 62 after ICU admission due to refractory shock and liver failure. Unfortunately, the autopsy could not be performed due to the lack of a negative pressure room required to execute safe procedures.Table 1 Daily clinical and laboratory variables during ICU stay
Variables Reference range Day 1 admission Day 3 Day 7 Day 10 Day14 Day21 Day 28 Day 40
SOFA 13 16 16 16 17 16 15 13
White blood counts (per μl) (4800–10,800) 5,700 11,710 9990 8760 17,070 27,930 23,750 10,710
Lymphocytes(per μl) (900–5200) 1000 700 800 400 800 1,200 800 1800
Neutrophils (per μl) (1900–8000) 4,300 10,300 8,500 7700 12,600 23,600 21,100 6300
N/L 4.3 14.7 10.6 19.2 15.7 19.6 26.3 3.5
CRP (mg/dl) (0–1) 17.7 31.9 18.6 14.3 25 17.1 8.65 8.80
Procalcitonin(ng/ml) (0–0.5) 1.9 13 7.92 6.91 13.1 10 6.06 5.75
Glycemia (mg/dl) (60–99) 124 96 96 70 92 96 86 96
D-dimer (mg/l) (0–0.5) 1.9 4.3 0.8 1 2.6 2.7 6.1 7.64
Ferritin (ng/ml) (26–388) 3216 3755 2012 1395 2128 2737 2084 –
NT-proBNP (pg/ml) (0–125) 653 2147 2869 2801 3028 12,541 26,116 19,692
PaO2/FiO2 190 197 208 174 130 161 189 260
SARS-CoV-2 RT-PCR Positive Positive Positive Positive Negative Negative after 24 h Negative
BAS/BAL Rhizopus spp. Rhizopus spp. Rhizopus spp. Rhizopus spp.
Pleural effusion Histo-pathology Negative
Microbio-logy Negative
SOFA sequential organ failure assessment, N/L neutrophils/lymphocyte ratio, CRP C-reactive protein
Fig. 1 Microbiological sample from bronchial aspirate: morphology of Rizhopus spp.. Lactophenol cotton blue preparation showed aseptate broad hyphae, sporangia and sporangiospores
Fig. 2 a Thoracic computed tomography (CT) scan showed buried cavitary lesions in the left lung; b cranial CT scan showed corpuscular material in the left maxillary sinus
Discussion
SARS-CoV-2 infection might alter the immune system by affecting T lymphocytes, particularly CD4+ and CD8+ T cells, which might be highly involved in the pathological process of COVID-19 infection [4]. The significant reduction of the absolute number of lymphocytes and specifically of T cells described in the most severe COVID-19 cases, is associated with the worst outcome and might expose patients to a higher risk of developing opportunistic infections [3, 4]. Mucormycosis is a fungal infection caused by a group of opportunistic moulds, i.e., mucormycetes [2, 3]. This infection is generally caused by an impairment of bronchial alveolar macrophages, but a role of T-cells was described as part of the adaptive immune system. Potenza et al., in a brief report on a group of haematological patients who suffered from mucormycosis described Mucorales-specific T-cells (CD4+ and CD8+), that were active against Mucorales by producing cytokines, such as IL-4, IL-10, IL-17 and IFN-γ, which could directly damage Mucorales hyphae [14]. The authors observed Mucorales-specific T-cells only in patients affected by invasive mucormycosis and they concluded that they could be a useful surrogate diagnostic marker of an invasive fungal disease and they might contribute to control the invasive fungal infection by Mucorales. We might speculate that lymphopenia could increase the risk of developing an invasive mucormycosis, while the recovery of lymphocytes count could improve the adaptive immune system and induce the production of Mucorales-specific T-cells, which might have a role in controlling the invasive infection. Unfortunately no lymphocyte subset typing could be performed by our haematological laboratory due to the fact thatat the beginning of the outbreak safe procedures to manage samples from patients affected by COVID-19 had not been implemented yet.
Different organs might be involved, the most frequently affected of which are the lungs, being the second most common manifestation (58%) with a mortality rate up to 80% due to its aggressive clinical course [2].
The high mortality rate described in pulmonary localization might be related to delays in diagnosis, to an unbalanced immune system and a poor host response, as well as to the complexity of the treatment that includes a combination of antifungal therapy and a high-risk surgical intervention [3, 4, 15, 16]. In the present case, a surgical intervention was not performed because a clinical improvement was observed in lung function, and surgeons considered the patient at extremely high risk for surgery.
Generally, mucormycosis affects immunocompromised patients: as a matter of fact, a recent systematic review by Jeong et al. showed that solid organ transplantations and neutropenia, commonly reported in patients affected by haematological malignancies, were the only independent risk factors for pulmonary mucormycosis [17]. Furthermore, SARS-CoV-2 infection itself might trigger an alteration of the immune system [4] and this is the first reported case of opportunistic co-infection caused by Rhizopus spp involving lungs with an extensive parenchymal damage.
Recently in a retrospective study, Koehler et al. analysed a cohort of patients admitted to ICU due to COVID-19 showing moderate to severe acute respiratory distress syndrome (ARDS) who developed invasive pulmonary aspergillosis as a consequence of the immune-paralysis related to SARS-CoV-2 infection [18]. Similarly, in the present case neither corticosteroids nor immunosuppressant therapies were administered, but the patient showed a severe form of COVID-19 with multiple organ dysfunctions and a significant and sustained lymphopenia with N/L ratio alteration; the latter has been recently described to be highly associated with the most severe clinical presentation and the worst outcome [4]. Therefore, it might be suggested that SARS-CoV-2 infection by itself can induce an immunosuppressive state that exposes the patient to the risk of developing opportunistic infections, such as moulds. These kind of infections by themselves are associated with the worst outcome, especially when the immune system response does not improve. However, when the immune system recovers, opportunistic infections might be controlled [16], as the present case shows, when an improvement of its clinical symptoms, specifically of the respiratory dysfunction, was observed. As a matter of fact, the patient’s oxygenation began to improve when lymphocytes increased and N/L ratio decreased, and the pulmonary cavitary lesion opened into the pleural space. Although moulds could not be isolated in the pleural effusion, surgical exploration to remove necrotic lesions should be considered to eradicate the mould infection and improve the patient’s outcome [15].
The European Confederation of Medical Mycology Mucormycosis Guidelines strongly suggest an early surgical treatment to remove the infected tissue (either through local debridement or complete resection) in addition to systemic antifungal treatment [11].
Which is the lesson learnt from the present case? When no improvement of clinical symptoms is shown following a wide spectrum empirical treatment, and when an impaired immune response induced by SARS-CoV-2 is observed, opportunistic infections, such as the mould infection shown in the present case, should be investigated as a potential causative agent. As a result, the antimicrobial therapy should include antifungal agents and, when data suggest a mucormycosis, a pharmacological treatment should be associated whenever possible with a surgical intervention to eradicate mould-associated necrotic lesions.
In conclusion, SARS-CoV-2 infection is regarded as a cause of severe immunosuppression that might compromise the host response and increase the risk to develop opportunistic infections, including those caused by moulds, leading to higher risk of negative outcomes in the case of delayed diagnosis and inadequate treatment.
Acknowledgements
This work was supported by a Rapid Response award for COVID-19, Canadian Institutes of Health Research (SR).
Author contributions
Collection and analysis of clinical data: DP, CL, DP, GPB, LC, RS, AB, DR. Analysis of microbiological data: SS, writing the manuscript: DP, CL, SS. Revision and final approval of the manuscript: DP, FB, SR, PT.
Funding
Open access funding provided by Università degli Studi di Sassari within the CRUI-CARE Agreement.
Compliance with ethical standards
Conflicts of interest
All authors declare that they have no conflicts of interest.
Ethics approval
All procedures performed were in accordance with the declaration of the ethical standards of the institutional research committee and with the 1964 Helsinki 387 Declaration and its later amendments. To collect data on patients admitted to our ICU with diagnosis of COVID-19, we asked the approval to our local ethics committee “comitato etico indipendente”, which was obtained on 06/05/2020 (Prot. PG/2020/9721). | UNK UNK, UNKNOWN FREQ. | DrugDosageText | CC BY | 33331988 | 20,421,010 | 2021-10 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Bacteraemia'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Brain abscess'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Central nervous system infection'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Ear infection'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Klebsiella infection'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Loss of consciousness'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Mastoiditis'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Meningitis'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Otitis media acute'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pneumonia'. | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | AMIKACIN, MEROPENEM, METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
What was the outcome of reaction 'Klebsiella infection'? | Community-acquired Klebsiella pneumoniae central nervous system infection after acute suppurative otitis.
Community-acquired Klebsiella pneumoniae (K. pneumoniae) central nervous system (CNS) infection combined with bacteremia is rarely identified worldwide. We received a 55-year-old woman on long-term corticosteroid therapy for Sjogren's syndrome. Onset began with acute suppurative otitis, followed by a severe headache and loss of consciousness. Cerebrospinal fluid (CSF) testing and brain imaging examinations were compatible with K. pneumoniae meningitis and likely brain abscesses, respectively. K. pneumoniae bacteremia was also found on blood cultures. Despite aggressive antibiotic and supportive therapy, the patient died after 2 day's therapy. Corticosteroid therapy may be a risk factor for a community-acquired K. pneumoniae infection. Appropriate antibiotics and abscess drainage are still recommended, despite the poor prognosis.
Background
Klebsiella pneumoniae (K. pneumoniae), a gram-negative, aerobic, rod-shaped bacterium, is usually hospital-acquired and occurs primarily in patients with impaired immune defenses [1]. Community-acquired infections caused by K. pneumoniae have mainly been reported in cases of invasive liver abscess syndrome (ILAS), meningitis, or endophthalmitis in Taiwan [2]. Although sporadic cases can occur elsewhere, community-acquired K. pneumoniae central nervous system (CNS) infections without liver abscesses are rarely seen [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. We report our recent experience with an adult case of community-acquired K. pneumoniae CNS infection associated with acute suppurative otitis.
Case presentation
A 55-year-old woman from Hebei province in Mainland China with a two-year history of Sjogren's syndrome was taking oral methylprednisolone 24 mg/day. Eight days prior to arriving at our hospital, the patient noticed a purulent discharge from her right ear but did not seek any diagnosis and treatment. Over the next six days, she developed a fever and headache, then was found unconscious by her family on her way to our emergency department (ED). On examination in our ED, the patient’s vital signs were normal, but she had altered mental status with a Glasgow coma scale of 5 (E1+V1+M3). On physical exam, she evidenced nuchal rigidity and had a positive right-sided Babinski sign.
Initial laboratory findings were as follows: white blood cell (WBC) count of 10.92 × 109/L with an elevated neutrophil ratio of 89.0 %, hemoglobin of 11.1 g/dL and a platelet count of 302 × 109/L. The C-reactive protein was >160 mg/L, liver and renal function tests were normal. Procalcitonin was 4.5 ng/mL. Lumbar puncture yielded pale yellow, cloudy cerebral spinal fluid (CSF) with an opening pressure of >330mmH2O. CSF results were as follows: WBC count, 1.15 × 109/L, with a predominance of polymorphonuclear leukocytes; total protein, 2.48 g/L; and glucose, <0.11 mmol/L. Head and temporal bone CT scans revealed right mastoiditis, while an abdominal ultrasound and CT scan of the chest, abdomen and pelvis were normal. The patient was diagnosed as having purulent meningitis and right mastoiditis complicated by underlying Sjogren's syndrome and was treated with 2 g of cefatriaxone and 1 g of vancomycin every 12 h, as well as mannitol to help reduce intracranial pressure. The patient’s level of consciousness did not recover, and she was transferred to the ED intensive care unit (EICU) that same day.
On hospital day two, K. pneumoniae grew from both her CSF and blood cultures and was found to be sensitive to all tested antibiotics. Antibiotics were adjusted to meropenem 2 g every 8 h and amikacin 0.4 g every 12 h, while the dose of methylprednisolone was gradually reduced to 10 mg daily to help control infection. However, the patient’s condition worsened, and she was endotracheally intubated for airway protection.
Because of her continuous fever and altered mental status, a contrast-enhanced head CT scan and lumbar puncture were again performed on hospital day seven. The contrast-enhanced head CT demonstrated acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement (see Fig. 1: contrast-enhanced head CT). CSF pressure decreased to 220 mmH2O, and her CSF WBC count dropped to 0.16 × 109/L, with a CSF protein of 3.55 g/L. We next planned a contrast-enhanced head MRI and consult neurosurgery for potential operative options. However, the patient’s relatives refused further treatment due to her already poor prognosis. The patient was extubated, and an ambulance was arranged to take her back home. She died on her way home secondary to a lack of spontaneous breathing.Fig. 1 Acute, multi-layered, low-density masses in the patient’s bilateral frontal, parietal, occipital and right temporal lobes without abnormal enhancement.
Fig. 1
Discussion and conclusion
Cases of community-acquired K. pneumoniae meningitis are exceedingly rare. In mainland China, the only such patient reported so far had ILAS [12]. Reviewing the literature to date, this patient is the first one we have found with community-acquired K. pneumoniae CNS infection and bacteremia with no neurosurgical or ENT (ear, nose and throat) surgical history or implants or procedures. We now review the current literature about community-acquired K. pneumoniae meningitis and try to analyze the reasons of our patient’s demise.
A total of eight patients with community-acquired K. pneumoniae meningitis not associated with liver pathology have been reported in the Caribbean, as well as in Italy, Singapore, Taiwan, the United Kingdom, and the United States [3,4,9,[13], [14], [15]]. However, not all of them had brain abscesses or bacteremia. Two of them had uncontrolled diabetes and another two had chronic alcoholic diseases. In our case, long-term corticosteroid therapy may explain the patient’s increased risk for developing meningitis with K. pneumoniae.
The primary infection for our patient was likely suppurative otitis, which is similar to four other patients reported in the literature with preceding infections of endophthalmitis, otitis and sphenoid sinusitis [3,4,8,14]. However, K. pneumoniae was also cultured in our patient’s blood, and the low-density CT scan showed likely multiple brain abscesses. This is the first patient shown to have K. pneumoniae bacteremia associated with meningitis, since the direct spread of organisms from a contiguous site (such as sinusitis) usually causes a solitary brain abscess [16]. The multiple brain abscesses in this case may be from the hematogenous spread of bacteria instead of direct spread from the suppurative otitis. Common conditions leading to hematogenous seeding of the brain usually involve chronic pulmonary infections, skin infections, pelvic and intraabdominal infections, bacterial endocarditis, esophageal dilation or the endoscopic sclerosis of esophageal varices [16,17]. Although the pathogenesis of this case is still unclear, the CNS infection combined with bacteremia could be a risk factor for the patient’s poor prognosis.
Besides supportive treatment, the main therapy for patients with K. pneumoniae meningitis is timely antibiotics. The seven surviving patients with K. pneumoniae meningitis in the literature were treated with a third-generation cephalosporin combined with pefloxacin or gentamicin or amikacin, and only one case used meropenem [3,4,9,[13], [14], [15]] (sensitivity results were not provided in these cases, however). In our case, K. pneumoniae cultured from the CSF was susceptible to all of the remaining antibiotics tested, meropenem and amikacin included. Those two were chosen based on in vitro sensitivity, but, unfortunately, the ending was still tragic.
Drainage of the patient’s brain abscesses was another point of consideration. Once an abscess has formed, surgical excision or drainage remains the treatment of choice [16]. Only one patient in the literature infected with K. pneumoniae has had a brain abscess, but this patient still died two weeks after lateral and ventricular drainage [12]. Other potential treatments of K. pneumoniae CNS infections, such as intrathecal antibiotics or brain abscess resection were not mentioned. Three patients with endophthalmitis, sphenoid sinusitis or liver abscesses, respectively, survived after eliminating the focus of infection [4,14], but these foci were outside the CNS. Because our patient had an already quite poor prognosis, her relatives chose to give up further efforts such as those mentioned above.
We also suspect that the strain of K. pneumoniae in this case may have had special virulence factors. K. pneumoniae strains related to ILAS harbors capsular serotypes K1 or K2, which are more virulent than those with non-K1/K2 serotypes. Besides capsular serotypes, the hypermucoviscosity phenotype, lipopolysaccharide, siderophores, and pili also contribute to the pathogenesis of K. pneumoniae [1,2]. Unfortunately, the molecule gene of K. pneumoniae in this case was not tested, but future studies may attempt to analyze the genetic code of K. pneumoniae for virulence factors leading to community-acquired meningitis without ILAS.
In conclusion, a community-acquired K. pneumoniae CNS infection without ILAS is a rare disease. Corticosteroid therapy was likely a risk factor for infection in our case. Appropriate antibiotic use and abscess drainage should be attempted for treating this condition. Our patient’s death was likely a combination of K. pneumoniae bacteremia, presence of intracranial lesions and potential virulence factors for this strain of K. pneumoniae. This case is reported to help identify this rare disease and promote a further progress on understanding its pathogenesis and treatment.
Statement
The patient’s next to kin has given their consent to publish the patients’ cases for this study.
Ethics approval and consent to participate
The Institutional Review Board (IRB) of Peking Union Medical College Hospital has reviewed the study and has determined that this is a retrospective study and the design is scientifically and is up to the ethics standards. The IRB thus approve the study.
Consent for publication
Written consent for publication has been obtained from all the study participants and the patients reported in this article.
Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Funding
No funding
Referee suggestions
(1) Bianca Lee, Department of Internal Medicine, Nassau University Medical Center, 2201 Hempstead Turnpike, East Meadow, NY, 11554, USA. blee5@numc.edu
(2) Yi-Tsung Lin, Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Number 201, Section 2, Shih-Pai Road, Beitou District, Taipei 11217, Taiwan.ytlin8@vghtpe.gov.tw
(3) A. G. Habib · P. A. Tambyah, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore, Singapore e-mail: mdcpat@nus.edu.sg
Author contributions
All authors contributed to the study conception and design. Case collection and description were performed by Ruixue Sun and Jun Xu. The first draft of the manuscript was written by Ruixue Sun and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no conflicts of interests.
Acknowledgement
Not applicable | Fatal | ReactionOutcome | CC BY-NC-ND | 33335834 | 18,799,620 | 2021 |
What was the administration route of drug 'AZITHROMYCIN ANHYDROUS'? | Mitigating arrhythmia risk in Hydroxychloroquine and Azithromycin treated COVID-19 patients using arrhythmia risk management plan.
To assess cardiac safety in COVID-19 patients treated with the combination of Hydroxychloroquine and Azithromycin using arrhythmia risk management plan.
We retrospectively examined arrhythmia safety of treatment with Hydroxychloroquine and Azithromycin in the setting of pre-defined arrhythmia risk management plan. The data was analyzed using R statistical package version 4.0.0. A two-tailed p-value<0.05 was considered significant. 81 patients were included from March 23rd to May 10th 2020. The median age was 59 years, 58.0% were female. The majority of the study population (82.7%) had comorbidities, 98.8% had radiological signs of pneumonia. Fourteen patients (17.3%) experienced QTc ≥ 480 ms and 16 patients (19.8%) had an increase of QTc ≥ 60 ms. Seven patients (8.6%) had QTc prolongation of ≥ 500 ms. The treatment was discontinued in 4 patients (4.9%). None of the patients developed ventricular tachycardia. The risk factors significantly associated with QTc ≥ 500 ms were hypokalemia (p = 0.032) and use of diuretics during the treatment (p = 0.020). Three patients (3.7%) died, the cause of death was bacterial superinfection with septic shock in two patients, and disseminated intravascular coagulation with multiple organ failure in one patient. None of these deaths were associated with cardiac arrhythmias.
We recorded a low incidence of QTc prolongation ≥ 500 ms and no ventricular tachycardia events in COVID-19 patients treated with Hydroxychloroquine and Azithromycin using cardiac arrhythmia risk management plan.
1 Introduction
Since the beginning of COVID-19 spread in December 2019, numerous treatment options have been studied worldwide. Based upon limited clinical data in case series, Hydroxychloroquine was recommended for treatment of hospitalized COVID-19 patients in several countries, and a number of national guidelines reported incorporating recommendations regarding use of this drug in the setting of COVID-19. Furthermore, on 28th March 2020 Hydroxychloroquine was authorized for emergency use by the U.S. Food and Drug Administration (FDA) during the COVID-19 pandemic [1]. Additional administration of macrolide antibiotic Azithromycin, seemed to reinforce the efficacy of Hydroxychloroquine [2].
Hydroxychloroquine in combination with Azithromycin or alone was used for COVID-19 treatment worldwide for almost 3 months. However, multiple large studies showed that the use of Hydroxychloroquine, either monotherapy or in combination with Azithromycin, did not improve clinical status as compared to the standard care [3], [4]. Based on emerging scientific data, emergency use authorization for these drugs was revoked on 15th June 2020. Furthermore, Solidarity trial results showed that Hydroxychloroquine produced little or no reduction in the mortality of hospitalized COVID-19 patients when compared to standard of care and on 4th July 2020 World Health Organization (WHO) discontinued the trial’s Hydroxychloroquine arm [5].
Despite the recall of Hydroxychloroquine by FDA and WHO, as of 1st November 2020, according to U.S. National Library of Medicine and European Union Clinical Trials Register, there are at least 45 interventional clinical trials for COVID-19 treatment with Hydroxychloroquine and Azithromycin that are active around the world. The drugs are studied in the setting of confirmed COVID-19 infection as well as in proactive prophylaxis. Although FDA and European Medicines Agency had warned to take precautions when using Hydroxychloroquine and Azithromycin [6], [7], no definite cardiac safety protocols have been established yet. Here we report QT interval prolongation and arrhythmia safety results in COVID-19 patients treated with the combination of HCQ and AZI using close monitoring and arrhythmia risk management plan.
2 Methods
We retrospectively examined the cardiac safety of treatment using HCQ and AZI in consecutive adult patients with COVID-19 infection confirmed by qPCR treated in Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania. Every patient consented to the treatment plan by signing an informed consent form and the study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution's human research committee. The data was acquired from the electronic medical records accessed through Vilnius University Hospital Santaros Klinikos Biobank. The study was approved by the regional ethics committee.
HCQ-AZI consisted of 5 days of oral Azithromycin (once daily; initial dose 500 mg on the first day followed by 250 mg during the next 4 days) and 10 days of oral Hydroxychloroquine 200 mg three times daily.
The cardiac arrhythmia risk management plan was designed by a multidisciplinary team which included a cardiologist – arrhythmia specialist. It was applied during HCQ-AZI treatment for each patient and was as follows: 1) QT prolonging concomitant drugs (assessed using “CredibleMeds®” database [8]) were discontinued if possible; 2) ECG recording and QTcF (Fridericia) calculation was performed daily; 3) K+ and Mg2+ were replaced if abnormal (hypokalemia and hypomagnesemia were defined as K+ levels < 3.5 mmol/L and Mg2+ < 0.65 mmol/L in blood serum, respectively); 4) if QTcF reached 480 ms during HCQ-AZI treatment, K+ and Mg2+ were replaced to reach maximum normal values, 5) if QTcF remained in the interval of 480 – 499 ms regardless of K+ and Mg2+ replacement, the risk–benefit of continuing HCQ-AZI was reviewed individually, 6) HCQ-AZI was discontinued in patients with QTcF ≥ 500 ms with the exception of patients treated in intensive care unit (ICU) who had continuous ECG monitoring and cardioversion equipment readily available at bedside. The HCQ-AZI dosing was not individually modified in any cases.
In this study, we focused on QT prolongation and arrhythmias associated with HCQ and AZI use. ECG was reviewed and QT was measured in each ECG using the tangent method by 2 cardiologists and 2 resident doctors trained in QT measurement. QT was corrected using the Fridericia formula.
Our primary end point was arrhythmia safety during HCQ and AZI treatment measured as the number of patients with QTcF prolongation ≥ 500 ms within the period of 14 days from the start of HCQ-AZI treatment. Our secondary end points were change in QTcF ≥ 60 ms, QTcF prolongation ≥ 480 ms, the number of ventricular tachycardia cases and cardiac mortality.
2.1 Statistical analysis
Frequency with percentage based on the total cohort was evaluated for categorical parameters while median (min – max) estimate was used for continuous variables. Univariate logistic regression model was used to evaluate odds ratio for QTcF prolongation. Factors found to be significant in univariate logistic regression analysis were entered into multivariate logistic regression model with forward model selection process. A two-tailed p-value<0.05 was considered to be significant. Statistical analysis was performed using R statistical package version 4.0.0.
3 Results
3.1 Demographics of the COVID-19 patients
81 consecutively hospitalized patients had been treated with HCQ and AZI combination from March 23rd to May 10th 2020 and were enrolled into the study (Table 1). The median age was 59 years (35 – 87), 58.0% (n = 47) were female. The largest patient group according to age was the 60–69 years old group (24.7%). The median baseline Cumulative Illness Rating scale (CIRS) score [9] was 4 (0 – 15). The majority of the study population (82.7%) had comorbidities and half of the patients (50.6%) had cardiological diseases: 50.6% had arterial hypertension, 22.2% had coronary artery disease and 11.1% had a history of atrial fibrillation. 33 patients (40.8%) were taking 1–2 and 10 patients (12.3%) 3–4 concomitant drugs.Table 1 Demographics of the COVID-19 patients.
Parameter Subgroup Statistics Total Cohort (N = 81)
Age Median (min–max) 59 (35 – 87)
18–44 n (%) 12 (14.8)
45–49 n (%) 12 (14.8)
50–59 n (%) 17 (21.0)
60–69 n (%) 20 (24.7)
70–79 n (%) 15 (18.5)
≥80 n (%) 5 (6.2)
Sex Female n (%) 47 (58.0)
Male n (%) 34 (42.0)
CIRS Median (min–max) 4 (0 – 15)
Comorbidities Cardiological +/- other n (%) 41 (50.6%)
Non-cardiological n (%) 26 (32.1%)
None n (%) 14 (17.3%)
Number of concomitant medications Median (min–max) 1 (0 – 4)
None n (%) 38 (46.9)
1–2 n (%) 33 (40.8)
3–4 n (%) 10 (12.3)
Antihypertensive medications n (%) 39 (48.1)
Antidiabetic medications n (%) 11 (13.6)
Antipsychotics n (%) 4 (4.9)
Antidepressants n (%) 5 (6.2)
Anticoagulants n (%) 13 (16.0)
Antiaggregants n (%) 3 (3.7)
Beta-mimetics n (%) 5 (6.2)
3.2 Clinical data and laboratory findings of the COVID-19 patients
The median time from symptom onset to hospitalization and treatment with HCQ-AZI were both 7 days (-1 – 42) (Table 2). The most common clinical symptoms were cough (84%) and fever (75.3%). Uncommon symptoms included diarrhea (13.6%), rhinitis (9.9%) and nausea (3.7%). 80 patients (98.8%) had radiological signs of pneumonia. The median baseline NEWS score was 2 (0 – 13). On admission, 34 patients (42.0%) required low-flow oxygen, 2 patients (2.5%) had to be on invasive ventilation and 1 patient (1.3%) was connected to an extracorporeal membrane oxygenation (ECMO) to sustain oxygen saturation above 92%. 3 patients (3.7%) were admitted directly to ICU. Two-thirds of the patients (67.9%) had electrolyte imbalance during the follow-up period.Table 2 Clinical data and laboratory findings of the COVID-19 patients.
Parameter Statistics Total Cohort (N = 81)
Days from symptom onset to hospitalization Median (min–max) 7 (-1 – 42)
Days from symptom onset to treatment initiation Median (min–max) 7 (1 – 42)
Baseline NEWS score Median (min–max) 2 (0 – 13)
Need for low-flow oxygen on admission n (%) 34 (42.0)
Need for invasive ventilation on admission n (%) 2 (2.5)
Need for extracorporeal membrane oxygenation on admission n (%) 1 (1.3)
Radiologically confirmed pneumonia n (%) 80 (98.8)
Additional antibiotics prescribed n (%) 49 (60.5)
Laboratory findings
Baseline absolute lymphocyte count (109/L) Median (min–max) 1.14 (0.42–2.64)
Baseline CRP (mg/l) Median (min–max) 33 (0.34 – 249.4)
Baseline Ferritin (µg/l) (n = 69) Median (min–max) 356 (4.2–2678)
Baseline Interleukin-6 (ng/l) (n = 64) Median (min–max) 15.3 (2–124)
Any electrolyte imbalance n (%) 55 (67.9)
Ca2+ < 1.05 (mmol/l) n (%) 47 (58.0)
K+ < 3.5 (mmol/l) n (%) 11 (13.6)
Mg2+ < 0.65 (mmol/l) n (%) 5 (6.2)
Symptoms
Cough n (%) 68 (84.0)
Rhinitis n (%) 8 (9.9)
Diarrhea n (%) 11 (13.6)
Nausea/vomiting n (%) 3 (3.7)
Fever (>38 °C) n (%) 43 (53.1)
Use of other QT prolonging drugs during hospitalization
At least 1 drug n (%) 42 (51.9)
Known risk of TdP* n (%) 13 (16.0)
Possible risk of TdP** n (%) 8 (9.9)
Conditional risk of TdP*** n (%) 71 (87.7)
* These drugs prolong the QT interval and are clearly associated with a known risk of TdP, even when taken as recommended.
** These drugs can cause QT prolongation but currently lack evidence for a risk of TdP when taken as recommended.
*** These drugs are associated with TdP but only under certain conditions of their use (e.g. excessive dose, in patients with conditions such as hypokalemia, or when taken with interacting drugs) or by creating conditions that facilitate or induce TdP (e.g. by inhibiting metabolism of a QT-prolonging drug or by causing an electrolyte disturbance that induces TdP).
3.3 Cardiotoxicity of HCQ-AZI treatment
More than half of the patients (51.9%) were prescribed at least one additional QT interval prolonging drug during the hospitalization, the majority of these drugs (87.7%) being in the “conditional risk of TdP” group according to “CredibleMeds®” (Table 2).
The median baseline QTcF was 416 ms (365 – 498). The median QTcF was rising daily and the peak of 436 ms (333 – 483) was observed on the 10th day of the HCQ-AZI treatment (Fig. 1a). The highest median ΔQTcF was observed on the 8th day (Fig. 1b).Fig. 1 Daily QTcF change in COVID-19 patients: a) daily QTcF and b) ΔQTcF distributions.
Seven patients (8.6%) had QTcF prolongation of ≥ 500 ms during the 14-day period from the initiation of the treatment (Table 3). Four of these cases were observed during and three immediately after the administration of HCQ-AZI. HCQ-AZI was discontinued in 4 patients (4.9%): one and three in 480–499 ms and ≥ 500 ms groups, respectively. None of the patients developed ventricular tachycardia. The risk factors significantly associated with QTcF ≥ 500 ms were hypokalemia (p = 0.032) and the use of diuretics during the treatment (p = 0.020), the odds ratios (95% CI) were 6.188 (1.168–32.774) and 7.778 (1.388–43.595), respectively. Multivariate logistic regression analysis was not performed due to strong dependence between hypokalemia and the use of diuretics (phi = 0.4).Table 3 Patients with QTcF ≥ 480 ms.
Age Sex1 CIRS Comorbidities CM prolonging QT K+ < 3.5 mmol/l First QTcF ≥ 480 ms Day of first QTcF ≥ 480 ms Cumulated HCQ/AZI dosage until first prolonged QTcF (mg) Had QTcF ≥ 500 ms HCQ/AZI discontinued Ventricular tachycardia
40 s F 7 Hypertension No Yes 516 7 4000/1500 Day 7 Day 7 No
60 s F 8 Hypertension; coronary heart disease; atrial fibrillation, obesity Ranolazine Yes 498 1 0/0 No No No
80 s F 7 Hypertension; coronary heart disease; atrial fibrillation Omeprazole; Metoclopramide No 482 2 600/750 No No No
60 s F 6 Diabetes mellitus; hypertension; coronary heart disease; atrial fibrillation; obesity Metoclopramide No 487 3 1400/750 No Day 4 No
80 s F 6 Diabetes mellitus; hypertension Furosemide No 492 3 1200/750 Day 5 Day 5 No
70 s F 7 Ovarian cancer; hypertension; coronary heart disease; atrial fibrillation No No 486 1 0/0 No No No
60 s M 5 Hypertension; coronary heart disease; obesity Piperacillin-Tazobactam Yes 513 13 6000/1500 Day 13 No No
70 s F 15 Chronic myeloid leukemia; diabetes mellitus; hypertension; coronary heart disease Sertraline; Dasatinib, Metoclopramide; Omeprazole No 492 1 0/0 Day 8 Day 8 No
50 s M 15 Renal cell carcinoma; diabetes mellitus; hypertension; coronary heart disease; chronic atrial fibrillation; chronic kidney disease Piperacillin-Tazobactam; Furosemide; Quetiapine; Fluconazole; Propofol; Metoclopramide; Haloperidol; Esomeprazol No 483 10 6000/1500 No No No
50 s M 5 Diabetes mellitus; hypertension; coronary heart disease; obesity Amiodarone; Furosemide; Omeprazol; Propofol; Metoclopramide Yes 493 5 3000/1500 No No No
70 s* M 6 Prostate cancer; hypertension Amiodarone; Haloperidol; Piperacillin-Tazobactam; Furosemide No 509 14 6000/1500 Day 14 No No
40 s M 0 None None Yes 480 2 1200/750 No No No
50 s M 4 Hypertension Furosemide; Propofol No 489 4 2400/1000 Day 13 No No
50 s M 8 Diabetes mellitus; coronary heart disease; hypertension; obesity Furosemide; Propofol Yes 496 4 2400/1000 Day 6 No No
* Subject died on day 16 due to multiple organ failure.
1 F: Female, M: Male.
14 patients (17.3%) experienced QTcF ≥ 480 ms (Table 3) and 16 patients (19.8%) had a change of QTcF ≥ 60 ms. Higher baseline NEWS score, presence of cardiological comorbidities, higher number of concomitant medications, hypokalemia, use of diuretics during the treatment and higher baseline QTcF were associated with QTcF prolongation ≥ 480 ms in the univariate logistic regression model (Table 4). On multivariate analysis, cardiological comorbidities (p = 0.034) and hypokalemia (p = 0.008) were found to be independent factors for QTcF ≥ 480 ms interval prolongation (Table 4, Fig. 2).Table 4 Logistic regression analysis of predictors for QTcF prolongation (≥480 ms) in COVID-19 patients.
Parameters Univariate model Multivariate model
Odds ratio P-value Odds ratio P-value
Estimate 95% CI Estimate 95% CI
Older age 1.043 0.995–1.093 0.081 ni
Male sex 1.482 0.466–4.706 0.505 ni
Higher baseline NEWS score 1.323 1.047–1.672 0.019 n-cs
Presence of cardiological comorbidity 18.107 2.237–146.55 0.007 10.311 1.186–89.604 0.034
Higher number of concomitant medications1 2.017 1.214–3.352 0.007 ni
Presence of hypocalcemia during treatment 0.675 0.212–2.144 0.505 ni
Presence of hypomagnesemia during treatment 3.556 0.536–23.593 0.189 ni
Presence of hypokalemia during treatment 9.300 2.301–37.588 0.002 8.116 1.718–38.347 0.008
Use of diuretics during treatment 6.814 1.968–23.587 0.002 n-cs
Higher baseline QTcF 1.030 1.005–1.055 0.017 n-cs
ni: not included. n-cs: non-clinically significant. CI: confidence interval.
