{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "36ff74fd-7380-4adc-87ca-6d6d7047632d", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/data/miniconda3/envs/llm_env/lib/python3.9/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", " from .autonotebook import tqdm as notebook_tqdm\n" ] } ], "source": [ "import re\n", "import json\n", "import torch\n", "import tiktoken\n", "import transformers\n", "from tqdm import tqdm\n", "\n", "from transformers import set_seed\n", "from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline\n", "from Levenshtein import distance as levenshtein_distance\n", "\n", "# Setting seed for reproduciblity of output\n", "# set_seed(42)\n", "# debug = False" ] }, { "cell_type": "code", "execution_count": 2, "id": "7c0500ff-6df7-41cc-92e1-7a71a396b378", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Tue Apr 30 13:49:31 2024 \n", "+---------------------------------------------------------------------------------------+\n", "| NVIDIA-SMI 545.23.08 Driver Version: 545.23.08 CUDA Version: 12.3 |\n", "|-----------------------------------------+----------------------+----------------------+\n", "| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n", "| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n", "| | | MIG M. |\n", "|=========================================+======================+======================|\n", "| 0 Tesla V100-SXM2-16GB Off | 00000000:00:1E.0 Off | 0 |\n", "| N/A 33C P0 47W / 300W | 5264MiB / 16384MiB | 0% Default |\n", "| | | N/A |\n", "+-----------------------------------------+----------------------+----------------------+\n", " \n", "+---------------------------------------------------------------------------------------+\n", "| Processes: |\n", "| GPU GI CI PID Type Process name GPU Memory |\n", "| ID ID Usage |\n", "|=======================================================================================|\n", "| 0 N/A N/A 2017768 C .../miniconda3/envs/llm_env/bin/python 5262MiB |\n", "+---------------------------------------------------------------------------------------+\n" ] } ], "source": [ "!nvidia-smi" ] }, { "cell_type": "code", "execution_count": 3, "id": "34d6032b-d615-490e-9b88-6854b089a73f", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n" ] } ], "source": [ "model_name_or_path = \"/data/mistral_gptq_model/\" #\"TheBloke/Mistral-7B-Instruct-v0.1-GPTQ\"\n", "\n", "model = AutoModelForCausalLM.from_pretrained(model_name_or_path,\n", " device_map=\"cuda:0\", #change this based on available instance\n", " trust_remote_code=False,\n", " revision=\"main\")\n", "\n", "tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)" ] }, { "cell_type": "code", "execution_count": 6, "id": "77f02223-5cef-49aa-9f12-3517ea4eaf9e", "metadata": {}, "outputs": [], "source": [ "debug = False" ] }, { "cell_type": "code", "execution_count": 7, "id": "0fccd68e-3e3c-4d52-9bf8-c31cf7c4f681", "metadata": {}, "outputs": [], "source": [ "# ui_map = {'Subject ID': 'SUBJID', 'Age': 'AGE', 'Race': 'RACE', 'Country': 'COUNTRY', 'Gender': 'GENDER', 'MH': 'MH', 'Date': 'DATE', 'Contact': 'CONTACT', 'Patient Related SiteID': 'Patient Related SiteID', 'CCI/CBI (Redact)': 'CCI/CBI (Redact)', 'Pre-Clinical Study Day': 'PRECLINICAL STUDY DAY', 'Circumstances': 'CIRCUMSTANCES', 'Non-Participant': 'NON-PARTICIPANTS', 'Weight': 'WEIGHT', 'Height': 'HEIGHT', 'BMI': 'BMI', 'AE/SAE': 'AE/SAE', 'Smoking Habits': 'SMOKING HABITS', 'Drinking Habits': 'DRINKING HABITS', 'Images': 'Images', 'Others-Patient Related': 'Others-Patient Related', 'Others-Study Related': 'Others-Study Related', 'Signature': 'Signature', 'Name': 'NAME', 'Study SiteID': 'Study SiteID', 'Ethnicity': 'ETHNICITY', 'Post-Clinical Study Day': 'Post-Clinical Study Day', 'Address': 'ADDRESS', 'Address': 'ADDRESS', 'BMI': 'BMI', 'Contact': 'CONTACT', 'Country': 'COUNTRY', 'Ethnicity': 'ETHNICITY', 'Gender': 'GENDER', 'Height': 'HEIGHT', 'Name': 'NAME', 'Race': 'RACE', 'Weight': 'WEIGHT', 'Social Security Numbers': 'Social Security Numbers', 'Account Numbers': 'Account Numbers', 'Certificate/License Numbers': 'Certificate/License Numbers', 'Circumstances': 'CIRCUMSTANCES', 'Drinking Habits': 'DRINKING HABITS', 'Medical History': 'MH', 'Non-Participant': 'NON-PARTICIPANTS', 'Pre-Clinical Study Day': 'PRECLINICAL STUDY DAY', 'Site ID': 'Study SiteID', 'Smoking Habits': 'SMOKING HABITS', 'Subject ID': 'SUBJID', 'Medical Record Numbers': 'Medical Record Numbers', 'Health Insurance Beneficiary Numbers': 'Health Insurance Beneficiary Numbers', 'Date': 'DATE', 'Vehicle Identifiers': 'Vehicle Identifiers', 'Device Attributes Or Serial Numbers': 'Device Attributes Or Serial Numbers', 'Website URL': 'Website URL', 'IP Address': 'IP Address', 'Age': 'AGE'}\n", "ui_map = {\"Name\": \"NAME\", \"Age\": \"AGE\", \"Gender\": \"GENDER\", \"Subject_id\": \"SUBJID\", \"Patient_id\": \"SUBJID\", \"Date\": \"DATE\", \"Study_dates\": \"DATE\", \"Medical_history\": \"MH\", \"Height\": \"HEIGHT\", \"Weight\": \"WEIGHT\", \"Phone_number\": \"CONTACT\", \"Email\": \"CONTACT\", \"Address\": \"ADDRESS\", \"Country\": \"COUNTRY\", \"Ethnicity\": \"RACE\",\n", " \"Non-participant\": \"NON_PARTICIPANT\", \"Non_participant\": \"NON_PARTICIPANT\", \"Smoking_habit\": \"SMOKING HABITS\", \"Drinking_habit\": \"DRINKING HABITS\", \"Study_day\": \"PRECLINICAL STUDY DAY\", \"Site_id\": \"Patient Related SiteID\", \"Circumstances\": \"CIRCUMSTANCES\"}\n", "# \"Location\": \"ORG\"," ] }, { "cell_type": "code", "execution_count": 8, "id": "e11d602e-446b-47e9-aec9-255661099373", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'Name': 'NAME', 'Age': 'AGE', 'Gender': 'GENDER', 'Subject_id': 'SUBJID', 'Patient_id': 'SUBJID', 'Date': 'DATE', 'Study_dates': 'DATE', 'Medical_history': 'MH', 'Height': 'HEIGHT', 'Weight': 'WEIGHT', 'Phone_number': 'CONTACT', 'Email': 'CONTACT', 'Address': 'ADDRESS', 'Country': 'COUNTRY', 'Ethnicity': 'RACE', 'Non-participant': 'NON_PARTICIPANT', 'Non_participant': 'NON_PARTICIPANT', 'Smoking_habit': 'SMOKING HABITS', 'Drinking_habit': 'DRINKING HABITS', 'Study_day': 'PRECLINICAL STUDY DAY', 'Site_id': 'Patient Related SiteID', 'Circumstances': 'CIRCUMSTANCES'}\n" ] } ], "source": [ "print(ui_map)" ] }, { "cell_type": "code", "execution_count": 9, "id": "8a440a10-9e8c-434b-9832-ca2a0700d384", "metadata": {}, "outputs": [], "source": [ "ui_model_key_mapping = {\"Name\": \"NAME\", \"Age\": \"AGE\", \"Gender\": \"GENDER\", \"Subject_id\": \"SUBJID\", \"Patient_id\": \"SUBJID\", \"Site_id\": \"SITEID\", \"Date\": \"DATE\", \"Medical_history\": \"MH\", \"Medical_condition\": \"MH\", \"Height\": \"HEIGHT\", \"Weight\": \"WEIGHT\", \"Phone_number\": \"CONTACT\", \"Email\": \"CONTACT\", \"Race\": \"RACE\", \"Ethnicity\": \"RACE\", \"Address\": \"ADDRESS\", \"Non-participant\": \"NON_PARTICIPANT\"}\n", "ui_hippa_mapping = {\"Name\": \"Name\", \"Age\": \"Age\", \"Gender\": \"Gender\", \"Address\": \"Address\", \"Country\": \"Country\", \"Location\": \"Location\", \"Date\": \"Date\",\n", " \"Email\": \"Contact\", \"Social_Security_Number\": \"Social Security Numbers\", \"Account_Number\": \"Account Numbers\", \"Medical_Record_Number\": \"Medical Record Numbers\", \"Health_Insurance_Beneficiary_Number\": \"Health Insurance Beneficiary Numbers\", \"Certificate_License_Number\": \"Certificate/License Numbers\", \"IP_Address\": \"IP Address\", \"URL\": \"Website URL\", \"Vehicle_Identifier\": \"Vehicle Identifiers\", \"Other_ID\": \"Device Attributes Or Serial Numbers\"}" ] }, { "cell_type": "code", "execution_count": 10, "id": "eadc500f-bb94-4a16-b0e2-8c99d6372524", "metadata": {}, "outputs": [], "source": [ "try:\n", " encoding = tiktoken.get_encoding(\"cl100k_base\")\n", "except:\n", " encoding = None\n", "\n", "def get_token_len(text):\n", " text = str(text)\n", " if encoding:\n", " encoded_vector = encoding.encode(text)\n", " else:\n", " encoded_vector = text.split(' ')\n", " return len(encoded_vector)" ] }, { "cell_type": "code", "execution_count": 11, "id": "17b4eaa6-7b95-4a33-b6cf-38ba619fe92a", "metadata": {}, "outputs": [], "source": [ "def search_phrase_in_paragraph(phrase, paragraph):\n", " # Search for the phrase in the paragraph using regular expression\n", " # print('===', paragraph)\n", " # print(phrase, paragraph)\n", " try:\n", " if phrase.startswith(\"[\") and phrase.endswith(\"]\"): \n", " phrase = re.escape(\"[\") + phrase + re.escape(\"]\")\n", " matches = re.finditer(phrase, paragraph, re.IGNORECASE)\n", " result = []\n", " \n", " for match in matches:\n", " # Get the start and end character positions of the phrase\n", " # if debug:\n", " # print(match)\n", " \n", " start = match.start()\n", " end = match.end()\n", " if match.group().startswith(' '):\n", " start = start + 1\n", " result.append((start, end))\n", " except:\n", " result = []\n", " return result" ] }, { "cell_type": "code", "execution_count": 12, "id": "28d0108c-2f87-4ecd-8ebf-80e80584323a", "metadata": {}, "outputs": [], "source": [ "def check_90_percent_match(phrase, paragraph):\n", " # Calculate the minimum required similarity score for a 90% match\n", " min_similarity_score = len(phrase) * 0.99\n", " \n", " # Initialize a list to store the start and end character positions of matches\n", " result = []\n", " \n", " # Iterate over the paragraph to find potential matches\n", " for i in range(len(paragraph) - len(phrase) + 1):\n", " substring = paragraph[i:i + len(phrase)]\n", " try:\n", " if paragraph[i-1] != ' ' and paragraph[i + len(phrase)] != ' ':\n", " continue\n", " except IndexError:\n", " continue\n", " # Calculate the Levenshtein distance between the phrase and the substring\n", " distance = levenshtein_distance(phrase, substring)\n", " \n", " # If the distance is less than the minimum required similarity score, it's a match\n", " if distance <= 2:\n", " start = i\n", " end = i + len(phrase)\n", " result.append((start, end))\n", " \n", " return result" ] }, { "cell_type": "code", "execution_count": 13, "id": "35a50521-ede3-4063-a46b-1133fd2fc265", "metadata": {}, "outputs": [], "source": [ "def merge_overlapping_tuples(tuples):\n", " if not tuples:\n", " return []\n", "\n", " # Sort the list of tuples by their start index\n", " sorted_tuples = sorted(tuples, key=lambda x: x[0])\n", "\n", " merged = []\n", " current_start, current_end = sorted_tuples[0]\n", "\n", " for start, end in sorted_tuples[1:]:\n", " if start <= current_end:\n", " current_end = max(current_end, end)\n", " else:\n", " merged.append((current_start, current_end))\n", " current_start, current_end = start, end\n", "\n", " # Append the last merged tuple\n", " merged.append((current_start, current_end))\n", "\n", " return merged\n", "\n", "# # Unit Test the function\n", "# input_tuples = [(10, 17), (11, 18), (12, 19), (56, 63), (57, 64), (58, 65)]\n", "# output_tuples = merge_overlapping_tuples(input_tuples)\n", "# print(output_tuples)" ] }, { "cell_type": "code", "execution_count": 14, "id": "4e29a412-bc44-4392-8b32-4b797680aec0", "metadata": {}, "outputs": [], "source": [ "def get_entity_position(phrase, paragraph):\n", " # Search for the phrase in the paragraph\n", " if len(phrase) <=2:\n", " phrase = ' '+phrase+' '\n", " elif len(phrase) <= 4:\n", " phrase = ' '+phrase\n", " positions = search_phrase_in_paragraph(phrase, paragraph)\n", "\n", " # Check for a 90% match of the phrase in the paragraph\n", " if len(positions) == 0:\n", " if len(phrase) > 6:\n", " positions = check_90_percent_match(phrase, paragraph)\n", " if positions:\n", " positions = merge_overlapping_tuples(positions)\n", " return positions" ] }, { "cell_type": "code", "execution_count": 15, "id": "9e03c942-1809-4fcf-b527-d676f1961228", "metadata": {}, "outputs": [], "source": [ "def get_word_from_character(span, text): \n", " # Returns the word based on character position\n", "\n", " text = text.strip()\n", " words_text = text.split(\" \")\n", " words_text = [w for w in words_text if w != \"\"]\n", " wrd_index = []\n", " char_counter = 0\n", " for idx, word in enumerate(words_text):\n", " wrd_index.append((char_counter, char_counter + len(word)))\n", " char_counter += len(word) + 1\n", "\n", " left_indices = [wrd_ind[0] for wrd_ind in wrd_index]\n", " right_indices = [wrd_ind[1] for wrd_ind in wrd_index]\n", "\n", " start_char, end_char = span\n", " assert not start_char > end_char\n", "\n", " nearest_start = left_indices[0]\n", " nearest_end = right_indices[0]\n", " nearest_start_diff = 989989\n", " nearest_end_diff = 989989\n", " start_wordloc = -1\n", " end_wordloc = -1\n", "\n", " for wordloc, indx in enumerate(left_indices):\n", " diff = abs(indx - start_char)\n", " if diff <= nearest_start_diff and start_char >= indx:\n", " nearest_start_diff = diff\n", " nearest_start = indx\n", " start_wordloc = wordloc\n", "\n", " for wordloc, indx in enumerate(right_indices):\n", " diff = abs(indx - end_char)\n", "\n", " if diff <= nearest_end_diff and end_char <= indx:\n", " nearest_end_diff = diff\n", " nearest_end = indx\n", " end_wordloc = wordloc\n", "\n", " if end_wordloc == -1 or end_wordloc == start_wordloc:\n", " end_wordloc = start_wordloc\n", " return (start_wordloc, end_wordloc + 1)" ] }, { "cell_type": "code", "execution_count": 16, "id": "257dcb66-6849-4e5d-9cf2-f36eee936419", "metadata": {}, "outputs": [], "source": [ "ui_mapping = {\"Name\": \"NAME\", \"Age\": \"AGE\", \"Gender\": \"GENDER\", \"Subject_id\": \"SUBJID\", \"Patient_id\": \"SUBJID\", \"Date\": \"DATE\", \"Study_dates\": \"DATE\", \"Medical_history\": \"MH\", \"Height\": \"HEIGHT\", \"Weight\": \"WEIGHT\", \"Phone_number\": \"CONTACT\", \"Email\": \"CONTACT\", \"Address\": \"ADDRESS\", \"Country\": \"COUNTRY\", \"Ethnicity\": \"RACE\",\n", " \"Non-participant\": \"NON_PARTICIPANT\", \"Non_participant\": \"NON_PARTICIPANT\", \"Smoking_habit\": \"SMOKING HABITS\", \"Drinking_habit\": \"DRINKING HABITS\", \"Study_day\": \"PRECLINICAL STUDY DAY\", \"Site_id\": \"Patient Related SiteID\", \"Circumstances\": \"CIRCUMSTANCES\"}\n", "def extract_values(text, paragraph):\n", " lines = text.split('\\n') # Split input text into lines\n", " result = [] # Initialize an empty list to store the output\n", " value_list = []\n", " coref_subjid = \"\"\n", " for line in lines:\n", " if line.strip() == \"End-Answer\":\n", " break\n", " parts = line.split(': ') # Split each line into key and value\n", " if len(parts) == 2:\n", " key, values = parts[0].strip(), parts[1].strip() # Extract key and values\n", " ui_key = ui_mapping.get(key, None)\n", " if ui_key:\n", " if (' | ' in values) or (ui_key in ['ADDRESS', 'DATE']):\n", " values = values.split('|') # Split values by comma\n", " else:\n", " values = values.split(', ')\n", " for value in values:\n", " value = value.strip()\n", " if value.strip() == '':\n", " continue\n", " value_positions = get_entity_position(value, paragraph)\n", " for entity_position in value_positions:\n", " word_positions = get_word_from_character((entity_position[0], entity_position[1]), paragraph)\n", " result.append((paragraph[entity_position[0]: entity_position[1]], ui_key, entity_position[0], entity_position[1], word_positions[0], word_positions[1])) # Append tuples to the result list\n", " return result" ] }, { "cell_type": "code", "execution_count": 17, "id": "c672acdb-3196-45d0-820b-9f20f0531799", "metadata": {}, "outputs": [], "source": [ "from auto_gptq import exllama_set_max_input_length\n", "model = exllama_set_max_input_length(model, 4096)" ] }, { "cell_type": "markdown", "id": "5ef8322c-e03e-4480-a588-65bc9f60db10", "metadata": {}, "source": [ "### Values with pipe" ] }, { "cell_type": "code", "execution_count": 26, "id": "5e8a1549-561f-4c8f-9fb9-2888b58ed6b4", "metadata": {}, "outputs": [], "source": [ "# General entities\n", "# {\"Name\": \"NAME\", \"Age\": \"AGE\", \"Gender\": \"GENDER\", \"Date\": \"DATE\", \"Location\": \"ADDRESS\", \"Country\": \"COUNTRY\"}\n", "system_prompt = '''An entity is a person(name), age, gender, title, named organization, city, state, country, address, zipcode, dates.\n", "All entities should strictly get extracted, but don't extract incorrect entities.\n", "Entity present in any part of the paragraph should be detected.\n", "Names/Person : Names of persons mentioned in the input sentence paragraph\n", "Gender: Male, Female, male, female, he, she, him, her should be detected as Gender.\n", "Age: Age of a person should get detected. Some examples are 12 yo, 15 years old, 11 months old, 79 years, 51 yrs.\n", "Date: All dates should get detected as Dates. Some examples are 10-1-2021, 23-01-2022, 27th June, June 29, 10-09-2007, 4/5/67, 7/67, 31-12-2000. \n", "start_date, end_date should get detected as Date itself.\n", "Study_day: Day 2, Day 5, day 9, day 11 of study, 3rd day of study. Study day shouldn't get detected as Date.\n", "If there are multiple values for same entity they should be extracted with | separation.\n", "Random text should not get detected as any of above entities.\n", "All above entities should strictly be extracted.\n", "URLs shouldn't get detected as Location or Address.\n", "'''\n", "# Countries like China, US, France, India should always get detected as country. Countries like China, US, France, India, South Korea should always get detected as Country. A country can't be a Subject_id.\n", "example1 = '''Essentially , Mr. Cornea is a 60 year old male from Canada who noted the onset of dark urine during early January . Subtyping was unsuccessful on Day 1, but showed infection with H3 subtype on Day 2, and with H1pdm09 on Day 6.\n", "Answer: \n", "Name: Mr. Cornea\n", "Age: 60 year old\n", "Gender: male\n", "Country: Canada\n", "Date: early January\n", "Study_day: Day 1 | Day 2 | Day 6\n", "End-Answer\n", "'''\n", "\n", "example2 = '''In the 205 subjects with post-ofatumumab HAHA results (G2 MSD ECL assay; Section 3.5), one subject (Subject 257) tested positive for HAHA on 10-1-2021, 23-01-2022, and 185 subjects had all negative post-ofatumumab HAHA results with at least one ofatumumab plasma concentration low enough (<200 µg/mL) for the negative HAHA results to be considered conclusive. Subject 257 tested positive for HAHA at the six-month follow-up visit (titer = 16); samples at all other time points were negative.\n", "2 patients were released from the study due to inactivity. On Day 3, the study was terminated.\n", "Answer:\n", "Subject_id: 257 | 257\n", "Date: 10-1-2021 | 23-01-2022\n", "Numper_of_patients: 2 | 205 | 185\n", "Study_day: Day 3\n", "End-Answer\n", "'''\n", "example3 = '''Subject 169: 54 years, Female, Pneumocystis jirovecii pneumonia, 61 days after last dose, resolved. The subject(CS3A Subject 1833-2303) was hospitalized for severe bronchopneumonia. Bronchoalveolar lavage and polymerase chain reaction (PCR) for Pneumocystis were positive for Pneumocystis jiroveci. The subject was hospitalized on 27th June and discharged on June 29. Two patients were on the study from 10-09-2007.\n", "Answer:\n", "Subject_id: 169 | 1833-2303\n", "Age: 54 years\n", "Gender: Female\n", "Race: White\n", "Date: 27th June | June 29 | 10-09-2007\n", "Medical_history: Pneumocystis jirovecii pneumonia\n", "End-Answer\n", "'''\n", "\n", "example4 = '''Saw Dr Oakley 4/5/67 - she was happy with results of ETT at Clarkfield. To f/u 7/67. No CP's since last admit. On Day 521, the patient was discharged.\n", "Another patient died of cardio-respiratory arrest on Day 8.\n", "Answer:\n", "Name: Dr Oakley\n", "Date: 4/5/67 | 7/67\n", "Gender: she\n", "Location: Clarkfield\n", "Study_day: Day 521 | Day 8\n", "End-Answer\n", "'''\n", "\n", "example5 = '''Patient 12367/2134 Oseltamivir 100 mg IV q12h for 5 days S295H/Y Patient 12367/2144 was a 57 year old male enrolled on 31-12-2000, on Day 5(03-Jan-2021) the patient was admitted.\n", "Answer:\n", "Patient_id: 12367/2134 | 12367/2144\n", "Gender: male\n", "Date: 31-12-2000 | 03-Jan-2021\n", "Study_day: Day 5\n", "Age: 57 year old\n", "End-Answer'''\n", "\n", "example6 = '''918 Abdominal pain 2013-02-02 (8) Y Y 918-07 Abdominal pain Gastrointestinal disorders 2013-02-03 (9) Y Grade II Medical / other treatment Not suspected Suspected Recovered/resolved.\n", "The events resolved afterwards on Day 4 (pyrexia, vomiting) (02-Feb-2013) and Day 21 (asthenia) (19-Feb-2013).'\n", "Answer:\n", "Site_id: 918\n", "Subject_id: 918-07\n", "Study_day: Day 4 | Day 21\n", "Date: 2013-02-02 | 2013-02-03 | 02-Feb-2013 | 19-Feb-2013\n", "Medical_history: Abdominal pain | Gastrointestinal disorders\n", "End-Answer'''\n", "\n", "example7 = '''The medical history included throat cancer beginning in 2011 and ending in 2011, a broken hip beginning in 2011 and ending in 2011, insertion of two stents in 2012, coronary artery bypass surgery on 25-Jun-2014, a right cerebrovascular accident on 25-Jun-2014, aortic valve replacement on 25-Jun-2014, bilateral cataracts on 15-Aug-2014.\n", "Answer:\n", "Medical_history: throat cancer | broken hip | insertion of two stents | coronary artery bypass surgery | right cerebrovascular accident | aortic valve replacement | bilateral cataracts\n", "Date: 25-Jun-2014 | 25-Jun-2014 | 25-Jun-2014 | 15-Aug-2014\n", "End-Answer'''\n", "\n", "example8 = '''1083\\nPatient [AUS02T-0215-787043] - Death\\n(disseminated intravascular coagulation),\\nSAE (haemorrhage intracranial,\\ndisseminated intravascular coagulation) .\n", "Answer:\n", "Subject_id: [AUS02T-0215-787043]\n", "End-Answer'''\n", "\n", "example9 = '''Pneumococcal vaccine E01-643, E03-285 E05-829. He underwent an ECHO and endoscopy at Ingree and Ot of Weamanshy Medical Center(www.wmchospital.com) on April 28 . In one patient the SAE had a fatal\\noutcome.\n", "Answer:\n", "Subject_id: E01-643 | E03-285 | E05-829\n", "Gender: He\n", "Location: Weamanshy Medical Center\n", "URL: www.wmchospital.com\n", "Date: April 28\n", "Numper_of_patients: one patient\n", "End-Answer'''\n", "\n", "# example9 = '''He underwent an ECHO and endoscopy at Ingree and Ot of Weamanshy Medical Center(www.wmchospital.com) on April 28 . In one patient the SAE had a fatal\\noutcome.\n", "# Answer:\n", "# Gender: He\n", "# Location: Weamanshy Medical Center\n", "# URL: www.wmchospital.com\n", "# Date: April 28\n", "# Numper_of_patients: one patient\n", "# End-Answer'''\n", "\n", "example10 = '''On Day 1 one blood sample was put in the centrifuge 1 minute late for 2 Subjects (ABC/1000/00009 and [X1101-2009-008]. The patient had never smoked and was a former drinker. The subject was accompanied by his mother in China. He then completed study treatment with a total of 19 doses of IV oseltamivir (10 doses standard treatment plus 9 doses of treatment extension). \n", "Answer: \n", "Subject_id: ABC/1000/00009 | [X1101-2009-008]\n", "Smoking_habit: never smoked\n", "Drinking_habit: former drinker\n", "Country: China\n", "Gender: his | He\n", "Non_participant: mother\n", "End-Answer'''\n", "\n", "example11 = '''Patient ois/51317-31580/10653 was a 54-year-old Hispanic female. A virus was identified in patients in the United States, South Korea, China. The virus spread rapidly throughout the US and the world.\n", "Answer:\n", "Subject_id: ois/51317-31580/10653\n", "Age: 54-year-old\n", "Race: Hispanic\n", "Gender: female\n", "Country: United States | South Korea | China | US\n", "End-Answer'''\n", "\n", "example12 = '''ATC classes are presented alphabetically; preferred terms are sorted within ATC class alphabetically. - A medication can appear with more than one ATC class. Program: CIGG013A/CIGG013A1101J/report/pgm_saf/t_cmd03.sas, 18:33 14OCT2014 Final. https://www.google.com/\n", "Study_dates: 18 Nov 2009-28 Jun 2010\n", "Answer:\n", "Program: CIGG013A/CIGG013A1101J/report/pgm_saf/t_cmd03.sas\n", "Date: 14OCT2014 | 18 Nov 2009 | 28 Jun 2010\n", "URL: https://www.google.com/\n", "End-Answer\n", "'''\n", "\n", "example13 = '''3. Appendix 6 ICH Guidelines for Clinical Safety Data Management. Linking to a website (https://www.antennahouse.com/)：external-destination=\"https://www.antennahouse.com/\"\n", "Answer:\n", "Appendix: Appendix 6 ICH Guidelines for Clinical Safety Data Management\n", "URL: https://www.antennahouse.com/\n", "End-Answer\n", "'''\n", "\n", "example14 = '''Subject in South Korea is reaching Cycle 37 Day 1 soon on Oct 23. \n", "Answer:\n", "Country: South Korea\n", "Cycle: Cycle 37\n", "Study_day: Day 1\n", "Date: Oct 23\n", "End-Answer\n", "'''\n", "whole_task = '''Given the paragraph below identify list of possible entities.\n", "Paragraph:'''\n", "\n", "example_list = [example1, example11, example2, example3, example4, example14, example5, example6, example7, example8, example9, example10, example12]\n", "\n", "final_prompt = ''\n", "for ex in example_list:\n", " final_prompt = final_prompt+''+whole_task+'\\n'\n", " final_prompt = final_prompt+''+ex+'\\n'\n", "\n", "final_prompt = system_prompt+'\\n'+final_prompt" ] }, { "cell_type": "code", "execution_count": 27, "id": "50aa0a75-1924-40f7-97ec-83463b99a179", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "An entity is a person(name), age, gender, title, named organization, city, state, country, address, zipcode, dates.\n", "All entities should strictly get extracted, but don't extract incorrect entities.\n", "Entity present in any part of the paragraph should be detected.\n", "Names/Person : Names of persons mentioned in the input sentence paragraph\n", "Gender: Male, Female, male, female, he, she, him, her should be detected as Gender.\n", "Age: Age of a person should get detected. Some examples are 12 yo, 15 years old, 11 months old, 79 years, 51 yrs.\n", "Date: All dates should get detected as Dates. Some examples are 10-1-2021, 23-01-2022, 27th June, June 29, 10-09-2007, 4/5/67, 7/67, 31-12-2000. \n", "start_date, end_date should get detected as Date itself.\n", "Study_day: Day 2, Day 5, day 9, day 11 of study, 3rd day of study. Study day shouldn't get detected as Date.\n", "If there are multiple values for same entity they should be extracted with | separation.\n", "Random text should not get detected as any of above entities.\n", "All above entities should strictly be extracted.\n", "URLs shouldn't get detected as Location or Address.\n", "\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Essentially , Mr. Cornea is a 60 year old male from Canada who noted the onset of dark urine during early January . Subtyping was unsuccessful on Day 1, but showed infection with H3 subtype on Day 2, and with H1pdm09 on Day 6.\n", "Answer: \n", "Name: Mr. Cornea\n", "Age: 60 year old\n", "Gender: male\n", "Country: Canada\n", "Date: early January\n", "Study_day: Day 1 | Day 2 | Day 6\n", "End-Answer\n", "\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Patient ois/51317-31580/10653 was a 54-year-old Hispanic female. A virus was identified in patients in the United States, South Korea, China. The virus spread rapidly throughout the US and the world.\n", "Answer:\n", "Subject_id: ois/51317-31580/10653\n", "Age: 54-year-old\n", "Race: Hispanic\n", "Gender: female\n", "Country: United States | South Korea | China | US\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "In the 205 subjects with post-ofatumumab HAHA results (G2 MSD ECL assay; Section 3.5), one subject (Subject 257) tested positive for HAHA on 10-1-2021, 23-01-2022, and 185 subjects had all negative post-ofatumumab HAHA results with at least one ofatumumab plasma concentration low enough (<200 µg/mL) for the negative HAHA results to be considered conclusive. Subject 257 tested positive for HAHA at the six-month follow-up visit (titer = 16); samples at all other time points were negative.\n", "2 patients were released from the study due to inactivity. On Day 3, the study was terminated.\n", "Answer:\n", "Subject_id: 257 | 257\n", "Date: 10-1-2021 | 23-01-2022\n", "Numper_of_patients: 2 | 205 | 185\n", "Study_day: Day 3\n", "End-Answer\n", "\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Subject 169: 54 years, Female, Pneumocystis jirovecii pneumonia, 61 days after last dose, resolved. The subject(CS3A Subject 1833-2303) was hospitalized for severe bronchopneumonia. Bronchoalveolar lavage and polymerase chain reaction (PCR) for Pneumocystis were positive for Pneumocystis jiroveci. The subject was hospitalized on 27th June and discharged on June 29. Two patients were on the study from 10-09-2007.\n", "Answer:\n", "Subject_id: 169 | 1833-2303\n", "Age: 54 years\n", "Gender: Female\n", "Race: White\n", "Date: 27th June | June 29 | 10-09-2007\n", "Medical_history: Pneumocystis jirovecii pneumonia\n", "End-Answer\n", "\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Saw Dr Oakley 4/5/67 - she was happy with results of ETT at Clarkfield. To f/u 7/67. No CP's since last admit. On Day 521, the patient was discharged.\n", "Another patient died of cardio-respiratory arrest on Day 8.\n", "Answer:\n", "Name: Dr Oakley\n", "Date: 4/5/67 | 7/67\n", "Gender: she\n", "Location: Clarkfield\n", "Study_day: Day 521 | Day 8\n", "End-Answer\n", "\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Subject in South Korea is reaching Cycle 37 Day 1 soon on Oct 23. \n", "Answer:\n", "Country: South Korea\n", "Cycle: Cycle 37\n", "Study_day: Day 1\n", "Date: Oct 23\n", "End-Answer\n", "\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Patient 12367/2134 Oseltamivir 100 mg IV q12h for 5 days S295H/Y Patient 12367/2144 was a 57 year old male enrolled on 31-12-2000, on Day 5(03-Jan-2021) the patient was admitted.\n", "Answer:\n", "Patient_id: 12367/2134 | 12367/2144\n", "Gender: male\n", "Date: 31-12-2000 | 03-Jan-2021\n", "Study_day: Day 5\n", "Age: 57 year old\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "918 Abdominal pain 2013-02-02 (8) Y Y 918-07 Abdominal pain Gastrointestinal disorders 2013-02-03 (9) Y Grade II Medical / other treatment Not suspected Suspected Recovered/resolved.\n", "The events resolved afterwards on Day 4 (pyrexia, vomiting) (02-Feb-2013) and Day 21 (asthenia) (19-Feb-2013).'\n", "Answer:\n", "Site_id: 918\n", "Subject_id: 918-07\n", "Study_day: Day 4 | Day 21\n", "Date: 2013-02-02 | 2013-02-03 | 02-Feb-2013 | 19-Feb-2013\n", "Medical_history: Abdominal pain | Gastrointestinal disorders\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "The medical history included throat cancer beginning in 2011 and ending in 2011, a broken hip beginning in 2011 and ending in 2011, insertion of two stents in 2012, coronary artery bypass surgery on 25-Jun-2014, a right cerebrovascular accident on 25-Jun-2014, aortic valve replacement on 25-Jun-2014, bilateral cataracts on 15-Aug-2014.\n", "Answer:\n", "Medical_history: throat cancer | broken hip | insertion of two stents | coronary artery bypass surgery | right cerebrovascular accident | aortic valve replacement | bilateral cataracts\n", "Date: 25-Jun-2014 | 25-Jun-2014 | 25-Jun-2014 | 15-Aug-2014\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "1083\n", "Patient [AUS02T-0215-787043] - Death\n", "(disseminated intravascular coagulation),\n", "SAE (haemorrhage intracranial,\n", "disseminated intravascular coagulation) .\n", "Answer:\n", "Subject_id: [AUS02T-0215-787043]\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "Pneumococcal vaccine E01-643, E03-285 E05-829. He underwent an ECHO and endoscopy at Ingree and Ot of Weamanshy Medical Center(www.wmchospital.com) on April 28 . In one patient the SAE had a fatal\n", "outcome.\n", "Answer:\n", "Subject_id: E01-643 | E03-285 | E05-829\n", "Gender: He\n", "Location: Weamanshy Medical Center\n", "URL: www.wmchospital.com\n", "Date: April 28\n", "Numper_of_patients: one patient\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "On Day 1 one blood sample was put in the centrifuge 1 minute late for 2 Subjects (ABC/1000/00009 and [X1101-2009-008]. The patient had never smoked and was a former drinker. The subject was accompanied by his mother in China. He then completed study treatment with a total of 19 doses of IV oseltamivir (10 doses standard treatment plus 9 doses of treatment extension). \n", "Answer: \n", "Subject_id: ABC/1000/00009 | [X1101-2009-008]\n", "Smoking_habit: never smoked\n", "Drinking_habit: former drinker\n", "Country: China\n", "Gender: his | He\n", "Non_participant: mother\n", "End-Answer\n", "Given the paragraph below identify list of possible entities.\n", "Paragraph:\n", "ATC classes are presented alphabetically; preferred terms are sorted within ATC class alphabetically. - A medication can appear with more than one ATC class. Program: CIGG013A/CIGG013A1101J/report/pgm_saf/t_cmd03.sas, 18:33 14OCT2014 Final. https://www.google.com/\n", "Study_dates: 18 Nov 2009-28 Jun 2010\n", "Answer:\n", "Program: CIGG013A/CIGG013A1101J/report/pgm_saf/t_cmd03.sas\n", "Date: 14OCT2014 | 18 Nov 2009 | 28 Jun 2010\n", "URL: https://www.google.com/\n", "End-Answer\n", "\n", "\n" ] } ], "source": [ "print(final_prompt)" ] }, { "cell_type": "markdown", "id": "412ed1b7-149d-4fea-bb13-752c492395dc", "metadata": {}, "source": [ "### HIPPA prompts" ] }, { "cell_type": "code", "execution_count": 20, "id": "9faae550-64b9-4598-bdec-270f5ae98b32", "metadata": {}, "outputs": [], "source": [ "# # def get_prompts_hippa():\n", "# system_prompt = ''' An entity is a Social Security Number, Account Number, Medical Record Number, Health Insurance Beneficiary Number, Certificate License Number, IP Address, Website URL, Vehicle Identifier.\n", "# General entities like Name, Date, Address, Phone_no, Email should also get detected.\n", "# Name should be a proper noun, not a phrase.\n", "# Any word with which gender can get revealed should get detected as Gender, for ex. he, she, He, She, him, her\n", "# Name can't be Other_ID\n", "# A Other_ID can't be a medical condition.\n", "# The output should be present in key value pair as provided in examples below.\n", "# The keys should be Social_Security_Number, Account_Number, Medical_Record_Number, Health_Insurance_Beneficiary_Number, Certificate_License_Number, IP_Address, URL, Vehicle_Identifier, Other_ID.\n", "# Medical_Record_Number should be numeric or alpha-numeric.\n", "# If there are multiple values for same entity they should be extracted with |(pipe) separator. All the entities should be extracted strictly if present in paragraph.\n", "# '''\n", "\n", "# example1 = '''John Smith is scheduled for a medical appointment on September 20th at 123 Main Street. You can reach him at 555-123-4567 or via email at john.smith@email.com.\n", "# Answer: \n", "# Name: John Smith\n", "# Date: September 20th\n", "# Address: 123 Main Street\n", "# Phone_no: 555-123-4567\n", "# Email: john.smith@email.com\n", "# End-Answer\n", "# '''\n", "\n", "# example2 = '''Dr. Sarah Miller has a valid medical license (Certificate License Number: MD12345) and practices at 789 Oak Avenue.\n", "# Answer:\n", "# Name: Dr. Sarah Miller, Mr. No\n", "# Certificate_License_Number: MD12345\n", "# Adddress: 789 Oak Avenue\n", "# End-Answer\n", "# '''\n", "# example3 = '''The accident involved Vehicle ID: ABC123 and occurred on August 15, 2023.\n", "# Answer:\n", "# Vehicle_Identifier: ABC123\n", "# Date: August 15, 2023\n", "# End-Answer\n", "# '''\n", "\n", "# example4 = '''Visit our website at www.healthcareclinic.com for more information on our services.\n", "# 6/7/1999 12:00:00 AM\n", "# 02/15/96\n", "# 3-1-95\n", "# Answer:\n", "# URL: www.healthcareclinic.com\n", "# Date: 6/7/1999, 02/15/96, 3-1-95\n", "# End-Answer\n", "# '''\n", "# example5 = '''113416550\n", "# PRGH\n", "# 13523357\n", "# 630190\n", "# Answer:\n", "# Other_ID: 113416550, 13523357, 630190\n", "# Location: PRGH\n", "# End-Answer\n", "# '''\n", "\n", "# example6 = '''Abdominal CT on April 14 , showed a 12 x 8 x 8 cm mass in the region of the left liver , and appeared to be from the lesser curvature of the stomach or left liver .\n", "# Answer:\n", "# Date: April 14\n", "# End-Answer\n", "# '''\n", "\n", "# example7 = '''229937784\n", "# Answer:\n", "# Other_ID: 229937784\n", "# End-Answer\n", "# '''\n", "\n", "# example8 = '''Mr. John suffers from a bipolar mood disorder , together with hypercholesterolemia and vertebral basilar insufficiency .\n", "# Answer:\n", "# Name: Mr. John\n", "# Medical_Condition: bipolar mood disorder, hypercholesterolemia, vertebral basilar insufficiency\n", "# End-Answer\n", "# '''\n", "# example9 = '''ATC classes are presented alphabetically; preferred terms are sorted within ATC class alphabetically. - A medication can appear with more than one ATC class. Program: CIGG013A/CIGG013A1101J/report/pgm_saf/t_cmd03.sas, 18:33 14OCT2014 Final\n", "# Answer:\n", "# Program: CIGG013A/CIGG013A1101J/report/pgm_saf/t_cmd03.sas\n", "# Date: 14OCT2014\n", "# End-Answer\n", "# '''\n", "\n", "# whole_task = '''Given the paragraph below identify list of possible entities.