1 Parameter was not included into multivariate analysis due to strong relation with subject’s comorbidities.
Fig. 2 Forest plot of univariate and multivariate analysis for risk factors associated with QTcF interval prolongation ≥ 480 ms.
During the course of HCQ-AZI treatment minority of patients presented with atrial fibrillation (3.7%) or complete bundle branch block (1.2%). PR and QRS duration dynamics were analyzed but no statistically significant changes were observed. According to acquired data, HCQ-AZI treatment did not have a significant impact on atrioventricular or intraventricular conduction.
3.4 Outcomes of the COVID-19 patients
11 patients (13.6%) were transferred to ICU and 3 patients (3.7%) were connected to ECMO. Cytokine adsorbtion using CytoSorb® filters was applied in 7 cases (8.6%) and interleukin-6-receptor inhibitor Tocilizumab was administered in 4 patients (4.9%). 78 patients (96.3%) were discharged from the hospital and three patients (3.7%) died. The lethal outcomes were considered to be indirectly related to COVID-19: two patients died due to bacterial superinfection, septic shock and multiple organ failure; one patient’s cause of death was disseminated intravascular coagulation, systemic inflammatory response syndrome and multiple organ failure.
4 Discussion
After promising initial results [2] and worldwide empirical administration of HCQ and AZI to treat COVID-19 patients, detailed arrhythmia risk mitigation guidelines have not been published. In order to reduce the risk of QTc prolongation and cardiac adverse events, we implemented a simplified HCQ-AZI arrhythmia risk management plan. With this approach, fourteen patients (17.3%) had QTc prolongation of ≥ 480 ms at least once. Among them only seven (8.6%) experienced extreme prolongation of QTc ≥ 500 ms with no observed ventricular tachycardia episodes.
During randomized trial from Brazil of low-dose chloroquine (CQ) for 5 days vs. high-dose CQ for 10 days, alarming prolongation of QTc ≥ 500 ms was documented in 4/36 (11.1%) vs. 7/37 (18.9%) and ventricular tachycardia in 0/36 vs. 2/37 (2.7%) patients, respectively [10]. Many of these patients had severe COVID-19 infection, serious comorbidities or were elderly. Severe infection and concomitant medications with QT prolonging potential may have been the reason of early timing (1–4 day of treatment) of extreme QTc prolongation or arrhythmia. For example, 89.6% of patients were taking Oseltamivir for suspected influenza infection, which may have contributed to QT prolongation [11]. In a retrospective HCQ and AZI treatment cohort of 90 patients with COVID-19, 11 of 53 (21%) subjects developed QTc ≥ 500 ms and 7 of 53 (13%) had ΔQTc ≥ 60 ms [12]. One case of torsades de pointes (TdP) which happened three days after discontinuation of treatment may indicate delayed risk possibly due to long half-life of HCQ [13].
The Wuhan group presented 416 patients with COVID-19. A cardiac injury defined as blood levels of cardiac biomarkers (hs-TNI) above the 99th percentile upper reference limit occurred in 82 subjects (19.7%) and were associated with worse outcome [14]. However, a much higher extent of cardiac injury was observed in a prospective cohort of 100 COVID-19 patients for whom cardiac magnetic resonance was performed [15]. Puntmann et al. detected cardiac involvement in 78 individuals (78%) and an ongoing myocardial inflammation in 60 of them (60%), defined as abnormal native T1 and T2 measures. Interestingly, these findings had no relation to preexisting diseases, severity and course of the acute illness or the time from diagnosis. Although remote clinical outcomes of these lesions remain unclear, the high prevalence of perimyocarditis raises many practical questions. For instance, there is a need of consensus on safe timing to return to competitive sports after COVID-19 infection. In the series of 26 athletes, 4 (15%) presented with signs of myocarditis in cardiac magnetic resonance imaging [16]. 8 other patients (30.8%) showed late gadolinium enhancement without T2 elevation which is compatible with prior myocardial injury. These magnetic resonance findings demonstrate that cardiac injury in COVID-19 patients is frequent. It is feasible, that electrical conduction system of the heart can be additionally adversely affected by viral myocarditis.
QT prolongation is a well-known side effect of HCQ and AZI. A larger retrospective cohort study of 251 subjects showed prolongation of QTc ≥ 500 ms in 15.9% of subjects with 1 case of TdP [17]. The peak value of ΔQTc in this study was reached at the end of the 5-day HCQ and AZI treatment scheme. Similarly, in our cohort the peak mean ΔQTcF was observed at the end of the treatment (day 8) (Fig. 1b). Application of ECG telemonitoring during the last days and immediately after the treatment may thus be indicated. Importantly, using our risk management plan QTc ≥ 500 ms was observed less frequently (8.6%), despite longer duration of HCQ-AZI treatment compared to both studies mentioned above.
To the best of our knowledge, a well-known link between hypokalemia and QTc prolongation has not been demonstrated in COVID-19 population treated with HCQ and AZI. Low potassium levels were associated with extreme prolongation of QTcF ≥ 500 ms (p = 0.032) in our cohort. Hypokalemia may be aggravated by the ability of SARS-CoV-2 to degrade angiotensin-converting enzyme 2 and increase the action of angiotensin I/II and renin–angiotensin–aldosterone system resulting in a challenging renal K+ loss [18]. The study found a positive association between the degree of hypokalemia and the severity of COVID-19. The resolution of urine K+ loss appeared to be a sensitive biomarker of good prognosis. The importance of electrolyte testing and correction aiming to prevent cardiac arrhythmias was further highlighted in a large observational study by Arshad et al. [19] Authors emphasized that stringently applied electrolyte protocols were effective in controlling adverse events.
The generalizability of our findings may be limited to patients hospitalized and monitored daily in a tertiary level university hospital. Therefore, it may not be applicable to other populations where such monitoring cannot be implemented. Our risk management plan aimed to prevent emerging cardiac arrhythmias. Structural myocardial damage related to COVID-19, such as myocarditis, may have a significant impact on prognosis. It was not evaluated in the mitigation protocol and its extent in our population remains unknown. However, a simple to follow protocol with no routine co-administration of other QT prolonging drugs, good daily ECG recording and electrolyte testing compliance resulted in few cardiac adverse events compared to other cohorts.
In conclusion, there was a low incidence of extreme QTc prolongation ≥ 500 ms and no ventricular tachycardia events in COVID-19 patients treated with HCQ and AZI in the setting of cardiac arrhythmia risk management plan.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The data was acquired from the electronic medical records accessed through Vilnius University Hospital Santaros Klinikos Biobank. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33335973 | 18,690,268 | 2021-02 |
What was the administration route of drug 'HYDROXYCHLOROQUINE SULFATE'? | Mitigating arrhythmia risk in Hydroxychloroquine and Azithromycin treated COVID-19 patients using arrhythmia risk management plan.
To assess cardiac safety in COVID-19 patients treated with the combination of Hydroxychloroquine and Azithromycin using arrhythmia risk management plan.
We retrospectively examined arrhythmia safety of treatment with Hydroxychloroquine and Azithromycin in the setting of pre-defined arrhythmia risk management plan. The data was analyzed using R statistical package version 4.0.0. A two-tailed p-value<0.05 was considered significant. 81 patients were included from March 23rd to May 10th 2020. The median age was 59 years, 58.0% were female. The majority of the study population (82.7%) had comorbidities, 98.8% had radiological signs of pneumonia. Fourteen patients (17.3%) experienced QTc ≥ 480 ms and 16 patients (19.8%) had an increase of QTc ≥ 60 ms. Seven patients (8.6%) had QTc prolongation of ≥ 500 ms. The treatment was discontinued in 4 patients (4.9%). None of the patients developed ventricular tachycardia. The risk factors significantly associated with QTc ≥ 500 ms were hypokalemia (p = 0.032) and use of diuretics during the treatment (p = 0.020). Three patients (3.7%) died, the cause of death was bacterial superinfection with septic shock in two patients, and disseminated intravascular coagulation with multiple organ failure in one patient. None of these deaths were associated with cardiac arrhythmias.
We recorded a low incidence of QTc prolongation ≥ 500 ms and no ventricular tachycardia events in COVID-19 patients treated with Hydroxychloroquine and Azithromycin using cardiac arrhythmia risk management plan.
1 Introduction
Since the beginning of COVID-19 spread in December 2019, numerous treatment options have been studied worldwide. Based upon limited clinical data in case series, Hydroxychloroquine was recommended for treatment of hospitalized COVID-19 patients in several countries, and a number of national guidelines reported incorporating recommendations regarding use of this drug in the setting of COVID-19. Furthermore, on 28th March 2020 Hydroxychloroquine was authorized for emergency use by the U.S. Food and Drug Administration (FDA) during the COVID-19 pandemic [1]. Additional administration of macrolide antibiotic Azithromycin, seemed to reinforce the efficacy of Hydroxychloroquine [2].
Hydroxychloroquine in combination with Azithromycin or alone was used for COVID-19 treatment worldwide for almost 3 months. However, multiple large studies showed that the use of Hydroxychloroquine, either monotherapy or in combination with Azithromycin, did not improve clinical status as compared to the standard care [3], [4]. Based on emerging scientific data, emergency use authorization for these drugs was revoked on 15th June 2020. Furthermore, Solidarity trial results showed that Hydroxychloroquine produced little or no reduction in the mortality of hospitalized COVID-19 patients when compared to standard of care and on 4th July 2020 World Health Organization (WHO) discontinued the trial’s Hydroxychloroquine arm [5].
Despite the recall of Hydroxychloroquine by FDA and WHO, as of 1st November 2020, according to U.S. National Library of Medicine and European Union Clinical Trials Register, there are at least 45 interventional clinical trials for COVID-19 treatment with Hydroxychloroquine and Azithromycin that are active around the world. The drugs are studied in the setting of confirmed COVID-19 infection as well as in proactive prophylaxis. Although FDA and European Medicines Agency had warned to take precautions when using Hydroxychloroquine and Azithromycin [6], [7], no definite cardiac safety protocols have been established yet. Here we report QT interval prolongation and arrhythmia safety results in COVID-19 patients treated with the combination of HCQ and AZI using close monitoring and arrhythmia risk management plan.
2 Methods
We retrospectively examined the cardiac safety of treatment using HCQ and AZI in consecutive adult patients with COVID-19 infection confirmed by qPCR treated in Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania. Every patient consented to the treatment plan by signing an informed consent form and the study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution's human research committee. The data was acquired from the electronic medical records accessed through Vilnius University Hospital Santaros Klinikos Biobank. The study was approved by the regional ethics committee.
HCQ-AZI consisted of 5 days of oral Azithromycin (once daily; initial dose 500 mg on the first day followed by 250 mg during the next 4 days) and 10 days of oral Hydroxychloroquine 200 mg three times daily.
The cardiac arrhythmia risk management plan was designed by a multidisciplinary team which included a cardiologist – arrhythmia specialist. It was applied during HCQ-AZI treatment for each patient and was as follows: 1) QT prolonging concomitant drugs (assessed using “CredibleMeds®” database [8]) were discontinued if possible; 2) ECG recording and QTcF (Fridericia) calculation was performed daily; 3) K+ and Mg2+ were replaced if abnormal (hypokalemia and hypomagnesemia were defined as K+ levels < 3.5 mmol/L and Mg2+ < 0.65 mmol/L in blood serum, respectively); 4) if QTcF reached 480 ms during HCQ-AZI treatment, K+ and Mg2+ were replaced to reach maximum normal values, 5) if QTcF remained in the interval of 480 – 499 ms regardless of K+ and Mg2+ replacement, the risk–benefit of continuing HCQ-AZI was reviewed individually, 6) HCQ-AZI was discontinued in patients with QTcF ≥ 500 ms with the exception of patients treated in intensive care unit (ICU) who had continuous ECG monitoring and cardioversion equipment readily available at bedside. The HCQ-AZI dosing was not individually modified in any cases.
In this study, we focused on QT prolongation and arrhythmias associated with HCQ and AZI use. ECG was reviewed and QT was measured in each ECG using the tangent method by 2 cardiologists and 2 resident doctors trained in QT measurement. QT was corrected using the Fridericia formula.
Our primary end point was arrhythmia safety during HCQ and AZI treatment measured as the number of patients with QTcF prolongation ≥ 500 ms within the period of 14 days from the start of HCQ-AZI treatment. Our secondary end points were change in QTcF ≥ 60 ms, QTcF prolongation ≥ 480 ms, the number of ventricular tachycardia cases and cardiac mortality.
2.1 Statistical analysis
Frequency with percentage based on the total cohort was evaluated for categorical parameters while median (min – max) estimate was used for continuous variables. Univariate logistic regression model was used to evaluate odds ratio for QTcF prolongation. Factors found to be significant in univariate logistic regression analysis were entered into multivariate logistic regression model with forward model selection process. A two-tailed p-value<0.05 was considered to be significant. Statistical analysis was performed using R statistical package version 4.0.0.
3 Results
3.1 Demographics of the COVID-19 patients
81 consecutively hospitalized patients had been treated with HCQ and AZI combination from March 23rd to May 10th 2020 and were enrolled into the study (Table 1). The median age was 59 years (35 – 87), 58.0% (n = 47) were female. The largest patient group according to age was the 60–69 years old group (24.7%). The median baseline Cumulative Illness Rating scale (CIRS) score [9] was 4 (0 – 15). The majority of the study population (82.7%) had comorbidities and half of the patients (50.6%) had cardiological diseases: 50.6% had arterial hypertension, 22.2% had coronary artery disease and 11.1% had a history of atrial fibrillation. 33 patients (40.8%) were taking 1–2 and 10 patients (12.3%) 3–4 concomitant drugs.Table 1 Demographics of the COVID-19 patients.
Parameter Subgroup Statistics Total Cohort (N = 81)
Age Median (min–max) 59 (35 – 87)
18–44 n (%) 12 (14.8)
45–49 n (%) 12 (14.8)
50–59 n (%) 17 (21.0)
60–69 n (%) 20 (24.7)
70–79 n (%) 15 (18.5)
≥80 n (%) 5 (6.2)
Sex Female n (%) 47 (58.0)
Male n (%) 34 (42.0)
CIRS Median (min–max) 4 (0 – 15)
Comorbidities Cardiological +/- other n (%) 41 (50.6%)
Non-cardiological n (%) 26 (32.1%)
None n (%) 14 (17.3%)
Number of concomitant medications Median (min–max) 1 (0 – 4)
None n (%) 38 (46.9)
1–2 n (%) 33 (40.8)
3–4 n (%) 10 (12.3)
Antihypertensive medications n (%) 39 (48.1)
Antidiabetic medications n (%) 11 (13.6)
Antipsychotics n (%) 4 (4.9)
Antidepressants n (%) 5 (6.2)
Anticoagulants n (%) 13 (16.0)
Antiaggregants n (%) 3 (3.7)
Beta-mimetics n (%) 5 (6.2)
3.2 Clinical data and laboratory findings of the COVID-19 patients
The median time from symptom onset to hospitalization and treatment with HCQ-AZI were both 7 days (-1 – 42) (Table 2). The most common clinical symptoms were cough (84%) and fever (75.3%). Uncommon symptoms included diarrhea (13.6%), rhinitis (9.9%) and nausea (3.7%). 80 patients (98.8%) had radiological signs of pneumonia. The median baseline NEWS score was 2 (0 – 13). On admission, 34 patients (42.0%) required low-flow oxygen, 2 patients (2.5%) had to be on invasive ventilation and 1 patient (1.3%) was connected to an extracorporeal membrane oxygenation (ECMO) to sustain oxygen saturation above 92%. 3 patients (3.7%) were admitted directly to ICU. Two-thirds of the patients (67.9%) had electrolyte imbalance during the follow-up period.Table 2 Clinical data and laboratory findings of the COVID-19 patients.
Parameter Statistics Total Cohort (N = 81)
Days from symptom onset to hospitalization Median (min–max) 7 (-1 – 42)
Days from symptom onset to treatment initiation Median (min–max) 7 (1 – 42)
Baseline NEWS score Median (min–max) 2 (0 – 13)
Need for low-flow oxygen on admission n (%) 34 (42.0)
Need for invasive ventilation on admission n (%) 2 (2.5)
Need for extracorporeal membrane oxygenation on admission n (%) 1 (1.3)
Radiologically confirmed pneumonia n (%) 80 (98.8)
Additional antibiotics prescribed n (%) 49 (60.5)
Laboratory findings
Baseline absolute lymphocyte count (109/L) Median (min–max) 1.14 (0.42–2.64)
Baseline CRP (mg/l) Median (min–max) 33 (0.34 – 249.4)
Baseline Ferritin (µg/l) (n = 69) Median (min–max) 356 (4.2–2678)
Baseline Interleukin-6 (ng/l) (n = 64) Median (min–max) 15.3 (2–124)
Any electrolyte imbalance n (%) 55 (67.9)
Ca2+ < 1.05 (mmol/l) n (%) 47 (58.0)
K+ < 3.5 (mmol/l) n (%) 11 (13.6)
Mg2+ < 0.65 (mmol/l) n (%) 5 (6.2)
Symptoms
Cough n (%) 68 (84.0)
Rhinitis n (%) 8 (9.9)
Diarrhea n (%) 11 (13.6)
Nausea/vomiting n (%) 3 (3.7)
Fever (>38 °C) n (%) 43 (53.1)
Use of other QT prolonging drugs during hospitalization
At least 1 drug n (%) 42 (51.9)
Known risk of TdP* n (%) 13 (16.0)
Possible risk of TdP** n (%) 8 (9.9)
Conditional risk of TdP*** n (%) 71 (87.7)
* These drugs prolong the QT interval and are clearly associated with a known risk of TdP, even when taken as recommended.
** These drugs can cause QT prolongation but currently lack evidence for a risk of TdP when taken as recommended.
*** These drugs are associated with TdP but only under certain conditions of their use (e.g. excessive dose, in patients with conditions such as hypokalemia, or when taken with interacting drugs) or by creating conditions that facilitate or induce TdP (e.g. by inhibiting metabolism of a QT-prolonging drug or by causing an electrolyte disturbance that induces TdP).
3.3 Cardiotoxicity of HCQ-AZI treatment
More than half of the patients (51.9%) were prescribed at least one additional QT interval prolonging drug during the hospitalization, the majority of these drugs (87.7%) being in the “conditional risk of TdP” group according to “CredibleMeds®” (Table 2).
The median baseline QTcF was 416 ms (365 – 498). The median QTcF was rising daily and the peak of 436 ms (333 – 483) was observed on the 10th day of the HCQ-AZI treatment (Fig. 1a). The highest median ΔQTcF was observed on the 8th day (Fig. 1b).Fig. 1 Daily QTcF change in COVID-19 patients: a) daily QTcF and b) ΔQTcF distributions.
Seven patients (8.6%) had QTcF prolongation of ≥ 500 ms during the 14-day period from the initiation of the treatment (Table 3). Four of these cases were observed during and three immediately after the administration of HCQ-AZI. HCQ-AZI was discontinued in 4 patients (4.9%): one and three in 480–499 ms and ≥ 500 ms groups, respectively. None of the patients developed ventricular tachycardia. The risk factors significantly associated with QTcF ≥ 500 ms were hypokalemia (p = 0.032) and the use of diuretics during the treatment (p = 0.020), the odds ratios (95% CI) were 6.188 (1.168–32.774) and 7.778 (1.388–43.595), respectively. Multivariate logistic regression analysis was not performed due to strong dependence between hypokalemia and the use of diuretics (phi = 0.4).Table 3 Patients with QTcF ≥ 480 ms.
Age Sex1 CIRS Comorbidities CM prolonging QT K+ < 3.5 mmol/l First QTcF ≥ 480 ms Day of first QTcF ≥ 480 ms Cumulated HCQ/AZI dosage until first prolonged QTcF (mg) Had QTcF ≥ 500 ms HCQ/AZI discontinued Ventricular tachycardia
40 s F 7 Hypertension No Yes 516 7 4000/1500 Day 7 Day 7 No
60 s F 8 Hypertension; coronary heart disease; atrial fibrillation, obesity Ranolazine Yes 498 1 0/0 No No No
80 s F 7 Hypertension; coronary heart disease; atrial fibrillation Omeprazole; Metoclopramide No 482 2 600/750 No No No
60 s F 6 Diabetes mellitus; hypertension; coronary heart disease; atrial fibrillation; obesity Metoclopramide No 487 3 1400/750 No Day 4 No
80 s F 6 Diabetes mellitus; hypertension Furosemide No 492 3 1200/750 Day 5 Day 5 No
70 s F 7 Ovarian cancer; hypertension; coronary heart disease; atrial fibrillation No No 486 1 0/0 No No No
60 s M 5 Hypertension; coronary heart disease; obesity Piperacillin-Tazobactam Yes 513 13 6000/1500 Day 13 No No
70 s F 15 Chronic myeloid leukemia; diabetes mellitus; hypertension; coronary heart disease Sertraline; Dasatinib, Metoclopramide; Omeprazole No 492 1 0/0 Day 8 Day 8 No
50 s M 15 Renal cell carcinoma; diabetes mellitus; hypertension; coronary heart disease; chronic atrial fibrillation; chronic kidney disease Piperacillin-Tazobactam; Furosemide; Quetiapine; Fluconazole; Propofol; Metoclopramide; Haloperidol; Esomeprazol No 483 10 6000/1500 No No No
50 s M 5 Diabetes mellitus; hypertension; coronary heart disease; obesity Amiodarone; Furosemide; Omeprazol; Propofol; Metoclopramide Yes 493 5 3000/1500 No No No
70 s* M 6 Prostate cancer; hypertension Amiodarone; Haloperidol; Piperacillin-Tazobactam; Furosemide No 509 14 6000/1500 Day 14 No No
40 s M 0 None None Yes 480 2 1200/750 No No No
50 s M 4 Hypertension Furosemide; Propofol No 489 4 2400/1000 Day 13 No No
50 s M 8 Diabetes mellitus; coronary heart disease; hypertension; obesity Furosemide; Propofol Yes 496 4 2400/1000 Day 6 No No
* Subject died on day 16 due to multiple organ failure.
1 F: Female, M: Male.
14 patients (17.3%) experienced QTcF ≥ 480 ms (Table 3) and 16 patients (19.8%) had a change of QTcF ≥ 60 ms. Higher baseline NEWS score, presence of cardiological comorbidities, higher number of concomitant medications, hypokalemia, use of diuretics during the treatment and higher baseline QTcF were associated with QTcF prolongation ≥ 480 ms in the univariate logistic regression model (Table 4). On multivariate analysis, cardiological comorbidities (p = 0.034) and hypokalemia (p = 0.008) were found to be independent factors for QTcF ≥ 480 ms interval prolongation (Table 4, Fig. 2).Table 4 Logistic regression analysis of predictors for QTcF prolongation (≥480 ms) in COVID-19 patients.
Parameters Univariate model Multivariate model
Odds ratio P-value Odds ratio P-value
Estimate 95% CI Estimate 95% CI
Older age 1.043 0.995–1.093 0.081 ni
Male sex 1.482 0.466–4.706 0.505 ni
Higher baseline NEWS score 1.323 1.047–1.672 0.019 n-cs
Presence of cardiological comorbidity 18.107 2.237–146.55 0.007 10.311 1.186–89.604 0.034
Higher number of concomitant medications1 2.017 1.214–3.352 0.007 ni
Presence of hypocalcemia during treatment 0.675 0.212–2.144 0.505 ni
Presence of hypomagnesemia during treatment 3.556 0.536–23.593 0.189 ni
Presence of hypokalemia during treatment 9.300 2.301–37.588 0.002 8.116 1.718–38.347 0.008
Use of diuretics during treatment 6.814 1.968–23.587 0.002 n-cs
Higher baseline QTcF 1.030 1.005–1.055 0.017 n-cs
ni: not included. n-cs: non-clinically significant. CI: confidence interval.
1 Parameter was not included into multivariate analysis due to strong relation with subject’s comorbidities.
Fig. 2 Forest plot of univariate and multivariate analysis for risk factors associated with QTcF interval prolongation ≥ 480 ms.
During the course of HCQ-AZI treatment minority of patients presented with atrial fibrillation (3.7%) or complete bundle branch block (1.2%). PR and QRS duration dynamics were analyzed but no statistically significant changes were observed. According to acquired data, HCQ-AZI treatment did not have a significant impact on atrioventricular or intraventricular conduction.
3.4 Outcomes of the COVID-19 patients
11 patients (13.6%) were transferred to ICU and 3 patients (3.7%) were connected to ECMO. Cytokine adsorbtion using CytoSorb® filters was applied in 7 cases (8.6%) and interleukin-6-receptor inhibitor Tocilizumab was administered in 4 patients (4.9%). 78 patients (96.3%) were discharged from the hospital and three patients (3.7%) died. The lethal outcomes were considered to be indirectly related to COVID-19: two patients died due to bacterial superinfection, septic shock and multiple organ failure; one patient’s cause of death was disseminated intravascular coagulation, systemic inflammatory response syndrome and multiple organ failure.
4 Discussion
After promising initial results [2] and worldwide empirical administration of HCQ and AZI to treat COVID-19 patients, detailed arrhythmia risk mitigation guidelines have not been published. In order to reduce the risk of QTc prolongation and cardiac adverse events, we implemented a simplified HCQ-AZI arrhythmia risk management plan. With this approach, fourteen patients (17.3%) had QTc prolongation of ≥ 480 ms at least once. Among them only seven (8.6%) experienced extreme prolongation of QTc ≥ 500 ms with no observed ventricular tachycardia episodes.
During randomized trial from Brazil of low-dose chloroquine (CQ) for 5 days vs. high-dose CQ for 10 days, alarming prolongation of QTc ≥ 500 ms was documented in 4/36 (11.1%) vs. 7/37 (18.9%) and ventricular tachycardia in 0/36 vs. 2/37 (2.7%) patients, respectively [10]. Many of these patients had severe COVID-19 infection, serious comorbidities or were elderly. Severe infection and concomitant medications with QT prolonging potential may have been the reason of early timing (1–4 day of treatment) of extreme QTc prolongation or arrhythmia. For example, 89.6% of patients were taking Oseltamivir for suspected influenza infection, which may have contributed to QT prolongation [11]. In a retrospective HCQ and AZI treatment cohort of 90 patients with COVID-19, 11 of 53 (21%) subjects developed QTc ≥ 500 ms and 7 of 53 (13%) had ΔQTc ≥ 60 ms [12]. One case of torsades de pointes (TdP) which happened three days after discontinuation of treatment may indicate delayed risk possibly due to long half-life of HCQ [13].
The Wuhan group presented 416 patients with COVID-19. A cardiac injury defined as blood levels of cardiac biomarkers (hs-TNI) above the 99th percentile upper reference limit occurred in 82 subjects (19.7%) and were associated with worse outcome [14]. However, a much higher extent of cardiac injury was observed in a prospective cohort of 100 COVID-19 patients for whom cardiac magnetic resonance was performed [15]. Puntmann et al. detected cardiac involvement in 78 individuals (78%) and an ongoing myocardial inflammation in 60 of them (60%), defined as abnormal native T1 and T2 measures. Interestingly, these findings had no relation to preexisting diseases, severity and course of the acute illness or the time from diagnosis. Although remote clinical outcomes of these lesions remain unclear, the high prevalence of perimyocarditis raises many practical questions. For instance, there is a need of consensus on safe timing to return to competitive sports after COVID-19 infection. In the series of 26 athletes, 4 (15%) presented with signs of myocarditis in cardiac magnetic resonance imaging [16]. 8 other patients (30.8%) showed late gadolinium enhancement without T2 elevation which is compatible with prior myocardial injury. These magnetic resonance findings demonstrate that cardiac injury in COVID-19 patients is frequent. It is feasible, that electrical conduction system of the heart can be additionally adversely affected by viral myocarditis.
QT prolongation is a well-known side effect of HCQ and AZI. A larger retrospective cohort study of 251 subjects showed prolongation of QTc ≥ 500 ms in 15.9% of subjects with 1 case of TdP [17]. The peak value of ΔQTc in this study was reached at the end of the 5-day HCQ and AZI treatment scheme. Similarly, in our cohort the peak mean ΔQTcF was observed at the end of the treatment (day 8) (Fig. 1b). Application of ECG telemonitoring during the last days and immediately after the treatment may thus be indicated. Importantly, using our risk management plan QTc ≥ 500 ms was observed less frequently (8.6%), despite longer duration of HCQ-AZI treatment compared to both studies mentioned above.
To the best of our knowledge, a well-known link between hypokalemia and QTc prolongation has not been demonstrated in COVID-19 population treated with HCQ and AZI. Low potassium levels were associated with extreme prolongation of QTcF ≥ 500 ms (p = 0.032) in our cohort. Hypokalemia may be aggravated by the ability of SARS-CoV-2 to degrade angiotensin-converting enzyme 2 and increase the action of angiotensin I/II and renin–angiotensin–aldosterone system resulting in a challenging renal K+ loss [18]. The study found a positive association between the degree of hypokalemia and the severity of COVID-19. The resolution of urine K+ loss appeared to be a sensitive biomarker of good prognosis. The importance of electrolyte testing and correction aiming to prevent cardiac arrhythmias was further highlighted in a large observational study by Arshad et al. [19] Authors emphasized that stringently applied electrolyte protocols were effective in controlling adverse events.
The generalizability of our findings may be limited to patients hospitalized and monitored daily in a tertiary level university hospital. Therefore, it may not be applicable to other populations where such monitoring cannot be implemented. Our risk management plan aimed to prevent emerging cardiac arrhythmias. Structural myocardial damage related to COVID-19, such as myocarditis, may have a significant impact on prognosis. It was not evaluated in the mitigation protocol and its extent in our population remains unknown. However, a simple to follow protocol with no routine co-administration of other QT prolonging drugs, good daily ECG recording and electrolyte testing compliance resulted in few cardiac adverse events compared to other cohorts.
In conclusion, there was a low incidence of extreme QTc prolongation ≥ 500 ms and no ventricular tachycardia events in COVID-19 patients treated with HCQ and AZI in the setting of cardiac arrhythmia risk management plan.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The data was acquired from the electronic medical records accessed through Vilnius University Hospital Santaros Klinikos Biobank. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33335973 | 18,690,267 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Lymphadenopathy'. | Interim [18F]FDG PET/CT can predict response to anti-PD-1 treatment in metastatic melanoma.
In an attempt to identify biomarkers that can reliably predict long-term outcomes to immunotherapy in metastatic melanoma, we investigated the prognostic role of [18F]FDG PET/CT, performed at baseline and early during the course of anti-PD-1 treatment.
Twenty-five patients with stage IV melanoma, scheduled for treatment with PD-1 inhibitors, were enrolled in the study (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients; nivolumab/ipilimumab, 13 patients). [18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of PD-1 blockade administration (interim PET/CT). Seventeen patients underwent also a third PET/CT scan after administration of four cycles of treatment. Evaluation of patients' response by means of PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria and the PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT). Response to treatment was classified into 4 categories: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD). Patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD. Moreover, patterns of [18F]FDG uptake suggestive of radiologic immune-related adverse events (irAEs) were documented. Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause.
Median follow-up from interim PET/CT was 24.2 months (19.3-41.7 months). According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), and 6 of them had no-MB (PMD). With regard to PFS, no significant difference was observed between patients with MB and no-MB on interim PET/CT according to the EORTC criteria (p = 0.088). In contrary, according to the PERCIMT criteria, patients demonstrating MB had a significantly longer PFS than those showing no-MB (p = 0.045). The emergence of radiologic irAEs (n = 11 patients) was not associated with a significant survival benefit. Regarding the sub-cohort undergoing also a third PET/CT, 14/17 patients (82%) showed concordant responses and 3/17 (18%) had a mismatch of response assessment between interim and late PET/CT.
PET/CT-based response of metastatic melanoma to PD-1 blockade after application of the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Introduction
Metastatic melanoma is a highly aggressive tumor, largely refractory to existing therapies, and associated with a very poor prognosis [1]. While until lately the treatment options for metastatic melanoma were limited, the recent development and introduction in clinical practice of several novel immunotherapeutic agents as well as of targeted therapy with BRAF and MEK inhibitors have revolutionized the systemic treatment of the disease, leading to unprecedented response and survival rates of melanoma patients [2].
The main form of immunotherapy applied in this new era of melanoma management involves immune checkpoint blockade. This immunomodulatory approach activates the immune system against tumors through the binding of the cytotoxic T lymphocyte–associated protein 4 (CTLA-4) and/or the programmed cell death protein 1 (PD-1), both of which are expressed by T cells [3, 4]. The monoclonal antibody ipilimumab, which acts by blocking CTLA-4, is considered a landmark agent in this context, being the first immunotherapeutic drug demonstrating a clear benefit in survival of patients with advanced melanoma, which led to its approval by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2011 [5]. A few years later, a second class of immune checkpoint inhibitors (ICIs), the PD-1 inhibitors nivolumab and pembrolizumab, were also approved for the treatment of melanoma, after having shown survival benefit in these patients [6–8]. Moreover, the anti-PD-1 monoclonal antibodies have shown superiority over ipilimumab, leading to their application both as single agents and in combination with ipilimumab, which is nowadays seldom used as monotherapy [9–13].
Despite these dramatic improvements, a significant amount of patients—approximately 40–45%—show no response to immunotherapy [14]. Additionally, the mechanism of action of these agents (which is markedly different from usual cytotoxic approaches—notably by generating inflammations rather than direct lysis) can pose relevant challenges in the interpretation of treatment response by conventional imaging approaches [15]. Furthermore, several patients experience a “new class” of cumulative, dose-dependent, and sometimes life-threatening side effects, the immune-related adverse events (irAEs), which scope is wide [16]; importantly, the occurrence of such irAEs may be of prognostic value, revealing a response to immunotherapy [17, 18]. These issues raise the question of how to evaluate the response to ICIs in a reliable fashion and early in the course of treatment. This information would help discriminate responders from non-responders, offering significant therapeutic and prognostic implications in the entire spectrum of patient management. Unfortunately, there exist at present only few reliable predictors of long-term response to immunotherapy.
[18F]FDG PET/CT is considered the elective imaging technique in detecting metastatic disease in advanced melanoma [19–23]. Moreover, a growing amount of recently published literature has highlighted the potential role of the modality in the prediction of treatment response to ICIs in melanoma, rendering it an attractive tool for the monitoring of immunotherapy [24–31].
In quest of identifying reliable biomarkers for the prediction of long-term outcomes to immunotherapy, we aim in the present prospective study to assess the value of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment.