\n", "# Paragraph:'''\n", "\n", "# example_list = [example1, example2, example3, example4, example5, example6, example7, example8]\n", "\n", "# final_prompt = ''\n", "# for ex in example_list:\n", "# final_prompt = final_prompt+''+whole_task+'\\n'\n", "# final_prompt = final_prompt+''+ex+'\\n'\n", "\n", "# final_prompt = system_prompt+'\\n'+final_prompt\n", "\n", "# # return final_prompt, whole_task" ] }, { "cell_type": "code", "execution_count": 20, "id": "1f1af643-bd98-4897-a376-0830bef3079b", "metadata": {}, "outputs": [], "source": [ "import pysbd" ] }, { "cell_type": "code", "execution_count": 28, "id": "28a5979c-be66-44a2-86bf-4c0535b15ac6", "metadata": {}, "outputs": [], "source": [ "seg = pysbd.Segmenter(language='en', clean=False)\n", "# seg.segment(input_paragraph)" ] }, { "cell_type": "code", "execution_count": 29, "id": "9fd1c502-3717-4835-b6f4-92ff8a6468a8", "metadata": {}, "outputs": [], "source": [ "def remove_overlapping_entities(entities):\n", " \"\"\"\n", " Remove overlapping entities from a list, giving priority to entities that appear earlier in the list.\n", "\n", " :param entities: List of entities, where each entity is represented as a tuple (start_pos, end_pos).\n", " :return: A new list of entities with no overlaps.\n", " \"\"\"\n", " non_overlapping = []\n", "\n", " for current_entity in entities:\n", " overlap = False\n", " for added_entity in non_overlapping:\n", " # Check if the current entity overlaps with any already added entity\n", " if (not (current_entity[3] <= added_entity[2] or current_entity[2] >= added_entity[3])) or ((current_entity[4] == added_entity[4] or current_entity[5] == added_entity[5])):\n", " overlap = True\n", " break\n", " \n", " if not overlap:\n", " non_overlapping.append(current_entity)\n", "\n", " return non_overlapping" ] }, { "cell_type": "code", "execution_count": 30, "id": "d8700227-e3f9-434f-b334-ea25ab480b94", "metadata": {}, "outputs": [], "source": [ "def basic_postprocessing(entity_list):\n", " filtered_list = []\n", " for entity in entity_list:\n", " if (entity[1] == \"SUBJID\" or entity[1] == \"Patient Related SiteID\") and ((not any(char.isdigit() for char in entity[0])) or (len(entity[0].strip()) < 3)):\n", " continue\n", " if (entity[1] == \"DATE\") and ((not any(char.isdigit() for char in entity[0])) or (len(entity[0].strip()) < 3)):\n", " continue\n", " elif entity[1] == \"DATE\" and 'day' in entity[0].lower():\n", " temp_entity = (entity[0], 'PRECLINICAL STUDY DAY', entity[2], entity[3], entity[4], entity[5])\n", " filtered_list.append(temp_entity)\n", " continue\n", " elif entity[1] == \"ADDRESS\" and len(entity[0].strip().split(' ')) < 2:\n", " continue\n", " filtered_list.append(entity)\n", " return filtered_list" ] }, { "cell_type": "markdown", "id": "24b6e061-dd98-4de0-8180-badf70bdb3fc", "metadata": {}, "source": [ "### Run this to just check the model output" ] }, { "cell_type": "code", "execution_count": 31, "id": "6f593837-3fc5-451c-bff7-784d9ec24d45", "metadata": {}, "outputs": [], "source": [ "# import re\n", "# import gc\n", "# import torch\n", "\n", "# sentence = 'Mr. John Wick is a 77 year old male with squamous cell carcinoma of the lung .'\n", "# sent_token_len = get_token_len(sentence)\n", "# final_prompt_to_model = final_prompt+'\\n'+whole_task+'\\n'+sentence+\"\\nAnswer:\"\n", "# input_ids = tokenizer(final_prompt_to_model, return_tensors='pt').input_ids.cuda()\n", "# output = model.generate(inputs=input_ids, temperature=1, do_sample=False, max_new_tokens=sent_token_len*3, eos_token_id=16981)\n", "# model_generated_output = tokenizer.decode(output[0][input_ids[0].shape[0]:])\n", "# print('\\n')\n", "# print(model_generated_output)\n", "# del output\n", "# del input_ids\n", "# torch.cuda.empty_cache()\n", "# gc.collect()" ] }, { "cell_type": "code", "execution_count": 32, "id": "4414cda5-058c-46ba-a525-17b90bd3c39f", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "['Samaresh Yadav joined Gramener on 23rd Nov 2024.\\n']\n", "\n", "Name: Samaresh Yadav\n", "Date: 23rd Nov 2024\n", "End-Answer\n", "\n", "\n", "\n", "[('Samaresh Yadav', 'NAME', 0, 14, 0, 2), ('23rd Nov 2024', 'DATE', 34, 47, 5, 8)]\n" ] } ], "source": [ "import re\n", "import gc\n", "import torch\n", "\n", "import os\n", "from datasets import load_dataset\n", "os.environ['https_proxy'] = 'http://eu-chst-proxy.eu.novartis.net:2011'\n", "os.environ['https_proxy'] = 'http://eu-nlms-proxy.eu.novartis.net:2011'\n", "input_paragraph = '''Samaresh Yadav joined Gramener on 23rd Nov 2024.\n", "'''\n", "# sentences = re.split(r'(?<=[.;:!?])\\s', input_paragraph)\n", "if len(input_paragraph.strip()) < 30:\n", " sentences = [input_paragraph]\n", "else:\n", " sentences = seg.segment(input_paragraph)\n", "print(sentences)\n", "para_value_list = []\n", "for sentence in sentences:\n", " sent_value_list = []\n", " sent_token_len = get_token_len(sentence)\n", " final_prompt_to_model = final_prompt+'\\n'+whole_task+'\\n'+sentence.strip()+\"\\nAnswer:\"\n", " input_ids = tokenizer(final_prompt_to_model, return_tensors='pt').input_ids.cuda()\n", " output = model.generate(inputs=input_ids, temperature=1, do_sample=False, max_new_tokens=sent_token_len*3, eos_token_id=16981)\n", " model_generated_output = tokenizer.decode(output[0][input_ids[0].shape[0]:])\n", " print(model_generated_output)\n", " sent_value_list = extract_values(model_generated_output, input_paragraph)\n", " para_value_list.extend(sent_value_list)\n", " del output\n", " del input_ids\n", " torch.cuda.empty_cache()\n", " gc.collect()\n", "print('\\n\\n')\n", "para_value_list = remove_overlapping_entities(para_value_list)\n", "para_value_list = basic_postprocessing(para_value_list)\n", "print(para_value_list)" ] }, { "cell_type": "code", "execution_count": 34, "id": "cb89de2f-2f79-4510-9970-72534b270be1", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Downloading readme: 100%|██████████| 666/666 [00:00<00:00, 3.03MB/s]\n", "Downloading data files: 0%| | 0/3 [00:00 0:\n", "# for i in tags.find_all():\n", "# tag = i[1]\n", "# start = i[2]\n", "# end = i[3]\n", "# text_patch = (modified_text[offset: start], tag)\n", "# text_patches.append(text_patch)\n", "# offset = end\n", "\n", "# text_patches.append((modified_text[offset:], ''))" ] }, { "cell_type": "code", "execution_count": 51, "id": "7753ea5e-3844-4075-979e-52eac475abf1", "metadata": {}, "outputs": [], "source": [ "def get_redacted_text(text_patches):\n", " redacted_text = ''\n", " for patch_tuple in text_patches:\n", " text = patch_tuple[0]\n", " text_label = patch_tuple[1]\n", " if text_label == 'ORG':\n", " text_label = ''\n", " if text_label != '':\n", " temp_text = text+'['+text_label+']'\n", " else:\n", " temp_text = text\n", " redacted_text += temp_text\n", "\n", " return redacted_text" ] }, { "cell_type": "code", "execution_count": 52, "id": "a88522ae-276b-47bd-8df4-6c31164a523f", "metadata": {}, "outputs": [], "source": [ "def get_key(i):\n", " return i[0]" ] }, { "cell_type": "code", "execution_count": 53, "id": "3d4c6654-b970-41ee-90b7-61188191d312", "metadata": {}, "outputs": [], "source": [ "def run_model_on_excel(input_para):\n", " output_entities = evaluate_json_paragraph(input_para)\n", " paragraph = input_para\n", " modified_text = paragraph\n", " offset = 0\n", " text_patches = []\n", " \n", " output_entities = sorted(output_entities, key=lambda x:x[2])\n", " print('~~~~~~~', output_entities)\n", " if len(output_entities) > 0:\n", " for i in output_entities:\n", " if i[0].strip() == '':\n", " continue\n", " tag = i[1]\n", " start = i[2]\n", " end = i[3]\n", " text_patch = (modified_text[offset: start], tag)\n", " text_patches.append(text_patch)\n", " offset = end\n", " \n", " text_patches.append((modified_text[offset:], ''))\n", " print(text_patches)\n", " redacted_text = get_redacted_text(text_patches)\n", " return redacted_text" ] }, { "cell_type": "code", "execution_count": 69, "id": "85ed8bd3-9088-4a05-b9bd-cdaa101dc804", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "\n", "\n", "----- [('14th Aug', 'DATE', 748, 756, 124, 127)]\n", "~~~~~~~ [('14th Aug', 'DATE', 748, 756, 124, 127)]\n", "[(\"ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. may lead to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLOU064C12302 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently. As per IQVIA assessment completed on \", 'DATE'), (' , total 12 studies impacted out of 28 ongoing studies. Impacted studies to initiate risk assessment at study level. Studies that are not impacted as per IQVIA assessment can close the risk in FIRST .', '')]\n" ] } ], "source": [ "# temp_ = '''Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\n", "# running Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\n", "# on the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\n", "# below). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\n", "# longer supported by the manufacturer. resulting in Blood volumes may change for some trials\n", "# '''\n", "# temp_ = '''ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. may lead to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLOU064C12302 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently. As per IQVIA assessment completed on 14th Aug , total 12 studies impacted out of 28 ongoing studies. Impacted studies to initiate risk assessment at study level. Studies that are not impacted as per IQVIA assessment can close the risk in FIRST .'''\n", "\n", "output = run_model_on_excel(temp_)\n", "# print(output)" ] }, { "cell_type": "code", "execution_count": 66, "id": "8604f9da-a6a9-4b08-a083-2177deb3c887", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'Scientific update - Sci Update [SUBJID] [MH], [MH], [MH] Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning [MH], [MH], [MH] screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials\\n'" ] }, "execution_count": 66, "metadata": {}, "output_type": "execute_result" } ], "source": [ "output" ] }, { "cell_type": "code", "execution_count": 70, "id": "690a4796-31f4-4fb1-a181-7cc3bbcde2c5", "metadata": { "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Slowness of RAVE/Rave not working may lead to causing issues in reviewing the data, sending reports, performing SDV, applying PI signs resulting in which is imapcting the overall activities for DBL and may hamper the timelines', '')]\n", "\n", "\n", "\n", "----- [('Day 1', 'PRECLINICAL STUDY DAY', 90, 95, 16, 18), ('2001003', 'SUBJID', 79, 86, 14, 15)]\n", "~~~~~~~ [('2001003', 'SUBJID', 79, 86, 14, 15), ('Day 1', 'PRECLINICAL STUDY DAY', 90, 95, 16, 18)]\n", "[('Data Manager from Aixial missed to update the correct unit for SAA for patient ', 'SUBJID'), (' at ', 'PRECLINICAL STUDY DAY'), (' (one timepoint) led to this timepoint to be registered with the units mg/dL instead of mg/L resulting in a 10-fold higher value in the respective data outputs and CSR.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Clario assigned a Patient data for Blinded Expediated review by accident .This was a one-time misinterpretation of a ticket between PM team and the Read Operations Team as Clario Read Operations teams combines one ticket to multiple patients which created confusion with the PM. may lead to A patient was incorrectly assigned by Clario PM for expedited review even though Site had not requested it. This is the first instance of this kind of mistake. resulting in After readings were complete, the Assistant Clinical Project Manager forwarded the expedited review's findings to the appropriate site. Clario submitted QER report(QER ID: 2023-QER-0262). \\nTo prevent the occurrence of the event, internal discussion between the Read Operations team and the Project Management team within Clario is to be held, with a potential outcome of one SalesForce ticket per patient.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Clario QC Technician failed to identify submitted anatomy (pelvis) in an image submission. led to inaccurate reconciliation in the EDRR resulting in erroneous queries and duplicate image submissions and reads for the same images.', '')]\n", "\n", "\n", "\n", "----- [('2002', 'DATE', 39, 43, 5, 6), ('2002', 'DATE', 160, 164, 23, 25), ('2002', 'DATE', 282, 286, 41, 43), ('2002', 'DATE', 356, 360, 54, 56), ('2002', 'DATE', 450, 454, 71, 72), ('16 Jan 2023', 'DATE', 330, 341, 50, 53)]\n", "~~~~~~~ [('2002', 'DATE', 39, 43, 5, 6), ('2002', 'DATE', 160, 164, 23, 25), ('2002', 'DATE', 282, 286, 41, 43), ('16 Jan 2023', 'DATE', 330, 341, 50, 53), ('2002', 'DATE', 356, 360, 54, 56), ('2002', 'DATE', 450, 454, 71, 72)]\n", "[('BTR released mistake notification that ', 'DATE'), (' sent screening X-ray to BTR. BTR recalled the mistake notification. led to BTR released mistake notification that ', 'DATE'), (' sent screening X-ray to BTR. BTR cancelled the mistake notification. Followingly NIRT released IMP shipment even if ', 'DATE'), (' has been deactivated for any screenings on ', 'DATE'), ('. resulting in ', 'DATE'), (' is ready to be closed. IMP has already been destroyed. New shipment has been released to ', 'DATE'), ('.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Late MRI eligibility result delivery may lead to screen failure resulting in delayed recruitment and site demotivation', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Due to low sample volume, 3 placebo subjects planned for T cell recall assay optimization/adaptation phase, need also to be used for clinical testing. may lead to This leads to unblinding of the principal scientist at the vendor cell carta. resulting in This leads to unblinding of the principal scientist at the vendor cell carta.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Low amount of PPV to perform T cell recall assay adaptation and sample analysis may lead to not able to perform all tests resulting in incomplete data set', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Failure to send over EDRR's in timely fashion after Initial IRT data transfer. led to delays in correcting errors and updating IRT database. resulting in delays of the Final IRT data transfer.\", '')]\n", "\n", "\n", "\n", "----- [('#13', 'SUBJID', 449, 452, 75, 76), ('#13', 'SUBJID', 883, 886, 143, 144), ('#13', 'SUBJID', 988, 991, 162, 164), ('#13', 'SUBJID', 1080, 1083, 178, 180)]\n", "~~~~~~~ [('#13', 'SUBJID', 449, 452, 75, 76), ('#13', 'SUBJID', 883, 886, 143, 144), ('#13', 'SUBJID', 988, 991, 162, 164), ('#13', 'SUBJID', 1080, 1083, 178, 180)]\n", "[(\"In the study's Cohort 1, 12 patients should be dosed at a 8:4 ratio. Since one patient (patient X) refused to continue with study assessments after T1, the CTT decided to replace this patient with a replacement patient (#13) in Cohort 1, in order to make sure there are sufficient patient #/proper assignment allocated for Cohort 1 data analysis. This replacement could only be performed by IQVIA IRT team to ensure the replacement patient (patient \", 'SUBJID'), (') got the same treatment assignment as patient X. IRT URS indicates that the trigger for this “replacement” is patient discontinuation. However patient X was not discontinued in IRT system as patient was being tracked for safety, and the IRT system was not set-up to capture discontinuation for a patient. Therefore the patient X discontinuation was not triggered via IRT system and change for replacement assignment for patient ', 'SUBJID'), (' was not performed as requested. led to the patient replacement in IRT did not occur and the patient ', 'SUBJID'), (' was dosed without being entered as a replacement for patient X resulting in the patient ', 'SUBJID'), (' in cohort 1 may not have been assigned to the correct treatment, which could imply there will be incorrect #patients assigned to treatments for cohort 1', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the supplier Cenduit, providing IRT services is currently not ready to support the bATR risk assessment procedures. may lead to Although procedures and controls are in place at the supplier to mitigate the risk of data integrity issues when processing data change requests, the delay in implementing bATR risk assessment procedures may prevent monitoring and controls from Novartis on IRT data entries and data changes. resulting in This will result in all Trials supported by Cenduit being unable to apply bATR risk assessment procedures until the supplier is ready.', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the supplier Cenduit, providing IRT services is currently not ready to support the bATR risk assessment procedures. may lead to Although procedures and controls are in place at the supplier to mitigate the risk of data integrity issues when processing data change requests, the delay in implementing bATR risk assessment procedures may prevent monitoring and controls from Novartis on IRT data entries and data changes. resulting in This will result in all Trials supported by Cenduit being unable to apply bATR risk assessment procedures until the supplier is ready.', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the supplier Cenduit, providing IRT services is currently not ready to support the bATR risk assessment procedures. may lead to Although procedures and controls are in place at the supplier to mitigate the risk of data integrity issues when processing data change requests, the delay in implementing bATR risk assessment procedures may prevent monitoring and controls from Novartis on IRT data entries and data changes. resulting in This will result in all Trials supported by Cenduit being unable to apply bATR risk assessment procedures until the supplier is ready.', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the supplier Cenduit, providing IRT services is currently not ready to support the bATR risk assessment procedures. may lead to Although procedures and controls are in place at the supplier to mitigate the risk of data integrity issues when processing data change requests, the delay in implementing bATR risk assessment procedures may prevent monitoring and controls from Novartis on IRT data entries and data changes. resulting in This will result in all Trials supported by Cenduit being unable to apply bATR risk assessment procedures until the supplier is ready.', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the supplier Cenduit, providing IRT services is currently not ready to support the bATR risk assessment procedures. may lead to Although