Materials and methods
Patients
Twenty-five patients (12 males, 13 females; mean age 54.7 years) with unresectable, stage IV melanoma undergoing immunotherapy with PD-1 inhibitors applied either as monotherapy (pembrolizumab, nivolumab) or as combination treatment with CTLA-4 inhibitors (nivolumab/ipilimumab) were enrolled in the study (Table 1). Pembrolizumab was administered intravenously at a dose of 2 mg/kg every 3 weeks, and nivolumab was administered intravenously at a dose of 3 mg/kg every 2 weeks. The combination ICI therapy was administered as an induction of 4 cycles of nivolumab (1 mg/kg) and ipilimumab (3 mg/kg) every 3 weeks, followed by single-agent nivolumab administration (3 mg/kg) every 2 weeks. The included patients had not received chemotherapy for at least 1 month prior to the initial PET/CT studies. None of the patients had a history of diabetes. Patients gave written informed consent to participate in the study and to have their medical records released. The study was approved by the Ethical Committee of the University of Heidelberg and the Federal Agency for Radiation Protection (Bundesamt für Strahlenschutz).Table 1 Patient characteristics
Patient number Age Gender LDH at baseline (U/I) Previous treatment ICI treatment EORTC PET response PERCIMT PET response Radiologic signs of irAEs Progression PFS (months)
1 56 F 770 Chemosaturation with melphalan, pembrolizumab, gemcitabine/treosulfan Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) Thyroiditis Yes 1.5
2 34 F 247 Pembrolizumab, IMCgp100 Nivolumab/ipilimumab CMR (MB) CMR (MB) Bone marrow activation, colitis No 17.9
3 46 F 166 Vemurafenib/cobimetinib Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Duodenitis Yes 4.0
4 54 M 218 Nivolumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Colitis Yes 3.2
5 53 M 275 No Nivolumab/ipilimumab SMD (MB) SMD (MB) - Yes 0.4
6 50 F 344 Vemurafenib/cobimetinib Nivolumab/ipilimumab SMD (MB) SMD (MB) Bone marrow activation, lymphadenopathy No 19.3
7 59 M 204 No Nivolumab/ipilimumab PMR (MB) PMR (MB) Sarcoid-like reaction, arthitis No 38.7
8 44 F 340 Dabrafenib/trametinib Nivolumab/ipilimumab PMR (MB) PMR (MB) Bone marrow activation, colitis No 18.3
9 60 F 269 No Nivolumab/ipilimumab PMR (MB) PMR (MB) – No 21.2
10 48 F 246 No Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Lymphadenopathy No 24.2
11 55 F 224 No Nivolumab SMD (MB) SMD (MB) - Yes 6.2
12 84 F 195 No Pembrolizumab SMD (MB) SMD (MB) - Yes 11.0
13 79 M 205 No Pembrolizumab PMD (no-MB) PMD (no-MB) Arthritis Yes 2.0
14 20 F 186 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 17.8
15 52 M 275 No Pembrolizumab PMR (MB) PMR (MB) - No 42.1
16 52 M 170 No Pembrolizumab PMD (no-MB) SMD (MB) - Yes 14.6
17 53 F 260 No Nivolumab SMD (MB) SMD (MB) - Yes 9.6
18 65 M 201 Ipilimumab Pembrolizumab SMD (MB) SMD (MB) - Yes 1.9
19 67 F 290 No Pembrolizumab PMD (no-MB) SMD (MB) Bone marrow activation, colitis Yes 5.9
20 55 M 200 Ipilimumab/nivolumab, dabrafenib/trametinib Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 1.5
21 58 M 183 Pembrolizumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Gastritis, colitis Yes 17.0
22 50 M 195 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 12.1
23 80 M 364 No Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 0.9
24 47 F 256 Dabrafenib Nivolumab SMD (MB) SMD (MB) - Yes 1.1
25 47 M 271 Ipilimumab, dabrafenib Nivolumab PMD (no-MB) SMD (MB) - Yes 2.7
F, female; M, male; LDH, lactate dehydrogenase; EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; irAEs, immune-related adverse events; PFS, progression-free survival
[18F]FDG PET/CT data acquisition
[18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of ICIs’ administration (interim PET/CT) in all 25 patients. Moreover, 17 patients of the cohort also had a third PET/CT scan within 2 weeks after administration of four cycles of treatment.
Patients underwent a whole body PET/CT after intravenous administration of maximum 250 MBq [18F]FDG 60 min post-injection (p.i.). Imaging was performed from the head to the feet with an image duration of 2 min per bed position. A dedicated PET/CT system (Biograph mCT, S128, Siemens Co., Erlangen, Germany) with an axial field of view of 21.6 cm with TruePoint and TrueV, operated in a three-dimensional mode was used. A low-dose attenuation CT (120 kV, 30 mA) was used for attenuation correction of the PET data and for image fusion. All PET images were attenuation-corrected and an image matrix of 400 × 400 pixels was used for iterative image reconstruction. Iterative image reconstruction was based on the ordered subset expectation maximization (OSEM) algorithm with two iterations and 21 subsets as well as time of flight (TOF).
[18F]FDG PET/CT data analysis
Data analysis consisted of visual (qualitative) assessment of the PET/CT scans and semi-quantitative evaluation based on standardized uptake value (SUV) calculations. PET/CT images were analyzed on an Aycan workstation by three nuclear medicine physicians (CS, DP, ADS). Images were interpreted by consensus. Visual analysis was based on the identification of sites of focal, non-physiologic [18F]FDG uptake above surrounding background activity, which were considered consistent with melanoma lesions.
Moreover, patterns of [18F]FDG uptake on interim PET/CT suggestive of radiologic manifestations of irAEs to immunotherapy were documented. Based on our experience and the published literature in the field [29, 32], we defined radiologic irAEs as sites of newly emerging, increased compared to baseline imaging, non-malignant [18F]FDG accumulation in organs known to exhibit immune-related signs on PET/CT. In particular, a new, diffusely enhanced tracer uptake in organs such as the gastrointestinal tract (mostly colon), the thyroid gland and the bone marrow, or, respectively, a new, relatively symmetrical, increased uptake in lymph nodes (e.g., mediastinal/hilar, inguinal) and in joints following ICIs were considered suggestive of radiologic irAEs in these organs. Semi-quantitative evaluation was based on volumes of interest (VOIs) and on subsequent calculation of SUVmean and SUVmax. VOIs were drawn using the pseudo-snake algorithm of the Pmod software (http://www.pmod.com/files/download/v31/doc/pbas/4729.htm) and were placed over melanoma lesions.
Response evaluation
Evaluation of patients’ response by means of [18F]FDG PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria [33] as well as the recently proposed PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT) [26]. Both criteria classify tumor response into 4 categories. The major difference between these criteria lies in the number of newly emerging lesions between baseline and follow-up PET/CT for the characterization of PMD: according to EORTC, the appearance of one new hypermetabolic lesion leads to patient classification to PMD, but according to PERCIMT, this requires the appearance of a minimum of four new lesions below 1 cm or respectively ≥ 3 new lesions of 1.0–1.5 cm or ≥ 2 new lesions of more than 1.5 cm. Another difference between the criteria is the role of SUV, which is central in EORTC, while it is not taken into account in PERCIMT (Table 2).Table 2 Summary of the EORTC and PERCIMT response criteria
EORTC PERCIMT
CMR Complete resolution of [18F]FDG uptake within the tumor volume Complete resolution of all pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
PMR Decrease in tumor SUV > 25% after more than 1 therapeutic cycle Complete resolution of some pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
SMD Increase in tumor SUV < 25% or decrease in SUV < 15% Neither PMD nor PMR/CMR
PMD Increase in tumor SUV > 25% or appearance of new lesions ≥ 4 new lesions of less than 1 cm in functional diameter or ≥ 3 new lesions between 1.0–1.5 cm in functional diameter or ≥ 2 new lesions of more than 1.5 cm in functional diameter
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; SUV, standardized uptake value
Stable disease (SD) represents a satisfactory outcome following immunotherapy, since—in contrast to conventional chemotherapy—it can be durable and survival rates related to SD are comparable to those associated with response [34–36]. Based on this, patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD.
Statistical analysis
Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause. Kaplan-Meier estimates were generated and median PFS estimated. Median follow-up time was determined by inverse Kaplan-Meier estimation. For univariate comparison of PFS, a log-rank test was used. Statistical analysis was performed using R version 4.0.2 (The R Foundation for Statistical Computing 2020) and R packages survival and prodlim. The results were considered significant for p values less than 0.05 (p < 0.05).
Results
Patient cohort
All included patients received treatment with anti-PD-1 agents, applied either as monotherapy (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients) or as combination therapy (nivolumab/ipilimumab, n = 13 patients). The mean baseline serum lactate dehydrogenase (LDH) was 262 U/l with four patients having pathologically high LDH levels and 21 of them normal levels. The detailed characteristics of the studied patients are presented in Table 1.
PET/CT findings
The findings of interim PET/CT were compared to those of the baseline scan and PET/CT-based response evaluation was performed for all 25 patients. According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), while six of them had no-MB (PMD) (Table 3) (Fig. 1).Table 3 Summary of the patients’ classifications in different response groups based on interim PET/CT and according to the EORTC and PERCIMT response criteria (n = 25 patients)
Metabolic benefit No-metabolic benefit
CMR PMR SMD PMD
EORTC 1 6 7 11
PERCIMT 1 6 12 6
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 1 Maximum intensity projection (MIP) [18F]FDG PET/CT images of a 34-year-old woman with metastatic melanoma before initiation of immunotherapy with nivolumab/ipilimumab (A) and after administration of two cycles of treatment (B). Baseline PET/CT image shows multiple lymph node, pulmonary, hepatic, adrenal, soft tissue, and osseous metastases (A). Interim PET/CT shows a complete metabolic remission (CMR) of all baseline lesions. The patient demonstrated metabolic benefit (MB) according to both the EORTC and PERCIMT criteria (B). Moreover, diffusely increased [18F]FDG uptake is observed in the ascending colon and the bone marrow on interim PET/CT. At the time of writing, the patient was still progression-free having reached a PFS of 17.9 months
With regard to the subgroup of 17 patients undergoing three PET/CT examinations (baseline, interim, late), the following results were revealed for interim PET/CT: 11 patients had MB (1 CMR, 5 PMR, and 5 SMD) and six patients had no-MB (PMD) according to EORTC, while 14 patients had MB (1 CMR, 5 PMR, and 8 SMD) and three of them had no-MB (PMD) according to PERCIMT. Respectively, on late PET/CT imaging, 13 patients had MB (3 CMR, 5 PMR, and 5 SMD) and four patients had no-MB (PMD) according to both EORTC and PERCIMT (Table 4). Two of the three patients exhibiting a mismatch between EORTC (PMD, no-MB) and PERCIMT (SMD, MB) on interim PET/CT, finally showed MB on the third examination based on both criteria (pseudoprogression) (Fig. 2). Respectively, one patient with early signs of PMD (no-MB) according to EORTC and SMD (MB) according to PERCIMT eventually exhibited PMD (no-MB) on the third examination based on both criteria.Table 4 Summary of the patients’ classifications in different response groups based on interim (after two cycles of ICIs) and late (after four cycles of ICIs) PET/CT (n = 17 patients)
EORTC PERCIMT
MB (late PET/CT) No-MB (late PET/CT) MB (late PET/CT) No-MB (late PET/CT)
CMR PMR SMD PMD CMR PMR SMD PMD
1.MB (interim PET/CT) CMR 1 0 0 0 1 0 0 0
PMR 0 5 0 0 0 5 0 0
SMD 1 0 0 4 2 0 5 1
No-MB (interim PET/CT) PMD 1 0 1 4 0 0 0 3
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; MB, metabolic benefit; no-MB, no metabolic benefit; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 2 Transaxial PET/CT (upper row, A–C) and low-dose CT (lower row, D–F) images at the cervical level of a 48-year-old female patient with advanced melanoma. The PET/CT (A) and CT (D) images obtained before immunotherapy show no pathologic lesions. Interim PET/CT performed after two cycles of nivolumab/ipilimumab shows a new [18F]FDG-avid lymph node (white arrow; B, E), suspicious of metastatic involvement. According to the EORTC criteria, the patient showed progressive metabolic disease (PMD), while according to PERCIMT, he had stable metabolic disease (SMD). A third PET/CT obtained after administration of four cycles of nivolumab/ipilimumab shows remission of the lesion (C, F), suggesting pseudoprogression of the cervical finding on interim PET/CT. The patient had a PFS of 24.2 months and was still progression-free at last contact
In total, 11 patients had PET/CT findings suggestive of irAEs on interim PET/CT, the majority of whom were under combination treatment: nine patients received nivolumab and ipilimumab, while two of them were under pembrolizumab (p = 0.015). In particular, the most common radiologic adverse event was a diffusely increased [18F]FDG uptake in the colon, defined as colitis, which was observed in five patients. One of these five patients also developed severe diarrhea, as clinical sign of treatment-induced colitis. Other gastrointestinal tract radiologic irAEs included gastritis (n = 1 patient) and duodenitis (n = 1 patient), defined as diffuse increased tracer uptake in the stomach and duodenum, respectively. Moreover, arthritis, defined as diffuse increased, periarticular, symmetrical tracer uptake in joints, was observed in two patients, and thyroiditis, a diffuse increased uptake in the thyroid gland, was observed in one patient. Furthermore, reactive, increased, symmetrical uptake in lymph nodes was observed in three patients, one of whom exhibited a sarcoid-like lymphadenopathy. Finally, diffuse increased bone marrow uptake was seen in four patients representing bone marrow activation in terms of a systemic immune response [37] (Table 1).
Survival analysis
Median follow-up (95% CI) of the patient cohort from interim PET/CT was 24.2 months (19.3–41.7 months). Patients receiving combination treatment (nivolumab/ipilimumab) had a median PFS of 17.8 months (4.0–NA), while those receiving PD-1 blockade monotherapy (nivolumab or pembrolizumab) had a median PFS of 4.3 months (1.9–NA) (p = 0.016) (Fig. 3).Fig. 3 Kaplan-Meier estimates of PFS according to the anti-PD-1 treatment applied. The numbers of patients at risk in each group and for the respective time points are shown below the plots. Combi, combination therapy (ipilimumab/nivolumab); Mono, monotherapy (nivolumab or pembrolizumab)
Based on the EORTC criteria, patients with MB on interim PET/CT had a median PFS of 14.0 months (6.2–NA), while those with no-MB had a median PFS of 4.0 months (2.0–NA) (p = 0.088) (Fig. 4A). Respectively, according to the PERCIMT criteria, patients with MB had a median PFS of 11.0 months (5.9–NA), while those with no-MB had a median PFS of 1.8 months (1.5–NA) (p = 0.045) (Fig. 4B).Fig. 4 Kaplan-Meier estimates of PFS according to the EORTC (A) and the PERCIMT (B) criteria. The numbers of patients at risk in each group and for the respective time points are shown below the plots. MB, metabolic benefit; no-MB, no metabolic benefit
The patient cohort was further dichotomized on the basis of the emergence of radiologic irAEs on interim PET/CT. Patients with irAEs had a median PFS of 17.0 months (4.0–NA), and patients without irAEs had a median PFS of 7.9 months (1.9–NA) (p = 0.128) (Fig. 5).Fig. 5 Kaplan-Meier estimates of PFS according to the emergence of radiologic irAEs on interim PET/CT. The numbers of patients at risk in each group and for the respective time points are shown below the plots
Finally, the potential correlation of baseline LDH with PFS was also investigated. However, pathologic levels of LDH had no adverse effect on survival of the cohort (p = 0.642) (Supplementary File 1).
Discussion
The therapeutic benefit of immune checkpoint blockade of PD-1 and CTLA-4 in the treatment of metastatic melanoma is variable [38]. Our understanding of how ICIs affect T cell evolution is incomplete [39], limiting the ability to derive full clinical benefit and, moreover, to predict responses from these drugs. However, tracking early response to immunotherapy is key for treatment options.
In this study, we investigated the role of interim PET/CT, performed after application of two cycles of anti-PD-1 treatment, in prediction of survival of metastatic melanoma patients. Our results showed that tumor response as classified by the PERCIMT criteria is significantly correlated with PFS, with metabolic responders demonstrating a significant survival benefit over non-responders. Respectively, the application of the EORTC criteria also led to a higher PFS for metabolic responders compared to non-responders but this difference was not statistically significant. These findings highlight the ability of [18F]FDG PET/CT—in particular, after application of the recently introduced PERCIMT criteria—to monitor and predict response to anti-PD-1 agents at an early but clinically relevant time point. This is of particular importance in clinical decision-making and prediction of outcome early during the course of immunotherapy, rendering PET/CT a potentially significant tool for the management of these patients. The main strengths of our study include its prospective nature, its rigorous protocol with imaging performed at strictly defined time points during treatment, the standardized for all patients PET/CT procedure, and the correlation with survival analysis.
Hitherto, a non-negligible number of studies have evaluated the efficacy of PET/CT in predicting treatment response of metastatic melanoma patients to ICIs. While most papers have focused on later time points during the course or after the end of treatment [26, 28, 30–32, 40], few of them also reported on application of the imaging modality early during immunotherapy. Our group previously showed in a cohort of 22 patients that PET/CT performed after two ipilimumab cycles—and after application of the EORTC criteria—correctly predicted treatment response after completion of the 4-cycle treatment in the majority (87%) of PMD patients and in all SMD patients [24]. In an expanded analysis of the ipilimumab patient cohort (n = 41 patients), the capacity of interim PET/CT in predicting clinical benefit to the agent was also highlighted. In that analysis, the performance of the PERCIMT criteria was superior to that of EORTC, which is in line with the results of the present study [27]. Furthermore, Cho et al. studied 20 melanoma patients treated with different ICIs (anti-PD-1 and anti-CTLA-4) with PET/CT at 3–4 weeks into therapy and found that a combination of changes in lesional dimensions along with changes in [18F]FDG uptake is a more accurate predictor of eventual response than each of these parameters alone. The authors proposed the PET/CT criteria for early prediction of response to immune checkpoint inhibitor therapy (PECRIT) criteria, based on a combination of the response evaluation criteria in solid tumors (RECIST) and PET response criteria in solid tumors (PERCIST) [25].
The current analysis represents the first study focusing on the role of interim PET/CT, performed as early as after application of two cycles of PD-1 blockade, in survival prediction of metastatic melanoma patients undergoing treatment with this class of ICIs. Taken together, the herein presented findings as well as those of previous studies in the field build further evidence on the potentially significant role of PET/CT performed early during the course of immunotherapy for prediction and stratification of response to treatment.
Novel patterns of response and progression, not previously seen with conventional therapies (such as cytotoxic or targeted anticancer regiments), have been described under immunotherapy and are attributed to the unique mechanism of action of these agents. In particular, phenomena such as pseudoprogression and irAEs may render response assessment to ICI challenging, questioning the utility of imaging modalities.
Evaluation of response to immunotherapy by means of PET/CT is primarily visual and subjective in nature. Considering the aforementioned challenges raised by the advent of immunotherapy for imaging interpretation, our group has recently introduced the PERCIMT criteria in metastatic melanoma, in an attempt to meet the need for reliable response assessment based on PET/CT. The cornerstone of these criteria is the finding that the absolute number of newly emerged [18F]FDG-avid lesions is more predictive of clinical outcome than SUV changes [26]. In specific, the application of a threshold of four newly emerged lesions on post-therapy PET/CT scan—with a decreasing cutoff of lesion number as the functional diameter of the lesions increases—in a cohort of 41 patients could predict clinical benefit to treatment with the agent ipilimumab better than the standard threshold of one new lesion or an increase in SUV, conventionally applied with the EORTC criteria. This was also confirmed in the present analysis with a significant correlation of metabolic response on interim PET with PFS only after application of PERCIMT.
Pseudoprogression, defined as an initial increase of tumor burden before the disease responds to therapy, has been initially described in melanoma patients undergoing ipilimumab therapy [41]. Since this phenomenon may be misclassified as progressive disease, the recently modified radiologic, immune-related response criteria (irRECIST, iRECIST) call for a 4-week re-assessment in order to overcome this limitation [42, 43]. In our study, the evaluation of pseudoprogression was partly feasible in the subgroup of 17 patients undergoing a third PET/CT after administration of four cycles of treatment. The comparison between interim and late PET/CT showed signs of pseudoprogression in 2/17 (11.8%) patients, who were characterized as PMD on interim PET/CT according to EORTC and “switched” to MB on late imaging. In contrary, no cases of pseudoprogression were observed after application of PERCIMT, highlighting the ability of the novel criteria in tackling this atypical response pattern. Moreover, the results of survival analysis, exhibiting a lower PFS for patients with early PMD (no-MB) compared to those with early MB, are another indirect proof of the rather low incidence of the phenomenon in this cohort. This is in line with previous results published in the literature documenting non-negligible, but not higher than 10% rates of the phenomenon in melanoma immunotherapy [36, 44, 45], and provides supporting evidence to the standpoint that an increase in tumor burden observed during ICI treatment more likely reflects true progression rather than pseudoprogression [41, 46, 47].
irAEs represent another source of false-positive findings on imaging. Radiologic manifestations of irAEs have been reported with variable incidences, reaching up to 31% of patients under ICIs [47–49]. Although the specific characteristics of individual patients play a significant role, the emergence of these toxicities is mainly dependent on the agents used, with the combination of an anti-CTLA-4 antibody and an anti-PD-1 antibody increasing both their incidence and severity [50]. In the herein studied cohort, 11/25 (44%) of the patients showed signs of irAEs on PET/CT. In line with data from the literature [50], the vast majority of these patients (9/11 patients) received combination treatment of nivolumab and ipilimumab. Colitis was the most frequent adverse event on PET/CT, observed in five patients; four of these patients received PD-1 inhibitors in combination with the anti-CTLA-4 regiment ipilimumab, which is known to often induce this reaction [51, 52]. We recognize that the diagnosis of colitis on PET imaging can be complicated, since its identification can be hampered by physiological metabolic activity in the colon. It is well known that enhanced colon [18F]FDG uptake of benign etiology is frequently observed in asymptomatic individuals [53, 54]. Moreover, several studies have shown that patients using the oral hypoglycemic drug metformin tend to have a diffusely increased tracer uptake in the colon [55–58]. In the present cohort, no patient had diabetes; thus, metformin can be ruled out as a cause for false-positive [18F]FDG accumulation in the colon. A further search in patients’ clinical history revealed that one of these five patients developed severe diarrhea during immunotherapy, most likely as a symptom of treatment-induced colitis, while the rest four patients did not have such symptoms. This higher incidence of “PET-colitis” under immunotherapy, compared to clinical signs of colitis, is in line with previously published results [59]. In this context, early recognition of radiologic irAEs could be potentially important since they may precede or correlate with clinical symptoms [37, 47], potentially leading to respective changes in management.
Another aspect pertaining to irAEs is that their emergence has been associated with a favorable efficacy of ICIs—mainly of PD-1 inhibitors—implying a potential predictive role of these events for response to ICI treatment [17, 18, 60]. Our analysis revealed that patients with radiologic signs of irAEs had a longer PFS than those without irAEs; however, this difference was non-significant. Apart from the relatively small cohort studied, an explanation for this finding may lie in the fact that most patients (82%) also received the CTLA-4 inhibitor ipilimumab in combination with the PD-1 inhibitor nivolumab; in a recently published meta-analysis of 30 studies including 4971 individuals, it was shown that no significant association between irAE development and a favorable benefit on survival is observed in ICI combination treatments, in contrary to anti-PD-1 monotherapies [60]. Another reason may lie in the nature of the observed irAEs, affecting the gastrointestinal tract in more than half of the cases (6/11 patients, 55%); data from the above-mentioned meta-analysis also highlight the lack of PFS benefit in patients presenting gastrointestinal irAEs [60].
Finally, the predictive role of baseline LDH before initiation of PD-1 inhibitors was investigated. Although serum LDH elevation is not specific for melanoma, it represents a poor prognostic factor and is one of the most influential factors associated with treatment response [61, 62]. An interesting finding of our analysis is the lack of any adverse effect of elevated baseline LDH on survival, which can be however attributed to the small number of patients with pathologic LDH (n = 4 patients, 16%).
We note some limitations in our study. Firstly, due to the strict inclusion criteria applied, the number of included patients was relatively low, not allowing us to draw more firm conclusions; ideally, further studies with larger patient cohorts would be required. Secondly, although the focus of the study was anti-PD-1 treatment, not all patients underwent exclusively PD-1 blockade, with several of them receiving combination therapy of PD-1 and CTLA-4 inhibitors. Although our patient cohort is too small to afford a PET/CT subanalysis based on a dichotomization of patients into those receiving anti-PD-1 monotherapy and those undergoing combined treatment, we highlight the similar approach followed in previous studies in the field [28, 63]. Finally, the vast majority of the PET/CT-positive, melanoma-consistent findings were not histopathologically confirmed. However, this is not usually possible in the clinical setting.
Conclusion
In an attempt to identify early and reliable biomarkers of survival prediction in immunotherapy of metastatic melanoma, we assessed the prognostic role of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment. Our results showed that tumor response as classified by the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Supplementary information
ESM 1 (DOCX 166 kb)
Funding
Open Access funding enabled and organized by Projekt DEAL.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all participants enrolled in the study.
This article is part of the Topical Collection on Oncology - General.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | IPILIMUMAB, NIVOLUMAB | DrugsGivenReaction | CC BY | 33336264 | 18,710,997 | 2021-06 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Interim [18F]FDG PET/CT can predict response to anti-PD-1 treatment in metastatic melanoma.
In an attempt to identify biomarkers that can reliably predict long-term outcomes to immunotherapy in metastatic melanoma, we investigated the prognostic role of [18F]FDG PET/CT, performed at baseline and early during the course of anti-PD-1 treatment.
Twenty-five patients with stage IV melanoma, scheduled for treatment with PD-1 inhibitors, were enrolled in the study (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients; nivolumab/ipilimumab, 13 patients). [18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of PD-1 blockade administration (interim PET/CT). Seventeen patients underwent also a third PET/CT scan after administration of four cycles of treatment. Evaluation of patients' response by means of PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria and the PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT). Response to treatment was classified into 4 categories: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD). Patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD. Moreover, patterns of [18F]FDG uptake suggestive of radiologic immune-related adverse events (irAEs) were documented. Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause.
Median follow-up from interim PET/CT was 24.2 months (19.3-41.7 months). According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), and 6 of them had no-MB (PMD). With regard to PFS, no significant difference was observed between patients with MB and no-MB on interim PET/CT according to the EORTC criteria (p = 0.088). In contrary, according to the PERCIMT criteria, patients demonstrating MB had a significantly longer PFS than those showing no-MB (p = 0.045). The emergence of radiologic irAEs (n = 11 patients) was not associated with a significant survival benefit. Regarding the sub-cohort undergoing also a third PET/CT, 14/17 patients (82%) showed concordant responses and 3/17 (18%) had a mismatch of response assessment between interim and late PET/CT.
PET/CT-based response of metastatic melanoma to PD-1 blockade after application of the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Introduction
Metastatic melanoma is a highly aggressive tumor, largely refractory to existing therapies, and associated with a very poor prognosis [1]. While until lately the treatment options for metastatic melanoma were limited, the recent development and introduction in clinical practice of several novel immunotherapeutic agents as well as of targeted therapy with BRAF and MEK inhibitors have revolutionized the systemic treatment of the disease, leading to unprecedented response and survival rates of melanoma patients [2].
The main form of immunotherapy applied in this new era of melanoma management involves immune checkpoint blockade. This immunomodulatory approach activates the immune system against tumors through the binding of the cytotoxic T lymphocyte–associated protein 4 (CTLA-4) and/or the programmed cell death protein 1 (PD-1), both of which are expressed by T cells [3, 4]. The monoclonal antibody ipilimumab, which acts by blocking CTLA-4, is considered a landmark agent in this context, being the first immunotherapeutic drug demonstrating a clear benefit in survival of patients with advanced melanoma, which led to its approval by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2011 [5]. A few years later, a second class of immune checkpoint inhibitors (ICIs), the PD-1 inhibitors nivolumab and pembrolizumab, were also approved for the treatment of melanoma, after having shown survival benefit in these patients [6–8]. Moreover, the anti-PD-1 monoclonal antibodies have shown superiority over ipilimumab, leading to their application both as single agents and in combination with ipilimumab, which is nowadays seldom used as monotherapy [9–13].
Despite these dramatic improvements, a significant amount of patients—approximately 40–45%—show no response to immunotherapy [14]. Additionally, the mechanism of action of these agents (which is markedly different from usual cytotoxic approaches—notably by generating inflammations rather than direct lysis) can pose relevant challenges in the interpretation of treatment response by conventional imaging approaches [15]. Furthermore, several patients experience a “new class” of cumulative, dose-dependent, and sometimes life-threatening side effects, the immune-related adverse events (irAEs), which scope is wide [16]; importantly, the occurrence of such irAEs may be of prognostic value, revealing a response to immunotherapy [17, 18]. These issues raise the question of how to evaluate the response to ICIs in a reliable fashion and early in the course of treatment. This information would help discriminate responders from non-responders, offering significant therapeutic and prognostic implications in the entire spectrum of patient management. Unfortunately, there exist at present only few reliable predictors of long-term response to immunotherapy.
[18F]FDG PET/CT is considered the elective imaging technique in detecting metastatic disease in advanced melanoma [19–23]. Moreover, a growing amount of recently published literature has highlighted the potential role of the modality in the prediction of treatment response to ICIs in melanoma, rendering it an attractive tool for the monitoring of immunotherapy [24–31].
In quest of identifying reliable biomarkers for the prediction of long-term outcomes to immunotherapy, we aim in the present prospective study to assess the value of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment.
Materials and methods
Patients
Twenty-five patients (12 males, 13 females; mean age 54.7 years) with unresectable, stage IV melanoma undergoing immunotherapy with PD-1 inhibitors applied either as monotherapy (pembrolizumab, nivolumab) or as combination treatment with CTLA-4 inhibitors (nivolumab/ipilimumab) were enrolled in the study (Table 1). Pembrolizumab was administered intravenously at a dose of 2 mg/kg every 3 weeks, and nivolumab was administered intravenously at a dose of 3 mg/kg every 2 weeks. The combination ICI therapy was administered as an induction of 4 cycles of nivolumab (1 mg/kg) and ipilimumab (3 mg/kg) every 3 weeks, followed by single-agent nivolumab administration (3 mg/kg) every 2 weeks. The included patients had not received chemotherapy for at least 1 month prior to the initial PET/CT studies. None of the patients had a history of diabetes. Patients gave written informed consent to participate in the study and to have their medical records released. The study was approved by the Ethical Committee of the University of Heidelberg and the Federal Agency for Radiation Protection (Bundesamt für Strahlenschutz).Table 1 Patient characteristics
Patient number Age Gender LDH at baseline (U/I) Previous treatment ICI treatment EORTC PET response PERCIMT PET response Radiologic signs of irAEs Progression PFS (months)
1 56 F 770 Chemosaturation with melphalan, pembrolizumab, gemcitabine/treosulfan Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) Thyroiditis Yes 1.5
2 34 F 247 Pembrolizumab, IMCgp100 Nivolumab/ipilimumab CMR (MB) CMR (MB) Bone marrow activation, colitis No 17.9
3 46 F 166 Vemurafenib/cobimetinib Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Duodenitis Yes 4.0
4 54 M 218 Nivolumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Colitis Yes 3.2
5 53 M 275 No Nivolumab/ipilimumab SMD (MB) SMD (MB) - Yes 0.4
6 50 F 344 Vemurafenib/cobimetinib Nivolumab/ipilimumab SMD (MB) SMD (MB) Bone marrow activation, lymphadenopathy No 19.3
7 59 M 204 No Nivolumab/ipilimumab PMR (MB) PMR (MB) Sarcoid-like reaction, arthitis No 38.7
8 44 F 340 Dabrafenib/trametinib Nivolumab/ipilimumab PMR (MB) PMR (MB) Bone marrow activation, colitis No 18.3
9 60 F 269 No Nivolumab/ipilimumab PMR (MB) PMR (MB) – No 21.2
10 48 F 246 No Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Lymphadenopathy No 24.2
11 55 F 224 No Nivolumab SMD (MB) SMD (MB) - Yes 6.2
12 84 F 195 No Pembrolizumab SMD (MB) SMD (MB) - Yes 11.0
13 79 M 205 No Pembrolizumab PMD (no-MB) PMD (no-MB) Arthritis Yes 2.0
14 20 F 186 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 17.8
15 52 M 275 No Pembrolizumab PMR (MB) PMR (MB) - No 42.1
16 52 M 170 No Pembrolizumab PMD (no-MB) SMD (MB) - Yes 14.6
17 53 F 260 No Nivolumab SMD (MB) SMD (MB) - Yes 9.6
18 65 M 201 Ipilimumab Pembrolizumab SMD (MB) SMD (MB) - Yes 1.9
19 67 F 290 No Pembrolizumab PMD (no-MB) SMD (MB) Bone marrow activation, colitis Yes 5.9
20 55 M 200 Ipilimumab/nivolumab, dabrafenib/trametinib Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 1.5
21 58 M 183 Pembrolizumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Gastritis, colitis Yes 17.0
22 50 M 195 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 12.1
23 80 M 364 No Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 0.9
24 47 F 256 Dabrafenib Nivolumab SMD (MB) SMD (MB) - Yes 1.1
25 47 M 271 Ipilimumab, dabrafenib Nivolumab PMD (no-MB) SMD (MB) - Yes 2.7
F, female; M, male; LDH, lactate dehydrogenase; EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; irAEs, immune-related adverse events; PFS, progression-free survival
[18F]FDG PET/CT data acquisition
[18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of ICIs’ administration (interim PET/CT) in all 25 patients. Moreover, 17 patients of the cohort also had a third PET/CT scan within 2 weeks after administration of four cycles of treatment.
Patients underwent a whole body PET/CT after intravenous administration of maximum 250 MBq [18F]FDG 60 min post-injection (p.i.). Imaging was performed from the head to the feet with an image duration of 2 min per bed position. A dedicated PET/CT system (Biograph mCT, S128, Siemens Co., Erlangen, Germany) with an axial field of view of 21.6 cm with TruePoint and TrueV, operated in a three-dimensional mode was used. A low-dose attenuation CT (120 kV, 30 mA) was used for attenuation correction of the PET data and for image fusion. All PET images were attenuation-corrected and an image matrix of 400 × 400 pixels was used for iterative image reconstruction. Iterative image reconstruction was based on the ordered subset expectation maximization (OSEM) algorithm with two iterations and 21 subsets as well as time of flight (TOF).
[18F]FDG PET/CT data analysis
Data analysis consisted of visual (qualitative) assessment of the PET/CT scans and semi-quantitative evaluation based on standardized uptake value (SUV) calculations. PET/CT images were analyzed on an Aycan workstation by three nuclear medicine physicians (CS, DP, ADS). Images were interpreted by consensus. Visual analysis was based on the identification of sites of focal, non-physiologic [18F]FDG uptake above surrounding background activity, which were considered consistent with melanoma lesions.
Moreover, patterns of [18F]FDG uptake on interim PET/CT suggestive of radiologic manifestations of irAEs to immunotherapy were documented. Based on our experience and the published literature in the field [29, 32], we defined radiologic irAEs as sites of newly emerging, increased compared to baseline imaging, non-malignant [18F]FDG accumulation in organs known to exhibit immune-related signs on PET/CT. In particular, a new, diffusely enhanced tracer uptake in organs such as the gastrointestinal tract (mostly colon), the thyroid gland and the bone marrow, or, respectively, a new, relatively symmetrical, increased uptake in lymph nodes (e.g., mediastinal/hilar, inguinal) and in joints following ICIs were considered suggestive of radiologic irAEs in these organs. Semi-quantitative evaluation was based on volumes of interest (VOIs) and on subsequent calculation of SUVmean and SUVmax. VOIs were drawn using the pseudo-snake algorithm of the Pmod software (http://www.pmod.com/files/download/v31/doc/pbas/4729.htm) and were placed over melanoma lesions.