procedures and controls are in place at the supplier to mitigate the risk of data integrity issues when processing data change requests, the delay in implementing bATR risk assessment procedures may prevent monitoring and controls from Novartis on IRT data entries and data changes. resulting in This will result in all Trials supported by Cenduit being unable to apply bATR risk assessment procedures until the supplier is ready.', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 119, 130, 17, 18)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the supplier Cenduit, providing IRT services is currently not ready to support the bATR risk assessment procedures. may lead to Although procedures and controls are in place at the supplier to mitigate the risk of data integrity issues when processing data change requests, the delay in implementing bATR risk assessment procedures may prevent monitoring and controls from Novartis on IRT data entries and data changes. resulting in This will result in all Trials supported by Cenduit being unable to apply bATR risk assessment procedures until the supplier is ready.', '')]\n", "\n", "\n", "\n", "----- [('South Korea', 'COUNTRY', 11, 22, 2, 4), ('Day 1', 'PRECLINICAL STUDY DAY', 44, 49, 8, 10)]\n", "~~~~~~~ [('South Korea', 'COUNTRY', 11, 22, 2, 4), ('Day 1', 'PRECLINICAL STUDY DAY', 44, 49, 8, 10)]\n", "[('Subject in ', 'COUNTRY'), (' is reaching Cycle 37 ', 'PRECLINICAL STUDY DAY'), (' soon. Subject is currently on Cycle 32. IRT system needs to be updated to add additional cycles in order to be ready for the subject who will reach Cycle 37 soon as at the moment the system has up to Cycle 36. Although the preparations for the URS update is initiated, this update was shown in the past to be complicated and lengthy which may lead to the event that the URS/UAT is delayed and the system is not ready to add the additional cycles after Cycle 36 which would be resulting in no possibility to dispense IMP after Cycles 36. The situation was discussed with GCS and a mitigation plan was added.', '')]\n", "\n", "\n", "\n", "----- [('2044', 'Patient Related SiteID', 32, 36, 5, 6)]\n", "~~~~~~~ [('2044', 'Patient Related SiteID', 32, 36, 5, 6)]\n", "[('Unblinded SN accidently at site ', 'Patient Related SiteID'), (', miss-understood the process led to Treatment information was provided to site team via IRT system resulting in Potential exclusion from extension if PI opened the unblinding email', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Lack of automated process for handling withdrawals of consent (WoC) led to delayed notification of 3rd party external service providers to exclude biological samples from analysis resulting in biological sample analysis (data generation) without valid consent', '')]\n", "\n", "\n", "\n", "----- [('end of Nov 2023', 'DATE', 265, 280, 39, 43), ('60 days', 'PRECLINICAL STUDY DAY', 327, 334, 50, 53)]\n", "~~~~~~~ [('end of Nov 2023', 'DATE', 265, 280, 39, 43), ('60 days', 'PRECLINICAL STUDY DAY', 327, 334, 50, 53)]\n", "[('Reagent backlog for anti-TGA testing at Viracor may lead to Vendor unable to test samples for anti-thyroglobulin Ab and will continue testing for CU-index and anti-tyroperoxidase Ab. Samples will be held in storage until reagent availability resumes, estimated by ', 'DATE'), ('. resulting in Samples stability is limited to ', 'PRECLINICAL STUDY DAY'), (' for this analyte. Samples may be tested out of stability.', '')]\n", "\n", "\n", "\n", "----- [('September 2023', 'DATE', 54, 68, 9, 11)]\n", "~~~~~~~ [('September 2023', 'DATE', 54, 68, 9, 11)]\n", "[('Contract end date amendment needed to be initiated in ', 'DATE'), (' with Mindmaze (formally GaitUP) - Exhibit A was added from NIBR quality review in this amendment although not relevant as no change in scope led to Many confusions as this implied additional services not originally agreed nor supported my MindMaze resulting in Delay in contract amendment finalization - active work conducted under an expired agreement', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Vendor error in conversion of travel costs in China to be reimbursed led to Novartis Steer team identified unexpectedly high cost/patient for China. In reviewing these amounts team asked vendor to review conversion factor to determine if yuan/USD conversion contained errors Greenphire confirmed correct conversion factor was not used resulting in Novartis charged more than anticipated per patient, resulting in greater than anticipated invoices and costs', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"-ICON lacks internal regulatory support and need improvement in the management of their subcontractors\\n-Novartis' bidding document did not specifically instruct the central lab to make the change to CE mark for EU sites when tube requirements are added to the bid based on assay validation, however all other central laboratories have confirmed that they do make the change. led to During the assessment of the initial issue, Novartis discovered that ICON central laboratories were not ensuring that tube requested in the Novartis bidding document were adjusted to comply with regulatory requirements regarding CE Marking of medical devices (blood collection tubes) in the EU. The tubes in scope of this issue are sterile tubes from known manufacturers resulting in Regulatory noncompliance\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"-ICON lacks internal regulatory support and need improvement in the management of their subcontractors\\n-Novartis' bidding document did not specifically instruct the central lab to make the change to CE mark for EU sites when tube requirements are added to the bid based on assay validation, however all other central laboratories have confirmed that they do make the change. led to During the assessment of the initial issue, Novartis discovered that ICON central laboratories were not ensuring that tube requested in the Novartis bidding document were adjusted to comply with regulatory requirements regarding CE Marking of medical devices (blood collection tubes) in the EU. The tubes in scope of this issue are sterile tubes from known manufacturers resulting in Regulatory noncompliance\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Central lab aliquoted six tubes of sample led to not having the infrastructure to maintain sixth tube resulting in destruction of samples that would otherwise have been used for biomarker analysis', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"ICON kits included non-sterile tubes that were required to be destroyed that required all kits to be destroyed led to sites confirming that kits were destroyed when they weren't resulting in the site using kits that they reported as destroyed\", '')]\n", "\n", "\n", "\n", "----- [('Feb 2023', 'DATE', 104, 112, 17, 19), ('May 2023', 'DATE', 179, 187, 29, 31), ('5002', 'SUBJID', 334, 338, 56, 57), ('5002', 'SUBJID', 514, 518, 88, 89), ('05-May-2023', 'DATE', 522, 533, 90, 91), ('14-Jun-2023', 'DATE', 577, 588, 99, 100), ('14-Jun-2023', 'DATE', 876, 887, 148, 150), ('04-May-2023', 'DATE', 821, 832, 139, 141), ('5004', 'Patient Related SiteID', 618, 622, 105, 106), ('5004', 'Patient Related SiteID', 813, 817, 137, 139), ('5004', 'Patient Related SiteID', 936, 940, 159, 161)]\n", "~~~~~~~ [('Feb 2023', 'DATE', 104, 112, 17, 19), ('May 2023', 'DATE', 179, 187, 29, 31), ('5002', 'SUBJID', 334, 338, 56, 57), ('5002', 'SUBJID', 514, 518, 88, 89), ('05-May-2023', 'DATE', 522, 533, 90, 91), ('14-Jun-2023', 'DATE', 577, 588, 99, 100), ('5004', 'Patient Related SiteID', 618, 622, 105, 106), ('5004', 'Patient Related SiteID', 813, 817, 137, 139), ('04-May-2023', 'DATE', 821, 832, 139, 141), ('14-Jun-2023', 'DATE', 876, 887, 148, 150), ('5004', 'Patient Related SiteID', 936, 940, 159, 161)]\n", "[('Novartis and ICON communicated changes of blood collection tube from 8mL Li-Heparin to 6 K2EDTA tube in ', 'DATE'), (', requested the old kits destruction and provided replaced kits in ', 'DATE'), ('. These were done before FPFV of the study. However, both sites still used incorrect 8 mL Heparin tube for 36 samples from 3 participants: \\n•\\tFor ', 'SUBJID'), (', the two kits with issues were not included in the ICON destruction memo, this may be an error from ICON missing these kits in memo. However, ICON shipped replacement kits to ', 'SUBJID'), (' on ', 'DATE'), (' and it should have arrived before the FPFV ', 'DATE'), (' (1 months 9 days). \\n\\n•\\tFor ', 'Patient Related SiteID'), (', all kits with issues were included in the ICON destruction memo. Therefore, this was the site missing action that did not destruct old kits following memo. ICON shipped replacement kits to ', 'Patient Related SiteID'), (' on ', 'DATE'), (' and it have should arrived before the FPFV ', 'DATE'), (' (1 months 10 days). led to two sites (5002 and ', 'Patient Related SiteID'), (') have used old kit, 8 mL Heparin tube instead of new kit, 6 mL K2EDTA tube PK collection tube. resulting in Samples cannot be measured if cross validation in human plasma (heparin) fails.\\n\\nThe stability must be established retrospectively in human plasma (heparin) otherwise concentration data cannot be released.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Scientific update - Sci Update 510 Hepatitis B Virus, DNA and Hepatitis C Virus, RNA Change in Test Platform may lead to Effective immediately, ICON Laboratory Services (ILS) will start to transition studies currently\\nrunning Hepatitis B Virus, DNA and Hepatitis C Virus, RNA screen with confirmatory testing\\non the Roche Cobas AmpliPrep/TaqMan to the Hologic Panther platform (see Tables 1 and 2\\nbelow). The change in test platform is required as the Roche Cobas AmpliPrep/TaqMan is no\\nlonger supported by the manufacturer. resulting in Blood volumes may change for some trials', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('PDA and cortisol samples of groups B4a and B4b were not shipped on time by the clinic to the BA lab, and annotation file was not uploaded led to those samples could not be reconciled and analyzed on time resulting in delay in final data delivery', '')]\n", "\n", "\n", "\n", "----- [('delayed Cycle 1 dosimetry results delivery', 'DATE', 52, 94, 11, 18)]\n", "~~~~~~~ [('delayed Cycle 1 dosimetry results delivery', 'DATE', 52, 94, 11, 18)]\n", "[('A late query raised by the imaging vendor led to a ', 'DATE'), (' resulting in the report to the local IRB of an unanticipated problem resulting in risk to patient and worsening relationship with the site.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Invicro does not have clear communications to sites on the Phantom site qualification process may lead to sites not understanding qualification process for SPECT/CT and PET scanner qualifications resulting in delay in SIVs', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. led to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLOU064C12302 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. led to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLOU064C12302 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently.\", '')]\n", "\n", "\n", "\n", "----- [('14th Aug', 'DATE', 742, 750, 123, 126)]\n", "~~~~~~~ [('14th Aug', 'DATE', 742, 750, 123, 126)]\n", "[(\"ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. led to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLXH254X2102 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently. As per IQVIA assessment completed on \", 'DATE'), (' , total 12 studies impacted out of 28 ongoing studies. Impacted studies to initiate risk assessment at study level. Studies that are not impacted as per IQVIA assessment can close the risk in FIRST .', '')]\n", "\n", "\n", "\n", "----- [('14th Aug', 'DATE', 748, 756, 124, 127)]\n", "~~~~~~~ [('14th Aug', 'DATE', 748, 756, 124, 127)]\n", "[(\"ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. may lead to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLOU064C12302 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently. As per IQVIA assessment completed on \", 'DATE'), (' , total 12 studies impacted out of 28 ongoing studies. Impacted studies to initiate risk assessment at study level. Studies that are not impacted as per IQVIA assessment can close the risk in FIRST .', '')]\n", "\n", "\n", "\n", "----- [('14th Aug', 'DATE', 748, 756, 124, 127)]\n", "~~~~~~~ [('14th Aug', 'DATE', 748, 756, 124, 127)]\n", "[(\"ECG is usually transmitted via device to vendor database through LAN, however in exceptional situation when device transmission fails site has an option to upload ECG tracing into vendor portal for overread. Hence ECG upload into portal is alternative fallback option and only site should have access to it. However, in certain cases it was observed that CRA had performed site activity to upload ECG and CRA's were given access to upload ECG as default by vendor. may lead to led to potential GCP deviation as multiple ECG upload by CRA into vendor portal was observed in CLOU064C12302 study resulting in risk of potential ECG upload by CRA in 28 ongoing ECG studies with IQVIA under investigation currently. As per IQVIA assessment completed on \", 'DATE'), (' , total 12 studies impacted out of 28 ongoing studies. Impacted studies to initiate risk assessment at study level. Studies that are not impacted as per IQVIA assessment can close the risk in FIRST .', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Priority 2 drug dispensation not working in IRT system due to programming error led to sites with insufficient stock for Priority 1 dispensation being unable to get kits automatically assigned from the system resulting in two patients got IMP manually dispensed.', '')]\n", "\n", "\n", "\n", "----- [('07-Mar-2022', 'DATE', 126, 137, 18, 19)]\n", "~~~~~~~ [('07-Mar-2022', 'DATE', 126, 137, 18, 19)]\n", "[('SOP-8114563 Risk-Based Business Audit Trail Review (bATR) of Electronic Patient Data in Clinical Trials was made effective on ', 'DATE'), ('., however the controls available at supplier, providing Data Management services is currently not assessed for the bATR requirements may lead to the delay in implementing bATR risk assessment from Novartis on Data Management activities outsourced to the supplier resulting in all trials supported by supplier being unable to apply bATR risk assessment until the supplier is able to support the Novartis requirements.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delays in setting up the ECG vendor, IQVIA, may lead to missing electronic collection of ECG data starting from FSFT resulting in increased labor burdens and costs on sites or CTTs, including loading of paper data', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Change in data collection strategy for blinded ESR data from CRF to Third party vendor. may lead to Data inconsistencies for Date, Visit name and ESR value between Rave CRF and Kayentis Web portal. To avoid such inconsistencies data reconciliation by independent -DM and unblinded CSE will be preformed. resulting in Data analysis.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Complexity of ESR Kayentis web portal (replacing existing ESR eCRF page) may lead to unblinded personnel not speaking English well entering ESR data incorrectly into the portal before the local language version becomes available resulting in errors in ESR data (secondary endpoint)', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Implementation of Protocol Amendment may lead to significant costs, additional resourcing needed for patient management, retraining of eCOA data capture resulting in impact to timelines of startup, rebuild of device that is already live and redo of translation work on PROs and patient facing screens.', '')]\n", "\n", "\n", "\n", "----- [('China', 'COUNTRY', 61, 66, 10, 11), ('China', 'COUNTRY', 194, 199, 32, 34)]\n", "~~~~~~~ [('China', 'COUNTRY', 61, 66, 10, 11), ('China', 'COUNTRY', 194, 199, 32, 34)]\n", "[('Kayentis ePRO device, mobile phone, not qualified for use in ', 'COUNTRY'), (' country led to long time spent on the qualification resulting in ePRO devices not available for scheduled RIS date Oct 2023. ', 'COUNTRY'), (' Country at risk to start', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Per the initial Kayentis vendor client specifications, the eDiary will be opened for a maximum of 109 times (106 daily sessions + 3 additional ones as buffer) post activation at Screening 2 (to cover Day -20 to Day 85) which may lead to the eDiary running out of daily sessions in case there was an unexpected deviation from the proposed visit schedule as per protocol resulting in patients not being able to record all required daily sessions until reaching Week 12. However, to not collect such remaining eDiary data has no impact on the primary/secondary endpoints.', '')]\n", "\n", "\n", "\n", "----- [('Hungarian', 'COUNTRY', 330, 339, 49, 51), ('Russian', 'COUNTRY', 344, 351, 50, 52), ('Slovak', 'COUNTRY', 393, 399, 54, 55)]\n", "~~~~~~~ [('Hungarian', 'COUNTRY', 330, 339, 49, 51), ('Russian', 'COUNTRY', 344, 351, 50, 52), ('Slovak', 'COUNTRY', 393, 399, 54, 55)]\n", "[('Protocol Amendment addition of FACT-RNT may lead to delays in translation work for languages not yet developed resulting in startup delays of achieving FPFV for specific countries, rebuild of device that is live in PROD, resubmission to countries with further patient facing screens and retesting/UAT #2. \\nLanguages at risk: \\n1.\\t', 'COUNTRY'), (' \\n2.\\t', 'COUNTRY'), (' \\n3.\\tJapanese \\n4.\\tChinese Traditional \\n5.\\t', 'COUNTRY'), ('', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Import of record details are required also for starting the Import License application for the Lab Kits. Labcorp did not request this details during the startup phase so this information were unavailable to the team. may lead to There have been delays to acquire the IOR contacts details from the site and to share it with the Central lab to initiate IL activities. resulting in Delay in obtaining IL for lab kits, which could lead to delay of FPFV', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Low stock of tubes due to high demand, unavailability or tube shortages. may lead to Tube replacement during study due to low stock - mitigation plan available. resulting in May impact blood or tissue collections at sites, increase site's frustration or may cause a potential loss of patient to the trial.