Response evaluation
Evaluation of patients’ response by means of [18F]FDG PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria [33] as well as the recently proposed PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT) [26]. Both criteria classify tumor response into 4 categories. The major difference between these criteria lies in the number of newly emerging lesions between baseline and follow-up PET/CT for the characterization of PMD: according to EORTC, the appearance of one new hypermetabolic lesion leads to patient classification to PMD, but according to PERCIMT, this requires the appearance of a minimum of four new lesions below 1 cm or respectively ≥ 3 new lesions of 1.0–1.5 cm or ≥ 2 new lesions of more than 1.5 cm. Another difference between the criteria is the role of SUV, which is central in EORTC, while it is not taken into account in PERCIMT (Table 2).Table 2 Summary of the EORTC and PERCIMT response criteria
EORTC PERCIMT
CMR Complete resolution of [18F]FDG uptake within the tumor volume Complete resolution of all pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
PMR Decrease in tumor SUV > 25% after more than 1 therapeutic cycle Complete resolution of some pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
SMD Increase in tumor SUV < 25% or decrease in SUV < 15% Neither PMD nor PMR/CMR
PMD Increase in tumor SUV > 25% or appearance of new lesions ≥ 4 new lesions of less than 1 cm in functional diameter or ≥ 3 new lesions between 1.0–1.5 cm in functional diameter or ≥ 2 new lesions of more than 1.5 cm in functional diameter
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; SUV, standardized uptake value
Stable disease (SD) represents a satisfactory outcome following immunotherapy, since—in contrast to conventional chemotherapy—it can be durable and survival rates related to SD are comparable to those associated with response [34–36]. Based on this, patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD.
Statistical analysis
Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause. Kaplan-Meier estimates were generated and median PFS estimated. Median follow-up time was determined by inverse Kaplan-Meier estimation. For univariate comparison of PFS, a log-rank test was used. Statistical analysis was performed using R version 4.0.2 (The R Foundation for Statistical Computing 2020) and R packages survival and prodlim. The results were considered significant for p values less than 0.05 (p < 0.05).
Results
Patient cohort
All included patients received treatment with anti-PD-1 agents, applied either as monotherapy (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients) or as combination therapy (nivolumab/ipilimumab, n = 13 patients). The mean baseline serum lactate dehydrogenase (LDH) was 262 U/l with four patients having pathologically high LDH levels and 21 of them normal levels. The detailed characteristics of the studied patients are presented in Table 1.
PET/CT findings
The findings of interim PET/CT were compared to those of the baseline scan and PET/CT-based response evaluation was performed for all 25 patients. According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), while six of them had no-MB (PMD) (Table 3) (Fig. 1).Table 3 Summary of the patients’ classifications in different response groups based on interim PET/CT and according to the EORTC and PERCIMT response criteria (n = 25 patients)
Metabolic benefit No-metabolic benefit
CMR PMR SMD PMD
EORTC 1 6 7 11
PERCIMT 1 6 12 6
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 1 Maximum intensity projection (MIP) [18F]FDG PET/CT images of a 34-year-old woman with metastatic melanoma before initiation of immunotherapy with nivolumab/ipilimumab (A) and after administration of two cycles of treatment (B). Baseline PET/CT image shows multiple lymph node, pulmonary, hepatic, adrenal, soft tissue, and osseous metastases (A). Interim PET/CT shows a complete metabolic remission (CMR) of all baseline lesions. The patient demonstrated metabolic benefit (MB) according to both the EORTC and PERCIMT criteria (B). Moreover, diffusely increased [18F]FDG uptake is observed in the ascending colon and the bone marrow on interim PET/CT. At the time of writing, the patient was still progression-free having reached a PFS of 17.9 months
With regard to the subgroup of 17 patients undergoing three PET/CT examinations (baseline, interim, late), the following results were revealed for interim PET/CT: 11 patients had MB (1 CMR, 5 PMR, and 5 SMD) and six patients had no-MB (PMD) according to EORTC, while 14 patients had MB (1 CMR, 5 PMR, and 8 SMD) and three of them had no-MB (PMD) according to PERCIMT. Respectively, on late PET/CT imaging, 13 patients had MB (3 CMR, 5 PMR, and 5 SMD) and four patients had no-MB (PMD) according to both EORTC and PERCIMT (Table 4). Two of the three patients exhibiting a mismatch between EORTC (PMD, no-MB) and PERCIMT (SMD, MB) on interim PET/CT, finally showed MB on the third examination based on both criteria (pseudoprogression) (Fig. 2). Respectively, one patient with early signs of PMD (no-MB) according to EORTC and SMD (MB) according to PERCIMT eventually exhibited PMD (no-MB) on the third examination based on both criteria.Table 4 Summary of the patients’ classifications in different response groups based on interim (after two cycles of ICIs) and late (after four cycles of ICIs) PET/CT (n = 17 patients)
EORTC PERCIMT
MB (late PET/CT) No-MB (late PET/CT) MB (late PET/CT) No-MB (late PET/CT)
CMR PMR SMD PMD CMR PMR SMD PMD
1.MB (interim PET/CT) CMR 1 0 0 0 1 0 0 0
PMR 0 5 0 0 0 5 0 0
SMD 1 0 0 4 2 0 5 1
No-MB (interim PET/CT) PMD 1 0 1 4 0 0 0 3
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; MB, metabolic benefit; no-MB, no metabolic benefit; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 2 Transaxial PET/CT (upper row, A–C) and low-dose CT (lower row, D–F) images at the cervical level of a 48-year-old female patient with advanced melanoma. The PET/CT (A) and CT (D) images obtained before immunotherapy show no pathologic lesions. Interim PET/CT performed after two cycles of nivolumab/ipilimumab shows a new [18F]FDG-avid lymph node (white arrow; B, E), suspicious of metastatic involvement. According to the EORTC criteria, the patient showed progressive metabolic disease (PMD), while according to PERCIMT, he had stable metabolic disease (SMD). A third PET/CT obtained after administration of four cycles of nivolumab/ipilimumab shows remission of the lesion (C, F), suggesting pseudoprogression of the cervical finding on interim PET/CT. The patient had a PFS of 24.2 months and was still progression-free at last contact
In total, 11 patients had PET/CT findings suggestive of irAEs on interim PET/CT, the majority of whom were under combination treatment: nine patients received nivolumab and ipilimumab, while two of them were under pembrolizumab (p = 0.015). In particular, the most common radiologic adverse event was a diffusely increased [18F]FDG uptake in the colon, defined as colitis, which was observed in five patients. One of these five patients also developed severe diarrhea, as clinical sign of treatment-induced colitis. Other gastrointestinal tract radiologic irAEs included gastritis (n = 1 patient) and duodenitis (n = 1 patient), defined as diffuse increased tracer uptake in the stomach and duodenum, respectively. Moreover, arthritis, defined as diffuse increased, periarticular, symmetrical tracer uptake in joints, was observed in two patients, and thyroiditis, a diffuse increased uptake in the thyroid gland, was observed in one patient. Furthermore, reactive, increased, symmetrical uptake in lymph nodes was observed in three patients, one of whom exhibited a sarcoid-like lymphadenopathy. Finally, diffuse increased bone marrow uptake was seen in four patients representing bone marrow activation in terms of a systemic immune response [37] (Table 1).
Survival analysis
Median follow-up (95% CI) of the patient cohort from interim PET/CT was 24.2 months (19.3–41.7 months). Patients receiving combination treatment (nivolumab/ipilimumab) had a median PFS of 17.8 months (4.0–NA), while those receiving PD-1 blockade monotherapy (nivolumab or pembrolizumab) had a median PFS of 4.3 months (1.9–NA) (p = 0.016) (Fig. 3).Fig. 3 Kaplan-Meier estimates of PFS according to the anti-PD-1 treatment applied. The numbers of patients at risk in each group and for the respective time points are shown below the plots. Combi, combination therapy (ipilimumab/nivolumab); Mono, monotherapy (nivolumab or pembrolizumab)
Based on the EORTC criteria, patients with MB on interim PET/CT had a median PFS of 14.0 months (6.2–NA), while those with no-MB had a median PFS of 4.0 months (2.0–NA) (p = 0.088) (Fig. 4A). Respectively, according to the PERCIMT criteria, patients with MB had a median PFS of 11.0 months (5.9–NA), while those with no-MB had a median PFS of 1.8 months (1.5–NA) (p = 0.045) (Fig. 4B).Fig. 4 Kaplan-Meier estimates of PFS according to the EORTC (A) and the PERCIMT (B) criteria. The numbers of patients at risk in each group and for the respective time points are shown below the plots. MB, metabolic benefit; no-MB, no metabolic benefit
The patient cohort was further dichotomized on the basis of the emergence of radiologic irAEs on interim PET/CT. Patients with irAEs had a median PFS of 17.0 months (4.0–NA), and patients without irAEs had a median PFS of 7.9 months (1.9–NA) (p = 0.128) (Fig. 5).Fig. 5 Kaplan-Meier estimates of PFS according to the emergence of radiologic irAEs on interim PET/CT. The numbers of patients at risk in each group and for the respective time points are shown below the plots
Finally, the potential correlation of baseline LDH with PFS was also investigated. However, pathologic levels of LDH had no adverse effect on survival of the cohort (p = 0.642) (Supplementary File 1).
Discussion
The therapeutic benefit of immune checkpoint blockade of PD-1 and CTLA-4 in the treatment of metastatic melanoma is variable [38]. Our understanding of how ICIs affect T cell evolution is incomplete [39], limiting the ability to derive full clinical benefit and, moreover, to predict responses from these drugs. However, tracking early response to immunotherapy is key for treatment options.
In this study, we investigated the role of interim PET/CT, performed after application of two cycles of anti-PD-1 treatment, in prediction of survival of metastatic melanoma patients. Our results showed that tumor response as classified by the PERCIMT criteria is significantly correlated with PFS, with metabolic responders demonstrating a significant survival benefit over non-responders. Respectively, the application of the EORTC criteria also led to a higher PFS for metabolic responders compared to non-responders but this difference was not statistically significant. These findings highlight the ability of [18F]FDG PET/CT—in particular, after application of the recently introduced PERCIMT criteria—to monitor and predict response to anti-PD-1 agents at an early but clinically relevant time point. This is of particular importance in clinical decision-making and prediction of outcome early during the course of immunotherapy, rendering PET/CT a potentially significant tool for the management of these patients. The main strengths of our study include its prospective nature, its rigorous protocol with imaging performed at strictly defined time points during treatment, the standardized for all patients PET/CT procedure, and the correlation with survival analysis.
Hitherto, a non-negligible number of studies have evaluated the efficacy of PET/CT in predicting treatment response of metastatic melanoma patients to ICIs. While most papers have focused on later time points during the course or after the end of treatment [26, 28, 30–32, 40], few of them also reported on application of the imaging modality early during immunotherapy. Our group previously showed in a cohort of 22 patients that PET/CT performed after two ipilimumab cycles—and after application of the EORTC criteria—correctly predicted treatment response after completion of the 4-cycle treatment in the majority (87%) of PMD patients and in all SMD patients [24]. In an expanded analysis of the ipilimumab patient cohort (n = 41 patients), the capacity of interim PET/CT in predicting clinical benefit to the agent was also highlighted. In that analysis, the performance of the PERCIMT criteria was superior to that of EORTC, which is in line with the results of the present study [27]. Furthermore, Cho et al. studied 20 melanoma patients treated with different ICIs (anti-PD-1 and anti-CTLA-4) with PET/CT at 3–4 weeks into therapy and found that a combination of changes in lesional dimensions along with changes in [18F]FDG uptake is a more accurate predictor of eventual response than each of these parameters alone. The authors proposed the PET/CT criteria for early prediction of response to immune checkpoint inhibitor therapy (PECRIT) criteria, based on a combination of the response evaluation criteria in solid tumors (RECIST) and PET response criteria in solid tumors (PERCIST) [25].
The current analysis represents the first study focusing on the role of interim PET/CT, performed as early as after application of two cycles of PD-1 blockade, in survival prediction of metastatic melanoma patients undergoing treatment with this class of ICIs. Taken together, the herein presented findings as well as those of previous studies in the field build further evidence on the potentially significant role of PET/CT performed early during the course of immunotherapy for prediction and stratification of response to treatment.
Novel patterns of response and progression, not previously seen with conventional therapies (such as cytotoxic or targeted anticancer regiments), have been described under immunotherapy and are attributed to the unique mechanism of action of these agents. In particular, phenomena such as pseudoprogression and irAEs may render response assessment to ICI challenging, questioning the utility of imaging modalities.
Evaluation of response to immunotherapy by means of PET/CT is primarily visual and subjective in nature. Considering the aforementioned challenges raised by the advent of immunotherapy for imaging interpretation, our group has recently introduced the PERCIMT criteria in metastatic melanoma, in an attempt to meet the need for reliable response assessment based on PET/CT. The cornerstone of these criteria is the finding that the absolute number of newly emerged [18F]FDG-avid lesions is more predictive of clinical outcome than SUV changes [26]. In specific, the application of a threshold of four newly emerged lesions on post-therapy PET/CT scan—with a decreasing cutoff of lesion number as the functional diameter of the lesions increases—in a cohort of 41 patients could predict clinical benefit to treatment with the agent ipilimumab better than the standard threshold of one new lesion or an increase in SUV, conventionally applied with the EORTC criteria. This was also confirmed in the present analysis with a significant correlation of metabolic response on interim PET with PFS only after application of PERCIMT.
Pseudoprogression, defined as an initial increase of tumor burden before the disease responds to therapy, has been initially described in melanoma patients undergoing ipilimumab therapy [41]. Since this phenomenon may be misclassified as progressive disease, the recently modified radiologic, immune-related response criteria (irRECIST, iRECIST) call for a 4-week re-assessment in order to overcome this limitation [42, 43]. In our study, the evaluation of pseudoprogression was partly feasible in the subgroup of 17 patients undergoing a third PET/CT after administration of four cycles of treatment. The comparison between interim and late PET/CT showed signs of pseudoprogression in 2/17 (11.8%) patients, who were characterized as PMD on interim PET/CT according to EORTC and “switched” to MB on late imaging. In contrary, no cases of pseudoprogression were observed after application of PERCIMT, highlighting the ability of the novel criteria in tackling this atypical response pattern. Moreover, the results of survival analysis, exhibiting a lower PFS for patients with early PMD (no-MB) compared to those with early MB, are another indirect proof of the rather low incidence of the phenomenon in this cohort. This is in line with previous results published in the literature documenting non-negligible, but not higher than 10% rates of the phenomenon in melanoma immunotherapy [36, 44, 45], and provides supporting evidence to the standpoint that an increase in tumor burden observed during ICI treatment more likely reflects true progression rather than pseudoprogression [41, 46, 47].
irAEs represent another source of false-positive findings on imaging. Radiologic manifestations of irAEs have been reported with variable incidences, reaching up to 31% of patients under ICIs [47–49]. Although the specific characteristics of individual patients play a significant role, the emergence of these toxicities is mainly dependent on the agents used, with the combination of an anti-CTLA-4 antibody and an anti-PD-1 antibody increasing both their incidence and severity [50]. In the herein studied cohort, 11/25 (44%) of the patients showed signs of irAEs on PET/CT. In line with data from the literature [50], the vast majority of these patients (9/11 patients) received combination treatment of nivolumab and ipilimumab. Colitis was the most frequent adverse event on PET/CT, observed in five patients; four of these patients received PD-1 inhibitors in combination with the anti-CTLA-4 regiment ipilimumab, which is known to often induce this reaction [51, 52]. We recognize that the diagnosis of colitis on PET imaging can be complicated, since its identification can be hampered by physiological metabolic activity in the colon. It is well known that enhanced colon [18F]FDG uptake of benign etiology is frequently observed in asymptomatic individuals [53, 54]. Moreover, several studies have shown that patients using the oral hypoglycemic drug metformin tend to have a diffusely increased tracer uptake in the colon [55–58]. In the present cohort, no patient had diabetes; thus, metformin can be ruled out as a cause for false-positive [18F]FDG accumulation in the colon. A further search in patients’ clinical history revealed that one of these five patients developed severe diarrhea during immunotherapy, most likely as a symptom of treatment-induced colitis, while the rest four patients did not have such symptoms. This higher incidence of “PET-colitis” under immunotherapy, compared to clinical signs of colitis, is in line with previously published results [59]. In this context, early recognition of radiologic irAEs could be potentially important since they may precede or correlate with clinical symptoms [37, 47], potentially leading to respective changes in management.
Another aspect pertaining to irAEs is that their emergence has been associated with a favorable efficacy of ICIs—mainly of PD-1 inhibitors—implying a potential predictive role of these events for response to ICI treatment [17, 18, 60]. Our analysis revealed that patients with radiologic signs of irAEs had a longer PFS than those without irAEs; however, this difference was non-significant. Apart from the relatively small cohort studied, an explanation for this finding may lie in the fact that most patients (82%) also received the CTLA-4 inhibitor ipilimumab in combination with the PD-1 inhibitor nivolumab; in a recently published meta-analysis of 30 studies including 4971 individuals, it was shown that no significant association between irAE development and a favorable benefit on survival is observed in ICI combination treatments, in contrary to anti-PD-1 monotherapies [60]. Another reason may lie in the nature of the observed irAEs, affecting the gastrointestinal tract in more than half of the cases (6/11 patients, 55%); data from the above-mentioned meta-analysis also highlight the lack of PFS benefit in patients presenting gastrointestinal irAEs [60].
Finally, the predictive role of baseline LDH before initiation of PD-1 inhibitors was investigated. Although serum LDH elevation is not specific for melanoma, it represents a poor prognostic factor and is one of the most influential factors associated with treatment response [61, 62]. An interesting finding of our analysis is the lack of any adverse effect of elevated baseline LDH on survival, which can be however attributed to the small number of patients with pathologic LDH (n = 4 patients, 16%).
We note some limitations in our study. Firstly, due to the strict inclusion criteria applied, the number of included patients was relatively low, not allowing us to draw more firm conclusions; ideally, further studies with larger patient cohorts would be required. Secondly, although the focus of the study was anti-PD-1 treatment, not all patients underwent exclusively PD-1 blockade, with several of them receiving combination therapy of PD-1 and CTLA-4 inhibitors. Although our patient cohort is too small to afford a PET/CT subanalysis based on a dichotomization of patients into those receiving anti-PD-1 monotherapy and those undergoing combined treatment, we highlight the similar approach followed in previous studies in the field [28, 63]. Finally, the vast majority of the PET/CT-positive, melanoma-consistent findings were not histopathologically confirmed. However, this is not usually possible in the clinical setting.
Conclusion
In an attempt to identify early and reliable biomarkers of survival prediction in immunotherapy of metastatic melanoma, we assessed the prognostic role of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment. Our results showed that tumor response as classified by the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Supplementary information
ESM 1 (DOCX 166 kb)
Funding
Open Access funding enabled and organized by Projekt DEAL.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all participants enrolled in the study.
This article is part of the Topical Collection on Oncology - General.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | IPILIMUMAB, NIVOLUMAB | DrugsGivenReaction | CC BY | 33336264 | 18,710,997 | 2021-06 |
What was the administration route of drug 'IPILIMUMAB'? | Interim [18F]FDG PET/CT can predict response to anti-PD-1 treatment in metastatic melanoma.
In an attempt to identify biomarkers that can reliably predict long-term outcomes to immunotherapy in metastatic melanoma, we investigated the prognostic role of [18F]FDG PET/CT, performed at baseline and early during the course of anti-PD-1 treatment.
Twenty-five patients with stage IV melanoma, scheduled for treatment with PD-1 inhibitors, were enrolled in the study (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients; nivolumab/ipilimumab, 13 patients). [18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of PD-1 blockade administration (interim PET/CT). Seventeen patients underwent also a third PET/CT scan after administration of four cycles of treatment. Evaluation of patients' response by means of PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria and the PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT). Response to treatment was classified into 4 categories: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD). Patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD. Moreover, patterns of [18F]FDG uptake suggestive of radiologic immune-related adverse events (irAEs) were documented. Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause.
Median follow-up from interim PET/CT was 24.2 months (19.3-41.7 months). According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), and 6 of them had no-MB (PMD). With regard to PFS, no significant difference was observed between patients with MB and no-MB on interim PET/CT according to the EORTC criteria (p = 0.088). In contrary, according to the PERCIMT criteria, patients demonstrating MB had a significantly longer PFS than those showing no-MB (p = 0.045). The emergence of radiologic irAEs (n = 11 patients) was not associated with a significant survival benefit. Regarding the sub-cohort undergoing also a third PET/CT, 14/17 patients (82%) showed concordant responses and 3/17 (18%) had a mismatch of response assessment between interim and late PET/CT.
PET/CT-based response of metastatic melanoma to PD-1 blockade after application of the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Introduction
Metastatic melanoma is a highly aggressive tumor, largely refractory to existing therapies, and associated with a very poor prognosis [1]. While until lately the treatment options for metastatic melanoma were limited, the recent development and introduction in clinical practice of several novel immunotherapeutic agents as well as of targeted therapy with BRAF and MEK inhibitors have revolutionized the systemic treatment of the disease, leading to unprecedented response and survival rates of melanoma patients [2].
The main form of immunotherapy applied in this new era of melanoma management involves immune checkpoint blockade. This immunomodulatory approach activates the immune system against tumors through the binding of the cytotoxic T lymphocyte–associated protein 4 (CTLA-4) and/or the programmed cell death protein 1 (PD-1), both of which are expressed by T cells [3, 4]. The monoclonal antibody ipilimumab, which acts by blocking CTLA-4, is considered a landmark agent in this context, being the first immunotherapeutic drug demonstrating a clear benefit in survival of patients with advanced melanoma, which led to its approval by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2011 [5]. A few years later, a second class of immune checkpoint inhibitors (ICIs), the PD-1 inhibitors nivolumab and pembrolizumab, were also approved for the treatment of melanoma, after having shown survival benefit in these patients [6–8]. Moreover, the anti-PD-1 monoclonal antibodies have shown superiority over ipilimumab, leading to their application both as single agents and in combination with ipilimumab, which is nowadays seldom used as monotherapy [9–13].
Despite these dramatic improvements, a significant amount of patients—approximately 40–45%—show no response to immunotherapy [14]. Additionally, the mechanism of action of these agents (which is markedly different from usual cytotoxic approaches—notably by generating inflammations rather than direct lysis) can pose relevant challenges in the interpretation of treatment response by conventional imaging approaches [15]. Furthermore, several patients experience a “new class” of cumulative, dose-dependent, and sometimes life-threatening side effects, the immune-related adverse events (irAEs), which scope is wide [16]; importantly, the occurrence of such irAEs may be of prognostic value, revealing a response to immunotherapy [17, 18]. These issues raise the question of how to evaluate the response to ICIs in a reliable fashion and early in the course of treatment. This information would help discriminate responders from non-responders, offering significant therapeutic and prognostic implications in the entire spectrum of patient management. Unfortunately, there exist at present only few reliable predictors of long-term response to immunotherapy.
[18F]FDG PET/CT is considered the elective imaging technique in detecting metastatic disease in advanced melanoma [19–23]. Moreover, a growing amount of recently published literature has highlighted the potential role of the modality in the prediction of treatment response to ICIs in melanoma, rendering it an attractive tool for the monitoring of immunotherapy [24–31].
In quest of identifying reliable biomarkers for the prediction of long-term outcomes to immunotherapy, we aim in the present prospective study to assess the value of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment.
Materials and methods
Patients
Twenty-five patients (12 males, 13 females; mean age 54.7 years) with unresectable, stage IV melanoma undergoing immunotherapy with PD-1 inhibitors applied either as monotherapy (pembrolizumab, nivolumab) or as combination treatment with CTLA-4 inhibitors (nivolumab/ipilimumab) were enrolled in the study (Table 1). Pembrolizumab was administered intravenously at a dose of 2 mg/kg every 3 weeks, and nivolumab was administered intravenously at a dose of 3 mg/kg every 2 weeks. The combination ICI therapy was administered as an induction of 4 cycles of nivolumab (1 mg/kg) and ipilimumab (3 mg/kg) every 3 weeks, followed by single-agent nivolumab administration (3 mg/kg) every 2 weeks. The included patients had not received chemotherapy for at least 1 month prior to the initial PET/CT studies. None of the patients had a history of diabetes. Patients gave written informed consent to participate in the study and to have their medical records released. The study was approved by the Ethical Committee of the University of Heidelberg and the Federal Agency for Radiation Protection (Bundesamt für Strahlenschutz).Table 1 Patient characteristics
Patient number Age Gender LDH at baseline (U/I) Previous treatment ICI treatment EORTC PET response PERCIMT PET response Radiologic signs of irAEs Progression PFS (months)
1 56 F 770 Chemosaturation with melphalan, pembrolizumab, gemcitabine/treosulfan Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) Thyroiditis Yes 1.5
2 34 F 247 Pembrolizumab, IMCgp100 Nivolumab/ipilimumab CMR (MB) CMR (MB) Bone marrow activation, colitis No 17.9
3 46 F 166 Vemurafenib/cobimetinib Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Duodenitis Yes 4.0
4 54 M 218 Nivolumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Colitis Yes 3.2
5 53 M 275 No Nivolumab/ipilimumab SMD (MB) SMD (MB) - Yes 0.4
6 50 F 344 Vemurafenib/cobimetinib Nivolumab/ipilimumab SMD (MB) SMD (MB) Bone marrow activation, lymphadenopathy No 19.3
7 59 M 204 No Nivolumab/ipilimumab PMR (MB) PMR (MB) Sarcoid-like reaction, arthitis No 38.7
8 44 F 340 Dabrafenib/trametinib Nivolumab/ipilimumab PMR (MB) PMR (MB) Bone marrow activation, colitis No 18.3
9 60 F 269 No Nivolumab/ipilimumab PMR (MB) PMR (MB) – No 21.2
10 48 F 246 No Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Lymphadenopathy No 24.2
11 55 F 224 No Nivolumab SMD (MB) SMD (MB) - Yes 6.2
12 84 F 195 No Pembrolizumab SMD (MB) SMD (MB) - Yes 11.0
13 79 M 205 No Pembrolizumab PMD (no-MB) PMD (no-MB) Arthritis Yes 2.0
14 20 F 186 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 17.8
15 52 M 275 No Pembrolizumab PMR (MB) PMR (MB) - No 42.1
16 52 M 170 No Pembrolizumab PMD (no-MB) SMD (MB) - Yes 14.6
17 53 F 260 No Nivolumab SMD (MB) SMD (MB) - Yes 9.6
18 65 M 201 Ipilimumab Pembrolizumab SMD (MB) SMD (MB) - Yes 1.9
19 67 F 290 No Pembrolizumab PMD (no-MB) SMD (MB) Bone marrow activation, colitis Yes 5.9
20 55 M 200 Ipilimumab/nivolumab, dabrafenib/trametinib Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 1.5
21 58 M 183 Pembrolizumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Gastritis, colitis Yes 17.0
22 50 M 195 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 12.1
23 80 M 364 No Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 0.9
24 47 F 256 Dabrafenib Nivolumab SMD (MB) SMD (MB) - Yes 1.1
25 47 M 271 Ipilimumab, dabrafenib Nivolumab PMD (no-MB) SMD (MB) - Yes 2.7
F, female; M, male; LDH, lactate dehydrogenase; EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; irAEs, immune-related adverse events; PFS, progression-free survival
[18F]FDG PET/CT data acquisition
[18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of ICIs’ administration (interim PET/CT) in all 25 patients. Moreover, 17 patients of the cohort also had a third PET/CT scan within 2 weeks after administration of four cycles of treatment.
Patients underwent a whole body PET/CT after intravenous administration of maximum 250 MBq [18F]FDG 60 min post-injection (p.i.). Imaging was performed from the head to the feet with an image duration of 2 min per bed position. A dedicated PET/CT system (Biograph mCT, S128, Siemens Co., Erlangen, Germany) with an axial field of view of 21.6 cm with TruePoint and TrueV, operated in a three-dimensional mode was used. A low-dose attenuation CT (120 kV, 30 mA) was used for attenuation correction of the PET data and for image fusion. All PET images were attenuation-corrected and an image matrix of 400 × 400 pixels was used for iterative image reconstruction. Iterative image reconstruction was based on the ordered subset expectation maximization (OSEM) algorithm with two iterations and 21 subsets as well as time of flight (TOF).
[18F]FDG PET/CT data analysis
Data analysis consisted of visual (qualitative) assessment of the PET/CT scans and semi-quantitative evaluation based on standardized uptake value (SUV) calculations. PET/CT images were analyzed on an Aycan workstation by three nuclear medicine physicians (CS, DP, ADS). Images were interpreted by consensus. Visual analysis was based on the identification of sites of focal, non-physiologic [18F]FDG uptake above surrounding background activity, which were considered consistent with melanoma lesions.
Moreover, patterns of [18F]FDG uptake on interim PET/CT suggestive of radiologic manifestations of irAEs to immunotherapy were documented. Based on our experience and the published literature in the field [29, 32], we defined radiologic irAEs as sites of newly emerging, increased compared to baseline imaging, non-malignant [18F]FDG accumulation in organs known to exhibit immune-related signs on PET/CT. In particular, a new, diffusely enhanced tracer uptake in organs such as the gastrointestinal tract (mostly colon), the thyroid gland and the bone marrow, or, respectively, a new, relatively symmetrical, increased uptake in lymph nodes (e.g., mediastinal/hilar, inguinal) and in joints following ICIs were considered suggestive of radiologic irAEs in these organs. Semi-quantitative evaluation was based on volumes of interest (VOIs) and on subsequent calculation of SUVmean and SUVmax. VOIs were drawn using the pseudo-snake algorithm of the Pmod software (http://www.pmod.com/files/download/v31/doc/pbas/4729.htm) and were placed over melanoma lesions.
Response evaluation
Evaluation of patients’ response by means of [18F]FDG PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria [33] as well as the recently proposed PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT) [26]. Both criteria classify tumor response into 4 categories. The major difference between these criteria lies in the number of newly emerging lesions between baseline and follow-up PET/CT for the characterization of PMD: according to EORTC, the appearance of one new hypermetabolic lesion leads to patient classification to PMD, but according to PERCIMT, this requires the appearance of a minimum of four new lesions below 1 cm or respectively ≥ 3 new lesions of 1.0–1.5 cm or ≥ 2 new lesions of more than 1.5 cm. Another difference between the criteria is the role of SUV, which is central in EORTC, while it is not taken into account in PERCIMT (Table 2).Table 2 Summary of the EORTC and PERCIMT response criteria
EORTC PERCIMT
CMR Complete resolution of [18F]FDG uptake within the tumor volume Complete resolution of all pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
PMR Decrease in tumor SUV > 25% after more than 1 therapeutic cycle Complete resolution of some pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
SMD Increase in tumor SUV < 25% or decrease in SUV < 15% Neither PMD nor PMR/CMR
PMD Increase in tumor SUV > 25% or appearance of new lesions ≥ 4 new lesions of less than 1 cm in functional diameter or ≥ 3 new lesions between 1.0–1.5 cm in functional diameter or ≥ 2 new lesions of more than 1.5 cm in functional diameter
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; SUV, standardized uptake value
Stable disease (SD) represents a satisfactory outcome following immunotherapy, since—in contrast to conventional chemotherapy—it can be durable and survival rates related to SD are comparable to those associated with response [34–36]. Based on this, patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD.
Statistical analysis
Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause. Kaplan-Meier estimates were generated and median PFS estimated. Median follow-up time was determined by inverse Kaplan-Meier estimation. For univariate comparison of PFS, a log-rank test was used. Statistical analysis was performed using R version 4.0.2 (The R Foundation for Statistical Computing 2020) and R packages survival and prodlim. The results were considered significant for p values less than 0.05 (p < 0.05).
Results
Patient cohort
All included patients received treatment with anti-PD-1 agents, applied either as monotherapy (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients) or as combination therapy (nivolumab/ipilimumab, n = 13 patients). The mean baseline serum lactate dehydrogenase (LDH) was 262 U/l with four patients having pathologically high LDH levels and 21 of them normal levels. The detailed characteristics of the studied patients are presented in Table 1.
PET/CT findings
The findings of interim PET/CT were compared to those of the baseline scan and PET/CT-based response evaluation was performed for all 25 patients. According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), while six of them had no-MB (PMD) (Table 3) (Fig. 1).Table 3 Summary of the patients’ classifications in different response groups based on interim PET/CT and according to the EORTC and PERCIMT response criteria (n = 25 patients)
Metabolic benefit No-metabolic benefit
CMR PMR SMD PMD
EORTC 1 6 7 11
PERCIMT 1 6 12 6
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 1 Maximum intensity projection (MIP) [18F]FDG PET/CT images of a 34-year-old woman with metastatic melanoma before initiation of immunotherapy with nivolumab/ipilimumab (A) and after administration of two cycles of treatment (B). Baseline PET/CT image shows multiple lymph node, pulmonary, hepatic, adrenal, soft tissue, and osseous metastases (A). Interim PET/CT shows a complete metabolic remission (CMR) of all baseline lesions. The patient demonstrated metabolic benefit (MB) according to both the EORTC and PERCIMT criteria (B). Moreover, diffusely increased [18F]FDG uptake is observed in the ascending colon and the bone marrow on interim PET/CT. At the time of writing, the patient was still progression-free having reached a PFS of 17.9 months
With regard to the subgroup of 17 patients undergoing three PET/CT examinations (baseline, interim, late), the following results were revealed for interim PET/CT: 11 patients had MB (1 CMR, 5 PMR, and 5 SMD) and six patients had no-MB (PMD) according to EORTC, while 14 patients had MB (1 CMR, 5 PMR, and 8 SMD) and three of them had no-MB (PMD) according to PERCIMT. Respectively, on late PET/CT imaging, 13 patients had MB (3 CMR, 5 PMR, and 5 SMD) and four patients had no-MB (PMD) according to both EORTC and PERCIMT (Table 4). Two of the three patients exhibiting a mismatch between EORTC (PMD, no-MB) and PERCIMT (SMD, MB) on interim PET/CT, finally showed MB on the third examination based on both criteria (pseudoprogression) (Fig. 2). Respectively, one patient with early signs of PMD (no-MB) according to EORTC and SMD (MB) according to PERCIMT eventually exhibited PMD (no-MB) on the third examination based on both criteria.Table 4 Summary of the patients’ classifications in different response groups based on interim (after two cycles of ICIs) and late (after four cycles of ICIs) PET/CT (n = 17 patients)
EORTC PERCIMT
MB (late PET/CT) No-MB (late PET/CT) MB (late PET/CT) No-MB (late PET/CT)
CMR PMR SMD PMD CMR PMR SMD PMD
1.MB (interim PET/CT) CMR 1 0 0 0 1 0 0 0
PMR 0 5 0 0 0 5 0 0
SMD 1 0 0 4 2 0 5 1
No-MB (interim PET/CT) PMD 1 0 1 4 0 0 0 3
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; MB, metabolic benefit; no-MB, no metabolic benefit; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 2 Transaxial PET/CT (upper row, A–C) and low-dose CT (lower row, D–F) images at the cervical level of a 48-year-old female patient with advanced melanoma. The PET/CT (A) and CT (D) images obtained before immunotherapy show no pathologic lesions. Interim PET/CT performed after two cycles of nivolumab/ipilimumab shows a new [18F]FDG-avid lymph node (white arrow; B, E), suspicious of metastatic involvement. According to the EORTC criteria, the patient showed progressive metabolic disease (PMD), while according to PERCIMT, he had stable metabolic disease (SMD). A third PET/CT obtained after administration of four cycles of nivolumab/ipilimumab shows remission of the lesion (C, F), suggesting pseudoprogression of the cervical finding on interim PET/CT. The patient had a PFS of 24.2 months and was still progression-free at last contact
In total, 11 patients had PET/CT findings suggestive of irAEs on interim PET/CT, the majority of whom were under combination treatment: nine patients received nivolumab and ipilimumab, while two of them were under pembrolizumab (p = 0.015). In particular, the most common radiologic adverse event was a diffusely increased [18F]FDG uptake in the colon, defined as colitis, which was observed in five patients. One of these five patients also developed severe diarrhea, as clinical sign of treatment-induced colitis. Other gastrointestinal tract radiologic irAEs included gastritis (n = 1 patient) and duodenitis (n = 1 patient), defined as diffuse increased tracer uptake in the stomach and duodenum, respectively. Moreover, arthritis, defined as diffuse increased, periarticular, symmetrical tracer uptake in joints, was observed in two patients, and thyroiditis, a diffuse increased uptake in the thyroid gland, was observed in one patient. Furthermore, reactive, increased, symmetrical uptake in lymph nodes was observed in three patients, one of whom exhibited a sarcoid-like lymphadenopathy. Finally, diffuse increased bone marrow uptake was seen in four patients representing bone marrow activation in terms of a systemic immune response [37] (Table 1).