\", '')]\n", "\n", "\n", "\n", "----- [('countries where import licenses are required', 'COUNTRY', 286, 330, 48, 54)]\n", "~~~~~~~ [('countries where import licenses are required', 'COUNTRY', 286, 330, 48, 54)]\n", "[('Low stock of tubs. due to high demand, unavailability or tub shortages. may lead to Tube replacement during study due to low stock - mitigation plans available: 1) Replace the tub by another one with similar characteristics. This strategy requires to apply for a new import license in ', 'COUNTRY'), (\". 2) Allow the shipment of tubs with short expiration dates. This option allows to skip the requirement of new import license. resulting in May impact blood or tissue collections at sites, may lead increase site's frustration or may cause a potential loss of patient to the trial.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Data entry errors of Mayo PPV-23 results in the LabCorp database were discovered as a result of LabCorp's QE-00479007 and related CAPA (see I021), led to updated results received after DBL, resulting in need to re-run the PPV-23 outputs\", '')]\n", "\n", "\n", "\n", "----- [('current SOW 5 update', 'DATE', 207, 227, 32, 37), ('malignancy', 'MH', 298, 308, 49, 51), ('Malignancy', 'MH', 476, 486, 78, 80), ('malignancy', 'MH', 580, 590, 94, 96), ('malignancy', 'MH', 634, 644, 102, 103), ('malignancy', 'MH', 744, 754, 119, 121)]\n", "~~~~~~~ [('current SOW 5 update', 'DATE', 207, 227, 32, 37), ('malignancy', 'MH', 298, 308, 49, 51), ('Malignancy', 'MH', 476, 486, 78, 80), ('malignancy', 'MH', 580, 590, 94, 96), ('malignancy', 'MH', 634, 644, 102, 103), ('malignancy', 'MH', 744, 754, 119, 121)]\n", "[('Due to Labcorp issues During database loading for SOW A5 and the recent identification of errors for the calculation of multiple flagging (absolute lymphocyte count , absolute neutrophile count) led to the ', 'DATE'), (' is dleayed and another SOW (6) update will be needed to implement the ', 'MH'), (' Kit, updated Leukapheresis kit and solve the Flag EX issues. The SOW 6 release that will include the corrections, will take additional 6 weeks. resulting in Delay for ', 'MH'), (' kit implementation. Possibility to lose the PBMC samples that should be collected in case of ', 'MH'), (' (probability low due to low probability of ', 'MH'), ('). Sites can t rely on absolute lymphocyte count , absolute neutrophile count flagging. In case of ', 'MH'), (', Sites will need to use bulk supplies for collection qPCR and Tissue blocks samples .', '')]\n", "\n", "\n", "\n", "----- [('12-Oct-2023', 'DATE', 53, 64, 9, 10), ('20-Oct-2023', 'DATE', 130, 141, 22, 23)]\n", "~~~~~~~ [('12-Oct-2023', 'DATE', 53, 64, 9, 10), ('20-Oct-2023', 'DATE', 130, 141, 22, 23)]\n", "[('Lab manual release date from LabCorp was planned for ', 'DATE'), ('. Now is delayed due to longer than expected vendor QC process to ', 'DATE'), (' may lead to delay in US site study start-up milestone resulting in delay in FPFV in US (IRB submission/approval at site).', '')]\n", "\n", "\n", "\n", "----- [('12-Oct-2023', 'DATE', 53, 64, 9, 10), ('20-Oct-2023', 'DATE', 84, 95, 14, 15)]\n", "~~~~~~~ [('12-Oct-2023', 'DATE', 53, 64, 9, 10), ('20-Oct-2023', 'DATE', 84, 95, 14, 15)]\n", "[('Lab manual release date from LabCorp was planned for ', 'DATE'), ('. Now is delayed to ', 'DATE'), ('. Lab manual is required by Novartis Data Management to draft and finalize DTS which has led to not finalizing DTS on time. As per the DTS timelines, Novartis was supposed to complete vendor review by this date. LabCorp DTS is Priority1 DTS resulting in patient enrollment delay and it is dependent on this lab data, further delay is a huge risk.', '')]\n", "\n", "\n", "\n", "----- [('Ivory Coast', 'COUNTRY', 475, 486, 69, 72), ('Rwanda', 'COUNTRY', 491, 497, 72, 73)]\n", "~~~~~~~ [('Ivory Coast', 'COUNTRY', 475, 486, 69, 72), ('Rwanda', 'COUNTRY', 491, 497, 72, 73)]\n", "[('LabCorp states that there is a delay in receiving the information from the manufacturer, however, further investigation suggests an inadequate process for retrieving the information in a timely manner as required for submission. may lead to Labcorp delay in providing Certificate of Analysis (CoA) and Certificate of Compliance (CoC) that are required for submission of import license for Sub-Saharan countries. resulting in Potential delay in import license submission for ', 'COUNTRY'), (' and ', 'COUNTRY'), ('.', '')]\n", "\n", "\n", "\n", "----- [('December 11th, 2023', 'DATE', 362, 381, 53, 57)]\n", "~~~~~~~ [('December 11th, 2023', 'DATE', 362, 381, 53, 57)]\n", "[('Manufacturer change to Mesoscale Discovery Proinflammatory Panel-1. may lead to Labcorp are updating existing test codes related to LLOQs and reference range intervals for Mesoscale Discovery Proinflammatory Panel. For ongoing studies, changes will not require the use of new test codes, however, Reference Range Intervals and LLOQ will be updated on or after ', 'DATE'), ('. resulting in May lead to issues due to changes to existing test codes, required for LLOQs and reference range intervals.', '')]\n", "\n", "\n", "\n", "----- [('December 11th, 2023', 'DATE', 362, 381, 53, 57)]\n", "~~~~~~~ [('December 11th, 2023', 'DATE', 362, 381, 53, 57)]\n", "[('Manufacturer change to Mesoscale Discovery Proinflammatory Panel-1. may lead to Labcorp are updating existing test codes related to LLOQs and reference range intervals for Mesoscale Discovery Proinflammatory Panel. For ongoing studies, changes will not require the use of new test codes, however, Reference Range Intervals and LLOQ will be updated on or after ', 'DATE'), ('. resulting in May lead to issues due to changes to existing test codes, required for LLOQs and reference range intervals.', '')]\n", "\n", "\n", "\n", "----- [('December 11th, 2023', 'DATE', 362, 381, 53, 57)]\n", "~~~~~~~ [('December 11th, 2023', 'DATE', 362, 381, 53, 57)]\n", "[('Manufacturer change to Mesoscale Discovery Proinflammatory Panel-1. may lead to Labcorp are updating existing test codes related to LLOQs and reference range intervals for Mesoscale Discovery Proinflammatory Panel. For ongoing studies, changes will not require the use of new test codes, however, Reference Range Intervals and LLOQ will be updated on or after ', 'DATE'), ('. resulting in May lead to issues due to changes to existing test codes, required for LLOQs and reference range intervals.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('A misunderstanding over whether the Marken contract for IMP supply direct to patient (in case of pandemic/healthcare disruption) had been finalized led to KOM being held prior to contract finalization resulting in a discussion with Marken occurring without a contract being in place', '')]\n", "\n", "\n", "\n", "----- [('10OCT2023', 'DATE', 6, 15, 2, 3)]\n", "~~~~~~~ [('10OCT2023', 'DATE', 6, 15, 2, 3)]\n", "[('As of ', 'DATE'), (', there is still a significant number of open DCF (72) in MedAvante, and in addition 30 VDMRF have been raised (requiring site /PM/ WCG DM/ NVS DM signatures before the update can be performed in Virgil) may lead to delay in the planned date of resolution of all DCF (planned on 18OCT2023) resulting in potential delay of the study DBL (planned on27OCT2023)', '')]\n", "\n", "\n", "\n", "----- [('11DEC23', 'DATE', 114, 121, 18, 19)]\n", "~~~~~~~ [('11DEC23', 'DATE', 114, 121, 18, 19)]\n", "[('Biological samples (PCSK9, ADA) not analyzed and results not available on time for the last transfer scheduled on ', 'DATE'), (' for the iDBL may lead to database not locked as per expected timelines resulting in FDA submission impacted', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay of the delivery of LCZ696 suspension materials to the sites may lead to delay of administration of LCZ696 suspension to the specific patients resulting in negative influence on patient safety', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Late supply shipment leading to delayed SIVs may lead to delayed SIVs resulting in delayed recruitment', '')]\n", "\n", "\n", "\n", "----- [('August 2023', 'DATE', 275, 286, 49, 51)]\n", "~~~~~~~ [('August 2023', 'DATE', 275, 286, 49, 51)]\n", "[('Concerning biomarker/ Navigate data, significant number of issues remaining opened in EDRR prior to iDBL, some of them open for many weeks may lead to high number of open queries (9 for Flow data) due to the belated resolvance of open issues provided in EDRR as far back as ', 'DATE'), (', resulting in which would lead to a not completely reconciled data included in database lock, and not reconciled data subsequently excluded from the primary analysis.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CRO failed to transfer eTMF to NVS within the agreed timelines following CSR finalization led to delay in receipt of quality document post study completion resulting in delay in overall study deliverables. Study leads are advised to ensure eTMF transfer from the Third Party to Novartis are performed in accordance with agreed timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect sample identifications (IDs) used on the CRO PK sample labels and Clin-base System resulting in a mismatch vs. the correct NVS sample identification IDs located in the Pre-Advance Sample Log Table (SLT). led to Delay in the PK annotation file reconciliation which delayed sample analysis. resulting in A delay in timelines resulted. The discrepancy was identified during sample reconciliation in preparation for sample analysis. The annotation file was updated to include the correct sample numbers by the Parexel Clinical Laboratory Coordinator (CLC). The sample IDs in Parexel Clin-Base were also updated to correctly match the NVS sample IDs within the Pre-Advance SLT. This enabled the pre-advance reconciliation to initiate sample analysis and to support the pharmacokinetic (PK) and clinical data merge post-analysis.', '')]\n", "\n", "\n", "\n", "----- [('01May2023', 'DATE', 233, 242, 31, 32)]\n", "~~~~~~~ [('01May2023', 'DATE', 233, 242, 31, 32)]\n", "[('The project manager from the vendor Parexel contracted for ancillary supplies has been significantly delayed in providing meeting minutes, updated timelines, availability of supplies and documentation since initiating the service on ', 'DATE'), ('. may lead to Delay in provision of documents and supplies. resulting in This could cause a delay to Site Initiation Visits and overall study operation.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Missed internal processes at Calyx led to led to incorrect queries being opened to sites after study end resulting in resulting in questions from sites and deletion of the queries', '')]\n", "\n", "\n", "\n", "----- [('01Dec2023', 'DATE', 89, 98, 14, 15), ('12 weeks', 'DATE', 167, 175, 26, 28), ('1Dec2023', 'DATE', 559, 567, 90, 91)]\n", "~~~~~~~ [('01Dec2023', 'DATE', 89, 98, 14, 15), ('12 weeks', 'DATE', 167, 175, 26, 28), ('1Dec2023', 'DATE', 559, 567, 90, 91)]\n", "[('Annual Permit application for South African sites required, permits are Due to expire by ', 'DATE'), ('. \\nSigned re-applications are required from sites. It can take up to ', 'DATE'), (\" from submission of application to approval. Re-applications have been sent to Country Leads to send to sites and follow up mid-September but no responses yet received may lead to delay in submitting the application and it not being approved before the expiry date resulting in sites not being able to export samples if the permit expires and application hasn't been approved before \", 'DATE'), ('.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delays in revisions to original Q2 task order to match changes in PA1 and PA2 led to vast discrepancy in spending in certain vendor budget categories resulting in vendor PO being overspent or requiring large increase at time when trial is working towards closeout', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Due to sites using the wrong lab kits to collect PK/IG samples, containers were logged incorrectly when received at Q2. As a result of inadequate sample tracking and regular PK reconciliation, this led to primary PK samples being shipped from Q2 to the analysis lab before the issues could be resolved resulting in irresolvable PK reconciliation issues in preadvance.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('During finalization of the BDR, it was noted in AFs of PP there were 63 samples with a comment of Unexpected Item that require further investigation. The comment section do not provide further clarification to the integrity of the samples, even if they were sent for analysis to the third party laboratory. may lead to There are analysed samples with comments in PP that require further clarification to avoid any sample integrity issues: 63 samples: Unexpected item; 1 sample: Sample has been through a freeze-thaw; 4 samples: ER00559232 resulting in If there are doubts on the integrity of these samples, the BDR could not be released and require revision.\\n\\nThis would have an impact on the analysis of the CSR and could possibly result in an amendment.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('During finalization of the BDR, it was noted in PreAdvance there were 63 samples with a comment of Unexpected Item that require further investigation. The comment section do not provide further clarification to the integrity of the samples, even if they were sent for analysis to the third party laboratory. led to There are analysed samples with comments in PP that require further clarification to avoid any sample integrity issues: 63 samples: Unexpected item; 1 sample: Sample has been through a freeze-thaw; 4 samples: ER00559232 resulting in BDRs for this study have not yet been finalized, and cannot be finalized until all samples are deemed viable or discarded from statistical analysis.\\n\\nIf there are doubts on the integrity of these samples, the BDR could not be released, would require revision and the database may need to be unlocked.\\n\\nThis would have an impact on the analysis of the CSR and could possibly result in an amendment.', '')]\n", "\n", "\n", "\n", "----- [('Oct 2023', 'DATE', 3, 11, 1, 3)]\n", "~~~~~~~ [('Oct 2023', 'DATE', 3, 11, 1, 3)]\n", "[('In ', 'DATE'), (', Q2 Solutions will move to another analyzer, and this change requires to update associated codes. Process update not initiated in due time may lead to codes not updated when transition to new analyzer will be effective resulting in samples not being analyzed and results not being provided to site until update is implemented.', '')]\n", "\n", "\n", "\n", "----- [('Oct 2023', 'DATE', 3, 11, 1, 3)]\n", "~~~~~~~ [('Oct 2023', 'DATE', 3, 11, 1, 3)]\n", "[('In ', 'DATE'), (', Q2 Solutions will move to another system to analyze the samples, and this requires to update associated codes. However, update process not initiated in due time may lead to codes not being available at time of transition to the new system resulting in Q2 not being able to analyze the samples, and delays in providing results to sites.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Late supply shipment may lead to delayed SIVs resulting in delayed recruitment', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Overly complex trial kit design with multiple parts, kits per visit, and options for visits to be completed on one day or over two days may lead to sites routinely choosing the wrong kit or incorrectly completing kits resulting in high number of Q2 queries and PreAdvance issues, which may delay final DBL timelines or result in lost samples if not all issues can be reconciled', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Q2 Solutions Site Portal system deficiencies may lead to sites unable to access and respond to queries resulting in site staff to provide their responses via email.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Site inadvertently used incorrect patient lab kit and collected samples for more than that required by the protocol or covered by the ICF led to More data than required by the protocol/covered by ICF were collected and may not be possible to remove from vendor database. resulting in Additional data collected and resulted.', '')]\n", "\n", "\n", "\n", "----- [('12Jan23', 'DATE', 273, 280, 47, 49)]\n", "~~~~~~~ [('12Jan23', 'DATE', 273, 280, 47, 49)]\n", "[('The protocol was not correctly implemented by Q2 in the central lab database which led to wrong PK kits and requisitions forms with incorrect timepoints for the 3 study treatment led to a confusion with the PK samples and costs incorrectly charged in the invoice sent on ', 'DATE'), (' LED 0509987 for the DB modification and creation of new PK kits resulting in the delay of invoice approval as the first invoice was rejected in January but was re-sent in September with no updates and again rejected.', '')]\n", "\n", "\n", "\n", "----- [('10Apr2023', 'SUBJID', 82, 91, 13, 14)]\n", "~~~~~~~ [('10Apr2023', 'SUBJID', 82, 91, 13, 14)]\n", "[('The subject declined collection of this sample within the separate Genetic ICF on ', 'SUBJID'), (' during screening, date of consent was entered in EDC in error this led to site 5012 collecting an optional DNA sample for subject without consent, resulting in sample being sent to central laboratory (Q2), but not being tested.', '')]\n", "\n", "\n", "\n", "----- [('CAIN457Q12301', 'SUBJID', 413, 426, 73, 75)]\n", "~~~~~~~ [('CAIN457Q12301', 'SUBJID', 413, 426, 73, 75)]\n", "[('Study team informed about a vendor audit fail for Quality metrics, whereby data for anther study who also used the SF 36 questionnaire, it was reported that Quality Metrics switched values when reporting the results. may lead to following this find, NVS was informed not to allow direct transfer of data from QM to NVS. Data had to go via a 3rd party. In this instance, Clario are receiving the data first. For ', 'SUBJID'), (\", it is flagged that the review of data should be in-depth, to check for any scoring inconsistences. resulting in delayed final data, due to set up of programming on Clario's side to send and receive data from Quality Metrics. Delayed database lock timelines possible.