Survival analysis
Median follow-up (95% CI) of the patient cohort from interim PET/CT was 24.2 months (19.3–41.7 months). Patients receiving combination treatment (nivolumab/ipilimumab) had a median PFS of 17.8 months (4.0–NA), while those receiving PD-1 blockade monotherapy (nivolumab or pembrolizumab) had a median PFS of 4.3 months (1.9–NA) (p = 0.016) (Fig. 3).Fig. 3 Kaplan-Meier estimates of PFS according to the anti-PD-1 treatment applied. The numbers of patients at risk in each group and for the respective time points are shown below the plots. Combi, combination therapy (ipilimumab/nivolumab); Mono, monotherapy (nivolumab or pembrolizumab)
Based on the EORTC criteria, patients with MB on interim PET/CT had a median PFS of 14.0 months (6.2–NA), while those with no-MB had a median PFS of 4.0 months (2.0–NA) (p = 0.088) (Fig. 4A). Respectively, according to the PERCIMT criteria, patients with MB had a median PFS of 11.0 months (5.9–NA), while those with no-MB had a median PFS of 1.8 months (1.5–NA) (p = 0.045) (Fig. 4B).Fig. 4 Kaplan-Meier estimates of PFS according to the EORTC (A) and the PERCIMT (B) criteria. The numbers of patients at risk in each group and for the respective time points are shown below the plots. MB, metabolic benefit; no-MB, no metabolic benefit
The patient cohort was further dichotomized on the basis of the emergence of radiologic irAEs on interim PET/CT. Patients with irAEs had a median PFS of 17.0 months (4.0–NA), and patients without irAEs had a median PFS of 7.9 months (1.9–NA) (p = 0.128) (Fig. 5).Fig. 5 Kaplan-Meier estimates of PFS according to the emergence of radiologic irAEs on interim PET/CT. The numbers of patients at risk in each group and for the respective time points are shown below the plots
Finally, the potential correlation of baseline LDH with PFS was also investigated. However, pathologic levels of LDH had no adverse effect on survival of the cohort (p = 0.642) (Supplementary File 1).
Discussion
The therapeutic benefit of immune checkpoint blockade of PD-1 and CTLA-4 in the treatment of metastatic melanoma is variable [38]. Our understanding of how ICIs affect T cell evolution is incomplete [39], limiting the ability to derive full clinical benefit and, moreover, to predict responses from these drugs. However, tracking early response to immunotherapy is key for treatment options.
In this study, we investigated the role of interim PET/CT, performed after application of two cycles of anti-PD-1 treatment, in prediction of survival of metastatic melanoma patients. Our results showed that tumor response as classified by the PERCIMT criteria is significantly correlated with PFS, with metabolic responders demonstrating a significant survival benefit over non-responders. Respectively, the application of the EORTC criteria also led to a higher PFS for metabolic responders compared to non-responders but this difference was not statistically significant. These findings highlight the ability of [18F]FDG PET/CT—in particular, after application of the recently introduced PERCIMT criteria—to monitor and predict response to anti-PD-1 agents at an early but clinically relevant time point. This is of particular importance in clinical decision-making and prediction of outcome early during the course of immunotherapy, rendering PET/CT a potentially significant tool for the management of these patients. The main strengths of our study include its prospective nature, its rigorous protocol with imaging performed at strictly defined time points during treatment, the standardized for all patients PET/CT procedure, and the correlation with survival analysis.
Hitherto, a non-negligible number of studies have evaluated the efficacy of PET/CT in predicting treatment response of metastatic melanoma patients to ICIs. While most papers have focused on later time points during the course or after the end of treatment [26, 28, 30–32, 40], few of them also reported on application of the imaging modality early during immunotherapy. Our group previously showed in a cohort of 22 patients that PET/CT performed after two ipilimumab cycles—and after application of the EORTC criteria—correctly predicted treatment response after completion of the 4-cycle treatment in the majority (87%) of PMD patients and in all SMD patients [24]. In an expanded analysis of the ipilimumab patient cohort (n = 41 patients), the capacity of interim PET/CT in predicting clinical benefit to the agent was also highlighted. In that analysis, the performance of the PERCIMT criteria was superior to that of EORTC, which is in line with the results of the present study [27]. Furthermore, Cho et al. studied 20 melanoma patients treated with different ICIs (anti-PD-1 and anti-CTLA-4) with PET/CT at 3–4 weeks into therapy and found that a combination of changes in lesional dimensions along with changes in [18F]FDG uptake is a more accurate predictor of eventual response than each of these parameters alone. The authors proposed the PET/CT criteria for early prediction of response to immune checkpoint inhibitor therapy (PECRIT) criteria, based on a combination of the response evaluation criteria in solid tumors (RECIST) and PET response criteria in solid tumors (PERCIST) [25].
The current analysis represents the first study focusing on the role of interim PET/CT, performed as early as after application of two cycles of PD-1 blockade, in survival prediction of metastatic melanoma patients undergoing treatment with this class of ICIs. Taken together, the herein presented findings as well as those of previous studies in the field build further evidence on the potentially significant role of PET/CT performed early during the course of immunotherapy for prediction and stratification of response to treatment.
Novel patterns of response and progression, not previously seen with conventional therapies (such as cytotoxic or targeted anticancer regiments), have been described under immunotherapy and are attributed to the unique mechanism of action of these agents. In particular, phenomena such as pseudoprogression and irAEs may render response assessment to ICI challenging, questioning the utility of imaging modalities.
Evaluation of response to immunotherapy by means of PET/CT is primarily visual and subjective in nature. Considering the aforementioned challenges raised by the advent of immunotherapy for imaging interpretation, our group has recently introduced the PERCIMT criteria in metastatic melanoma, in an attempt to meet the need for reliable response assessment based on PET/CT. The cornerstone of these criteria is the finding that the absolute number of newly emerged [18F]FDG-avid lesions is more predictive of clinical outcome than SUV changes [26]. In specific, the application of a threshold of four newly emerged lesions on post-therapy PET/CT scan—with a decreasing cutoff of lesion number as the functional diameter of the lesions increases—in a cohort of 41 patients could predict clinical benefit to treatment with the agent ipilimumab better than the standard threshold of one new lesion or an increase in SUV, conventionally applied with the EORTC criteria. This was also confirmed in the present analysis with a significant correlation of metabolic response on interim PET with PFS only after application of PERCIMT.
Pseudoprogression, defined as an initial increase of tumor burden before the disease responds to therapy, has been initially described in melanoma patients undergoing ipilimumab therapy [41]. Since this phenomenon may be misclassified as progressive disease, the recently modified radiologic, immune-related response criteria (irRECIST, iRECIST) call for a 4-week re-assessment in order to overcome this limitation [42, 43]. In our study, the evaluation of pseudoprogression was partly feasible in the subgroup of 17 patients undergoing a third PET/CT after administration of four cycles of treatment. The comparison between interim and late PET/CT showed signs of pseudoprogression in 2/17 (11.8%) patients, who were characterized as PMD on interim PET/CT according to EORTC and “switched” to MB on late imaging. In contrary, no cases of pseudoprogression were observed after application of PERCIMT, highlighting the ability of the novel criteria in tackling this atypical response pattern. Moreover, the results of survival analysis, exhibiting a lower PFS for patients with early PMD (no-MB) compared to those with early MB, are another indirect proof of the rather low incidence of the phenomenon in this cohort. This is in line with previous results published in the literature documenting non-negligible, but not higher than 10% rates of the phenomenon in melanoma immunotherapy [36, 44, 45], and provides supporting evidence to the standpoint that an increase in tumor burden observed during ICI treatment more likely reflects true progression rather than pseudoprogression [41, 46, 47].
irAEs represent another source of false-positive findings on imaging. Radiologic manifestations of irAEs have been reported with variable incidences, reaching up to 31% of patients under ICIs [47–49]. Although the specific characteristics of individual patients play a significant role, the emergence of these toxicities is mainly dependent on the agents used, with the combination of an anti-CTLA-4 antibody and an anti-PD-1 antibody increasing both their incidence and severity [50]. In the herein studied cohort, 11/25 (44%) of the patients showed signs of irAEs on PET/CT. In line with data from the literature [50], the vast majority of these patients (9/11 patients) received combination treatment of nivolumab and ipilimumab. Colitis was the most frequent adverse event on PET/CT, observed in five patients; four of these patients received PD-1 inhibitors in combination with the anti-CTLA-4 regiment ipilimumab, which is known to often induce this reaction [51, 52]. We recognize that the diagnosis of colitis on PET imaging can be complicated, since its identification can be hampered by physiological metabolic activity in the colon. It is well known that enhanced colon [18F]FDG uptake of benign etiology is frequently observed in asymptomatic individuals [53, 54]. Moreover, several studies have shown that patients using the oral hypoglycemic drug metformin tend to have a diffusely increased tracer uptake in the colon [55–58]. In the present cohort, no patient had diabetes; thus, metformin can be ruled out as a cause for false-positive [18F]FDG accumulation in the colon. A further search in patients’ clinical history revealed that one of these five patients developed severe diarrhea during immunotherapy, most likely as a symptom of treatment-induced colitis, while the rest four patients did not have such symptoms. This higher incidence of “PET-colitis” under immunotherapy, compared to clinical signs of colitis, is in line with previously published results [59]. In this context, early recognition of radiologic irAEs could be potentially important since they may precede or correlate with clinical symptoms [37, 47], potentially leading to respective changes in management.
Another aspect pertaining to irAEs is that their emergence has been associated with a favorable efficacy of ICIs—mainly of PD-1 inhibitors—implying a potential predictive role of these events for response to ICI treatment [17, 18, 60]. Our analysis revealed that patients with radiologic signs of irAEs had a longer PFS than those without irAEs; however, this difference was non-significant. Apart from the relatively small cohort studied, an explanation for this finding may lie in the fact that most patients (82%) also received the CTLA-4 inhibitor ipilimumab in combination with the PD-1 inhibitor nivolumab; in a recently published meta-analysis of 30 studies including 4971 individuals, it was shown that no significant association between irAE development and a favorable benefit on survival is observed in ICI combination treatments, in contrary to anti-PD-1 monotherapies [60]. Another reason may lie in the nature of the observed irAEs, affecting the gastrointestinal tract in more than half of the cases (6/11 patients, 55%); data from the above-mentioned meta-analysis also highlight the lack of PFS benefit in patients presenting gastrointestinal irAEs [60].
Finally, the predictive role of baseline LDH before initiation of PD-1 inhibitors was investigated. Although serum LDH elevation is not specific for melanoma, it represents a poor prognostic factor and is one of the most influential factors associated with treatment response [61, 62]. An interesting finding of our analysis is the lack of any adverse effect of elevated baseline LDH on survival, which can be however attributed to the small number of patients with pathologic LDH (n = 4 patients, 16%).
We note some limitations in our study. Firstly, due to the strict inclusion criteria applied, the number of included patients was relatively low, not allowing us to draw more firm conclusions; ideally, further studies with larger patient cohorts would be required. Secondly, although the focus of the study was anti-PD-1 treatment, not all patients underwent exclusively PD-1 blockade, with several of them receiving combination therapy of PD-1 and CTLA-4 inhibitors. Although our patient cohort is too small to afford a PET/CT subanalysis based on a dichotomization of patients into those receiving anti-PD-1 monotherapy and those undergoing combined treatment, we highlight the similar approach followed in previous studies in the field [28, 63]. Finally, the vast majority of the PET/CT-positive, melanoma-consistent findings were not histopathologically confirmed. However, this is not usually possible in the clinical setting.
Conclusion
In an attempt to identify early and reliable biomarkers of survival prediction in immunotherapy of metastatic melanoma, we assessed the prognostic role of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment. Our results showed that tumor response as classified by the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Supplementary information
ESM 1 (DOCX 166 kb)
Funding
Open Access funding enabled and organized by Projekt DEAL.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all participants enrolled in the study.
This article is part of the Topical Collection on Oncology - General.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY | 33336264 | 18,710,997 | 2021-06 |
What was the administration route of drug 'NIVOLUMAB'? | Interim [18F]FDG PET/CT can predict response to anti-PD-1 treatment in metastatic melanoma.
In an attempt to identify biomarkers that can reliably predict long-term outcomes to immunotherapy in metastatic melanoma, we investigated the prognostic role of [18F]FDG PET/CT, performed at baseline and early during the course of anti-PD-1 treatment.
Twenty-five patients with stage IV melanoma, scheduled for treatment with PD-1 inhibitors, were enrolled in the study (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients; nivolumab/ipilimumab, 13 patients). [18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of PD-1 blockade administration (interim PET/CT). Seventeen patients underwent also a third PET/CT scan after administration of four cycles of treatment. Evaluation of patients' response by means of PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria and the PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT). Response to treatment was classified into 4 categories: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD). Patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD. Moreover, patterns of [18F]FDG uptake suggestive of radiologic immune-related adverse events (irAEs) were documented. Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause.
Median follow-up from interim PET/CT was 24.2 months (19.3-41.7 months). According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), and 6 of them had no-MB (PMD). With regard to PFS, no significant difference was observed between patients with MB and no-MB on interim PET/CT according to the EORTC criteria (p = 0.088). In contrary, according to the PERCIMT criteria, patients demonstrating MB had a significantly longer PFS than those showing no-MB (p = 0.045). The emergence of radiologic irAEs (n = 11 patients) was not associated with a significant survival benefit. Regarding the sub-cohort undergoing also a third PET/CT, 14/17 patients (82%) showed concordant responses and 3/17 (18%) had a mismatch of response assessment between interim and late PET/CT.
PET/CT-based response of metastatic melanoma to PD-1 blockade after application of the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Introduction
Metastatic melanoma is a highly aggressive tumor, largely refractory to existing therapies, and associated with a very poor prognosis [1]. While until lately the treatment options for metastatic melanoma were limited, the recent development and introduction in clinical practice of several novel immunotherapeutic agents as well as of targeted therapy with BRAF and MEK inhibitors have revolutionized the systemic treatment of the disease, leading to unprecedented response and survival rates of melanoma patients [2].
The main form of immunotherapy applied in this new era of melanoma management involves immune checkpoint blockade. This immunomodulatory approach activates the immune system against tumors through the binding of the cytotoxic T lymphocyte–associated protein 4 (CTLA-4) and/or the programmed cell death protein 1 (PD-1), both of which are expressed by T cells [3, 4]. The monoclonal antibody ipilimumab, which acts by blocking CTLA-4, is considered a landmark agent in this context, being the first immunotherapeutic drug demonstrating a clear benefit in survival of patients with advanced melanoma, which led to its approval by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2011 [5]. A few years later, a second class of immune checkpoint inhibitors (ICIs), the PD-1 inhibitors nivolumab and pembrolizumab, were also approved for the treatment of melanoma, after having shown survival benefit in these patients [6–8]. Moreover, the anti-PD-1 monoclonal antibodies have shown superiority over ipilimumab, leading to their application both as single agents and in combination with ipilimumab, which is nowadays seldom used as monotherapy [9–13].
Despite these dramatic improvements, a significant amount of patients—approximately 40–45%—show no response to immunotherapy [14]. Additionally, the mechanism of action of these agents (which is markedly different from usual cytotoxic approaches—notably by generating inflammations rather than direct lysis) can pose relevant challenges in the interpretation of treatment response by conventional imaging approaches [15]. Furthermore, several patients experience a “new class” of cumulative, dose-dependent, and sometimes life-threatening side effects, the immune-related adverse events (irAEs), which scope is wide [16]; importantly, the occurrence of such irAEs may be of prognostic value, revealing a response to immunotherapy [17, 18]. These issues raise the question of how to evaluate the response to ICIs in a reliable fashion and early in the course of treatment. This information would help discriminate responders from non-responders, offering significant therapeutic and prognostic implications in the entire spectrum of patient management. Unfortunately, there exist at present only few reliable predictors of long-term response to immunotherapy.
[18F]FDG PET/CT is considered the elective imaging technique in detecting metastatic disease in advanced melanoma [19–23]. Moreover, a growing amount of recently published literature has highlighted the potential role of the modality in the prediction of treatment response to ICIs in melanoma, rendering it an attractive tool for the monitoring of immunotherapy [24–31].
In quest of identifying reliable biomarkers for the prediction of long-term outcomes to immunotherapy, we aim in the present prospective study to assess the value of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment.
Materials and methods
Patients
Twenty-five patients (12 males, 13 females; mean age 54.7 years) with unresectable, stage IV melanoma undergoing immunotherapy with PD-1 inhibitors applied either as monotherapy (pembrolizumab, nivolumab) or as combination treatment with CTLA-4 inhibitors (nivolumab/ipilimumab) were enrolled in the study (Table 1). Pembrolizumab was administered intravenously at a dose of 2 mg/kg every 3 weeks, and nivolumab was administered intravenously at a dose of 3 mg/kg every 2 weeks. The combination ICI therapy was administered as an induction of 4 cycles of nivolumab (1 mg/kg) and ipilimumab (3 mg/kg) every 3 weeks, followed by single-agent nivolumab administration (3 mg/kg) every 2 weeks. The included patients had not received chemotherapy for at least 1 month prior to the initial PET/CT studies. None of the patients had a history of diabetes. Patients gave written informed consent to participate in the study and to have their medical records released. The study was approved by the Ethical Committee of the University of Heidelberg and the Federal Agency for Radiation Protection (Bundesamt für Strahlenschutz).Table 1 Patient characteristics
Patient number Age Gender LDH at baseline (U/I) Previous treatment ICI treatment EORTC PET response PERCIMT PET response Radiologic signs of irAEs Progression PFS (months)
1 56 F 770 Chemosaturation with melphalan, pembrolizumab, gemcitabine/treosulfan Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) Thyroiditis Yes 1.5
2 34 F 247 Pembrolizumab, IMCgp100 Nivolumab/ipilimumab CMR (MB) CMR (MB) Bone marrow activation, colitis No 17.9
3 46 F 166 Vemurafenib/cobimetinib Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Duodenitis Yes 4.0
4 54 M 218 Nivolumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Colitis Yes 3.2
5 53 M 275 No Nivolumab/ipilimumab SMD (MB) SMD (MB) - Yes 0.4
6 50 F 344 Vemurafenib/cobimetinib Nivolumab/ipilimumab SMD (MB) SMD (MB) Bone marrow activation, lymphadenopathy No 19.3
7 59 M 204 No Nivolumab/ipilimumab PMR (MB) PMR (MB) Sarcoid-like reaction, arthitis No 38.7
8 44 F 340 Dabrafenib/trametinib Nivolumab/ipilimumab PMR (MB) PMR (MB) Bone marrow activation, colitis No 18.3
9 60 F 269 No Nivolumab/ipilimumab PMR (MB) PMR (MB) – No 21.2
10 48 F 246 No Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Lymphadenopathy No 24.2
11 55 F 224 No Nivolumab SMD (MB) SMD (MB) - Yes 6.2
12 84 F 195 No Pembrolizumab SMD (MB) SMD (MB) - Yes 11.0
13 79 M 205 No Pembrolizumab PMD (no-MB) PMD (no-MB) Arthritis Yes 2.0
14 20 F 186 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 17.8
15 52 M 275 No Pembrolizumab PMR (MB) PMR (MB) - No 42.1
16 52 M 170 No Pembrolizumab PMD (no-MB) SMD (MB) - Yes 14.6
17 53 F 260 No Nivolumab SMD (MB) SMD (MB) - Yes 9.6
18 65 M 201 Ipilimumab Pembrolizumab SMD (MB) SMD (MB) - Yes 1.9
19 67 F 290 No Pembrolizumab PMD (no-MB) SMD (MB) Bone marrow activation, colitis Yes 5.9
20 55 M 200 Ipilimumab/nivolumab, dabrafenib/trametinib Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 1.5
21 58 M 183 Pembrolizumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Gastritis, colitis Yes 17.0
22 50 M 195 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 12.1
23 80 M 364 No Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 0.9
24 47 F 256 Dabrafenib Nivolumab SMD (MB) SMD (MB) - Yes 1.1
25 47 M 271 Ipilimumab, dabrafenib Nivolumab PMD (no-MB) SMD (MB) - Yes 2.7
F, female; M, male; LDH, lactate dehydrogenase; EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; irAEs, immune-related adverse events; PFS, progression-free survival
[18F]FDG PET/CT data acquisition
[18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of ICIs’ administration (interim PET/CT) in all 25 patients. Moreover, 17 patients of the cohort also had a third PET/CT scan within 2 weeks after administration of four cycles of treatment.
Patients underwent a whole body PET/CT after intravenous administration of maximum 250 MBq [18F]FDG 60 min post-injection (p.i.). Imaging was performed from the head to the feet with an image duration of 2 min per bed position. A dedicated PET/CT system (Biograph mCT, S128, Siemens Co., Erlangen, Germany) with an axial field of view of 21.6 cm with TruePoint and TrueV, operated in a three-dimensional mode was used. A low-dose attenuation CT (120 kV, 30 mA) was used for attenuation correction of the PET data and for image fusion. All PET images were attenuation-corrected and an image matrix of 400 × 400 pixels was used for iterative image reconstruction. Iterative image reconstruction was based on the ordered subset expectation maximization (OSEM) algorithm with two iterations and 21 subsets as well as time of flight (TOF).
[18F]FDG PET/CT data analysis
Data analysis consisted of visual (qualitative) assessment of the PET/CT scans and semi-quantitative evaluation based on standardized uptake value (SUV) calculations. PET/CT images were analyzed on an Aycan workstation by three nuclear medicine physicians (CS, DP, ADS). Images were interpreted by consensus. Visual analysis was based on the identification of sites of focal, non-physiologic [18F]FDG uptake above surrounding background activity, which were considered consistent with melanoma lesions.
Moreover, patterns of [18F]FDG uptake on interim PET/CT suggestive of radiologic manifestations of irAEs to immunotherapy were documented. Based on our experience and the published literature in the field [29, 32], we defined radiologic irAEs as sites of newly emerging, increased compared to baseline imaging, non-malignant [18F]FDG accumulation in organs known to exhibit immune-related signs on PET/CT. In particular, a new, diffusely enhanced tracer uptake in organs such as the gastrointestinal tract (mostly colon), the thyroid gland and the bone marrow, or, respectively, a new, relatively symmetrical, increased uptake in lymph nodes (e.g., mediastinal/hilar, inguinal) and in joints following ICIs were considered suggestive of radiologic irAEs in these organs. Semi-quantitative evaluation was based on volumes of interest (VOIs) and on subsequent calculation of SUVmean and SUVmax. VOIs were drawn using the pseudo-snake algorithm of the Pmod software (http://www.pmod.com/files/download/v31/doc/pbas/4729.htm) and were placed over melanoma lesions.
Response evaluation
Evaluation of patients’ response by means of [18F]FDG PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria [33] as well as the recently proposed PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT) [26]. Both criteria classify tumor response into 4 categories. The major difference between these criteria lies in the number of newly emerging lesions between baseline and follow-up PET/CT for the characterization of PMD: according to EORTC, the appearance of one new hypermetabolic lesion leads to patient classification to PMD, but according to PERCIMT, this requires the appearance of a minimum of four new lesions below 1 cm or respectively ≥ 3 new lesions of 1.0–1.5 cm or ≥ 2 new lesions of more than 1.5 cm. Another difference between the criteria is the role of SUV, which is central in EORTC, while it is not taken into account in PERCIMT (Table 2).Table 2 Summary of the EORTC and PERCIMT response criteria
EORTC PERCIMT
CMR Complete resolution of [18F]FDG uptake within the tumor volume Complete resolution of all pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
PMR Decrease in tumor SUV > 25% after more than 1 therapeutic cycle Complete resolution of some pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
SMD Increase in tumor SUV < 25% or decrease in SUV < 15% Neither PMD nor PMR/CMR
PMD Increase in tumor SUV > 25% or appearance of new lesions ≥ 4 new lesions of less than 1 cm in functional diameter or ≥ 3 new lesions between 1.0–1.5 cm in functional diameter or ≥ 2 new lesions of more than 1.5 cm in functional diameter
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; SUV, standardized uptake value
Stable disease (SD) represents a satisfactory outcome following immunotherapy, since—in contrast to conventional chemotherapy—it can be durable and survival rates related to SD are comparable to those associated with response [34–36]. Based on this, patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD.
Statistical analysis
Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause. Kaplan-Meier estimates were generated and median PFS estimated. Median follow-up time was determined by inverse Kaplan-Meier estimation. For univariate comparison of PFS, a log-rank test was used. Statistical analysis was performed using R version 4.0.2 (The R Foundation for Statistical Computing 2020) and R packages survival and prodlim. The results were considered significant for p values less than 0.05 (p < 0.05).
Results
Patient cohort
All included patients received treatment with anti-PD-1 agents, applied either as monotherapy (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients) or as combination therapy (nivolumab/ipilimumab, n = 13 patients). The mean baseline serum lactate dehydrogenase (LDH) was 262 U/l with four patients having pathologically high LDH levels and 21 of them normal levels. The detailed characteristics of the studied patients are presented in Table 1.
PET/CT findings
The findings of interim PET/CT were compared to those of the baseline scan and PET/CT-based response evaluation was performed for all 25 patients. According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), while six of them had no-MB (PMD) (Table 3) (Fig. 1).Table 3 Summary of the patients’ classifications in different response groups based on interim PET/CT and according to the EORTC and PERCIMT response criteria (n = 25 patients)
Metabolic benefit No-metabolic benefit
CMR PMR SMD PMD
EORTC 1 6 7 11
PERCIMT 1 6 12 6
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 1 Maximum intensity projection (MIP) [18F]FDG PET/CT images of a 34-year-old woman with metastatic melanoma before initiation of immunotherapy with nivolumab/ipilimumab (A) and after administration of two cycles of treatment (B). Baseline PET/CT image shows multiple lymph node, pulmonary, hepatic, adrenal, soft tissue, and osseous metastases (A). Interim PET/CT shows a complete metabolic remission (CMR) of all baseline lesions. The patient demonstrated metabolic benefit (MB) according to both the EORTC and PERCIMT criteria (B). Moreover, diffusely increased [18F]FDG uptake is observed in the ascending colon and the bone marrow on interim PET/CT. At the time of writing, the patient was still progression-free having reached a PFS of 17.9 months
With regard to the subgroup of 17 patients undergoing three PET/CT examinations (baseline, interim, late), the following results were revealed for interim PET/CT: 11 patients had MB (1 CMR, 5 PMR, and 5 SMD) and six patients had no-MB (PMD) according to EORTC, while 14 patients had MB (1 CMR, 5 PMR, and 8 SMD) and three of them had no-MB (PMD) according to PERCIMT. Respectively, on late PET/CT imaging, 13 patients had MB (3 CMR, 5 PMR, and 5 SMD) and four patients had no-MB (PMD) according to both EORTC and PERCIMT (Table 4). Two of the three patients exhibiting a mismatch between EORTC (PMD, no-MB) and PERCIMT (SMD, MB) on interim PET/CT, finally showed MB on the third examination based on both criteria (pseudoprogression) (Fig. 2). Respectively, one patient with early signs of PMD (no-MB) according to EORTC and SMD (MB) according to PERCIMT eventually exhibited PMD (no-MB) on the third examination based on both criteria.Table 4 Summary of the patients’ classifications in different response groups based on interim (after two cycles of ICIs) and late (after four cycles of ICIs) PET/CT (n = 17 patients)
EORTC PERCIMT
MB (late PET/CT) No-MB (late PET/CT) MB (late PET/CT) No-MB (late PET/CT)
CMR PMR SMD PMD CMR PMR SMD PMD
1.MB (interim PET/CT) CMR 1 0 0 0 1 0 0 0
PMR 0 5 0 0 0 5 0 0
SMD 1 0 0 4 2 0 5 1
No-MB (interim PET/CT) PMD 1 0 1 4 0 0 0 3
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; MB, metabolic benefit; no-MB, no metabolic benefit; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 2 Transaxial PET/CT (upper row, A–C) and low-dose CT (lower row, D–F) images at the cervical level of a 48-year-old female patient with advanced melanoma. The PET/CT (A) and CT (D) images obtained before immunotherapy show no pathologic lesions. Interim PET/CT performed after two cycles of nivolumab/ipilimumab shows a new [18F]FDG-avid lymph node (white arrow; B, E), suspicious of metastatic involvement. According to the EORTC criteria, the patient showed progressive metabolic disease (PMD), while according to PERCIMT, he had stable metabolic disease (SMD). A third PET/CT obtained after administration of four cycles of nivolumab/ipilimumab shows remission of the lesion (C, F), suggesting pseudoprogression of the cervical finding on interim PET/CT. The patient had a PFS of 24.2 months and was still progression-free at last contact
In total, 11 patients had PET/CT findings suggestive of irAEs on interim PET/CT, the majority of whom were under combination treatment: nine patients received nivolumab and ipilimumab, while two of them were under pembrolizumab (p = 0.015). In particular, the most common radiologic adverse event was a diffusely increased [18F]FDG uptake in the colon, defined as colitis, which was observed in five patients. One of these five patients also developed severe diarrhea, as clinical sign of treatment-induced colitis. Other gastrointestinal tract radiologic irAEs included gastritis (n = 1 patient) and duodenitis (n = 1 patient), defined as diffuse increased tracer uptake in the stomach and duodenum, respectively. Moreover, arthritis, defined as diffuse increased, periarticular, symmetrical tracer uptake in joints, was observed in two patients, and thyroiditis, a diffuse increased uptake in the thyroid gland, was observed in one patient. Furthermore, reactive, increased, symmetrical uptake in lymph nodes was observed in three patients, one of whom exhibited a sarcoid-like lymphadenopathy. Finally, diffuse increased bone marrow uptake was seen in four patients representing bone marrow activation in terms of a systemic immune response [37] (Table 1).
Survival analysis
Median follow-up (95% CI) of the patient cohort from interim PET/CT was 24.2 months (19.3–41.7 months). Patients receiving combination treatment (nivolumab/ipilimumab) had a median PFS of 17.8 months (4.0–NA), while those receiving PD-1 blockade monotherapy (nivolumab or pembrolizumab) had a median PFS of 4.3 months (1.9–NA) (p = 0.016) (Fig. 3).Fig. 3 Kaplan-Meier estimates of PFS according to the anti-PD-1 treatment applied. The numbers of patients at risk in each group and for the respective time points are shown below the plots. Combi, combination therapy (ipilimumab/nivolumab); Mono, monotherapy (nivolumab or pembrolizumab)
Based on the EORTC criteria, patients with MB on interim PET/CT had a median PFS of 14.0 months (6.2–NA), while those with no-MB had a median PFS of 4.0 months (2.0–NA) (p = 0.088) (Fig. 4A). Respectively, according to the PERCIMT criteria, patients with MB had a median PFS of 11.0 months (5.9–NA), while those with no-MB had a median PFS of 1.8 months (1.5–NA) (p = 0.045) (Fig. 4B).Fig. 4 Kaplan-Meier estimates of PFS according to the EORTC (A) and the PERCIMT (B) criteria. The numbers of patients at risk in each group and for the respective time points are shown below the plots. MB, metabolic benefit; no-MB, no metabolic benefit
The patient cohort was further dichotomized on the basis of the emergence of radiologic irAEs on interim PET/CT. Patients with irAEs had a median PFS of 17.0 months (4.0–NA), and patients without irAEs had a median PFS of 7.9 months (1.9–NA) (p = 0.128) (Fig. 5).Fig. 5 Kaplan-Meier estimates of PFS according to the emergence of radiologic irAEs on interim PET/CT. The numbers of patients at risk in each group and for the respective time points are shown below the plots
Finally, the potential correlation of baseline LDH with PFS was also investigated. However, pathologic levels of LDH had no adverse effect on survival of the cohort (p = 0.642) (Supplementary File 1).
Discussion
The therapeutic benefit of immune checkpoint blockade of PD-1 and CTLA-4 in the treatment of metastatic melanoma is variable [38]. Our understanding of how ICIs affect T cell evolution is incomplete [39], limiting the ability to derive full clinical benefit and, moreover, to predict responses from these drugs. However, tracking early response to immunotherapy is key for treatment options.
In this study, we investigated the role of interim PET/CT, performed after application of two cycles of anti-PD-1 treatment, in prediction of survival of metastatic melanoma patients. Our results showed that tumor response as classified by the PERCIMT criteria is significantly correlated with PFS, with metabolic responders demonstrating a significant survival benefit over non-responders. Respectively, the application of the EORTC criteria also led to a higher PFS for metabolic responders compared to non-responders but this difference was not statistically significant. These findings highlight the ability of [18F]FDG PET/CT—in particular, after application of the recently introduced PERCIMT criteria—to monitor and predict response to anti-PD-1 agents at an early but clinically relevant time point. This is of particular importance in clinical decision-making and prediction of outcome early during the course of immunotherapy, rendering PET/CT a potentially significant tool for the management of these patients. The main strengths of our study include its prospective nature, its rigorous protocol with imaging performed at strictly defined time points during treatment, the standardized for all patients PET/CT procedure, and the correlation with survival analysis.
Hitherto, a non-negligible number of studies have evaluated the efficacy of PET/CT in predicting treatment response of metastatic melanoma patients to ICIs. While most papers have focused on later time points during the course or after the end of treatment [26, 28, 30–32, 40], few of them also reported on application of the imaging modality early during immunotherapy. Our group previously showed in a cohort of 22 patients that PET/CT performed after two ipilimumab cycles—and after application of the EORTC criteria—correctly predicted treatment response after completion of the 4-cycle treatment in the majority (87%) of PMD patients and in all SMD patients [24]. In an expanded analysis of the ipilimumab patient cohort (n = 41 patients), the capacity of interim PET/CT in predicting clinical benefit to the agent was also highlighted. In that analysis, the performance of the PERCIMT criteria was superior to that of EORTC, which is in line with the results of the present study [27]. Furthermore, Cho et al. studied 20 melanoma patients treated with different ICIs (anti-PD-1 and anti-CTLA-4) with PET/CT at 3–4 weeks into therapy and found that a combination of changes in lesional dimensions along with changes in [18F]FDG uptake is a more accurate predictor of eventual response than each of these parameters alone. The authors proposed the PET/CT criteria for early prediction of response to immune checkpoint inhibitor therapy (PECRIT) criteria, based on a combination of the response evaluation criteria in solid tumors (RECIST) and PET response criteria in solid tumors (PERCIST) [25].
The current analysis represents the first study focusing on the role of interim PET/CT, performed as early as after application of two cycles of PD-1 blockade, in survival prediction of metastatic melanoma patients undergoing treatment with this class of ICIs. Taken together, the herein presented findings as well as those of previous studies in the field build further evidence on the potentially significant role of PET/CT performed early during the course of immunotherapy for prediction and stratification of response to treatment.
Novel patterns of response and progression, not previously seen with conventional therapies (such as cytotoxic or targeted anticancer regiments), have been described under immunotherapy and are attributed to the unique mechanism of action of these agents. In particular, phenomena such as pseudoprogression and irAEs may render response assessment to ICI challenging, questioning the utility of imaging modalities.
Evaluation of response to immunotherapy by means of PET/CT is primarily visual and subjective in nature. Considering the aforementioned challenges raised by the advent of immunotherapy for imaging interpretation, our group has recently introduced the PERCIMT criteria in metastatic melanoma, in an attempt to meet the need for reliable response assessment based on PET/CT. The cornerstone of these criteria is the finding that the absolute number of newly emerged [18F]FDG-avid lesions is more predictive of clinical outcome than SUV changes [26]. In specific, the application of a threshold of four newly emerged lesions on post-therapy PET/CT scan—with a decreasing cutoff of lesion number as the functional diameter of the lesions increases—in a cohort of 41 patients could predict clinical benefit to treatment with the agent ipilimumab better than the standard threshold of one new lesion or an increase in SUV, conventionally applied with the EORTC criteria. This was also confirmed in the present analysis with a significant correlation of metabolic response on interim PET with PFS only after application of PERCIMT.
Pseudoprogression, defined as an initial increase of tumor burden before the disease responds to therapy, has been initially described in melanoma patients undergoing ipilimumab therapy [41]. Since this phenomenon may be misclassified as progressive disease, the recently modified radiologic, immune-related response criteria (irRECIST, iRECIST) call for a 4-week re-assessment in order to overcome this limitation [42, 43]. In our study, the evaluation of pseudoprogression was partly feasible in the subgroup of 17 patients undergoing a third PET/CT after administration of four cycles of treatment. The comparison between interim and late PET/CT showed signs of pseudoprogression in 2/17 (11.8%) patients, who were characterized as PMD on interim PET/CT according to EORTC and “switched” to MB on late imaging. In contrary, no cases of pseudoprogression were observed after application of PERCIMT, highlighting the ability of the novel criteria in tackling this atypical response pattern. Moreover, the results of survival analysis, exhibiting a lower PFS for patients with early PMD (no-MB) compared to those with early MB, are another indirect proof of the rather low incidence of the phenomenon in this cohort. This is in line with previous results published in the literature documenting non-negligible, but not higher than 10% rates of the phenomenon in melanoma immunotherapy [36, 44, 45], and provides supporting evidence to the standpoint that an increase in tumor burden observed during ICI treatment more likely reflects true progression rather than pseudoprogression [41, 46, 47].
irAEs represent another source of false-positive findings on imaging. Radiologic manifestations of irAEs have been reported with variable incidences, reaching up to 31% of patients under ICIs [47–49]. Although the specific characteristics of individual patients play a significant role, the emergence of these toxicities is mainly dependent on the agents used, with the combination of an anti-CTLA-4 antibody and an anti-PD-1 antibody increasing both their incidence and severity [50]. In the herein studied cohort, 11/25 (44%) of the patients showed signs of irAEs on PET/CT. In line with data from the literature [50], the vast majority of these patients (9/11 patients) received combination treatment of nivolumab and ipilimumab. Colitis was the most frequent adverse event on PET/CT, observed in five patients; four of these patients received PD-1 inhibitors in combination with the anti-CTLA-4 regiment ipilimumab, which is known to often induce this reaction [51, 52]. We recognize that the diagnosis of colitis on PET imaging can be complicated, since its identification can be hampered by physiological metabolic activity in the colon. It is well known that enhanced colon [18F]FDG uptake of benign etiology is frequently observed in asymptomatic individuals [53, 54]. Moreover, several studies have shown that patients using the oral hypoglycemic drug metformin tend to have a diffusely increased tracer uptake in the colon [55–58]. In the present cohort, no patient had diabetes; thus, metformin can be ruled out as a cause for false-positive [18F]FDG accumulation in the colon. A further search in patients’ clinical history revealed that one of these five patients developed severe diarrhea during immunotherapy, most likely as a symptom of treatment-induced colitis, while the rest four patients did not have such symptoms. This higher incidence of “PET-colitis” under immunotherapy, compared to clinical signs of colitis, is in line with previously published results [59]. In this context, early recognition of radiologic irAEs could be potentially important since they may precede or correlate with clinical symptoms [37, 47], potentially leading to respective changes in management.
Another aspect pertaining to irAEs is that their emergence has been associated with a favorable efficacy of ICIs—mainly of PD-1 inhibitors—implying a potential predictive role of these events for response to ICI treatment [17, 18, 60]. Our analysis revealed that patients with radiologic signs of irAEs had a longer PFS than those without irAEs; however, this difference was non-significant. Apart from the relatively small cohort studied, an explanation for this finding may lie in the fact that most patients (82%) also received the CTLA-4 inhibitor ipilimumab in combination with the PD-1 inhibitor nivolumab; in a recently published meta-analysis of 30 studies including 4971 individuals, it was shown that no significant association between irAE development and a favorable benefit on survival is observed in ICI combination treatments, in contrary to anti-PD-1 monotherapies [60]. Another reason may lie in the nature of the observed irAEs, affecting the gastrointestinal tract in more than half of the cases (6/11 patients, 55%); data from the above-mentioned meta-analysis also highlight the lack of PFS benefit in patients presenting gastrointestinal irAEs [60].
Finally, the predictive role of baseline LDH before initiation of PD-1 inhibitors was investigated. Although serum LDH elevation is not specific for melanoma, it represents a poor prognostic factor and is one of the most influential factors associated with treatment response [61, 62]. An interesting finding of our analysis is the lack of any adverse effect of elevated baseline LDH on survival, which can be however attributed to the small number of patients with pathologic LDH (n = 4 patients, 16%).
We note some limitations in our study. Firstly, due to the strict inclusion criteria applied, the number of included patients was relatively low, not allowing us to draw more firm conclusions; ideally, further studies with larger patient cohorts would be required. Secondly, although the focus of the study was anti-PD-1 treatment, not all patients underwent exclusively PD-1 blockade, with several of them receiving combination therapy of PD-1 and CTLA-4 inhibitors. Although our patient cohort is too small to afford a PET/CT subanalysis based on a dichotomization of patients into those receiving anti-PD-1 monotherapy and those undergoing combined treatment, we highlight the similar approach followed in previous studies in the field [28, 63]. Finally, the vast majority of the PET/CT-positive, melanoma-consistent findings were not histopathologically confirmed. However, this is not usually possible in the clinical setting.
Conclusion
In an attempt to identify early and reliable biomarkers of survival prediction in immunotherapy of metastatic melanoma, we assessed the prognostic role of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment. Our results showed that tumor response as classified by the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Supplementary information
ESM 1 (DOCX 166 kb)
Funding
Open Access funding enabled and organized by Projekt DEAL.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all participants enrolled in the study.
This article is part of the Topical Collection on Oncology - General.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY | 33336264 | 18,710,997 | 2021-06 |
What was the dosage of drug 'IPILIMUMAB'? | Interim [18F]FDG PET/CT can predict response to anti-PD-1 treatment in metastatic melanoma.
In an attempt to identify biomarkers that can reliably predict long-term outcomes to immunotherapy in metastatic melanoma, we investigated the prognostic role of [18F]FDG PET/CT, performed at baseline and early during the course of anti-PD-1 treatment.
Twenty-five patients with stage IV melanoma, scheduled for treatment with PD-1 inhibitors, were enrolled in the study (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients; nivolumab/ipilimumab, 13 patients). [18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of PD-1 blockade administration (interim PET/CT). Seventeen patients underwent also a third PET/CT scan after administration of four cycles of treatment. Evaluation of patients' response by means of PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria and the PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT). Response to treatment was classified into 4 categories: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD). Patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD. Moreover, patterns of [18F]FDG uptake suggestive of radiologic immune-related adverse events (irAEs) were documented. Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause.
Median follow-up from interim PET/CT was 24.2 months (19.3-41.7 months). According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), and 6 of them had no-MB (PMD). With regard to PFS, no significant difference was observed between patients with MB and no-MB on interim PET/CT according to the EORTC criteria (p = 0.088). In contrary, according to the PERCIMT criteria, patients demonstrating MB had a significantly longer PFS than those showing no-MB (p = 0.045). The emergence of radiologic irAEs (n = 11 patients) was not associated with a significant survival benefit. Regarding the sub-cohort undergoing also a third PET/CT, 14/17 patients (82%) showed concordant responses and 3/17 (18%) had a mismatch of response assessment between interim and late PET/CT.
PET/CT-based response of metastatic melanoma to PD-1 blockade after application of the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Introduction
Metastatic melanoma is a highly aggressive tumor, largely refractory to existing therapies, and associated with a very poor prognosis [1]. While until lately the treatment options for metastatic melanoma were limited, the recent development and introduction in clinical practice of several novel immunotherapeutic agents as well as of targeted therapy with BRAF and MEK inhibitors have revolutionized the systemic treatment of the disease, leading to unprecedented response and survival rates of melanoma patients [2].
The main form of immunotherapy applied in this new era of melanoma management involves immune checkpoint blockade. This immunomodulatory approach activates the immune system against tumors through the binding of the cytotoxic T lymphocyte–associated protein 4 (CTLA-4) and/or the programmed cell death protein 1 (PD-1), both of which are expressed by T cells [3, 4]. The monoclonal antibody ipilimumab, which acts by blocking CTLA-4, is considered a landmark agent in this context, being the first immunotherapeutic drug demonstrating a clear benefit in survival of patients with advanced melanoma, which led to its approval by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2011 [5]. A few years later, a second class of immune checkpoint inhibitors (ICIs), the PD-1 inhibitors nivolumab and pembrolizumab, were also approved for the treatment of melanoma, after having shown survival benefit in these patients [6–8]. Moreover, the anti-PD-1 monoclonal antibodies have shown superiority over ipilimumab, leading to their application both as single agents and in combination with ipilimumab, which is nowadays seldom used as monotherapy [9–13].
Despite these dramatic improvements, a significant amount of patients—approximately 40–45%—show no response to immunotherapy [14]. Additionally, the mechanism of action of these agents (which is markedly different from usual cytotoxic approaches—notably by generating inflammations rather than direct lysis) can pose relevant challenges in the interpretation of treatment response by conventional imaging approaches [15]. Furthermore, several patients experience a “new class” of cumulative, dose-dependent, and sometimes life-threatening side effects, the immune-related adverse events (irAEs), which scope is wide [16]; importantly, the occurrence of such irAEs may be of prognostic value, revealing a response to immunotherapy [17, 18]. These issues raise the question of how to evaluate the response to ICIs in a reliable fashion and early in the course of treatment. This information would help discriminate responders from non-responders, offering significant therapeutic and prognostic implications in the entire spectrum of patient management. Unfortunately, there exist at present only few reliable predictors of long-term response to immunotherapy.
[18F]FDG PET/CT is considered the elective imaging technique in detecting metastatic disease in advanced melanoma [19–23]. Moreover, a growing amount of recently published literature has highlighted the potential role of the modality in the prediction of treatment response to ICIs in melanoma, rendering it an attractive tool for the monitoring of immunotherapy [24–31].
In quest of identifying reliable biomarkers for the prediction of long-term outcomes to immunotherapy, we aim in the present prospective study to assess the value of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment.
Materials and methods
Patients
Twenty-five patients (12 males, 13 females; mean age 54.7 years) with unresectable, stage IV melanoma undergoing immunotherapy with PD-1 inhibitors applied either as monotherapy (pembrolizumab, nivolumab) or as combination treatment with CTLA-4 inhibitors (nivolumab/ipilimumab) were enrolled in the study (Table 1). Pembrolizumab was administered intravenously at a dose of 2 mg/kg every 3 weeks, and nivolumab was administered intravenously at a dose of 3 mg/kg every 2 weeks. The combination ICI therapy was administered as an induction of 4 cycles of nivolumab (1 mg/kg) and ipilimumab (3 mg/kg) every 3 weeks, followed by single-agent nivolumab administration (3 mg/kg) every 2 weeks. The included patients had not received chemotherapy for at least 1 month prior to the initial PET/CT studies. None of the patients had a history of diabetes. Patients gave written informed consent to participate in the study and to have their medical records released. The study was approved by the Ethical Committee of the University of Heidelberg and the Federal Agency for Radiation Protection (Bundesamt für Strahlenschutz).Table 1 Patient characteristics
Patient number Age Gender LDH at baseline (U/I) Previous treatment ICI treatment EORTC PET response PERCIMT PET response Radiologic signs of irAEs Progression PFS (months)
1 56 F 770 Chemosaturation with melphalan, pembrolizumab, gemcitabine/treosulfan Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) Thyroiditis Yes 1.5
2 34 F 247 Pembrolizumab, IMCgp100 Nivolumab/ipilimumab CMR (MB) CMR (MB) Bone marrow activation, colitis No 17.9
3 46 F 166 Vemurafenib/cobimetinib Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Duodenitis Yes 4.0
4 54 M 218 Nivolumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Colitis Yes 3.2
5 53 M 275 No Nivolumab/ipilimumab SMD (MB) SMD (MB) - Yes 0.4
6 50 F 344 Vemurafenib/cobimetinib Nivolumab/ipilimumab SMD (MB) SMD (MB) Bone marrow activation, lymphadenopathy No 19.3
7 59 M 204 No Nivolumab/ipilimumab PMR (MB) PMR (MB) Sarcoid-like reaction, arthitis No 38.7
8 44 F 340 Dabrafenib/trametinib Nivolumab/ipilimumab PMR (MB) PMR (MB) Bone marrow activation, colitis No 18.3
9 60 F 269 No Nivolumab/ipilimumab PMR (MB) PMR (MB) – No 21.2
10 48 F 246 No Nivolumab/ipilimumab PMD (no-MB) SMD (MB) Lymphadenopathy No 24.2
11 55 F 224 No Nivolumab SMD (MB) SMD (MB) - Yes 6.2
12 84 F 195 No Pembrolizumab SMD (MB) SMD (MB) - Yes 11.0
13 79 M 205 No Pembrolizumab PMD (no-MB) PMD (no-MB) Arthritis Yes 2.0
14 20 F 186 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 17.8
15 52 M 275 No Pembrolizumab PMR (MB) PMR (MB) - No 42.1
16 52 M 170 No Pembrolizumab PMD (no-MB) SMD (MB) - Yes 14.6
17 53 F 260 No Nivolumab SMD (MB) SMD (MB) - Yes 9.6
18 65 M 201 Ipilimumab Pembrolizumab SMD (MB) SMD (MB) - Yes 1.9
19 67 F 290 No Pembrolizumab PMD (no-MB) SMD (MB) Bone marrow activation, colitis Yes 5.9
20 55 M 200 Ipilimumab/nivolumab, dabrafenib/trametinib Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 1.5
21 58 M 183 Pembrolizumab Nivolumab/ipilimumab PMR (MB) PMR (MB) Gastritis, colitis Yes 17.0
22 50 M 195 No Nivolumab/ipilimumab PMD (no-MB) PMD (no-MB) - Yes 12.1
23 80 M 364 No Pembrolizumab PMD (no-MB) PMD (no-MB) - Yes 0.9
24 47 F 256 Dabrafenib Nivolumab SMD (MB) SMD (MB) - Yes 1.1
25 47 M 271 Ipilimumab, dabrafenib Nivolumab PMD (no-MB) SMD (MB) - Yes 2.7
F, female; M, male; LDH, lactate dehydrogenase; EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; irAEs, immune-related adverse events; PFS, progression-free survival
[18F]FDG PET/CT data acquisition
[18F]FDG PET/CT was performed before the start of treatment (baseline PET/CT) and after the initial two cycles of ICIs’ administration (interim PET/CT) in all 25 patients. Moreover, 17 patients of the cohort also had a third PET/CT scan within 2 weeks after administration of four cycles of treatment.
Patients underwent a whole body PET/CT after intravenous administration of maximum 250 MBq [18F]FDG 60 min post-injection (p.i.). Imaging was performed from the head to the feet with an image duration of 2 min per bed position. A dedicated PET/CT system (Biograph mCT, S128, Siemens Co., Erlangen, Germany) with an axial field of view of 21.6 cm with TruePoint and TrueV, operated in a three-dimensional mode was used. A low-dose attenuation CT (120 kV, 30 mA) was used for attenuation correction of the PET data and for image fusion. All PET images were attenuation-corrected and an image matrix of 400 × 400 pixels was used for iterative image reconstruction. Iterative image reconstruction was based on the ordered subset expectation maximization (OSEM) algorithm with two iterations and 21 subsets as well as time of flight (TOF).
[18F]FDG PET/CT data analysis
Data analysis consisted of visual (qualitative) assessment of the PET/CT scans and semi-quantitative evaluation based on standardized uptake value (SUV) calculations. PET/CT images were analyzed on an Aycan workstation by three nuclear medicine physicians (CS, DP, ADS). Images were interpreted by consensus. Visual analysis was based on the identification of sites of focal, non-physiologic [18F]FDG uptake above surrounding background activity, which were considered consistent with melanoma lesions.
Moreover, patterns of [18F]FDG uptake on interim PET/CT suggestive of radiologic manifestations of irAEs to immunotherapy were documented. Based on our experience and the published literature in the field [29, 32], we defined radiologic irAEs as sites of newly emerging, increased compared to baseline imaging, non-malignant [18F]FDG accumulation in organs known to exhibit immune-related signs on PET/CT. In particular, a new, diffusely enhanced tracer uptake in organs such as the gastrointestinal tract (mostly colon), the thyroid gland and the bone marrow, or, respectively, a new, relatively symmetrical, increased uptake in lymph nodes (e.g., mediastinal/hilar, inguinal) and in joints following ICIs were considered suggestive of radiologic irAEs in these organs. Semi-quantitative evaluation was based on volumes of interest (VOIs) and on subsequent calculation of SUVmean and SUVmax. VOIs were drawn using the pseudo-snake algorithm of the Pmod software (http://www.pmod.com/files/download/v31/doc/pbas/4729.htm) and were placed over melanoma lesions.
Response evaluation
Evaluation of patients’ response by means of [18F]FDG PET/CT was performed after application of the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria [33] as well as the recently proposed PET Response Evaluation Criteria for IMmunoTherapy (PERCIMT) [26]. Both criteria classify tumor response into 4 categories. The major difference between these criteria lies in the number of newly emerging lesions between baseline and follow-up PET/CT for the characterization of PMD: according to EORTC, the appearance of one new hypermetabolic lesion leads to patient classification to PMD, but according to PERCIMT, this requires the appearance of a minimum of four new lesions below 1 cm or respectively ≥ 3 new lesions of 1.0–1.5 cm or ≥ 2 new lesions of more than 1.5 cm. Another difference between the criteria is the role of SUV, which is central in EORTC, while it is not taken into account in PERCIMT (Table 2).Table 2 Summary of the EORTC and PERCIMT response criteria
EORTC PERCIMT
CMR Complete resolution of [18F]FDG uptake within the tumor volume Complete resolution of all pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
PMR Decrease in tumor SUV > 25% after more than 1 therapeutic cycle Complete resolution of some pre-existing [18F]FDG avid lesions. No new, [18F]FDG avid lesions.
SMD Increase in tumor SUV < 25% or decrease in SUV < 15% Neither PMD nor PMR/CMR
PMD Increase in tumor SUV > 25% or appearance of new lesions ≥ 4 new lesions of less than 1 cm in functional diameter or ≥ 3 new lesions between 1.0–1.5 cm in functional diameter or ≥ 2 new lesions of more than 1.5 cm in functional diameter
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease; SUV, standardized uptake value
Stable disease (SD) represents a satisfactory outcome following immunotherapy, since—in contrast to conventional chemotherapy—it can be durable and survival rates related to SD are comparable to those associated with response [34–36]. Based on this, patients were further grouped into two groups: those demonstrating metabolic benefit (MB), including patients with SMD, PMR, and CMR, and those demonstrating no MB (no-MB), including patients with PMD.
Statistical analysis
Progression-free survival (PFS) was measured from the date of interim PET/CT until disease progression or death from any cause. Kaplan-Meier estimates were generated and median PFS estimated. Median follow-up time was determined by inverse Kaplan-Meier estimation. For univariate comparison of PFS, a log-rank test was used. Statistical analysis was performed using R version 4.0.2 (The R Foundation for Statistical Computing 2020) and R packages survival and prodlim. The results were considered significant for p values less than 0.05 (p < 0.05).
Results
Patient cohort
All included patients received treatment with anti-PD-1 agents, applied either as monotherapy (pembrolizumab, n = 8 patients; nivolumab, n = 4 patients) or as combination therapy (nivolumab/ipilimumab, n = 13 patients). The mean baseline serum lactate dehydrogenase (LDH) was 262 U/l with four patients having pathologically high LDH levels and 21 of them normal levels. The detailed characteristics of the studied patients are presented in Table 1.
PET/CT findings
The findings of interim PET/CT were compared to those of the baseline scan and PET/CT-based response evaluation was performed for all 25 patients. According to the EORTC criteria, 14 patients showed MB (1 CMR, 6 PMR, and 7 SMD), while 11 patients showed no-MB (PMD). Respectively, the application of the PERCIMT criteria revealed that 19 patients had MB (1 CMR, 6 PMR, and 12 SMD), while six of them had no-MB (PMD) (Table 3) (Fig. 1).Table 3 Summary of the patients’ classifications in different response groups based on interim PET/CT and according to the EORTC and PERCIMT response criteria (n = 25 patients)
Metabolic benefit No-metabolic benefit
CMR PMR SMD PMD
EORTC 1 6 7 11
PERCIMT 1 6 12 6
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 1 Maximum intensity projection (MIP) [18F]FDG PET/CT images of a 34-year-old woman with metastatic melanoma before initiation of immunotherapy with nivolumab/ipilimumab (A) and after administration of two cycles of treatment (B). Baseline PET/CT image shows multiple lymph node, pulmonary, hepatic, adrenal, soft tissue, and osseous metastases (A). Interim PET/CT shows a complete metabolic remission (CMR) of all baseline lesions. The patient demonstrated metabolic benefit (MB) according to both the EORTC and PERCIMT criteria (B). Moreover, diffusely increased [18F]FDG uptake is observed in the ascending colon and the bone marrow on interim PET/CT. At the time of writing, the patient was still progression-free having reached a PFS of 17.9 months
With regard to the subgroup of 17 patients undergoing three PET/CT examinations (baseline, interim, late), the following results were revealed for interim PET/CT: 11 patients had MB (1 CMR, 5 PMR, and 5 SMD) and six patients had no-MB (PMD) according to EORTC, while 14 patients had MB (1 CMR, 5 PMR, and 8 SMD) and three of them had no-MB (PMD) according to PERCIMT. Respectively, on late PET/CT imaging, 13 patients had MB (3 CMR, 5 PMR, and 5 SMD) and four patients had no-MB (PMD) according to both EORTC and PERCIMT (Table 4). Two of the three patients exhibiting a mismatch between EORTC (PMD, no-MB) and PERCIMT (SMD, MB) on interim PET/CT, finally showed MB on the third examination based on both criteria (pseudoprogression) (Fig. 2). Respectively, one patient with early signs of PMD (no-MB) according to EORTC and SMD (MB) according to PERCIMT eventually exhibited PMD (no-MB) on the third examination based on both criteria.Table 4 Summary of the patients’ classifications in different response groups based on interim (after two cycles of ICIs) and late (after four cycles of ICIs) PET/CT (n = 17 patients)
EORTC PERCIMT
MB (late PET/CT) No-MB (late PET/CT) MB (late PET/CT) No-MB (late PET/CT)
CMR PMR SMD PMD CMR PMR SMD PMD
1.MB (interim PET/CT) CMR 1 0 0 0 1 0 0 0
PMR 0 5 0 0 0 5 0 0
SMD 1 0 0 4 2 0 5 1
No-MB (interim PET/CT) PMD 1 0 1 4 0 0 0 3
EORTC, European Organization for Research and Treatment of Cancer; PERCIMT, PET Response Evaluation Criteria for IMmunoTherapy; MB, metabolic benefit; no-MB, no metabolic benefit; CMR, complete metabolic response; PMR, partial metabolic response; SMD, stable metabolic disease; PMD, progressive metabolic disease
Fig. 2 Transaxial PET/CT (upper row, A–C) and low-dose CT (lower row, D–F) images at the cervical level of a 48-year-old female patient with advanced melanoma. The PET/CT (A) and CT (D) images obtained before immunotherapy show no pathologic lesions. Interim PET/CT performed after two cycles of nivolumab/ipilimumab shows a new [18F]FDG-avid lymph node (white arrow; B, E), suspicious of metastatic involvement. According to the EORTC criteria, the patient showed progressive metabolic disease (PMD), while according to PERCIMT, he had stable metabolic disease (SMD). A third PET/CT obtained after administration of four cycles of nivolumab/ipilimumab shows remission of the lesion (C, F), suggesting pseudoprogression of the cervical finding on interim PET/CT. The patient had a PFS of 24.2 months and was still progression-free at last contact
In total, 11 patients had PET/CT findings suggestive of irAEs on interim PET/CT, the majority of whom were under combination treatment: nine patients received nivolumab and ipilimumab, while two of them were under pembrolizumab (p = 0.015). In particular, the most common radiologic adverse event was a diffusely increased [18F]FDG uptake in the colon, defined as colitis, which was observed in five patients. One of these five patients also developed severe diarrhea, as clinical sign of treatment-induced colitis. Other gastrointestinal tract radiologic irAEs included gastritis (n = 1 patient) and duodenitis (n = 1 patient), defined as diffuse increased tracer uptake in the stomach and duodenum, respectively. Moreover, arthritis, defined as diffuse increased, periarticular, symmetrical tracer uptake in joints, was observed in two patients, and thyroiditis, a diffuse increased uptake in the thyroid gland, was observed in one patient. Furthermore, reactive, increased, symmetrical uptake in lymph nodes was observed in three patients, one of whom exhibited a sarcoid-like lymphadenopathy. Finally, diffuse increased bone marrow uptake was seen in four patients representing bone marrow activation in terms of a systemic immune response [37] (Table 1).
Survival analysis
Median follow-up (95% CI) of the patient cohort from interim PET/CT was 24.2 months (19.3–41.7 months). Patients receiving combination treatment (nivolumab/ipilimumab) had a median PFS of 17.8 months (4.0–NA), while those receiving PD-1 blockade monotherapy (nivolumab or pembrolizumab) had a median PFS of 4.3 months (1.9–NA) (p = 0.016) (Fig. 3).Fig. 3 Kaplan-Meier estimates of PFS according to the anti-PD-1 treatment applied. The numbers of patients at risk in each group and for the respective time points are shown below the plots. Combi, combination therapy (ipilimumab/nivolumab); Mono, monotherapy (nivolumab or pembrolizumab)
Based on the EORTC criteria, patients with MB on interim PET/CT had a median PFS of 14.0 months (6.2–NA), while those with no-MB had a median PFS of 4.0 months (2.0–NA) (p = 0.088) (Fig. 4A). Respectively, according to the PERCIMT criteria, patients with MB had a median PFS of 11.0 months (5.9–NA), while those with no-MB had a median PFS of 1.8 months (1.5–NA) (p = 0.045) (Fig. 4B).Fig. 4 Kaplan-Meier estimates of PFS according to the EORTC (A) and the PERCIMT (B) criteria. The numbers of patients at risk in each group and for the respective time points are shown below the plots. MB, metabolic benefit; no-MB, no metabolic benefit
The patient cohort was further dichotomized on the basis of the emergence of radiologic irAEs on interim PET/CT. Patients with irAEs had a median PFS of 17.0 months (4.0–NA), and patients without irAEs had a median PFS of 7.9 months (1.9–NA) (p = 0.128) (Fig. 5).Fig. 5 Kaplan-Meier estimates of PFS according to the emergence of radiologic irAEs on interim PET/CT. The numbers of patients at risk in each group and for the respective time points are shown below the plots
Finally, the potential correlation of baseline LDH with PFS was also investigated. However, pathologic levels of LDH had no adverse effect on survival of the cohort (p = 0.642) (Supplementary File 1).
Discussion
The therapeutic benefit of immune checkpoint blockade of PD-1 and CTLA-4 in the treatment of metastatic melanoma is variable [38]. Our understanding of how ICIs affect T cell evolution is incomplete [39], limiting the ability to derive full clinical benefit and, moreover, to predict responses from these drugs. However, tracking early response to immunotherapy is key for treatment options.
In this study, we investigated the role of interim PET/CT, performed after application of two cycles of anti-PD-1 treatment, in prediction of survival of metastatic melanoma patients. Our results showed that tumor response as classified by the PERCIMT criteria is significantly correlated with PFS, with metabolic responders demonstrating a significant survival benefit over non-responders. Respectively, the application of the EORTC criteria also led to a higher PFS for metabolic responders compared to non-responders but this difference was not statistically significant. These findings highlight the ability of [18F]FDG PET/CT—in particular, after application of the recently introduced PERCIMT criteria—to monitor and predict response to anti-PD-1 agents at an early but clinically relevant time point. This is of particular importance in clinical decision-making and prediction of outcome early during the course of immunotherapy, rendering PET/CT a potentially significant tool for the management of these patients. The main strengths of our study include its prospective nature, its rigorous protocol with imaging performed at strictly defined time points during treatment, the standardized for all patients PET/CT procedure, and the correlation with survival analysis.
Hitherto, a non-negligible number of studies have evaluated the efficacy of PET/CT in predicting treatment response of metastatic melanoma patients to ICIs. While most papers have focused on later time points during the course or after the end of treatment [26, 28, 30–32, 40], few of them also reported on application of the imaging modality early during immunotherapy. Our group previously showed in a cohort of 22 patients that PET/CT performed after two ipilimumab cycles—and after application of the EORTC criteria—correctly predicted treatment response after completion of the 4-cycle treatment in the majority (87%) of PMD patients and in all SMD patients [24]. In an expanded analysis of the ipilimumab patient cohort (n = 41 patients), the capacity of interim PET/CT in predicting clinical benefit to the agent was also highlighted. In that analysis, the performance of the PERCIMT criteria was superior to that of EORTC, which is in line with the results of the present study [27]. Furthermore, Cho et al. studied 20 melanoma patients treated with different ICIs (anti-PD-1 and anti-CTLA-4) with PET/CT at 3–4 weeks into therapy and found that a combination of changes in lesional dimensions along with changes in [18F]FDG uptake is a more accurate predictor of eventual response than each of these parameters alone. The authors proposed the PET/CT criteria for early prediction of response to immune checkpoint inhibitor therapy (PECRIT) criteria, based on a combination of the response evaluation criteria in solid tumors (RECIST) and PET response criteria in solid tumors (PERCIST) [25].
The current analysis represents the first study focusing on the role of interim PET/CT, performed as early as after application of two cycles of PD-1 blockade, in survival prediction of metastatic melanoma patients undergoing treatment with this class of ICIs. Taken together, the herein presented findings as well as those of previous studies in the field build further evidence on the potentially significant role of PET/CT performed early during the course of immunotherapy for prediction and stratification of response to treatment.
Novel patterns of response and progression, not previously seen with conventional therapies (such as cytotoxic or targeted anticancer regiments), have been described under immunotherapy and are attributed to the unique mechanism of action of these agents. In particular, phenomena such as pseudoprogression and irAEs may render response assessment to ICI challenging, questioning the utility of imaging modalities.
Evaluation of response to immunotherapy by means of PET/CT is primarily visual and subjective in nature. Considering the aforementioned challenges raised by the advent of immunotherapy for imaging interpretation, our group has recently introduced the PERCIMT criteria in metastatic melanoma, in an attempt to meet the need for reliable response assessment based on PET/CT. The cornerstone of these criteria is the finding that the absolute number of newly emerged [18F]FDG-avid lesions is more predictive of clinical outcome than SUV changes [26]. In specific, the application of a threshold of four newly emerged lesions on post-therapy PET/CT scan—with a decreasing cutoff of lesion number as the functional diameter of the lesions increases—in a cohort of 41 patients could predict clinical benefit to treatment with the agent ipilimumab better than the standard threshold of one new lesion or an increase in SUV, conventionally applied with the EORTC criteria. This was also confirmed in the present analysis with a significant correlation of metabolic response on interim PET with PFS only after application of PERCIMT.
Pseudoprogression, defined as an initial increase of tumor burden before the disease responds to therapy, has been initially described in melanoma patients undergoing ipilimumab therapy [41]. Since this phenomenon may be misclassified as progressive disease, the recently modified radiologic, immune-related response criteria (irRECIST, iRECIST) call for a 4-week re-assessment in order to overcome this limitation [42, 43]. In our study, the evaluation of pseudoprogression was partly feasible in the subgroup of 17 patients undergoing a third PET/CT after administration of four cycles of treatment. The comparison between interim and late PET/CT showed signs of pseudoprogression in 2/17 (11.8%) patients, who were characterized as PMD on interim PET/CT according to EORTC and “switched” to MB on late imaging. In contrary, no cases of pseudoprogression were observed after application of PERCIMT, highlighting the ability of the novel criteria in tackling this atypical response pattern. Moreover, the results of survival analysis, exhibiting a lower PFS for patients with early PMD (no-MB) compared to those with early MB, are another indirect proof of the rather low incidence of the phenomenon in this cohort. This is in line with previous results published in the literature documenting non-negligible, but not higher than 10% rates of the phenomenon in melanoma immunotherapy [36, 44, 45], and provides supporting evidence to the standpoint that an increase in tumor burden observed during ICI treatment more likely reflects true progression rather than pseudoprogression [41, 46, 47].
irAEs represent another source of false-positive findings on imaging. Radiologic manifestations of irAEs have been reported with variable incidences, reaching up to 31% of patients under ICIs [47–49]. Although the specific characteristics of individual patients play a significant role, the emergence of these toxicities is mainly dependent on the agents used, with the combination of an anti-CTLA-4 antibody and an anti-PD-1 antibody increasing both their incidence and severity [50]. In the herein studied cohort, 11/25 (44%) of the patients showed signs of irAEs on PET/CT. In line with data from the literature [50], the vast majority of these patients (9/11 patients) received combination treatment of nivolumab and ipilimumab. Colitis was the most frequent adverse event on PET/CT, observed in five patients; four of these patients received PD-1 inhibitors in combination with the anti-CTLA-4 regiment ipilimumab, which is known to often induce this reaction [51, 52]. We recognize that the diagnosis of colitis on PET imaging can be complicated, since its identification can be hampered by physiological metabolic activity in the colon. It is well known that enhanced colon [18F]FDG uptake of benign etiology is frequently observed in asymptomatic individuals [53, 54]. Moreover, several studies have shown that patients using the oral hypoglycemic drug metformin tend to have a diffusely increased tracer uptake in the colon [55–58]. In the present cohort, no patient had diabetes; thus, metformin can be ruled out as a cause for false-positive [18F]FDG accumulation in the colon. A further search in patients’ clinical history revealed that one of these five patients developed severe diarrhea during immunotherapy, most likely as a symptom of treatment-induced colitis, while the rest four patients did not have such symptoms. This higher incidence of “PET-colitis” under immunotherapy, compared to clinical signs of colitis, is in line with previously published results [59]. In this context, early recognition of radiologic irAEs could be potentially important since they may precede or correlate with clinical symptoms [37, 47], potentially leading to respective changes in management.