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Recent change in IVDR compliance for GLDH sample assay from compliant to non-compliant may lead to lack of CE marking for GLDH assay in EU countries resulting in inability to report GLDH from Q2 central lab to site in EU countries, could impact patient safety', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Omission of point assignments in the text for Q8 on the VISA-A PRO source document (created by RWS) led to the US country team inquiring on clarity for Q8 point assignments resulting in review of EDC system and PRO paper binders for accuracy and possible correction.', '')]\n", "\n", "\n", "\n", "----- [('NVS native speakers', 'COUNTRY', 114, 133, 21, 24), ('10-Oct-2023', 'DATE', 195, 206, 35, 36), ('10-Oct-2023', 'DATE', 379, 390, 62, 64), ('Fiona Martin', 'NAME', 308, 320, 50, 52), ('Aadarsh Patel', 'NAME', 361, 374, 59, 61)]\n", "~~~~~~~ [('NVS native speakers', 'COUNTRY', 114, 133, 21, 24), ('10-Oct-2023', 'DATE', 195, 206, 35, 36), ('Fiona Martin', 'NAME', 308, 320, 50, 52), ('Aadarsh Patel', 'NAME', 361, 374, 59, 61), ('10-Oct-2023', 'DATE', 379, 390, 62, 64)]\n", "[('RWS will not complete the In-Country Review Steps. This is a NVS process step where we have to locate the country ', 'COUNTRY'), (' to assist with In-Country Review Steps. may lead to CTT (on ', 'DATE'), (') decided to waive requirement for In-Country Review Steps for Translations and Licensing (T&L). VPM (', 'NAME'), (') to send an email communication to VSM (', 'NAME'), (') on ', 'DATE'), (' to document this CTT decision. resulting in There is a very limited impact for this decision.', '')]\n", "\n", "\n", "\n", "----- [('NVS native speakers', 'COUNTRY', 114, 133, 21, 24), ('10-Oct-2023', 'DATE', 195, 206, 35, 36), ('10-Oct-2023', 'DATE', 379, 390, 62, 64), ('Fiona Martin', 'NAME', 308, 320, 50, 52), ('Aadarsh Patel', 'NAME', 361, 374, 59, 61)]\n", "~~~~~~~ [('NVS native speakers', 'COUNTRY', 114, 133, 21, 24), ('10-Oct-2023', 'DATE', 195, 206, 35, 36), ('Fiona Martin', 'NAME', 308, 320, 50, 52), ('Aadarsh Patel', 'NAME', 361, 374, 59, 61), ('10-Oct-2023', 'DATE', 379, 390, 62, 64)]\n", "[('RWS will not complete the In-Country Review Steps. This is a NVS process step where we have to locate the country ', 'COUNTRY'), (' to assist with In-Country Review Steps. may lead to CTT (on ', 'DATE'), (') decided to waive requirement for In-Country Review Steps for Translations and Licensing (T&L). VPM (', 'NAME'), (') to send an email communication to VSM (', 'NAME'), (') on ', 'DATE'), (' to document this CTT decision. resulting in There is a very limited impact for this decision.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The ISRs failure on PK CFZ533 samples led to the investigation of original run as high dilution were performed. The conclusion is that those runs with high dilution should be reanalyzed. Nonetheless, the demonstrated stability does not allow the reanalysis of those samples resulting in the need to deactivate 59 data and to perform a new audit of dataset that that can impact timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Per GCP audit conducted Q1 2023 (#2042320) several computerized systems require validation (GeneMapper and SeqScape). may lead to Systems should be validation prior to start of GCP work with the vendor. resulting in Delay to validation of systems could impact testing timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('BA analysis delay may lead to late PK analysis delays/reconciliation issues resulting in late FIR and CSR.', '')]\n", "\n", "\n", "\n", "----- [('1001', 'Patient Related SiteID', 111, 115, 17, 18), ('1001', 'Patient Related SiteID', 510, 514, 82, 83)]\n", "~~~~~~~ [('1001', 'Patient Related SiteID', 111, 115, 17, 18), ('1001', 'Patient Related SiteID', 510, 514, 82, 83)]\n", "[('There have been multiple CE raters (individuals) for the Verasci datapoints that are no longer supporting site ', 'Patient Related SiteID'), (\" in India led to Greater variability than desired in CE raters Current CE rater trained in July has left the site, Raters will ideally rate the same patient's data/assessments throughout of the course of their study participation, and this is not possible in India resulting in Greater variability in results due to rater variability. Potential risk to the primary endpoint for study for site \", 'Patient Related SiteID'), ('.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Review of ePRO screens prior to finalization failed to identify an error in the EORTC QLQ-C30 question 30 response range led to Erroneous EORTC QLQ-C30 ePRO released by the vendor ePRO system, and completed by at least 43% of participants (baseline). resulting in Data integrity related to secondary endpoint', '')]\n", "\n", "\n", "\n", "----- [('5011', 'Patient Related SiteID', 9, 13, 2, 3), ('5011', 'Patient Related SiteID', 230, 234, 38, 39), ('5011', 'Patient Related SiteID', 459, 463, 80, 82), ('Austin Lambert', 'NAME', 111, 125, 19, 21), ('08/04/22', 'DATE', 262, 270, 44, 45), ('08/04/22', 'DATE', 420, 428, 73, 75), ('PETERSON', 'NAME', 323, 331, 55, 56), ('PETERSON', 'NAME', 408, 416, 71, 73)]\n", "~~~~~~~ [('5011', 'Patient Related SiteID', 9, 13, 2, 3), ('Austin Lambert', 'NAME', 111, 125, 19, 21), ('5011', 'Patient Related SiteID', 230, 234, 38, 39), ('08/04/22', 'DATE', 262, 270, 44, 45), ('PETERSON', 'NAME', 323, 331, 55, 56), ('PETERSON', 'NAME', 408, 416, 71, 73), ('08/04/22', 'DATE', 420, 428, 73, 75), ('5011', 'Patient Related SiteID', 459, 463, 80, 82)]\n", "[('The Site ', 'Patient Related SiteID'), (' reported that they did not receive any devices from YPrime to perform ePRO assessments. The CRA, ', 'NAME'), (\", confirmed that this was discussed at the time of SIV. However, according to YPrime's records, the Site \", 'Patient Related SiteID'), (' was delivered 3 devices on ', 'DATE'), (' at 2:30 pm. The shipment was received and signed by ', 'NAME'), ('. may lead to YPrime provided us a document signed as a proof of receipt by ', 'NAME'), (' on ', 'DATE'), (' at 2:30 pm. However, the Site ', 'Patient Related SiteID'), (' replied to us said \"Unfortunately, we do not have an employee by that name, there are many locations inside of our very large institution. I assure you we do not have this shipment within our department at Michigan medicine. I cannot speak as to where the driver left the package. This has happened before, and we have no way of locating these packages.\" Therefore, these 3 devices recoded as \"lost\", and CRA was requested to document this in Monitoring Visit Report. resulting in The site opted to use the BYOD feature.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"YPrime (ePRO vendor) implemented an incorrect EORTC QLQ-C30 screen in an effort to expedite implementation of correct screen. may lead to Work on configuration of CCF013 (correction of question 30 for the EORTC QLQ-C30 ePRO) started before CCF012 was complete. When CCF012 was completed, an IT bug in the system resulted in question 30 of EORTC QLQ-C30 ePRO being released without the text tags that accompany the numerical response scale ('very poor' and 'excellent') . resulting in 1 participant completed the EORTC QLQ-C30 ePRO with the missing response tags for question 30.\", '')]\n", "\n", "\n", "\n", "----- [('before the previous system update (CCF012) was complete.', 'DATE', 134, 190, 22, 30)]\n", "~~~~~~~ [('before the previous system update (CCF012) was complete.', 'DATE', 134, 190, 22, 30)]\n", "[('YPrime (ePRO vendor) started work on configuration of a system update (CCF013 - correction of question 30 for the EORTC QLQ-C30 ePRO) ', 'DATE'), (\" led to An IT bug in the system introduced when CCF012 was released. This resulted in question 30 of EORTC QLQ-C30 ePRO being released without the text tags that accompany the numerical response scale ('very poor' and 'excellent') . resulting in No impact to date - zero participants completed the incorrect ePRO.\", '')]\n", "\n", "\n", "\n", "----- [('27-Sep', 'DATE', 228, 234, 39, 41), ('04-Oct', 'DATE', 279, 285, 47, 48), ('10-Oct', 'DATE', 316, 322, 53, 54), ('17-Oct', 'DATE', 375, 381, 62, 63)]\n", "~~~~~~~ [('27-Sep', 'DATE', 228, 234, 39, 41), ('04-Oct', 'DATE', 279, 285, 47, 48), ('10-Oct', 'DATE', 316, 322, 53, 54), ('17-Oct', 'DATE', 375, 381, 62, 63)]\n", "[('YPrime User Acceptance Testing (UAT) extended due to product bugs found by YPrime Validation team on our custom report and web backup led to a delay of custom report testing resulting in original go live date being pushed out (', 'DATE'), (' original UAT date, tablet UAT was available ', 'DATE'), (', web backup UAT was available ', 'DATE'), (', and pending custom report availability still as of ', 'DATE'), (')', '')]\n", "\n", "\n", "\n", "----- [('PDR001 BDR', 'DATE', 177, 187, 33, 34)]\n", "~~~~~~~ [('PDR001 BDR', 'DATE', 177, 187, 33, 34)]\n", "[('1 ADA sample collected in error by the site, shipped to the BA lab and analysed led to sample results included in the final analysis resulting in data that was included in the ', 'DATE'), ('', '')]\n", "\n", "\n", "\n", "----- [('JDQ443', 'SUBJID', 283, 289, 40, 41), ('JDQ443', 'SUBJID', 449, 455, 63, 65), ('JDQ443', 'SUBJID', 490, 496, 70, 72), ('JDQ443', 'SUBJID', 519, 525, 74, 76), ('CRA', 'NAME', 229, 232, 31, 32), ('CRA', 'NAME', 693, 696, 101, 103), ('CRA', 'NAME', 713, 716, 103, 105), ('Estonia', 'COUNTRY', 299, 306, 42, 44), ('Estonia', 'COUNTRY', 611, 618, 89, 90), ('Estonia', 'COUNTRY', 832, 839, 121, 123)]\n", "~~~~~~~ [('CRA', 'NAME', 229, 232, 31, 32), ('JDQ443', 'SUBJID', 283, 289, 40, 41), ('Estonia', 'COUNTRY', 299, 306, 42, 44), ('JDQ443', 'SUBJID', 449, 455, 63, 65), ('JDQ443', 'SUBJID', 490, 496, 70, 72), ('JDQ443', 'SUBJID', 519, 525, 74, 76), ('Estonia', 'COUNTRY', 611, 618, 89, 90), ('CRA', 'NAME', 693, 696, 101, 103), ('CRA', 'NAME', 713, 716, 103, 105), ('Estonia', 'COUNTRY', 832, 839, 121, 123)]\n", "[('1.\\tCRA did not fulfill its obligation related to documents translation to local language what caused lack of GCP documents submission;\\n2.\\tInsufficient of CSM supervision over the process; \\n3.\\tDelay in escalation of inappropriate ', 'NAME'), (' performance by CSM; led to For clinical project C', 'SUBJID'), ('B12301 in ', 'COUNTRY'), (' a significant delay in GCP documents submission to EC and HA occuered.\\nThe following documents have not been submitted for EC/HA approval:\\n•\\t', 'SUBJID'), (' IB Ed 05\\n•\\tUpdated ICFs based on ', 'SUBJID'), (' IB Ed.5\\n•\\tERRATUM to ', 'SUBJID'), (' IB Edition 05\\nDelay was caused due to lack availablility of ICFs in local languages (', 'COUNTRY'), ('n and Russian). Process of document translation to local language is under ', 'NAME'), (' responsibility. ', 'NAME'), (' failed to full fill his obligation related to ICF translation resulting in Potential impact on patient safety.\\nIn ', 'COUNTRY'), (' there is only 1 site active and no active subjects at this site.\\nPatientsÂ’ safety was not affected.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('72h hospitalization requirement (after dosing), may lead to loss of interest from potential patients in joining the study, resulting in slower recruitment', '')]\n", "\n", "\n", "\n", "----- [('1014', 'Patient Related SiteID', 112, 116, 16, 18)]\n", "~~~~~~~ [('1014', 'Patient Related SiteID', 112, 116, 16, 18)]\n", "[('1014 site can not conduct Triacylglycerol lipase testing led to Missing lipase assessments were identified for ', 'Patient Related SiteID'), (' site resulting in In asciminib IB, pancreatic is one of serious ADRs considered expected for safety reporting purposes in adult population\\nLipase is more specific to detct pancreatis and required in each visit with chemistry test planned. \\nContinous missing lipase may significantly impact patient safety', '')]\n", "\n", "\n", "\n", "----- [('3Oct2023', 'DATE', 95, 103, 14, 15), ('batch failure', 'MH', 2, 15, 1, 3), ('Lu-PSMA-617', 'MH', 46, 57, 7, 8), ('clinical dose', 'MH', 114, 127, 18, 21), ('administration', 'MH', 151, 165, 24, 26), ('dose delay', 'MH', 200, 210, 31, 34)]\n", "~~~~~~~ [('batch failure', 'MH', 2, 15, 1, 3), ('Lu-PSMA-617', 'MH', 46, 57, 7, 8), ('3Oct2023', 'DATE', 95, 103, 14, 15), ('clinical dose', 'MH', 114, 127, 18, 21), ('administration', 'MH', 151, 165, 24, 26), ('dose delay', 'MH', 200, 210, 31, 34)]\n", "[('A ', 'MH'), (' occurred during production of ', 'MH'), (' at Ivrea Italy manufacturing site on ', 'DATE'), (' led to 1 ', 'MH'), (' to Japan impacted and ', 'MH'), (' to be rescheduled resulting in a ', 'MH'), (' by 6 weeks', '')]\n", "\n", "\n", "\n", "----- [('22Sep2023', 'DATE', 95, 104, 14, 15)]\n", "~~~~~~~ [('22Sep2023', 'DATE', 95, 104, 14, 15)]\n", "[('A batch failure occurred during production of Lu-PSMA-617 at Ivrea Italy manufacturing site on ', 'DATE'), (' led to 12 clinical doses impacted including 5 doses to China, 2 doses to Japan, 2 dose to Taiwan, 1 dose to Sweden, 1 dose to Austria, and 1 dose to UK resulting in delay of infusion for 12 patients', '')]\n", "\n", "\n", "\n", "----- [('22Sep2023', 'DATE', 95, 104, 14, 15), ('Two clinical doses', 'PRECLINICAL STUDY DAY', 114, 132, 17, 21)]\n", "~~~~~~~ [('22Sep2023', 'DATE', 95, 104, 14, 15), ('Two clinical doses', 'PRECLINICAL STUDY DAY', 114, 132, 17, 21)]\n", "[('A batch failure occurred during production of Lu-PSMA-617 at Ivrea Italy manufacturing site on ', 'DATE'), ('. led to ', 'PRECLINICAL STUDY DAY'), (' were impacted and administration needed to be rescheduled. resulting in One clinical dose was rescheduled from 26Sep2023 to 3Oct2023 administration date, and one clinical dose was rescheduled from 26Sep2023 to 17Oct2023 administration date.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"A delay in PPC 14, and lack of communication with the DM team may lead to a delay in data entry resulting in a delay in the Part A's manuscript deliverable.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('A high early dropout rate may lead to difficulties in measuring the primary/secondary endpoints resulting in UNK Best overall response for phase 2 or censoring of PFS due to no TA available', '')]\n", "\n", "\n", "\n", "----- [('29sep2023', 'DATE', 373, 382, 64, 66), ('03-Oct-2023', 'DATE', 386, 397, 66, 67)]\n", "~~~~~~~ [('29sep2023', 'DATE', 373, 382, 64, 66), ('03-Oct-2023', 'DATE', 386, 397, 66, 67)]\n", "[('A large number of data entries occurred around data cut off date that required queries, this combined with standard pre cut off screening activities led to a higher than average number of queries. may lead to Due to this cause along with open query, PI signature and uncoded terms count is also high. resulting in Potential delayed database lock. Note- Rave was slow from ', 'DATE'), (' to ', 'DATE'), (', has also impacted the timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('A significantly higher enrollment rate than expected may lead to may lead to insufficient material at digital vendor, Actigraph resulting in resulting in sites required to reduce screening and/or randomisation rates until new material has been allocated to the study specific depots.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Addition of an expansion cohort to this study may lead to delay of Lu administration to the patients due to the increase of Lu supply risk resulting in patients' safety\", '')]\n", "\n", "\n", "\n", "----- [('various countries', 'COUNTRY', 48, 65, 7, 9), ('various countries', 'COUNTRY', 206, 223, 31, 34)]\n", "~~~~~~~ [('various countries', 'COUNTRY', 48, 65, 7, 9), ('various countries', 'COUNTRY', 206, 223, 31, 34)]\n", "[('Additional quantities for IMP when requested by ', 'COUNTRY'), (' are causing delays, as it requires updates to import license and documentation may lead to Additional quantities for IMP when requested by ', 'COUNTRY'), (' are causing delays, as it requires updates to import license and documentation resulting in Delayed shipments of IMP into countries', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('All SOC anti-cancer drugs (bevacizumab, modified FOLFOX6, FOLFIRI, paclitaxel, ramucirumab and cabozantinib) provided locally by the study site, subsidiary, or designee as commercially available or centrally by Novartis, in each participating country according to local clinical practices and local regulations may lead to differences in trademarks, drug accountability and drug management resulting in potential impact to data integrity, subjectÂ’s safety and study primary and secondary endpoints.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('An amendment (update) to the digital recruitment campaign was approved in a staggered (by study) manner by the central IRB led to led to The NDR team missed to make the approved updates on the webpage resulting in and therefore some patients were not seeing the latest changes on the digital recruitment platform (webpage). It is related to the to the 1QEM record 98142 resulting in 24 potential patients have been directed to the website since May. All of those potential patients would have later spoken to the call center (if they proceeded through the online flow) and been given accurate information on the studies at that time.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('An amendment (update) to the digital recruitment campaign was approved in a staggered (by study) manner by the central IRB. led to The NDR team missed to make the approved updates on the webpage and therefore some patients were not seeing the latest changes on the digital recruitment platform (webpage). It is related to the to the 1QEM record 98142 resulting in 24 potential patients have been directed to the website since May. All of those potential patients would have later spoken to the call center (if they proceeded through the online flow) and been given accurate information on the studies at that time.', '')]\n", "\n", "\n", "\n", "----- [('Sept 2023', 'DATE', 74, 83, 11, 13)]\n", "~~~~~~~ [('Sept 2023', 'DATE', 74, 83, 11, 13)]\n", "[('Any further IMP dosing deviations at site following restart of enrolment (', 'DATE'), (') may lead to shortage of IMP supplies to screen further patients resulting in risk to delaying LPFT and downstream milestones.