Another aspect pertaining to irAEs is that their emergence has been associated with a favorable efficacy of ICIs—mainly of PD-1 inhibitors—implying a potential predictive role of these events for response to ICI treatment [17, 18, 60]. Our analysis revealed that patients with radiologic signs of irAEs had a longer PFS than those without irAEs; however, this difference was non-significant. Apart from the relatively small cohort studied, an explanation for this finding may lie in the fact that most patients (82%) also received the CTLA-4 inhibitor ipilimumab in combination with the PD-1 inhibitor nivolumab; in a recently published meta-analysis of 30 studies including 4971 individuals, it was shown that no significant association between irAE development and a favorable benefit on survival is observed in ICI combination treatments, in contrary to anti-PD-1 monotherapies [60]. Another reason may lie in the nature of the observed irAEs, affecting the gastrointestinal tract in more than half of the cases (6/11 patients, 55%); data from the above-mentioned meta-analysis also highlight the lack of PFS benefit in patients presenting gastrointestinal irAEs [60].
Finally, the predictive role of baseline LDH before initiation of PD-1 inhibitors was investigated. Although serum LDH elevation is not specific for melanoma, it represents a poor prognostic factor and is one of the most influential factors associated with treatment response [61, 62]. An interesting finding of our analysis is the lack of any adverse effect of elevated baseline LDH on survival, which can be however attributed to the small number of patients with pathologic LDH (n = 4 patients, 16%).
We note some limitations in our study. Firstly, due to the strict inclusion criteria applied, the number of included patients was relatively low, not allowing us to draw more firm conclusions; ideally, further studies with larger patient cohorts would be required. Secondly, although the focus of the study was anti-PD-1 treatment, not all patients underwent exclusively PD-1 blockade, with several of them receiving combination therapy of PD-1 and CTLA-4 inhibitors. Although our patient cohort is too small to afford a PET/CT subanalysis based on a dichotomization of patients into those receiving anti-PD-1 monotherapy and those undergoing combined treatment, we highlight the similar approach followed in previous studies in the field [28, 63]. Finally, the vast majority of the PET/CT-positive, melanoma-consistent findings were not histopathologically confirmed. However, this is not usually possible in the clinical setting.
Conclusion
In an attempt to identify early and reliable biomarkers of survival prediction in immunotherapy of metastatic melanoma, we assessed the prognostic role of interim [18F]FDG PET/CT performed after the first two cycles of anti-PD-1 treatment. Our results showed that tumor response as classified by the recently proposed PERCIMT criteria is significantly correlated with PFS. This highlights the potential ability of [18F]FDG PET/CT for early stratification of response to anti-PD-1 agents, a finding with possible significant clinical and financial implications. Further studies including larger numbers of patients are necessary to validate these results.
Supplementary information
ESM 1 (DOCX 166 kb)
Funding
Open Access funding enabled and organized by Projekt DEAL.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all participants enrolled in the study.
This article is part of the Topical Collection on Oncology - General.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | 3 MILLIGRAM/KILOGRAM, Q3WK | DrugDosageText | CC BY | 33336264 | 18,710,997 | 2021-06 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Concomitant disease aggravated'. | De novo lupus nephritis during treatment with belimumab.
In light of reports of de novo LN during belimumab (BLM) treatment, we sought to determine its frequency and contributing or protective factors in a real-life setting.
Patients with SLE who received BLM between 2011 and 2017 at five European academic practices were enrolled (n = 95) and followed longitudinally for a median time of 13.1 months [interquartile range (IQR): 6.0-34.7]; 52.6% were anti-dsDNA positive, 60.0% had low complement levels, and 69.5% had no renal involvement prior to/at BLM initiation [mean disease duration at baseline: 11.4 (9.3) years]. Age- and sex-matched patients with non-renal SLE who had similar serological profiles, but were not exposed to BLM, served as controls (median follow-up: 132.0 months; IQR: 98.3-151.2).
We observed 6/66 cases (9.1%) of biopsy-proven de novo LN (4/6 proliferative) among the non-renal BLM-treated SLE cases after a follow-up of 7.4 months (IQR: 2.7-22.2). Among controls, 2/66 cases (3.0%) of de novo LN (both proliferative) were observed after 21 and 50 months. BLM treatment was associated with an increased frequency and/or shorter time to de novo LN [hazard ratio (HR): 10.7; 95% CI: 1.7, 67.9; P = 0.012], while concomitant use of antimalarial agents along with BLM showed an opposing association (HR: 0.2; 95% CI: 0.03, 0.97; P = 0.046).
Addition of BLM to standard-of-care did not prevent LN in patients with active non-renal SLE, but a favourable effect of concomitant use of antimalarials was implicated. Studies of whether effects of B-cell activating factor inhibition on lymphocyte subsets contribute to LN susceptibility are warranted.
pmc Rheumatology key messages
Irrespective of prior renal involvement, belimumab treatment may not adequately protect against lupus nephritis.
Concomitant antimalarial therapy along with belimumab was implied to protect against development of lupus nephritis.
Our observations call for vigilance with regard to evolving renal disease during belimumab therapy.
Introduction
SLE is a chronic, multisystem autoimmune disease with unmet needs, such as delayed diagnosis, premature atherosclerosis, drug-associated organ damage and a prominent impairment of health-related quality of life [1]. The wide range of manifestations and serological findings pose challenges with regard to diagnosis and treatment. Today, standard-of-care (SoC) therapy includes glucocorticoids, antimalarials, immunosuppressants and biologic agents, e.g. belimumab (BLM) and rituximab (RTX). The selection of drugs is mainly based on the afflicted organ systems and the organ-specific or global disease activity [2]. LN is a manifestation of SLE with a potentially life-threatening course [3].
BLM is a recombinant human IgG1-λ monoclonal antibody that specifically binds the soluble form of B cell activating factor (BAFF). The efficacy of BLM has been demonstrated to date in five placebo-controlled phase III trials and several observational studies [4]. Although post hoc analysis of clinical trials of BLM showed superiority of BLM over placebo in preventing renal flares [5] and a systematic review suggested an overall promising effect of BAFF inhibition on renal outcomes [6], development of LN during BLM treatment has also been reported [7–10]. Clinical trials of BLM in LN, either as an add-on therapy to SoC or in combination with RTX, are underway [11–13] and the BLISS-LN trial recently demonstrated superiority of addition of BLM to SoC for active LN over SoC alone [14].
We herein report cases of de novo LN during treatment with BLM observed in our academic practices, and cases of LN flares in patients with a history of renal SLE at the time of BLM initiation. We further aimed at identifying factors or risk phenotypes that are associated with the development of LN, in order to contribute to optimized monitoring during treatment with BLM.
Methods
Patients
Patients, classified with SLE according to the 1982 ACR [15] and/or 2012 SLICC [16] criteria, receiving BLM 10 mg/kg intravenously at week 0, 2, 4 and thereafter every fourth week from its approval in 2011 until 31 December 2017 in the Day Care Units of four Swedish academic rheumatology centres (Linköping, Lund, Stockholm and Uppsala) and one academic centre in Leeds, UK, were followed longitudinally within the frame of observational research programmes, and were included in the present report (n = 95). BLM was given as an add-on to background SoC, with no change in SoC implemented unless clinically indicated. None of these patients were given cyclophosphamide, RTX or other B cell depleting agents during treatment with BLM. No patient selection was applied other than consent to participate in the study. Sixty-six of these patients (69.5%) had no history of renal involvement until BLM initiation. As a comparator group to the non-renal SLE cases exposed to BLM, we included 66 non-renal SLE cases from Linköping and Stockholm, individually matched for age and sex, with similar serological profiles (anti-dsDNA positivity, low complement protein 3 and/or 4), who were also followed longitudinally; no selection other than matched serology and age at baseline was applied. Kidney biopsy was performed in the case of a suspected new onset of LN during follow-up. Patient characteristics are detailed in Table 1.
Table 1 Patient characteristics
Item Belimumab-treated SLE Non-renal SLE comparators P-value
Total Non-renal
Background variables
Number of cases, n 95 66 66
Age, mean (s.d.), years 42.2 (14.2) 42.2 (15.2) 43.4 (16.0) 0.152
Females, n (%) 89 (93.7) 63 (95.5) 63 (95.5) NA
Current tobacco smoking, n (%) 11 (12.5); n = 88 9 (15.0); n = 60 14 (21.2) 0.367
Former tobacco smoking, n (%) 25 (28.4); n = 88 14 (23.3); n = 60 23 (34.8) 0.047
Caucasian, n (%) 86 (90.5) 59 (89.4) 64 (97.0) NA
African, n (%) 6 (6.3) 5 (7.6) 0 (0.0) NA
Asian, n (%) 2 (2.1) 2 (3.0) 2 (3.0) NA
Hispanic, n (%) 1 (1.1) 0 (0.0) 0 (0.0) NA
Diabetes until enrolment, n (%) 3 (3.2) 0 (0.0) 0 (0.0) NA
Hypertension until enrolment, n (%) 23 (24.2) 9 (13.6) 14 (21.2) 0.332
Disease variables at enrolment
Duration of SLE, mean (s.d.), years 11.4 (9.3) 10.5 (9.1) 9.8 (11.1)d 0.529
SLEDAI-2K score, mean (s.d.) 9.3 (5.9) 8.2 (4.7) 4.9 (3.7) <0.001
SDI score, median (IQR) 1 (0–1); n = 93 0 (0–1); n = 64 0 (0–2) 0.594
Serological activitya, n (%) 68 (71.6) 47 (71.2) 50 (75.8) 0.250
Anti-dsDNA positive, n (%) 50 (52.6) 33 (50.0) 34 (51.5) 1.000
Low complement, n (%) 57 (60.0) 40 (60.6) 41 (62.1) 1.000
Anti-Smith positive, n (%) 24 (25.3) 16 (24.2) 14 (21.2) 0.832
Main reasons for belimumab
General, n (%) 4 (4.2) 3 (4.5) NA NA
Mucocutaneous, n (%) 55 (57.9) 39 (59.1) NA NA
Musculoskeletal, n (%) 54 (56.8) 39 (59.1) NA NA
Haematological, n (%) 12 (12.6) 8 (12.1) NA NA
Cardiorespiratory, n (%) 6 (6.3) 4 (6.1) NA NA
Renal, n (%) 9 (9.5) 0 (0.0) NA NA
Neurological, n (%) 5 (9.5) 2 (3.0) NA NA
Immunological, n (%) 3 (3.2) 2 (3.0) NA NA
Ongoing concomitant treatments
Daily prednisolone doseb, mean (s.d.), mg 11.3 (9.4) 11.1 (9.4) 7.3 (12.1)e 0.004
Antimalarial agents, n (%) 67 (70.5) 45 (68.2) 36 (54.5) 0.137
Immunosuppressantsc, n (%) 58 (61.1) 40 (60.6) 21 (31.8) 0.002
Azathioprine, n (%) 27 (28.4) 17 (25.8) 6 (9.1) 0.013
Methotrexate, n (%) 14 (14.7) 11 (16.7) 8 (12.1) 0.629
Mycophenolate mofetil/sodium, n (%) 14 (14.7) 11 (16.7) 3 (4.5) 0.057
Other immunosuppressants, n (%) 4 (6.8) 2 (3.0) 5 (7.6) 0.375
In cases of missing values, the total number of available observations (n) is indicated. P-values are derived from comparisons between non-renal SLE patients who were treated with belimumab and individually matched for age and sex non-renal SLE comparators who were not treated with belimumab, using Wilcoxon’s signed rank test for continues variables and McNemar’s test for dichotomous variables, or the χ2 test in cases of missing values in one of the two groups. Significant P-values are indicated in bold. aAnti-dsDNA positivity and/or low complement levels. bAt the time of belimumab initiation or enrolment for the comparators. cExcluding antimalarial agents. dMedian (IQR): 6.4 (0.5–13.4) years. eMedian (IQR): 5.0 (0.0–10.0) mg. IQR: interquartile range; NA: not applicable or not available; SDI: SLICC/ACR Damage Index.
Definitions
We defined de novo LN as a new onset of significant proteinuria, defined as a urinary protein-to-creatinine ratio or protein excretion in 24-h urine collection corresponding to >0.5 g/day, combined with renal histology consistent with LN according to the WHO and/or 2003 International Society of Nephrology/Renal Pathology Society classification [17], in patients who previously had not met the ACR criterion for renal disorder [15].
Global SLE disease activity was evaluated using the SLEDAI-2K [18], and organ damage using the SLICC/ACR Damage Index (SDI) [19]. For SLEDAI-2K scores, laboratory and serological items were assessed based on results from routine tests at the local university hospital laboratories.
Statistics
Comparisons between matched non-renal SLE patients who received BLM vs those who did not were performed using Wilcoxon’s signed rank test for continuous and McNemar’s test for dichotomous variables. The occurrence of de novo LN or LN flares during follow-up was illustrated using Kaplan–Meier curves, and the pairwise log-rank (Mantel–Cox) test was employed to compare the de novo LN distributions between BLM exposed vs not exposed non-renal SLE patients. Contingency between unrelated dichotomous variables was tested using Fisher’s exact test. Proportional hazards (Cox) regression was used to investigate factors and disease phenotypes associated with LN development during therapy. P-values<0.05 were considered statistically significant. IBM SPSS version 25 software (IBM Corp., Armonk, NY, USA) was used for statistical analyses and GraphPad Prism 7 (GraphPad Software Inc., La Jolla, CA, USA) for construction of graphs.
Ethical considerations
The study complied with the ethical principles of the Declaration of Helsinki. Written informed consent was obtained from all patients. The study protocol was approved by regional ethics review boards.
Results
Outcome of cases without prior LN
As shown in Table 1, non-renal SLE patients who were selected for treatment with BLM had comparable serological profiles, disease duration and SDI scores but higher baseline SLEDAI-2K scores [mean (s.d.): 8.2 (4.7)] than age- and sex-matched non-renal SLE controls [4.9 (3.7); P < 0.001]. Accordingly, they were on higher daily prednisolone doses [11.1 (9.4) vs 7.3 (12.1) mg; P = 0.004] and a higher proportion within BLM-treated non-renal SLE patients used immunosuppressants (60.6%) compared with the controls (31.8%; P = 0.002), but the proportions of patients using antimalarial agents did not differ significantly (P = 0.137). Use of immunosuppressants and antimalarials for the controls during the entire follow-up period is delineated in Supplementary Figs S1 and S2 (available at Rheumatology online), respectively.
Six patients (9.1%) developed a biopsy-proven de novo LN in the BLM-treated non-renal SLE group after a median follow-up time of 7.4 (IQR: 2.7–22.2) months. Among the comparators, two individuals (3.0%) developed de novo LN, one class III and one class IV after 21 and 50 months, respectively.
In the six patients who developed de novo LN, all Caucasians, BLM was primarily initiated for active mucocutaneous and/or musculoskeletal disease. All had positive anti-dsDNA levels and were hypocomplementaemic at baseline. At BLM initiation, SLEDAI-2K scores ranged from 6 to 23, and the daily prednisolone dose from 7.5 to 30 mg. Only 2/6 patients were on concomitant treatment with antimalarials. The renal histopathology in 4/6 subjects was consistent with proliferative LN (class III or IV), whereas the two remaining cases showed membranous LN (class V) in combination with class II. Detailed information is shown in Supplementary Table S1, available at Rheumatology online.
As illustrated in Fig. 1A, non-renal SLE patients treated with BLM showed a higher frequency of and/or shorter time to de novo LN compared with non-renal SLE patients who did not receive BLM (hazard ratio (HR): 10.7; 95% CI: 1.7, 67.9; P = 0.012). This association between BLM treatment and de novo LN development remained significant after adjustment for SLEDAI-2K scores (HR: 8.3; 95% CI: 1.2, 57.0; P = 0.031), while no such association was seen for SLEDAI-2K scores as a co-variate in the same model (HR: 1.1; 95% CI: 0.9, 1.2; P = 0.362). The Kaplan–Meier curve in Fig. 1B illustrates the course of BLM-treated patients with and without a history of renal SLE at BLM initiation, as well as the non-renal comparators, until the time of LN development or the last available evaluation.
Fig. 1 Development of LN in BLM-treated patients and unexposed comparators
(A) Bar graph showing proportions of patients who developed de novo LN within the BLM-treated non-renal patient subgroup (red) and age- and sex-matched comparators not exposed to BLM (blue). The forest plot above illustrates the result from Cox regression analysis, with the dark blue circle representing the HR and the whiskers representing the 95% CI. (B) Kaplan–Meier curve illustrating the course of BLM-treated cases with (green) and without (red) a history of LN at the time of treatment initiation, and the non-renal SLE comparators (blue), until the time of LN development or the last available follow-up evaluation. BLM: belimumab; HR: hazard ratio.
Next, we selected patients not exposed to BLM with baseline SLEDAI-2K scores >4, which yielded a control group with comparable SLEDAI-2K scores [8.5 (3.2); n = 25] to the non-renal BLM group. None of the patients within this group had developed LN after a mean follow-up of 126.5 (37.8) months.
Outcome of cases with previous LN
Among the 29/95 BLM-treated patients with LN prior to enrolment, but quiescent renal disease at the time of BLM initiation, two cases (6.9%) of LN flare were observed after 1 and 9 months (Fig. 1B). One of these patients underwent a renal biopsy that showed a proliferative LN (class IV); prior to BLM treatment, this patient had a history of class IV nephritis that later shifted to class V in two subsequent biopsies. The second patient presented with heavy proteinuria, haematuria and hypertension, indicating renal flare. Therefore, a clinical decision was made not to wait for a biopsy and instead to promptly initiate induction therapy with pulsed cyclophosphamide.
Associations between anti-dsDNA seroconversion and LN development
Of patients with positive anti-dsDNA levels at baseline and available follow-up data, no seroconversion was observed among those who developed LN (n = 8) in the BLM-treated group (n = 46) or de novo LN (n = 6) in the BLM-treated non-renal SLE group (n = 30), while 15 and 13 patients seroconverted among those who did not develop LN (n = 38; P = 0.040) or de novo LN (n = 24; P = 0.024), respectively. Of patients with low complement levels at baseline, one among those who developed de novo LN showed normalization during follow-up; no significant association between C3/C4 normalization and LN development was observed.
Predictors of LN development
The following variables were investigated using univariable Cox regression analysis: age at baseline, SLE disease duration, baseline SLEDAI-2K score, anti-dsDNA positivity, low complement (C3 and/or C4), serological activity (anti-dsDNA positivity and/or hypocomplementaemia), anti-Smith positivity, SDI score, current or former tobacco smoking, daily prednisolone dose, use of antimalarial agents, concomitant use of immunosuppressants, comorbid hypertension and diabetes, and history of renal involvement when all BLM-treated cases were analysed. From these variables, only use of antimalarial agents was negatively associated with development of LN when all BLM-treated patients were considered (coefficient: −0.6; HR: 0.2; 95% CI: 0.05, 0.86; P = 0.031) and with de novo LN when non-renal cases were considered (coefficient: −1.7; HR: 0.2; 95% CI: 0.03, 0.97; P = 0.046).
Discussion
In our real-life setting of BLM-treated subjects, 9% of patients with no renal history developed de novo LN and 7% of patients with prior LN relapsed during treatment. Using age- and sex-matched non-renal comparators with similar serological profiles and a long follow-up, we showed that use of BLM was associated with an increased frequency of de novo LN. Interestingly, our data indicated that concomitant use of antimalarial agents along with BLM may be protective.
In 2014, de novo LN during BLM treatment was first reported in a serologically active middle-aged woman with relapsing serositis, resistant to conventional therapies, and unacceptable doses of corticosteroids [7]. Later, three patients who developed LN over the first year of BLM therapy were observed among 195 patients in 10 centres, mainly American [8]. Staveri et al. reported de novo LN shortly after BLM initiation in two women who had a moderately active non-renal SLE at baseline; one was anti-dsDNA negative [9]. Finally, one case of de novo LN was observed among 23 patients (4%) treated with BLM in a Spanish setting [10].
It is important to highlight that the majority of patients chosen for biologic therapy had a severe disease course, and had failed conventional disease-modifying non-biologic drugs, including the patients who developed de novo LN, of whom 5/6 had a long-standing disease (>7 years). A possible explanation for the development of de novo LN might be a more aggressive disease, as reflected by higher SLEDAI-2K scores and prednisolone doses in these patients; however, neither these features nor SDI scores, also a proxy for severe disease course, were associated with LN development. The non-renal SLE comparators were carefully selected to have similar serological profiles and age at enrolment, and were individually matched with the BLM-treated non-renal SLE patients. However, they had lower levels of disease activity, lower prednisolone doses and fewer patients required immunosuppressants. This reflects that the majority of patients in the comparator group were in a quiescent phase of their disease at the time of enrolment, but could also mirror an overall milder disease phenotype. Nevertheless, they were followed for a longer time compared with the BLM-treated patients, and the observed association between BLM and de novo LN was still present after adjustment for disease activity. Notably, in a subgroup of the comparators comprising 25 patients with comparable degree of activity to the BLM-treated group, none developed de novo LN during follow-up. The reasons behind the observed associations are not clear. Awareness of the steroid-sparing effects of belimumab may have contributed to rapid tapering of glucocorticoid doses, which in turn unveiled renal activity. Belimumab binds to the soluble counterpart of BAFF, a molecule implicated in the pathogenesis of LN [20, 21], and has been shown to alter absolute and relative numbers of B cell subsets, mainly B cells of early developmental stages [22, 23]. However, the long-term consequences of BAFF inhibition, e.g. regarding B cell subsets with regulatory properties, have yet to be determined. Such long-term effects on subsets of B cells could potentially increase the susceptibility of these patients to develop a more severe or organ-specific (renal) phenotype. Accumulating evidence indicates that B cells exert regulatory properties through production of IL-10 [24]. Hence, our recent observation of decreasing serum IL-10 levels during BLM treatment [25] may be suggestive of a regulatory B cell downregulation, collectively warranting granular survey of BLM-mediated effects on the B cell repertoire.
Another interesting finding was that concomitant use of antimalarial agents along with BLM was implied to be protective against the development of de novo LN or LN relapse. Although no firm conclusions can be drawn due to the relatively low number of patients and the known non-adherence of patients to antimalarials, this association is in line with the known beneficial effects of antimalarials that include prevention of renal flares [26] and is also of particular importance in light of a recent report that showed that decreasing levels of IgG and IgA anticardiolipin antibodies in BLM-treated patients were solely observed among those receiving antimalarials [27]. The mechanistic explanation for such a synergy remains to be elucidated. SLE patients using antimalarials have been shown to have lower BAFF levels compared with non-users [28]; while BLM binds to circulating BAFF, antimalarials are likely to hamper type I IFN-mediated BAFF excretion, potentially contributing to additive neutralization. Furthermore, antimalarials also bind nucleic acids, impeding Toll-like receptor activation and therefore innate immune responses, and inhibit loading of antigen into MHC and antigen presentation to T cells, both constituting further explanations for the additional benefit in patients in whom B cells are inhibited [29].
The observational design of our study constituted a limitation, yet the cases represent real-life use of BLM in our academic practices. The vast majority of study participants were of Caucasian origin, reflecting the patient population in Sweden and the UK, and the findings cannot be directly extrapolated to other populations, in particular African/African American or Asian patients. Another limitation was the relatively low number of patients who were enrolled and those who developed de novo LN and LN relapse, limiting the power in statistical analyses. Lastly, non-adherence assessment for background therapies was not performed.
Although firm conclusions cannot be drawn, our observations imply that BLM may not be sufficient for the prevention of LN and suggest close monitoring of BLM-treated patients for signs of evolving renal disease. Concomitant use of antimalarial agents may exert synergistic effects along with BLM with regard to renal outcomes, a finding that warrants corroboration in other settings. Investigation of the long-term effects of BAFF inhibition on B cell subsets with regulatory properties is merited.
Supplementary Material
keaa796_Supplementary_Data Click here for additional data file.
Acknowledgements
We thank Professor Elisabet Svenungsson for kindly reviewing the manuscript.
Funding: This work was supported by grants from the Alfred Österlund’s Foundation, the Anna-Greta Crafoord Foundation, the Greta and Johan Kock’s Foundation, the Gustafsson Foundation, the King Gustaf V and Queen Victoria’s Freemasons Foundation, the King Gustaf V’s 80-year Anniversary Foundation, the Professor Nanna Svartz Foundation, the Region Östergötland (ALF grants), the Selander Foundation, the Skåne University Hospital and the Medical Faculty of Lund University, the Swedish Research Council, the Swedish Rheumatism Association, and the Swedish Society of Medicine (the Ingegerd Johansson donation).
Disclosure statement: I.P. has received research funding from GlaxoSmithKline and Elli Lilly and Company, and honoraria from Gilead Sciences, GlaxoSmithKline and Novartis. E.M.V. has received honoraria and research grant support from Roche, GlaxoSmithKline and AstraZeneca. The other authors have declared no conflict of interest. The funders had no role in the design of the study; in the analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Data availability statement
The datasets used and analysed during the current study are available from the corresponding author upon reasonable request.
Supplementary data
Supplementary data are available at Rheumatology online. | BELIMUMAB, HYDROXYCHLOROQUINE, PREDNISOLONE | DrugsGivenReaction | CC BY-NC | 33341888 | 18,694,286 | 2021-09-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Condition aggravated'. | De novo lupus nephritis during treatment with belimumab.
In light of reports of de novo LN during belimumab (BLM) treatment, we sought to determine its frequency and contributing or protective factors in a real-life setting.
Patients with SLE who received BLM between 2011 and 2017 at five European academic practices were enrolled (n = 95) and followed longitudinally for a median time of 13.1 months [interquartile range (IQR): 6.0-34.7]; 52.6% were anti-dsDNA positive, 60.0% had low complement levels, and 69.5% had no renal involvement prior to/at BLM initiation [mean disease duration at baseline: 11.4 (9.3) years]. Age- and sex-matched patients with non-renal SLE who had similar serological profiles, but were not exposed to BLM, served as controls (median follow-up: 132.0 months; IQR: 98.3-151.2).
We observed 6/66 cases (9.1%) of biopsy-proven de novo LN (4/6 proliferative) among the non-renal BLM-treated SLE cases after a follow-up of 7.4 months (IQR: 2.7-22.2). Among controls, 2/66 cases (3.0%) of de novo LN (both proliferative) were observed after 21 and 50 months. BLM treatment was associated with an increased frequency and/or shorter time to de novo LN [hazard ratio (HR): 10.7; 95% CI: 1.7, 67.9; P = 0.012], while concomitant use of antimalarial agents along with BLM showed an opposing association (HR: 0.2; 95% CI: 0.03, 0.97; P = 0.046).
Addition of BLM to standard-of-care did not prevent LN in patients with active non-renal SLE, but a favourable effect of concomitant use of antimalarials was implicated. Studies of whether effects of B-cell activating factor inhibition on lymphocyte subsets contribute to LN susceptibility are warranted.
pmc Rheumatology key messages
Irrespective of prior renal involvement, belimumab treatment may not adequately protect against lupus nephritis.
Concomitant antimalarial therapy along with belimumab was implied to protect against development of lupus nephritis.
Our observations call for vigilance with regard to evolving renal disease during belimumab therapy.
Introduction
SLE is a chronic, multisystem autoimmune disease with unmet needs, such as delayed diagnosis, premature atherosclerosis, drug-associated organ damage and a prominent impairment of health-related quality of life [1]. The wide range of manifestations and serological findings pose challenges with regard to diagnosis and treatment. Today, standard-of-care (SoC) therapy includes glucocorticoids, antimalarials, immunosuppressants and biologic agents, e.g. belimumab (BLM) and rituximab (RTX). The selection of drugs is mainly based on the afflicted organ systems and the organ-specific or global disease activity [2]. LN is a manifestation of SLE with a potentially life-threatening course [3].
BLM is a recombinant human IgG1-λ monoclonal antibody that specifically binds the soluble form of B cell activating factor (BAFF). The efficacy of BLM has been demonstrated to date in five placebo-controlled phase III trials and several observational studies [4]. Although post hoc analysis of clinical trials of BLM showed superiority of BLM over placebo in preventing renal flares [5] and a systematic review suggested an overall promising effect of BAFF inhibition on renal outcomes [6], development of LN during BLM treatment has also been reported [7–10]. Clinical trials of BLM in LN, either as an add-on therapy to SoC or in combination with RTX, are underway [11–13] and the BLISS-LN trial recently demonstrated superiority of addition of BLM to SoC for active LN over SoC alone [14].
We herein report cases of de novo LN during treatment with BLM observed in our academic practices, and cases of LN flares in patients with a history of renal SLE at the time of BLM initiation. We further aimed at identifying factors or risk phenotypes that are associated with the development of LN, in order to contribute to optimized monitoring during treatment with BLM.
Methods
Patients
Patients, classified with SLE according to the 1982 ACR [15] and/or 2012 SLICC [16] criteria, receiving BLM 10 mg/kg intravenously at week 0, 2, 4 and thereafter every fourth week from its approval in 2011 until 31 December 2017 in the Day Care Units of four Swedish academic rheumatology centres (Linköping, Lund, Stockholm and Uppsala) and one academic centre in Leeds, UK, were followed longitudinally within the frame of observational research programmes, and were included in the present report (n = 95). BLM was given as an add-on to background SoC, with no change in SoC implemented unless clinically indicated. None of these patients were given cyclophosphamide, RTX or other B cell depleting agents during treatment with BLM. No patient selection was applied other than consent to participate in the study. Sixty-six of these patients (69.5%) had no history of renal involvement until BLM initiation. As a comparator group to the non-renal SLE cases exposed to BLM, we included 66 non-renal SLE cases from Linköping and Stockholm, individually matched for age and sex, with similar serological profiles (anti-dsDNA positivity, low complement protein 3 and/or 4), who were also followed longitudinally; no selection other than matched serology and age at baseline was applied. Kidney biopsy was performed in the case of a suspected new onset of LN during follow-up. Patient characteristics are detailed in Table 1.
Table 1 Patient characteristics
Item Belimumab-treated SLE Non-renal SLE comparators P-value
Total Non-renal
Background variables
Number of cases, n 95 66 66
Age, mean (s.d.), years 42.2 (14.2) 42.2 (15.2) 43.4 (16.0) 0.152
Females, n (%) 89 (93.7) 63 (95.5) 63 (95.5) NA
Current tobacco smoking, n (%) 11 (12.5); n = 88 9 (15.0); n = 60 14 (21.2) 0.367
Former tobacco smoking, n (%) 25 (28.4); n = 88 14 (23.3); n = 60 23 (34.8) 0.047
Caucasian, n (%) 86 (90.5) 59 (89.4) 64 (97.0) NA
African, n (%) 6 (6.3) 5 (7.6) 0 (0.0) NA
Asian, n (%) 2 (2.1) 2 (3.0) 2 (3.0) NA
Hispanic, n (%) 1 (1.1) 0 (0.0) 0 (0.0) NA
Diabetes until enrolment, n (%) 3 (3.2) 0 (0.0) 0 (0.0) NA
Hypertension until enrolment, n (%) 23 (24.2) 9 (13.6) 14 (21.2) 0.332
Disease variables at enrolment
Duration of SLE, mean (s.d.), years 11.4 (9.3) 10.5 (9.1) 9.8 (11.1)d 0.529
SLEDAI-2K score, mean (s.d.) 9.3 (5.9) 8.2 (4.7) 4.9 (3.7) <0.001
SDI score, median (IQR) 1 (0–1); n = 93 0 (0–1); n = 64 0 (0–2) 0.594
Serological activitya, n (%) 68 (71.6) 47 (71.2) 50 (75.8) 0.250
Anti-dsDNA positive, n (%) 50 (52.6) 33 (50.0) 34 (51.5) 1.000
Low complement, n (%) 57 (60.0) 40 (60.6) 41 (62.1) 1.000
Anti-Smith positive, n (%) 24 (25.3) 16 (24.2) 14 (21.2) 0.832
Main reasons for belimumab
General, n (%) 4 (4.2) 3 (4.5) NA NA
Mucocutaneous, n (%) 55 (57.9) 39 (59.1) NA NA
Musculoskeletal, n (%) 54 (56.8) 39 (59.1) NA NA
Haematological, n (%) 12 (12.6) 8 (12.1) NA NA
Cardiorespiratory, n (%) 6 (6.3) 4 (6.1) NA NA
Renal, n (%) 9 (9.5) 0 (0.0) NA NA
Neurological, n (%) 5 (9.5) 2 (3.0) NA NA
Immunological, n (%) 3 (3.2) 2 (3.0) NA NA
Ongoing concomitant treatments
Daily prednisolone doseb, mean (s.d.), mg 11.3 (9.4) 11.1 (9.4) 7.3 (12.1)e 0.004
Antimalarial agents, n (%) 67 (70.5) 45 (68.2) 36 (54.5) 0.137
Immunosuppressantsc, n (%) 58 (61.1) 40 (60.6) 21 (31.8) 0.002
Azathioprine, n (%) 27 (28.4) 17 (25.8) 6 (9.1) 0.013
Methotrexate, n (%) 14 (14.7) 11 (16.7) 8 (12.1) 0.629
Mycophenolate mofetil/sodium, n (%) 14 (14.7) 11 (16.7) 3 (4.5) 0.057
Other immunosuppressants, n (%) 4 (6.8) 2 (3.0) 5 (7.6) 0.375
In cases of missing values, the total number of available observations (n) is indicated. P-values are derived from comparisons between non-renal SLE patients who were treated with belimumab and individually matched for age and sex non-renal SLE comparators who were not treated with belimumab, using Wilcoxon’s signed rank test for continues variables and McNemar’s test for dichotomous variables, or the χ2 test in cases of missing values in one of the two groups. Significant P-values are indicated in bold. aAnti-dsDNA positivity and/or low complement levels. bAt the time of belimumab initiation or enrolment for the comparators. cExcluding antimalarial agents. dMedian (IQR): 6.4 (0.5–13.4) years. eMedian (IQR): 5.0 (0.0–10.0) mg. IQR: interquartile range; NA: not applicable or not available; SDI: SLICC/ACR Damage Index.
Definitions
We defined de novo LN as a new onset of significant proteinuria, defined as a urinary protein-to-creatinine ratio or protein excretion in 24-h urine collection corresponding to >0.5 g/day, combined with renal histology consistent with LN according to the WHO and/or 2003 International Society of Nephrology/Renal Pathology Society classification [17], in patients who previously had not met the ACR criterion for renal disorder [15].