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('As an amendment (update) to the digital recruitment campaign was approved in a staggered (by study) manner by the central IRB led to the NDR team missed to make the approved updates on the webpage resulting in and therefore some patients were not seeing the latest changes on the digital recruitment platform (webpage)', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('As an amendment (update) to the digital recruitment campaign was approved in a staggered (by study) manner by the central IRB led to the NDR team missed to make the approved updates on the webpage resulting in and therefore some patients were not seeing the latest changes on the digital recruitment platform (webpage)', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('As an amendment (update) to the digital recruitment campaign was approved in a staggered (by study) manner by the central IRB may lead to the NDR team missed to make the approved updates on the webpage resulting in and therefore some patients were not seeing the latest changes on the digital recruitment platform (webpage)', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('As an amendment (update) to the digital recruitment campaign was approved in a staggered (by study) manner by the central IRB, this led to the NDR team having missed to make the approved updates on the webpage resulting in some patients were not seeing the latest changes on the digital recruitment platform (webpage).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('B cell TBNK assay without LLoQ validated method used as a criterion for B cell recovery/recurrence for re-dosing patients with RTX in maintenance period may lead to to a push back from HAs in EU because not validated for purpose (as per new IVDR compliance rules) resulting in Delay the start of the trial in EU and delay trial overall timelines', '')]\n", "\n", "\n", "\n", "----- [('June 2023', 'DATE', 118, 127, 21, 23)]\n", "~~~~~~~ [('June 2023', 'DATE', 118, 127, 21, 23)]\n", "[('Based on a Powerpoint presentation for DBL meeting, a subject was listed as true WoC. This case had been discussed in ', 'DATE'), (', but at that time, discussion concluded that the subject discontinued from the study, it was not a true WoC. led to PK and IG KAZ samples from the WoC subject were analyzed after the WoC date resulting in PKS-PTM confirmed no data from this subject were used in any presentation and the data was flagged and excluded from PK data analysis due to important dose deviation versus actual dose.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Based on Novartis QUC398A12201 (14040) Xray Eligibility Results, study team identified site 1003 has wrongly uploaded Xrays not belonging to our study. may lead to Two incidents happened in late Aug/early Sep with wrong X-rays uploaded to the QUC398A12201 study. resulting in May lead to incorrect data collection. This needs to be properly documented (incl. action items such as re-training).', '')]\n", "\n", "\n", "\n", "----- [('Q3 2024', 'DATE', 303, 310, 55, 57), ('end Jan 2025', 'DATE', 355, 367, 65, 68), ('end Jan 2025', 'DATE', 617, 629, 112, 115), ('end January 2025', 'DATE', 473, 489, 86, 89)]\n", "~~~~~~~ [('Q3 2024', 'DATE', 303, 310, 55, 57), ('end Jan 2025', 'DATE', 355, 367, 65, 68), ('end January 2025', 'DATE', 473, 489, 86, 89), ('end Jan 2025', 'DATE', 617, 629, 112, 115)]\n", "[('Based on the current recruitment rate, the LPLV is expected to be extended till May 2025. This date is beyond the cut- off date, when all ongoing trials in EU countries have to transition to EU CTR. In case recruitment will not progress faster than expected, the team has to submit EU CTR transition by ', 'DATE'), (' the latest, to receive approval in time (by ', 'DATE'), ('). The issue does not impact RoW countries may lead to If transition to EU CTR not completed/approved by ', 'DATE'), (', the trial cannot proceed in EU resulting in Trial cannot be conducted in EU if not trnasitioned to EU CTR within the deadline ', 'DATE'), ('.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('bATR vendor reporting activity placed on hold per Novartis EDO communication to vendors. led to Reports from IRT and Clario (PROs) will not be available for data management review at database lock. resulting in Without reports, audit reviews will not be completed.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Because of additional discussions with the US FDA the last-minute decision was made that an iDBL is needed instead of a DB Snapshot to be able to provide PDRs (4 BDs before the planned DB Snapshot date). Some sites may not be able to sign off on all patient eCRFs led to iDBL delayed by more than the 5 BDs that were added to allow for PI signatures to occur resulting in team working during the year-end break and potentially missing the 120-day Safety Update submission deadline.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Biomarker data requires external input may lead to Setup of consultancy agreement with Dr Hsu resulting in Potential delays to analysis of biomarker data', '')]\n", "\n", "\n", "\n", "----- [('2002', 'DATE', 39, 43, 5, 6), ('2002', 'DATE', 161, 165, 23, 25), ('2002', 'DATE', 345, 349, 56, 58), ('2002', 'DATE', 535, 539, 91, 93), ('2002', 'DATE', 562, 566, 96, 98), ('16 Jan 2023', 'DATE', 209, 220, 31, 34), ('Jan 2023', 'DATE', 375, 383, 63, 66), ('2001', 'Patient Related SiteID', 484, 488, 81, 83), ('2001', 'Patient Related SiteID', 544, 548, 93, 95), ('2001', 'Patient Related SiteID', 571, 575, 98, 100)]\n", "~~~~~~~ [('2002', 'DATE', 39, 43, 5, 6), ('2002', 'DATE', 161, 165, 23, 25), ('16 Jan 2023', 'DATE', 209, 220, 31, 34), ('2002', 'DATE', 345, 349, 56, 58), ('Jan 2023', 'DATE', 375, 383, 63, 66), ('2001', 'Patient Related SiteID', 484, 488, 81, 83), ('2002', 'DATE', 535, 539, 91, 93), ('2001', 'Patient Related SiteID', 544, 548, 93, 95), ('2002', 'DATE', 562, 566, 96, 98), ('2001', 'Patient Related SiteID', 571, 575, 98, 100)]\n", "[('BTR released mistake notification that ', 'DATE'), (' sent screening X-ray to BTR. BTR cancelled the mistake notification. Followingly NIRT released IMP shipment even if ', 'DATE'), (' has been deactivated for any screenings on ', 'DATE'), ('. led to NIRT released IMP even if NIRT confirmed that no screenings are possible at this site. NIRT has been notified that ', 'DATE'), (' is going to be closed in ', 'DATE'), ('. Related issue is I061.\\nPM (CZE) contacted NIRT with information that IMP has been released also to ', 'Patient Related SiteID'), (' site. resulting in IMP, that is not needed at ', 'DATE'), (' and ', 'Patient Related SiteID'), (', released to ', 'DATE'), (' and ', 'Patient Related SiteID'), (', that have been ready to be closed.', '')]\n", "\n", "\n", "\n", "----- [('2002', 'DATE', 39, 43, 5, 6), ('2002', 'DATE', 161, 165, 23, 25), ('2002', 'DATE', 350, 354, 57, 59), ('2002', 'DATE', 453, 457, 78, 80), ('2002', 'DATE', 471, 475, 81, 83), ('16 Jan 2023', 'DATE', 209, 220, 31, 34), ('Jan 2023', 'DATE', 380, 388, 64, 67)]\n", "~~~~~~~ [('2002', 'DATE', 39, 43, 5, 6), ('2002', 'DATE', 161, 165, 23, 25), ('16 Jan 2023', 'DATE', 209, 220, 31, 34), ('2002', 'DATE', 350, 354, 57, 59), ('Jan 2023', 'DATE', 380, 388, 64, 67), ('2002', 'DATE', 453, 457, 78, 80), ('2002', 'DATE', 471, 475, 81, 83)]\n", "[('BTR released mistake notification that ', 'DATE'), (' sent screening X-ray to BTR. BTR cancelled the mistake notification. Followingly NIRT released IMP shipment even if ', 'DATE'), (' has been deactivated for any screenings on ', 'DATE'), ('. may lead to NIRT released IMP even if NIRT confirmed that no screenings are possible at this site. NIRT has been notified that ', 'DATE'), (' is going to be closed in ', 'DATE'), ('. Related issue is I061. resulting in IMP, that is not needed at ', 'DATE'), (', released to ', 'DATE'), (' that has been ready to be closed.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('C-SSRS not planned at every visit may lead to may lead to FDA comments which may require protocol amendment resulting in delay to the current FPFV timelines of 29th Jan 2024', '')]\n", "\n", "\n", "\n", "----- [('China', 'COUNTRY', 0, 5, 0, 1)]\n", "~~~~~~~ [('China', 'COUNTRY', 0, 5, 0, 1)]\n", "[('', 'COUNTRY'), (' country specific requirement of Manufacturer CoA for Teriflunomid may lead to is limiting the availability and usability of the Teriflunomid batches already procured and over encapsulated from open market resulting in resulting in a potential supply shortage in the country.', '')]\n", "\n", "\n", "\n", "----- [('v06', 'DATE', 145, 148, 26, 28)]\n", "~~~~~~~ [('v06', 'DATE', 145, 148, 26, 28)]\n", "[('Clinical hold due to study stopping criteria met during dosing of Cohort A5 (50 mg with 3 days DT) may lead to submission of protocol amendment ', 'DATE'), (' resulting in delayed LPLV or early termination if not approved by MHRA', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Competitive landscape in non-small cell lung cancer (NSCLC) may lead to slow recruitment at clinical sites in an indication that has become a business priority for the company resulting in inability to fill expansion cohort with NSCLC patients within a reasonable amount of time and early termination of the trial.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Concomitant use of SARS-CoV-2 (COVID-19) vaccines led to an interaction with study drug resulting in an impact on patient efficacy/safety and or clinical trial and data integrity', '')]\n", "\n", "\n", "\n", "----- [('June 2024', 'DATE', 50, 59, 9, 11)]\n", "~~~~~~~ [('June 2024', 'DATE', 50, 59, 9, 11)]\n", "[('Conversion factor from CSF1 to FMI is expected in ', 'DATE'), (', but the clinical dose may not be known at that time may lead to delays with the CMC process resulting in the need to re-manufacture radiolabeled material and late study start.', '')]\n", "\n", "\n", "\n", "----- [('teams', 'COUNTRY', 8, 13, 1, 2)]\n", "~~~~~~~ [('teams', 'COUNTRY', 8, 13, 1, 2)]\n", "[('Country ', 'COUNTRY'), (' to review placeholder report and ensure timely filing of the documents may lead to Low TMF completeness score in SENSE resulting in low TMF compliance', '')]\n", "\n", "\n", "\n", "----- [('Country teams', 'COUNTRY', 0, 13, 0, 2), ('country teams', 'COUNTRY', 131, 144, 24, 26), ('country teams', 'COUNTRY', 256, 269, 45, 48), ('country teams', 'COUNTRY', 310, 323, 54, 57)]\n", "~~~~~~~ [('Country teams', 'COUNTRY', 0, 13, 0, 2), ('country teams', 'COUNTRY', 131, 144, 24, 26), ('country teams', 'COUNTRY', 256, 269, 45, 48), ('country teams', 'COUNTRY', 310, 323, 54, 57)]\n", "[('', 'COUNTRY'), (' were going through a lot of change at the time of CSR appendices preparation and we were not getting a response from ', 'COUNTRY'), (' to complete eTRAC. led to the study team to consider eTRAC obsolete and collect the data needed directly from ', 'COUNTRY'), (' with the use of different trackers. The ', 'COUNTRY'), (' finally managed to complete eTRAC but only after the CSR was published. resulting in small discrepancies between the data provided in eTRAC and our trackers. The discrepancies found are minimal and have no impact on the quality of data and safety of patients so the study team decided to accept it.', '')]\n", "\n", "\n", "\n", "----- [('June 2020', 'DATE', 143, 152, 22, 25), ('USA', 'COUNTRY', 92, 95, 14, 15), ('Argentina', 'COUNTRY', 97, 106, 15, 16)]\n", "~~~~~~~ [('USA', 'COUNTRY', 92, 95, 14, 15), ('Argentina', 'COUNTRY', 97, 106, 15, 16), ('June 2020', 'DATE', 143, 152, 22, 25)]\n", "[('COVID-19 pandemic lasts longer than predicted led to Recruitment delays in few countries - ', 'COUNTRY'), (', ', 'COUNTRY'), (', vs. predictions in FPO scenario V. ', 'DATE'), (' resulting in Delay in meeting the interim milestone of our 48 patients commitment by April 2021', '')]\n", "\n", "\n", "\n", "----- [('March 2020- until now', 'DATE', 26, 47, 3, 7)]\n", "~~~~~~~ [('March 2020- until now', 'DATE', 26, 47, 3, 7)]\n", "[('COVID-19 pandemic ongoing ', 'DATE'), (' in many countries will delay site selection visits, planned SIVs for new sites in some regions and countries. led to Slower than predicted recruitment in FPO version June 2020 resulting in Extended recruitment timelines for week-48 patients (beyond April 2021)', '')]\n", "\n", "\n", "\n", "----- [('0005', 'Patient Related SiteID', 52, 56, 8, 9), ('wrong address', 'ADDRESS', 29, 42, 4, 6)]\n", "~~~~~~~ [('wrong address', 'ADDRESS', 29, 42, 4, 6), ('0005', 'Patient Related SiteID', 52, 56, 8, 9)]\n", "[('CRA accidentally entered the ', 'ADDRESS'), (' for site ', 'Patient Related SiteID'), (' into vendor tracker may lead to Heparin Tubes may not be delivered to site in time for FPFT resulting in FPFT cannot go ahead', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('CREDI migration to Novartis Clinical Vault may lead to CREDI migration to Novartis CLinical VAult may lead to risks associated with the TMF migration (e.g. missing documents) but could also impact on important study milestones being performed at the moment as the interim analysis 4 and interim CSR 2. resulting in Impact is uncertain but cannot be eliminated.', '')]\n", "\n", "\n", "\n", "----- [('C1D1', 'PRECLINICAL STUDY DAY', 30, 34, 4, 5)]\n", "~~~~~~~ [('C1D1', 'PRECLINICAL STUDY DAY', 30, 34, 4, 5)]\n", "[('Critical timepoints missed at ', 'PRECLINICAL STUDY DAY'), (' for participants therefore assessed as not evaluable may lead to protocol amendment to get evaluability criteria less stringent resulting in delay in entire study timelines', '')]\n", "\n", "\n", "\n", "----- [('1001001', 'SUBJID', 151, 158, 28, 29), ('1001001', 'SUBJID', 211, 218, 30, 31), ('1001001', 'SUBJID', 283, 290, 35, 36), ('1001001', 'SUBJID', 343, 350, 37, 38), ('1001001', 'SUBJID', 403, 410, 39, 40), ('1001001', 'SUBJID', 463, 470, 41, 42), ('1001001', 'SUBJID', 523, 530, 43, 44), ('1001001', 'SUBJID', 572, 579, 45, 46), ('1001001', 'SUBJID', 645, 652, 50, 51), ('1001001', 'SUBJID', 706, 713, 52, 53), ('1001001', 'SUBJID', 767, 774, 54, 55), ('1001001', 'SUBJID', 828, 835, 56, 57), ('1001001', 'SUBJID', 889, 896, 58, 59), ('1001001', 'SUBJID', 939, 946, 60, 61), ('1001001', 'SUBJID', 1005, 1012, 63, 64), ('1001001', 'SUBJID', 1071, 1078, 66, 67), ('1001001', 'SUBJID', 1137, 1144, 69, 70), ('1001001', 'SUBJID', 1203, 1210, 72, 73), ('08-Feb-2022', 'DATE', 1303, 1314, 75, 76), ('Neutropenia', 'MH', 2947, 2958, 142, 143), ('Neutropenia', 'MH', 3007, 3018, 144, 145), ('Neutropenia', 'MH', 3561, 3572, 168, 169), ('25-Mar-2022', 'DATE', 1502, 1513, 87, 88), ('07-Apr-2022', 'DATE', 967, 978, 61, 62), ('May-2022', 'DATE', 2374, 2382, 122, 123), ('08-Jul-2022', 'DATE', 1045, 1056, 64, 65), ('08-Jul-2022', 'DATE', 1099, 1110, 67, 68), ('12-Jul-2022', 'DATE', 1165, 1176, 70, 71), ('12-Jul-2022', 'DATE', 1369, 1380, 78, 79), ('12-Jul-2022', 'DATE', 2014, 2025, 109, 110), ('12-Jul-2022', 'DATE', 2131, 2142, 113, 114), ('27-Jul-2022', 'DATE', 1243, 1254, 73, 74), ('27-Jul-2022', 'DATE', 1953, 1964, 107, 108), ('27-Jul-2022', 'DATE', 2321, 2332, 120, 121), ('27-Jul-2022', 'DATE', 3078, 3089, 146, 147), ('leukopenia', 'MH', 1860, 1870, 101, 102), ('leukopenia', 'MH', 2060, 2070, 111, 112), ('leukopenia', 'MH', 2592, 2602, 130, 131), ('leukopenia', 'MH', 2653, 2663, 132, 133), ('leukopenia', 'MH', 2714, 2724, 134, 135), ('leukopenia', 'MH', 2775, 2785, 136, 137), ('leukopenia', 'MH', 2836, 2846, 138, 139), ('leukopenia', 'MH', 2897, 2907, 140, 141), ('leukopenia', 'MH', 3511, 3521, 166, 167), ('29-Jul-2022', 'DATE', 2187, 2198, 115, 116), ('1001005', 'SUBJID', 1329, 1336, 77, 78), ('1001005', 'SUBJID', 1407, 1414, 83, 84), ('1001005', 'SUBJID', 1462, 1469, 86, 87), ('1-07-Sep-2022', 'DATE', 1433, 1446, 84, 85), ('1001006', 'SUBJID', 1528, 1535, 89, 90), ('1001006', 'SUBJID', 1583, 1590, 91, 92), ('1001006', 'SUBJID', 1639, 1646, 93, 94), ('1001006', 'SUBJID', 1695, 1702, 95, 96), ('1001006', 'SUBJID', 1751, 1758, 97, 98), ('1001006', 'SUBJID', 1807, 1814, 99, 100), ('anaemia', 'MH', 2233, 2240, 117, 118), ('anaemia', 'MH', 3112, 3119, 148, 149), ('anaemia', 'MH', 3168, 3175, 150, 151), ('anaemia', 'MH', 3224, 3231, 152, 153), ('anaemia', 'MH', 3326, 3333, 157, 158), ('1001009', 'SUBJID', 1925, 1932, 106, 107), ('1001009', 'SUBJID', 2281, 2288, 119, 120), ('03-Aug-2022', 'DATE', 2507, 2518, 126, 127), ('03-Aug-2022', 'DATE', 3422, 3433, 162, 164), ('1001010', 'SUBJID', 2409, 2416, 124, 125), ('1001010', 'SUBJID', 2533, 2540, 128, 129), ('05-Jul-2022', 'DATE', 3410, 3421, 161, 162), ('5-Aug-2022', 'DATE', 3292, 3302, 155, 156), ('1001011', 'SUBJID', 3261, 3268, 154, 155), ('1001012', 'SUBJID', 3448, 3455, 163, 165), ('1001013', 'SUBJID', 3625, 3632, 173, 175), ('29-Nov-2022', 'DATE', 3587, 3598, 169, 172), ('1-03-Nov-2022', 'DATE', 3651, 3664, 174, 176), ('22-Nov-2022', 'DATE', 3665, 3676, 175, 178)]\n", "~~~~~~~ [('1001001', 'SUBJID', 151, 158, 28, 29), ('1001001', 'SUBJID', 211, 218, 30, 31), ('1001001', 'SUBJID', 283, 290, 35, 36), ('1001001', 'SUBJID', 343, 350, 37, 38), ('1001001', 'SUBJID', 403, 410, 39, 40), ('1001001', 'SUBJID', 463, 470, 41, 42), ('1001001', 'SUBJID', 523, 530, 43, 44), ('1001001', 'SUBJID', 572, 579, 45, 46), ('1001001', 'SUBJID', 645, 652, 50, 51), ('1001001', 'SUBJID', 706, 713, 52, 53), ('1001001', 'SUBJID', 767, 774, 54, 55), ('1001001', 'SUBJID', 828, 835, 56, 57), ('1001001', 'SUBJID', 889, 896, 58, 59), ('1001001', 'SUBJID', 939, 946, 60, 61), ('07-Apr-2022', 'DATE', 967, 978, 61, 62), ('1001001', 'SUBJID', 1005, 1012, 63, 64), ('08-Jul-2022', 'DATE', 1045, 1056, 64, 65), ('1001001', 'SUBJID', 1071, 1078, 66, 67), ('08-Jul-2022', 'DATE', 1099, 1110, 67, 68), ('1001001', 'SUBJID', 1137, 1144, 69, 70), ('12-Jul-2022', 'DATE', 1165, 1176, 70, 71), ('1001001', 'SUBJID', 1203, 1210, 72, 73), ('27-Jul-2022', 'DATE', 1243, 1254, 73, 74), ('08-Feb-2022', 'DATE', 1303, 1314, 75, 76), ('1001005', 'SUBJID', 1329, 1336, 77, 78), ('12-Jul-2022', 'DATE', 1369, 1380, 78, 79), ('1001005', 'SUBJID', 1407, 1414, 83, 84), ('1-07-Sep-2022', 'DATE', 1433, 1446, 84, 85), ('1001005', 'SUBJID', 1462, 1469, 86, 87), ('25-Mar-2022', 'DATE', 1502, 1513, 87, 88), ('1001006', 'SUBJID', 1528, 1535, 89, 90), ('1001006', 'SUBJID', 1583, 1590, 91, 92), ('1001006', 'SUBJID', 1639, 1646, 93, 94), ('1001006', 'SUBJID', 1695, 1702, 95, 96), ('1001006', 'SUBJID', 1751, 1758, 97, 98), ('1001006', 'SUBJID', 1807, 1814, 99, 100), ('leukopenia', 'MH', 1860, 1870, 101, 102), ('1001009', 'SUBJID', 1925, 1932, 106, 107), ('27-Jul-2022', 'DATE', 1953, 1964, 107, 108), ('12-Jul-2022', 'DATE', 2014, 2025, 109, 110), ('leukopenia', 'MH', 2060, 2070, 111, 112), ('12-Jul-2022', 'DATE', 2131, 2142, 113, 114), ('29-Jul-2022', 'DATE', 2187, 2198, 115, 116), ('anaemia', 'MH', 2233, 2240, 117, 118), ('1001009', 'SUBJID', 2281, 2288, 119, 120), ('27-Jul-2022', 'DATE', 2321, 2332, 120, 121), ('May-2022', 'DATE', 2374, 2382, 122, 123), ('1001010', 'SUBJID', 2409, 2416, 124, 125), ('03-Aug-2022', 'DATE', 2507, 2518, 126, 127), ('1001010', 'SUBJID', 2533, 2540, 128, 129), ('leukopenia', 'MH', 2592, 2602, 130, 131), ('leukopenia', 'MH', 2653, 2663, 132, 133), ('leukopenia', 'MH', 2714, 2724, 134, 135), ('leukopenia', 'MH', 2775, 2785, 136, 137), ('leukopenia', 'MH', 2836, 2846, 138, 139), ('leukopenia', 'MH', 2897, 2907, 140, 141), ('Neutropenia', 'MH', 2947, 2958, 142, 143), ('Neutropenia', 'MH', 3007, 3018, 144, 145), ('27-Jul-2022', 'DATE', 3078, 3089, 146, 147), ('anaemia', 'MH', 3112, 3119, 148, 149), ('anaemia', 'MH', 3168, 3175, 150, 151), ('anaemia', 'MH', 3224, 3231, 152, 153), ('1001011', 'SUBJID', 3261, 3268, 154, 155), ('5-Aug-2022', 'DATE', 3292, 3302, 155, 156), ('anaemia', 'MH', 3326, 3333, 157, 158), ('05-Jul-2022', 'DATE', 3410, 3421, 161, 162), ('03-Aug-2022', 'DATE', 3422, 3433, 162, 164), ('1001012', 'SUBJID', 3448, 3455, 163, 165), ('leukopenia', 'MH', 3511, 3521, 166, 167), ('Neutropenia', 'MH', 3561, 3572, 168, 169), ('29-Nov-2022', 'DATE', 3587, 3598, 169, 172), ('1001013', 'SUBJID', 3625, 3632, 173, 175), ('1-03-Nov-2022', 'DATE', 3651, 3664, 174, 176), ('22-Nov-2022', 'DATE', 3665, 3676, 175, 178)]\n", "[('Data has been updated by site post iDBL for AE, CM and MH page and was in scope for iDBL-week 24.