Global SLE disease activity was evaluated using the SLEDAI-2K [18], and organ damage using the SLICC/ACR Damage Index (SDI) [19]. For SLEDAI-2K scores, laboratory and serological items were assessed based on results from routine tests at the local university hospital laboratories.
Statistics
Comparisons between matched non-renal SLE patients who received BLM vs those who did not were performed using Wilcoxon’s signed rank test for continuous and McNemar’s test for dichotomous variables. The occurrence of de novo LN or LN flares during follow-up was illustrated using Kaplan–Meier curves, and the pairwise log-rank (Mantel–Cox) test was employed to compare the de novo LN distributions between BLM exposed vs not exposed non-renal SLE patients. Contingency between unrelated dichotomous variables was tested using Fisher’s exact test. Proportional hazards (Cox) regression was used to investigate factors and disease phenotypes associated with LN development during therapy. P-values<0.05 were considered statistically significant. IBM SPSS version 25 software (IBM Corp., Armonk, NY, USA) was used for statistical analyses and GraphPad Prism 7 (GraphPad Software Inc., La Jolla, CA, USA) for construction of graphs.
Ethical considerations
The study complied with the ethical principles of the Declaration of Helsinki. Written informed consent was obtained from all patients. The study protocol was approved by regional ethics review boards.
Results
Outcome of cases without prior LN
As shown in Table 1, non-renal SLE patients who were selected for treatment with BLM had comparable serological profiles, disease duration and SDI scores but higher baseline SLEDAI-2K scores [mean (s.d.): 8.2 (4.7)] than age- and sex-matched non-renal SLE controls [4.9 (3.7); P < 0.001]. Accordingly, they were on higher daily prednisolone doses [11.1 (9.4) vs 7.3 (12.1) mg; P = 0.004] and a higher proportion within BLM-treated non-renal SLE patients used immunosuppressants (60.6%) compared with the controls (31.8%; P = 0.002), but the proportions of patients using antimalarial agents did not differ significantly (P = 0.137). Use of immunosuppressants and antimalarials for the controls during the entire follow-up period is delineated in Supplementary Figs S1 and S2 (available at Rheumatology online), respectively.
Six patients (9.1%) developed a biopsy-proven de novo LN in the BLM-treated non-renal SLE group after a median follow-up time of 7.4 (IQR: 2.7–22.2) months. Among the comparators, two individuals (3.0%) developed de novo LN, one class III and one class IV after 21 and 50 months, respectively.
In the six patients who developed de novo LN, all Caucasians, BLM was primarily initiated for active mucocutaneous and/or musculoskeletal disease. All had positive anti-dsDNA levels and were hypocomplementaemic at baseline. At BLM initiation, SLEDAI-2K scores ranged from 6 to 23, and the daily prednisolone dose from 7.5 to 30 mg. Only 2/6 patients were on concomitant treatment with antimalarials. The renal histopathology in 4/6 subjects was consistent with proliferative LN (class III or IV), whereas the two remaining cases showed membranous LN (class V) in combination with class II. Detailed information is shown in Supplementary Table S1, available at Rheumatology online.
As illustrated in Fig. 1A, non-renal SLE patients treated with BLM showed a higher frequency of and/or shorter time to de novo LN compared with non-renal SLE patients who did not receive BLM (hazard ratio (HR): 10.7; 95% CI: 1.7, 67.9; P = 0.012). This association between BLM treatment and de novo LN development remained significant after adjustment for SLEDAI-2K scores (HR: 8.3; 95% CI: 1.2, 57.0; P = 0.031), while no such association was seen for SLEDAI-2K scores as a co-variate in the same model (HR: 1.1; 95% CI: 0.9, 1.2; P = 0.362). The Kaplan–Meier curve in Fig. 1B illustrates the course of BLM-treated patients with and without a history of renal SLE at BLM initiation, as well as the non-renal comparators, until the time of LN development or the last available evaluation.
Fig. 1 Development of LN in BLM-treated patients and unexposed comparators
(A) Bar graph showing proportions of patients who developed de novo LN within the BLM-treated non-renal patient subgroup (red) and age- and sex-matched comparators not exposed to BLM (blue). The forest plot above illustrates the result from Cox regression analysis, with the dark blue circle representing the HR and the whiskers representing the 95% CI. (B) Kaplan–Meier curve illustrating the course of BLM-treated cases with (green) and without (red) a history of LN at the time of treatment initiation, and the non-renal SLE comparators (blue), until the time of LN development or the last available follow-up evaluation. BLM: belimumab; HR: hazard ratio.
Next, we selected patients not exposed to BLM with baseline SLEDAI-2K scores >4, which yielded a control group with comparable SLEDAI-2K scores [8.5 (3.2); n = 25] to the non-renal BLM group. None of the patients within this group had developed LN after a mean follow-up of 126.5 (37.8) months.
Outcome of cases with previous LN
Among the 29/95 BLM-treated patients with LN prior to enrolment, but quiescent renal disease at the time of BLM initiation, two cases (6.9%) of LN flare were observed after 1 and 9 months (Fig. 1B). One of these patients underwent a renal biopsy that showed a proliferative LN (class IV); prior to BLM treatment, this patient had a history of class IV nephritis that later shifted to class V in two subsequent biopsies. The second patient presented with heavy proteinuria, haematuria and hypertension, indicating renal flare. Therefore, a clinical decision was made not to wait for a biopsy and instead to promptly initiate induction therapy with pulsed cyclophosphamide.
Associations between anti-dsDNA seroconversion and LN development
Of patients with positive anti-dsDNA levels at baseline and available follow-up data, no seroconversion was observed among those who developed LN (n = 8) in the BLM-treated group (n = 46) or de novo LN (n = 6) in the BLM-treated non-renal SLE group (n = 30), while 15 and 13 patients seroconverted among those who did not develop LN (n = 38; P = 0.040) or de novo LN (n = 24; P = 0.024), respectively. Of patients with low complement levels at baseline, one among those who developed de novo LN showed normalization during follow-up; no significant association between C3/C4 normalization and LN development was observed.
Predictors of LN development
The following variables were investigated using univariable Cox regression analysis: age at baseline, SLE disease duration, baseline SLEDAI-2K score, anti-dsDNA positivity, low complement (C3 and/or C4), serological activity (anti-dsDNA positivity and/or hypocomplementaemia), anti-Smith positivity, SDI score, current or former tobacco smoking, daily prednisolone dose, use of antimalarial agents, concomitant use of immunosuppressants, comorbid hypertension and diabetes, and history of renal involvement when all BLM-treated cases were analysed. From these variables, only use of antimalarial agents was negatively associated with development of LN when all BLM-treated patients were considered (coefficient: −0.6; HR: 0.2; 95% CI: 0.05, 0.86; P = 0.031) and with de novo LN when non-renal cases were considered (coefficient: −1.7; HR: 0.2; 95% CI: 0.03, 0.97; P = 0.046).
Discussion
In our real-life setting of BLM-treated subjects, 9% of patients with no renal history developed de novo LN and 7% of patients with prior LN relapsed during treatment. Using age- and sex-matched non-renal comparators with similar serological profiles and a long follow-up, we showed that use of BLM was associated with an increased frequency of de novo LN. Interestingly, our data indicated that concomitant use of antimalarial agents along with BLM may be protective.
In 2014, de novo LN during BLM treatment was first reported in a serologically active middle-aged woman with relapsing serositis, resistant to conventional therapies, and unacceptable doses of corticosteroids [7]. Later, three patients who developed LN over the first year of BLM therapy were observed among 195 patients in 10 centres, mainly American [8]. Staveri et al. reported de novo LN shortly after BLM initiation in two women who had a moderately active non-renal SLE at baseline; one was anti-dsDNA negative [9]. Finally, one case of de novo LN was observed among 23 patients (4%) treated with BLM in a Spanish setting [10].
It is important to highlight that the majority of patients chosen for biologic therapy had a severe disease course, and had failed conventional disease-modifying non-biologic drugs, including the patients who developed de novo LN, of whom 5/6 had a long-standing disease (>7 years). A possible explanation for the development of de novo LN might be a more aggressive disease, as reflected by higher SLEDAI-2K scores and prednisolone doses in these patients; however, neither these features nor SDI scores, also a proxy for severe disease course, were associated with LN development. The non-renal SLE comparators were carefully selected to have similar serological profiles and age at enrolment, and were individually matched with the BLM-treated non-renal SLE patients. However, they had lower levels of disease activity, lower prednisolone doses and fewer patients required immunosuppressants. This reflects that the majority of patients in the comparator group were in a quiescent phase of their disease at the time of enrolment, but could also mirror an overall milder disease phenotype. Nevertheless, they were followed for a longer time compared with the BLM-treated patients, and the observed association between BLM and de novo LN was still present after adjustment for disease activity. Notably, in a subgroup of the comparators comprising 25 patients with comparable degree of activity to the BLM-treated group, none developed de novo LN during follow-up. The reasons behind the observed associations are not clear. Awareness of the steroid-sparing effects of belimumab may have contributed to rapid tapering of glucocorticoid doses, which in turn unveiled renal activity. Belimumab binds to the soluble counterpart of BAFF, a molecule implicated in the pathogenesis of LN [20, 21], and has been shown to alter absolute and relative numbers of B cell subsets, mainly B cells of early developmental stages [22, 23]. However, the long-term consequences of BAFF inhibition, e.g. regarding B cell subsets with regulatory properties, have yet to be determined. Such long-term effects on subsets of B cells could potentially increase the susceptibility of these patients to develop a more severe or organ-specific (renal) phenotype. Accumulating evidence indicates that B cells exert regulatory properties through production of IL-10 [24]. Hence, our recent observation of decreasing serum IL-10 levels during BLM treatment [25] may be suggestive of a regulatory B cell downregulation, collectively warranting granular survey of BLM-mediated effects on the B cell repertoire.
Another interesting finding was that concomitant use of antimalarial agents along with BLM was implied to be protective against the development of de novo LN or LN relapse. Although no firm conclusions can be drawn due to the relatively low number of patients and the known non-adherence of patients to antimalarials, this association is in line with the known beneficial effects of antimalarials that include prevention of renal flares [26] and is also of particular importance in light of a recent report that showed that decreasing levels of IgG and IgA anticardiolipin antibodies in BLM-treated patients were solely observed among those receiving antimalarials [27]. The mechanistic explanation for such a synergy remains to be elucidated. SLE patients using antimalarials have been shown to have lower BAFF levels compared with non-users [28]; while BLM binds to circulating BAFF, antimalarials are likely to hamper type I IFN-mediated BAFF excretion, potentially contributing to additive neutralization. Furthermore, antimalarials also bind nucleic acids, impeding Toll-like receptor activation and therefore innate immune responses, and inhibit loading of antigen into MHC and antigen presentation to T cells, both constituting further explanations for the additional benefit in patients in whom B cells are inhibited [29].
The observational design of our study constituted a limitation, yet the cases represent real-life use of BLM in our academic practices. The vast majority of study participants were of Caucasian origin, reflecting the patient population in Sweden and the UK, and the findings cannot be directly extrapolated to other populations, in particular African/African American or Asian patients. Another limitation was the relatively low number of patients who were enrolled and those who developed de novo LN and LN relapse, limiting the power in statistical analyses. Lastly, non-adherence assessment for background therapies was not performed.
Although firm conclusions cannot be drawn, our observations imply that BLM may not be sufficient for the prevention of LN and suggest close monitoring of BLM-treated patients for signs of evolving renal disease. Concomitant use of antimalarial agents may exert synergistic effects along with BLM with regard to renal outcomes, a finding that warrants corroboration in other settings. Investigation of the long-term effects of BAFF inhibition on B cell subsets with regulatory properties is merited.
Supplementary Material
keaa796_Supplementary_Data Click here for additional data file.
Acknowledgements
We thank Professor Elisabet Svenungsson for kindly reviewing the manuscript.
Funding: This work was supported by grants from the Alfred Österlund’s Foundation, the Anna-Greta Crafoord Foundation, the Greta and Johan Kock’s Foundation, the Gustafsson Foundation, the King Gustaf V and Queen Victoria’s Freemasons Foundation, the King Gustaf V’s 80-year Anniversary Foundation, the Professor Nanna Svartz Foundation, the Region Östergötland (ALF grants), the Selander Foundation, the Skåne University Hospital and the Medical Faculty of Lund University, the Swedish Research Council, the Swedish Rheumatism Association, and the Swedish Society of Medicine (the Ingegerd Johansson donation).
Disclosure statement: I.P. has received research funding from GlaxoSmithKline and Elli Lilly and Company, and honoraria from Gilead Sciences, GlaxoSmithKline and Novartis. E.M.V. has received honoraria and research grant support from Roche, GlaxoSmithKline and AstraZeneca. The other authors have declared no conflict of interest. The funders had no role in the design of the study; in the analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Data availability statement
The datasets used and analysed during the current study are available from the corresponding author upon reasonable request.
Supplementary data
Supplementary data are available at Rheumatology online. | BELIMUMAB, PREDNISOLONE | DrugsGivenReaction | CC BY-NC | 33341888 | 18,694,287 | 2021-09-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Haematuria'. | De novo lupus nephritis during treatment with belimumab.
In light of reports of de novo LN during belimumab (BLM) treatment, we sought to determine its frequency and contributing or protective factors in a real-life setting.
Patients with SLE who received BLM between 2011 and 2017 at five European academic practices were enrolled (n = 95) and followed longitudinally for a median time of 13.1 months [interquartile range (IQR): 6.0-34.7]; 52.6% were anti-dsDNA positive, 60.0% had low complement levels, and 69.5% had no renal involvement prior to/at BLM initiation [mean disease duration at baseline: 11.4 (9.3) years]. Age- and sex-matched patients with non-renal SLE who had similar serological profiles, but were not exposed to BLM, served as controls (median follow-up: 132.0 months; IQR: 98.3-151.2).
We observed 6/66 cases (9.1%) of biopsy-proven de novo LN (4/6 proliferative) among the non-renal BLM-treated SLE cases after a follow-up of 7.4 months (IQR: 2.7-22.2). Among controls, 2/66 cases (3.0%) of de novo LN (both proliferative) were observed after 21 and 50 months. BLM treatment was associated with an increased frequency and/or shorter time to de novo LN [hazard ratio (HR): 10.7; 95% CI: 1.7, 67.9; P = 0.012], while concomitant use of antimalarial agents along with BLM showed an opposing association (HR: 0.2; 95% CI: 0.03, 0.97; P = 0.046).
Addition of BLM to standard-of-care did not prevent LN in patients with active non-renal SLE, but a favourable effect of concomitant use of antimalarials was implicated. Studies of whether effects of B-cell activating factor inhibition on lymphocyte subsets contribute to LN susceptibility are warranted.
pmc Rheumatology key messages
Irrespective of prior renal involvement, belimumab treatment may not adequately protect against lupus nephritis.
Concomitant antimalarial therapy along with belimumab was implied to protect against development of lupus nephritis.
Our observations call for vigilance with regard to evolving renal disease during belimumab therapy.
Introduction
SLE is a chronic, multisystem autoimmune disease with unmet needs, such as delayed diagnosis, premature atherosclerosis, drug-associated organ damage and a prominent impairment of health-related quality of life [1]. The wide range of manifestations and serological findings pose challenges with regard to diagnosis and treatment. Today, standard-of-care (SoC) therapy includes glucocorticoids, antimalarials, immunosuppressants and biologic agents, e.g. belimumab (BLM) and rituximab (RTX). The selection of drugs is mainly based on the afflicted organ systems and the organ-specific or global disease activity [2]. LN is a manifestation of SLE with a potentially life-threatening course [3].
BLM is a recombinant human IgG1-λ monoclonal antibody that specifically binds the soluble form of B cell activating factor (BAFF). The efficacy of BLM has been demonstrated to date in five placebo-controlled phase III trials and several observational studies [4]. Although post hoc analysis of clinical trials of BLM showed superiority of BLM over placebo in preventing renal flares [5] and a systematic review suggested an overall promising effect of BAFF inhibition on renal outcomes [6], development of LN during BLM treatment has also been reported [7–10]. Clinical trials of BLM in LN, either as an add-on therapy to SoC or in combination with RTX, are underway [11–13] and the BLISS-LN trial recently demonstrated superiority of addition of BLM to SoC for active LN over SoC alone [14].
We herein report cases of de novo LN during treatment with BLM observed in our academic practices, and cases of LN flares in patients with a history of renal SLE at the time of BLM initiation. We further aimed at identifying factors or risk phenotypes that are associated with the development of LN, in order to contribute to optimized monitoring during treatment with BLM.
Methods
Patients
Patients, classified with SLE according to the 1982 ACR [15] and/or 2012 SLICC [16] criteria, receiving BLM 10 mg/kg intravenously at week 0, 2, 4 and thereafter every fourth week from its approval in 2011 until 31 December 2017 in the Day Care Units of four Swedish academic rheumatology centres (Linköping, Lund, Stockholm and Uppsala) and one academic centre in Leeds, UK, were followed longitudinally within the frame of observational research programmes, and were included in the present report (n = 95). BLM was given as an add-on to background SoC, with no change in SoC implemented unless clinically indicated. None of these patients were given cyclophosphamide, RTX or other B cell depleting agents during treatment with BLM. No patient selection was applied other than consent to participate in the study. Sixty-six of these patients (69.5%) had no history of renal involvement until BLM initiation. As a comparator group to the non-renal SLE cases exposed to BLM, we included 66 non-renal SLE cases from Linköping and Stockholm, individually matched for age and sex, with similar serological profiles (anti-dsDNA positivity, low complement protein 3 and/or 4), who were also followed longitudinally; no selection other than matched serology and age at baseline was applied. Kidney biopsy was performed in the case of a suspected new onset of LN during follow-up. Patient characteristics are detailed in Table 1.
Table 1 Patient characteristics
Item Belimumab-treated SLE Non-renal SLE comparators P-value
Total Non-renal
Background variables
Number of cases, n 95 66 66
Age, mean (s.d.), years 42.2 (14.2) 42.2 (15.2) 43.4 (16.0) 0.152
Females, n (%) 89 (93.7) 63 (95.5) 63 (95.5) NA
Current tobacco smoking, n (%) 11 (12.5); n = 88 9 (15.0); n = 60 14 (21.2) 0.367
Former tobacco smoking, n (%) 25 (28.4); n = 88 14 (23.3); n = 60 23 (34.8) 0.047
Caucasian, n (%) 86 (90.5) 59 (89.4) 64 (97.0) NA
African, n (%) 6 (6.3) 5 (7.6) 0 (0.0) NA
Asian, n (%) 2 (2.1) 2 (3.0) 2 (3.0) NA
Hispanic, n (%) 1 (1.1) 0 (0.0) 0 (0.0) NA
Diabetes until enrolment, n (%) 3 (3.2) 0 (0.0) 0 (0.0) NA
Hypertension until enrolment, n (%) 23 (24.2) 9 (13.6) 14 (21.2) 0.332
Disease variables at enrolment
Duration of SLE, mean (s.d.), years 11.4 (9.3) 10.5 (9.1) 9.8 (11.1)d 0.529
SLEDAI-2K score, mean (s.d.) 9.3 (5.9) 8.2 (4.7) 4.9 (3.7) <0.001
SDI score, median (IQR) 1 (0–1); n = 93 0 (0–1); n = 64 0 (0–2) 0.594
Serological activitya, n (%) 68 (71.6) 47 (71.2) 50 (75.8) 0.250
Anti-dsDNA positive, n (%) 50 (52.6) 33 (50.0) 34 (51.5) 1.000
Low complement, n (%) 57 (60.0) 40 (60.6) 41 (62.1) 1.000
Anti-Smith positive, n (%) 24 (25.3) 16 (24.2) 14 (21.2) 0.832
Main reasons for belimumab
General, n (%) 4 (4.2) 3 (4.5) NA NA
Mucocutaneous, n (%) 55 (57.9) 39 (59.1) NA NA
Musculoskeletal, n (%) 54 (56.8) 39 (59.1) NA NA
Haematological, n (%) 12 (12.6) 8 (12.1) NA NA
Cardiorespiratory, n (%) 6 (6.3) 4 (6.1) NA NA
Renal, n (%) 9 (9.5) 0 (0.0) NA NA
Neurological, n (%) 5 (9.5) 2 (3.0) NA NA
Immunological, n (%) 3 (3.2) 2 (3.0) NA NA
Ongoing concomitant treatments
Daily prednisolone doseb, mean (s.d.), mg 11.3 (9.4) 11.1 (9.4) 7.3 (12.1)e 0.004
Antimalarial agents, n (%) 67 (70.5) 45 (68.2) 36 (54.5) 0.137
Immunosuppressantsc, n (%) 58 (61.1) 40 (60.6) 21 (31.8) 0.002
Azathioprine, n (%) 27 (28.4) 17 (25.8) 6 (9.1) 0.013
Methotrexate, n (%) 14 (14.7) 11 (16.7) 8 (12.1) 0.629
Mycophenolate mofetil/sodium, n (%) 14 (14.7) 11 (16.7) 3 (4.5) 0.057
Other immunosuppressants, n (%) 4 (6.8) 2 (3.0) 5 (7.6) 0.375
In cases of missing values, the total number of available observations (n) is indicated. P-values are derived from comparisons between non-renal SLE patients who were treated with belimumab and individually matched for age and sex non-renal SLE comparators who were not treated with belimumab, using Wilcoxon’s signed rank test for continues variables and McNemar’s test for dichotomous variables, or the χ2 test in cases of missing values in one of the two groups. Significant P-values are indicated in bold. aAnti-dsDNA positivity and/or low complement levels. bAt the time of belimumab initiation or enrolment for the comparators. cExcluding antimalarial agents. dMedian (IQR): 6.4 (0.5–13.4) years. eMedian (IQR): 5.0 (0.0–10.0) mg. IQR: interquartile range; NA: not applicable or not available; SDI: SLICC/ACR Damage Index.
Definitions
We defined de novo LN as a new onset of significant proteinuria, defined as a urinary protein-to-creatinine ratio or protein excretion in 24-h urine collection corresponding to >0.5 g/day, combined with renal histology consistent with LN according to the WHO and/or 2003 International Society of Nephrology/Renal Pathology Society classification [17], in patients who previously had not met the ACR criterion for renal disorder [15].
Global SLE disease activity was evaluated using the SLEDAI-2K [18], and organ damage using the SLICC/ACR Damage Index (SDI) [19]. For SLEDAI-2K scores, laboratory and serological items were assessed based on results from routine tests at the local university hospital laboratories.
Statistics
Comparisons between matched non-renal SLE patients who received BLM vs those who did not were performed using Wilcoxon’s signed rank test for continuous and McNemar’s test for dichotomous variables. The occurrence of de novo LN or LN flares during follow-up was illustrated using Kaplan–Meier curves, and the pairwise log-rank (Mantel–Cox) test was employed to compare the de novo LN distributions between BLM exposed vs not exposed non-renal SLE patients. Contingency between unrelated dichotomous variables was tested using Fisher’s exact test. Proportional hazards (Cox) regression was used to investigate factors and disease phenotypes associated with LN development during therapy. P-values<0.05 were considered statistically significant. IBM SPSS version 25 software (IBM Corp., Armonk, NY, USA) was used for statistical analyses and GraphPad Prism 7 (GraphPad Software Inc., La Jolla, CA, USA) for construction of graphs.
Ethical considerations
The study complied with the ethical principles of the Declaration of Helsinki. Written informed consent was obtained from all patients. The study protocol was approved by regional ethics review boards.
Results
Outcome of cases without prior LN
As shown in Table 1, non-renal SLE patients who were selected for treatment with BLM had comparable serological profiles, disease duration and SDI scores but higher baseline SLEDAI-2K scores [mean (s.d.): 8.2 (4.7)] than age- and sex-matched non-renal SLE controls [4.9 (3.7); P < 0.001]. Accordingly, they were on higher daily prednisolone doses [11.1 (9.4) vs 7.3 (12.1) mg; P = 0.004] and a higher proportion within BLM-treated non-renal SLE patients used immunosuppressants (60.6%) compared with the controls (31.8%; P = 0.002), but the proportions of patients using antimalarial agents did not differ significantly (P = 0.137). Use of immunosuppressants and antimalarials for the controls during the entire follow-up period is delineated in Supplementary Figs S1 and S2 (available at Rheumatology online), respectively.
Six patients (9.1%) developed a biopsy-proven de novo LN in the BLM-treated non-renal SLE group after a median follow-up time of 7.4 (IQR: 2.7–22.2) months. Among the comparators, two individuals (3.0%) developed de novo LN, one class III and one class IV after 21 and 50 months, respectively.
In the six patients who developed de novo LN, all Caucasians, BLM was primarily initiated for active mucocutaneous and/or musculoskeletal disease. All had positive anti-dsDNA levels and were hypocomplementaemic at baseline. At BLM initiation, SLEDAI-2K scores ranged from 6 to 23, and the daily prednisolone dose from 7.5 to 30 mg. Only 2/6 patients were on concomitant treatment with antimalarials. The renal histopathology in 4/6 subjects was consistent with proliferative LN (class III or IV), whereas the two remaining cases showed membranous LN (class V) in combination with class II. Detailed information is shown in Supplementary Table S1, available at Rheumatology online.
As illustrated in Fig. 1A, non-renal SLE patients treated with BLM showed a higher frequency of and/or shorter time to de novo LN compared with non-renal SLE patients who did not receive BLM (hazard ratio (HR): 10.7; 95% CI: 1.7, 67.9; P = 0.012). This association between BLM treatment and de novo LN development remained significant after adjustment for SLEDAI-2K scores (HR: 8.3; 95% CI: 1.2, 57.0; P = 0.031), while no such association was seen for SLEDAI-2K scores as a co-variate in the same model (HR: 1.1; 95% CI: 0.9, 1.2; P = 0.362). The Kaplan–Meier curve in Fig. 1B illustrates the course of BLM-treated patients with and without a history of renal SLE at BLM initiation, as well as the non-renal comparators, until the time of LN development or the last available evaluation.
Fig. 1 Development of LN in BLM-treated patients and unexposed comparators
(A) Bar graph showing proportions of patients who developed de novo LN within the BLM-treated non-renal patient subgroup (red) and age- and sex-matched comparators not exposed to BLM (blue). The forest plot above illustrates the result from Cox regression analysis, with the dark blue circle representing the HR and the whiskers representing the 95% CI. (B) Kaplan–Meier curve illustrating the course of BLM-treated cases with (green) and without (red) a history of LN at the time of treatment initiation, and the non-renal SLE comparators (blue), until the time of LN development or the last available follow-up evaluation. BLM: belimumab; HR: hazard ratio.
Next, we selected patients not exposed to BLM with baseline SLEDAI-2K scores >4, which yielded a control group with comparable SLEDAI-2K scores [8.5 (3.2); n = 25] to the non-renal BLM group. None of the patients within this group had developed LN after a mean follow-up of 126.5 (37.8) months.
Outcome of cases with previous LN
Among the 29/95 BLM-treated patients with LN prior to enrolment, but quiescent renal disease at the time of BLM initiation, two cases (6.9%) of LN flare were observed after 1 and 9 months (Fig. 1B). One of these patients underwent a renal biopsy that showed a proliferative LN (class IV); prior to BLM treatment, this patient had a history of class IV nephritis that later shifted to class V in two subsequent biopsies. The second patient presented with heavy proteinuria, haematuria and hypertension, indicating renal flare. Therefore, a clinical decision was made not to wait for a biopsy and instead to promptly initiate induction therapy with pulsed cyclophosphamide.
Associations between anti-dsDNA seroconversion and LN development
Of patients with positive anti-dsDNA levels at baseline and available follow-up data, no seroconversion was observed among those who developed LN (n = 8) in the BLM-treated group (n = 46) or de novo LN (n = 6) in the BLM-treated non-renal SLE group (n = 30), while 15 and 13 patients seroconverted among those who did not develop LN (n = 38; P = 0.040) or de novo LN (n = 24; P = 0.024), respectively. Of patients with low complement levels at baseline, one among those who developed de novo LN showed normalization during follow-up; no significant association between C3/C4 normalization and LN development was observed.
Predictors of LN development
The following variables were investigated using univariable Cox regression analysis: age at baseline, SLE disease duration, baseline SLEDAI-2K score, anti-dsDNA positivity, low complement (C3 and/or C4), serological activity (anti-dsDNA positivity and/or hypocomplementaemia), anti-Smith positivity, SDI score, current or former tobacco smoking, daily prednisolone dose, use of antimalarial agents, concomitant use of immunosuppressants, comorbid hypertension and diabetes, and history of renal involvement when all BLM-treated cases were analysed. From these variables, only use of antimalarial agents was negatively associated with development of LN when all BLM-treated patients were considered (coefficient: −0.6; HR: 0.2; 95% CI: 0.05, 0.86; P = 0.031) and with de novo LN when non-renal cases were considered (coefficient: −1.7; HR: 0.2; 95% CI: 0.03, 0.97; P = 0.046).
Discussion
In our real-life setting of BLM-treated subjects, 9% of patients with no renal history developed de novo LN and 7% of patients with prior LN relapsed during treatment. Using age- and sex-matched non-renal comparators with similar serological profiles and a long follow-up, we showed that use of BLM was associated with an increased frequency of de novo LN. Interestingly, our data indicated that concomitant use of antimalarial agents along with BLM may be protective.
In 2014, de novo LN during BLM treatment was first reported in a serologically active middle-aged woman with relapsing serositis, resistant to conventional therapies, and unacceptable doses of corticosteroids [7]. Later, three patients who developed LN over the first year of BLM therapy were observed among 195 patients in 10 centres, mainly American [8]. Staveri et al. reported de novo LN shortly after BLM initiation in two women who had a moderately active non-renal SLE at baseline; one was anti-dsDNA negative [9]. Finally, one case of de novo LN was observed among 23 patients (4%) treated with BLM in a Spanish setting [10].
It is important to highlight that the majority of patients chosen for biologic therapy had a severe disease course, and had failed conventional disease-modifying non-biologic drugs, including the patients who developed de novo LN, of whom 5/6 had a long-standing disease (>7 years). A possible explanation for the development of de novo LN might be a more aggressive disease, as reflected by higher SLEDAI-2K scores and prednisolone doses in these patients; however, neither these features nor SDI scores, also a proxy for severe disease course, were associated with LN development. The non-renal SLE comparators were carefully selected to have similar serological profiles and age at enrolment, and were individually matched with the BLM-treated non-renal SLE patients. However, they had lower levels of disease activity, lower prednisolone doses and fewer patients required immunosuppressants. This reflects that the majority of patients in the comparator group were in a quiescent phase of their disease at the time of enrolment, but could also mirror an overall milder disease phenotype. Nevertheless, they were followed for a longer time compared with the BLM-treated patients, and the observed association between BLM and de novo LN was still present after adjustment for disease activity. Notably, in a subgroup of the comparators comprising 25 patients with comparable degree of activity to the BLM-treated group, none developed de novo LN during follow-up. The reasons behind the observed associations are not clear. Awareness of the steroid-sparing effects of belimumab may have contributed to rapid tapering of glucocorticoid doses, which in turn unveiled renal activity. Belimumab binds to the soluble counterpart of BAFF, a molecule implicated in the pathogenesis of LN [20, 21], and has been shown to alter absolute and relative numbers of B cell subsets, mainly B cells of early developmental stages [22, 23]. However, the long-term consequences of BAFF inhibition, e.g. regarding B cell subsets with regulatory properties, have yet to be determined. Such long-term effects on subsets of B cells could potentially increase the susceptibility of these patients to develop a more severe or organ-specific (renal) phenotype. Accumulating evidence indicates that B cells exert regulatory properties through production of IL-10 [24]. Hence, our recent observation of decreasing serum IL-10 levels during BLM treatment [25] may be suggestive of a regulatory B cell downregulation, collectively warranting granular survey of BLM-mediated effects on the B cell repertoire.
Another interesting finding was that concomitant use of antimalarial agents along with BLM was implied to be protective against the development of de novo LN or LN relapse. Although no firm conclusions can be drawn due to the relatively low number of patients and the known non-adherence of patients to antimalarials, this association is in line with the known beneficial effects of antimalarials that include prevention of renal flares [26] and is also of particular importance in light of a recent report that showed that decreasing levels of IgG and IgA anticardiolipin antibodies in BLM-treated patients were solely observed among those receiving antimalarials [27]. The mechanistic explanation for such a synergy remains to be elucidated. SLE patients using antimalarials have been shown to have lower BAFF levels compared with non-users [28]; while BLM binds to circulating BAFF, antimalarials are likely to hamper type I IFN-mediated BAFF excretion, potentially contributing to additive neutralization. Furthermore, antimalarials also bind nucleic acids, impeding Toll-like receptor activation and therefore innate immune responses, and inhibit loading of antigen into MHC and antigen presentation to T cells, both constituting further explanations for the additional benefit in patients in whom B cells are inhibited [29].
The observational design of our study constituted a limitation, yet the cases represent real-life use of BLM in our academic practices. The vast majority of study participants were of Caucasian origin, reflecting the patient population in Sweden and the UK, and the findings cannot be directly extrapolated to other populations, in particular African/African American or Asian patients. Another limitation was the relatively low number of patients who were enrolled and those who developed de novo LN and LN relapse, limiting the power in statistical analyses. Lastly, non-adherence assessment for background therapies was not performed.
Although firm conclusions cannot be drawn, our observations imply that BLM may not be sufficient for the prevention of LN and suggest close monitoring of BLM-treated patients for signs of evolving renal disease. Concomitant use of antimalarial agents may exert synergistic effects along with BLM with regard to renal outcomes, a finding that warrants corroboration in other settings. Investigation of the long-term effects of BAFF inhibition on B cell subsets with regulatory properties is merited.
Supplementary Material
keaa796_Supplementary_Data Click here for additional data file.
Acknowledgements
We thank Professor Elisabet Svenungsson for kindly reviewing the manuscript.
Funding: This work was supported by grants from the Alfred Österlund’s Foundation, the Anna-Greta Crafoord Foundation, the Greta and Johan Kock’s Foundation, the Gustafsson Foundation, the King Gustaf V and Queen Victoria’s Freemasons Foundation, the King Gustaf V’s 80-year Anniversary Foundation, the Professor Nanna Svartz Foundation, the Region Östergötland (ALF grants), the Selander Foundation, the Skåne University Hospital and the Medical Faculty of Lund University, the Swedish Research Council, the Swedish Rheumatism Association, and the Swedish Society of Medicine (the Ingegerd Johansson donation).
Disclosure statement: I.P. has received research funding from GlaxoSmithKline and Elli Lilly and Company, and honoraria from Gilead Sciences, GlaxoSmithKline and Novartis. E.M.V. has received honoraria and research grant support from Roche, GlaxoSmithKline and AstraZeneca. The other authors have declared no conflict of interest. The funders had no role in the design of the study; in the analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Data availability statement
The datasets used and analysed during the current study are available from the corresponding author upon reasonable request.
Supplementary data
Supplementary data are available at Rheumatology online. | BELIMUMAB, HYDROXYCHLOROQUINE, PREDNISOLONE | DrugsGivenReaction | CC BY-NC | 33341888 | 18,694,286 | 2021-09-01 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.