\\nBelow is the details of records updated post iDBL.\\n\\n', 'SUBJID'), ('-Neutropenia-1-08-Feb-2022-24-Feb-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-2-25-Mar-2022-07-Apr-2022-End date was wrong in EDC\\n', 'SUBJID'), ('-Neutropenia-1-02-May-2022-31-May-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-2-31-May-2022-08-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-3-08-Jul-2022-12-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-2-12-Jul-2022-27-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-1-27-Jul-2022--Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-2-08-Feb-2022-07-Apr-2022-End date was wrong in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-1-07-Apr-2022-31-May-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-3-08-Jul-2022-12-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-2-12-Jul-2022-19-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-1-19-Jul-2022-27-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-2-29-Jul-2022--Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -1-', 'DATE'), ('-08-Jun-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -1-16-Jun-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -2-', 'DATE'), ('-12-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -1-', 'DATE'), ('-19-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -2-19-Jul-2022-', 'DATE'), ('-Missed in EDC\\n1001002-Neutropenia-1-11-Jan-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -1-14-Jun-2022-', 'DATE'), ('-End date was wrong in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('--Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -1-10-Mar-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-anaemia-1-14-Feb-2022-12-Apr-2022-Missed in MH\\n', 'SUBJID'), ('-anaemia-2-12-Apr-2022-10-May-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-1-10-May-2023-24-May-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-2-24-May-2023-07-Jun-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-1-07-Jun-2023-14-Feb-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-1-09-May-2023--Missed in EDC\\n1001009-', 'MH'), ('Â\\xa0-1-14-Jun-2022-12-Jul-2023-End date was wrong in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -1-', 'DATE'), ('-03-Aug-2022-Missed in EDC\\n1001009-leukopeniaÂ\\xa0-2-', 'DATE'), ('-27-Jul-2022-Missed in EDC\\n1001009-', 'MH'), ('Â\\xa0-1-29-Jul-2022-03-Aug-2022-Missed in EDC\\n1001009-anaemia-1-', 'DATE'), ('-13-Jul-2022-Missed in EDC\\n1001009-anaemia-1-', 'DATE'), ('-30-Jul-2022-Missed in EDC\\n1001009-', 'MH'), ('-1-03-Aug-2022-10-Aug-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -2-13-Jul-2022-', 'DATE'), ('-Missed in EDC\\n1001010-hyperuricemia-1-12-', 'DATE'), ('-24-May-2022-Missed in EDC\\n', 'SUBJID'), ('-hyperuricemia-1-07-Jun-2022-05-Jul-2022-Missed in EDC\\n1001010-hyperuricemia-3-09-Jul-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-hyperuricemia-1-12-Aug-2022--Missed in EDC\\n1001010-', 'MH'), ('Â\\xa0-1-17-Jun-2022-20-Jun-2022-Missed in EDC\\n1001010-', 'MH'), ('Â\\xa0-1-23-Jun-2022-27-Jun-2022-Missed in EDC\\n1001010-', 'MH'), ('Â\\xa0-2-27-Jun-2022-28-Jun-2022-Missed in EDC\\n1001010-', 'MH'), ('Â\\xa0-1-03-Aug-2022-05-Aug-2022-Missed in EDC\\n1001011-', 'MH'), ('Â\\xa0-1-14-Aug-2022-16-Aug-2022-Missed in EDC\\n1001011-', 'MH'), ('Â\\xa0-1-30-Aug-2022--Missed in EDC\\n1001011-', 'MH'), ('-3-30-Aug-2022-14-Sep-2022-Missed in EDC\\n1001011-', 'MH'), ('-2-14-Sep-2022--Missed in EDC\\n1001011-anaemia-1-28-Jun-2022-', 'DATE'), ('-Missed in EDC\\n1001011-', 'MH'), ('-1-28-Jul-2022-09-Aug-2022-Missed in EDC\\n1001011-', 'MH'), ('-1-30-Aug-2022-01-Sep-2022-Missed in EDC\\n1001011-', 'MH'), ('-2-01-Sep-2022--Missed in EDC\\n', 'SUBJID'), ('-Total Cholesteremia-1-0', 'DATE'), ('--Missed in EDC\\n1001012-', 'MH'), ('-2-03-Aug-2022- -Missed in EDC\\n1001012-Thrombocytopenia -1-', 'DATE'), ('-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -2-03-Aug-2022--Missed in EDC\\n1001013-', 'MH'), ('Â\\xa0-1-29-Nov-2022--Missed in EDC\\n1001013-', 'MH'), ('-3-08-Nov-2022-', 'DATE'), ('-End date was wrong in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('-', 'DATE'), ('-End date was wrong in EDC may lead to Source data was not reviewed to update the change in toxicity grade or dates in AE page during iDBL-week 24 resulting in Study team needs to access the output and TFLS impacted due to change in data post iDBL.\\n\\n 1. CTT need to assess whether this data is associated with primary or secondary end point analysis, if yes, whether this will have an impact to CSR outputs or CSR TFLS.\\n2. CTT needs to consider this as a critical data for iDBL:- If yes, need to be documented in Post lock issue form even if not going to have any impact in CSR\\n3. Whether this data is a critical safety data, if yes, any safety outputs are impacted:- then need a Post lock issue form.\\n\\nAE related output impacted due to data change or update', '')]\n", "\n", "\n", "\n", "----- [('1001001', 'SUBJID', 150, 157, 28, 29), ('1001001', 'SUBJID', 208, 215, 30, 31), ('1001001', 'SUBJID', 278, 285, 35, 36), ('1001001', 'SUBJID', 336, 343, 37, 38), ('1001001', 'SUBJID', 394, 401, 39, 40), ('1001001', 'SUBJID', 452, 459, 41, 42), ('1001001', 'SUBJID', 510, 517, 43, 44), ('1001001', 'SUBJID', 568, 575, 45, 46), ('1001001', 'SUBJID', 639, 646, 50, 51), ('1001001', 'SUBJID', 698, 705, 52, 53), ('1001001', 'SUBJID', 757, 764, 54, 55), ('1001001', 'SUBJID', 816, 823, 56, 57), ('1001001', 'SUBJID', 875, 882, 58, 59), ('1001001', 'SUBJID', 934, 941, 60, 61), ('1001001', 'SUBJID', 998, 1005, 63, 64), ('1001001', 'SUBJID', 1062, 1069, 66, 67), ('1001001', 'SUBJID', 1126, 1133, 69, 70), ('1001001', 'SUBJID', 1190, 1197, 72, 73), ('08-Feb-2022', 'DATE', 1286, 1297, 75, 76), ('25-Mar-2022', 'DATE', 1490, 1501, 87, 88), ('07-Apr-2022', 'DATE', 960, 971, 61, 62), ('Neutropenia', 'MH', 2920, 2931, 142, 143), ('Neutropenia', 'MH', 2978, 2989, 144, 145), ('Neutropenia', 'MH', 3567, 3578, 168, 169), ('08-Jul-2022', 'DATE', 1036, 1047, 64, 65), ('08-Jul-2022', 'DATE', 1088, 1099, 67, 68), ('12-Jul-2022', 'DATE', 1152, 1163, 70, 71), ('12-Jul-2022', 'DATE', 1350, 1361, 78, 79), ('12-Jul-2022', 'DATE', 1995, 2006, 109, 110), ('12-Jul-2022', 'DATE', 2108, 2119, 113, 114), ('27-Jul-2022', 'DATE', 1228, 1239, 73, 74), ('27-Jul-2022', 'DATE', 1936, 1947, 107, 108), ('27-Jul-2022', 'DATE', 2292, 2303, 120, 121), ('27-Jul-2022', 'DATE', 3056, 3067, 146, 147), ('00-Jan-1900', 'DATE', 1426, 1437, 84, 85), ('00-Jan-1900', 'DATE', 1813, 1824, 99, 100), ('00-Jan-1900', 'DATE', 2532, 2543, 128, 129), ('00-Jan-1900', 'DATE', 2886, 2897, 140, 141), ('00-Jan-1900', 'DATE', 3218, 3229, 152, 153), ('00-Jan-1900', 'DATE', 3284, 3295, 155, 156), ('00-Jan-1900', 'DATE', 3474, 3485, 164, 167), ('29-Jul-2022', 'DATE', 2054, 2065, 111, 112), ('29-Jul-2022', 'DATE', 2162, 2173, 115, 116), ('1001005', 'SUBJID', 1312, 1319, 77, 78), ('1001005', 'SUBJID', 1388, 1395, 83, 84), ('1001005', 'SUBJID', 1452, 1459, 86, 87), ('14-Jun-2022', 'DATE', 1860, 1871, 101, 102), ('1001006', 'SUBJID', 1516, 1523, 89, 90), ('1001006', 'SUBJID', 1569, 1576, 91, 92), ('1001006', 'SUBJID', 1623, 1630, 93, 94), ('1001006', 'SUBJID', 1677, 1684, 95, 96), ('1001006', 'SUBJID', 1731, 1738, 97, 98), ('anaemia', 'MH', 2208, 2215, 117, 118), ('anaemia', 'MH', 3090, 3097, 148, 149), ('anaemia', 'MH', 3144, 3151, 150, 151), ('anaemia', 'MH', 3318, 3325, 157, 158), ('1001009', 'SUBJID', 1910, 1917, 106, 107), ('1001009', 'SUBJID', 2254, 2261, 119, 120), ('03-Aug-2022', 'DATE', 2472, 2483, 126, 127), ('03-Aug-2022', 'DATE', 2756, 2767, 136, 137), ('03-Aug-2022', 'DATE', 3410, 3421, 161, 164), ('1001010', 'SUBJID', 2318, 2325, 122, 123), ('1001010', 'SUBJID', 2378, 2385, 124, 125), ('1001010', 'SUBJID', 2558, 2565, 130, 131), ('1001010', 'SUBJID', 2617, 2624, 132, 133), ('1001010', 'SUBJID', 2676, 2683, 134, 135), ('1001011', 'SUBJID', 2794, 2801, 138, 139), ('1001011', 'SUBJID', 3244, 3251, 154, 155), ('1001012', 'SUBJID', 3372, 3379, 160, 162), ('1001012', 'SUBJID', 3436, 3443, 163, 165), ('1001013', 'SUBJID', 3629, 3636, 173, 175), ('29-Nov-2022', 'DATE', 3591, 3602, 169, 172), ('03-Nov-2022', 'DATE', 3655, 3666, 174, 176), ('22-Nov-2022', 'DATE', 3667, 3678, 175, 178)]\n", "~~~~~~~ [('1001001', 'SUBJID', 150, 157, 28, 29), ('1001001', 'SUBJID', 208, 215, 30, 31), ('1001001', 'SUBJID', 278, 285, 35, 36), ('1001001', 'SUBJID', 336, 343, 37, 38), ('1001001', 'SUBJID', 394, 401, 39, 40), ('1001001', 'SUBJID', 452, 459, 41, 42), ('1001001', 'SUBJID', 510, 517, 43, 44), ('1001001', 'SUBJID', 568, 575, 45, 46), ('1001001', 'SUBJID', 639, 646, 50, 51), ('1001001', 'SUBJID', 698, 705, 52, 53), ('1001001', 'SUBJID', 757, 764, 54, 55), ('1001001', 'SUBJID', 816, 823, 56, 57), ('1001001', 'SUBJID', 875, 882, 58, 59), ('1001001', 'SUBJID', 934, 941, 60, 61), ('07-Apr-2022', 'DATE', 960, 971, 61, 62), ('1001001', 'SUBJID', 998, 1005, 63, 64), ('08-Jul-2022', 'DATE', 1036, 1047, 64, 65), ('1001001', 'SUBJID', 1062, 1069, 66, 67), ('08-Jul-2022', 'DATE', 1088, 1099, 67, 68), ('1001001', 'SUBJID', 1126, 1133, 69, 70), ('12-Jul-2022', 'DATE', 1152, 1163, 70, 71), ('1001001', 'SUBJID', 1190, 1197, 72, 73), ('27-Jul-2022', 'DATE', 1228, 1239, 73, 74), ('08-Feb-2022', 'DATE', 1286, 1297, 75, 76), ('1001005', 'SUBJID', 1312, 1319, 77, 78), ('12-Jul-2022', 'DATE', 1350, 1361, 78, 79), ('1001005', 'SUBJID', 1388, 1395, 83, 84), ('00-Jan-1900', 'DATE', 1426, 1437, 84, 85), ('1001005', 'SUBJID', 1452, 1459, 86, 87), ('25-Mar-2022', 'DATE', 1490, 1501, 87, 88), ('1001006', 'SUBJID', 1516, 1523, 89, 90), ('1001006', 'SUBJID', 1569, 1576, 91, 92), ('1001006', 'SUBJID', 1623, 1630, 93, 94), ('1001006', 'SUBJID', 1677, 1684, 95, 96), ('1001006', 'SUBJID', 1731, 1738, 97, 98), ('00-Jan-1900', 'DATE', 1813, 1824, 99, 100), ('14-Jun-2022', 'DATE', 1860, 1871, 101, 102), ('1001009', 'SUBJID', 1910, 1917, 106, 107), ('27-Jul-2022', 'DATE', 1936, 1947, 107, 108), ('12-Jul-2022', 'DATE', 1995, 2006, 109, 110), ('29-Jul-2022', 'DATE', 2054, 2065, 111, 112), ('12-Jul-2022', 'DATE', 2108, 2119, 113, 114), ('29-Jul-2022', 'DATE', 2162, 2173, 115, 116), ('anaemia', 'MH', 2208, 2215, 117, 118), ('1001009', 'SUBJID', 2254, 2261, 119, 120), ('27-Jul-2022', 'DATE', 2292, 2303, 120, 121), ('1001010', 'SUBJID', 2318, 2325, 122, 123), ('1001010', 'SUBJID', 2378, 2385, 124, 125), ('03-Aug-2022', 'DATE', 2472, 2483, 126, 127), ('00-Jan-1900', 'DATE', 2532, 2543, 128, 129), ('1001010', 'SUBJID', 2558, 2565, 130, 131), ('1001010', 'SUBJID', 2617, 2624, 132, 133), ('1001010', 'SUBJID', 2676, 2683, 134, 135), ('03-Aug-2022', 'DATE', 2756, 2767, 136, 137), ('1001011', 'SUBJID', 2794, 2801, 138, 139), ('00-Jan-1900', 'DATE', 2886, 2897, 140, 141), ('Neutropenia', 'MH', 2920, 2931, 142, 143), ('Neutropenia', 'MH', 2978, 2989, 144, 145), ('27-Jul-2022', 'DATE', 3056, 3067, 146, 147), ('anaemia', 'MH', 3090, 3097, 148, 149), ('anaemia', 'MH', 3144, 3151, 150, 151), ('00-Jan-1900', 'DATE', 3218, 3229, 152, 153), ('1001011', 'SUBJID', 3244, 3251, 154, 155), ('00-Jan-1900', 'DATE', 3284, 3295, 155, 156), ('anaemia', 'MH', 3318, 3325, 157, 158), ('1001012', 'SUBJID', 3372, 3379, 160, 162), ('03-Aug-2022', 'DATE', 3410, 3421, 161, 164), ('1001012', 'SUBJID', 3436, 3443, 163, 165), ('00-Jan-1900', 'DATE', 3474, 3485, 164, 167), ('Neutropenia', 'MH', 3567, 3578, 168, 169), ('29-Nov-2022', 'DATE', 3591, 3602, 169, 172), ('1001013', 'SUBJID', 3629, 3636, 173, 175), ('03-Nov-2022', 'DATE', 3655, 3666, 174, 176), ('22-Nov-2022', 'DATE', 3667, 3678, 175, 178)]\n", "[('Data has been updated by site post iDBL for AE, CM and MH page and was in scope for iDBL-week 24.\\nBelow is the details for records updated post iDBL\\n\\n', 'SUBJID'), ('-Neutropenia-08-Feb-2022-24-Feb-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-25-Mar-2022-07-Apr-2022-End date was wrong in EDC\\n', 'SUBJID'), ('-Neutropenia-02-May-2022-31-May-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-31-May-2022-08-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-08-Jul-2022-12-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-12-Jul-2022-27-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Neutropenia-27-Jul-2022-00-Jan-1900-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-08-Feb-2022-07-Apr-2022-End date was wrong in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-07-Apr-2022-31-May-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-08-Jul-2022-12-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-12-Jul-2022-19-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-19-Jul-2022-27-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-29-Jul-2022-00-Jan-1900-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('-08-Jun-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -16-Jun-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('-12-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('-19-Jul-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -19-Jul-2022-', 'DATE'), ('-Missed in EDC\\n1001002-Neutropenia-11-Jan-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -14-Jun-2022-', 'DATE'), ('-End date was wrong in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -07-Sep-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -10-Mar-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-anaemia-14-Feb-2022-12-Apr-2022-Missed in MH\\n', 'SUBJID'), ('-anaemia-12-Apr-2022-10-May-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-10-May-2023-24-May-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-24-May-2023-07-Jun-2023-Missed in EDC\\n', 'SUBJID'), ('-anaemia-07-Jun-2023-14-Feb-2023-Missed in EDC\\n1001006-anaemia-09-May-2023-', 'DATE'), ('-Missed in EDC\\n1001009-leukopeniaÂ\\xa0-', 'DATE'), ('-12-Jul-2023-End date was wrong in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('-03-Aug-2022-Missed in EDC\\n1001009-leukopeniaÂ\\xa0-', 'DATE'), ('-27-Jul-2022-Missed in EDC\\n1001009-leukopeniaÂ\\xa0-', 'DATE'), ('-03-Aug-2022-Missed in EDC\\n1001009-anaemia-', 'DATE'), ('-13-Jul-2022-Missed in EDC\\n1001009-anaemia-', 'DATE'), ('-30-Jul-2022-Missed in EDC\\n1001009-', 'MH'), ('-03-Aug-2022-10-Aug-2022-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -13-Jul-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-hyperuricemia-12-May-2022-24-May-2022-Missed in EDC\\n', 'SUBJID'), ('-hyperuricemia-07-Jun-2022-05-Jul-2022-Missed in EDC\\n1001010-hyperuricemia-09-Jul-2022-', 'DATE'), ('-Missed in EDC\\n1001010-hyperuricemia-12-Aug-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-17-Jun-2022-20-Jun-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-23-Jun-2022-27-Jun-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-27-Jun-2022-28-Jun-2022-Missed in EDC\\n1001010-leukopeniaÂ\\xa0-', 'DATE'), ('-05-Aug-2022-Missed in EDC\\n', 'SUBJID'), ('-leukopeniaÂ\\xa0-14-Aug-2022-16-Aug-2022-Missed in EDC\\n1001011-leukopeniaÂ\\xa0-30-Aug-2022-', 'DATE'), ('-Missed in EDC\\n1001011-', 'MH'), ('-30-Aug-2022-14-Sep-2022-Missed in EDC\\n1001011-', 'MH'), ('-14-Sep-2022-00-Jan-1900-Missed in EDC\\n1001011-anaemia-28-Jun-2022-', 'DATE'), ('-Missed in EDC\\n1001011-', 'MH'), ('-28-Jul-2022-09-Aug-2022-Missed in EDC\\n1001011-', 'MH'), ('-30-Aug-2022-01-Sep-2022-Missed in EDC\\n1001011-anaemia-01-Sep-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Total Cholesteremia-05-Aug-2022-', 'DATE'), ('-Missed in EDC\\n1001012-', 'MH'), ('-03-Aug-2022- -Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -05-Jul-2022-', 'DATE'), ('-Missed in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -03-Aug-2022-', 'DATE'), ('-Missed in EDC\\n1001013-leukopeniaÂ\\xa0-29-Nov-2022-00-Jan-1900-Missed in EDC\\n1001013-', 'MH'), ('-08-Nov-2022-', 'DATE'), ('-End date was wrong in EDC\\n', 'SUBJID'), ('-Thrombocytopenia -', 'DATE'), ('-', 'DATE'), ('-End date was wrong in EDC may lead to Source data was not reviewed to update the change in toxicity grade in AE or dates in AE during iDBL-week 24 resulting in Study team needs to access the output and TFLS impacted due to change in data post iDBL.\\n\\n1. CTT need to assess whether this data is associated with primary or secondary end point analysis, if yes, whether this will have an impact to CSR outputs or CSR TFLS.\\n2. CTT needs to consider this as a critical data for iDBL:- If yes, need to be documented in Post lock issue form even if not going to have any impact in CSR\\n3. Whether this data is a critical safety data, if yes, any safety outputs are impacted:- then need a Post lock issue form.\\n\\nAE related output impacted due to data change or update', '')]\n", "\n", "\n", "\n", "----- [('2nd August 2023', 'DATE', 126, 141, 20, 23), ('24Aug2023', 'DATE', 254, 263, 41, 42)]\n", "~~~~~~~ [('2nd August 2023', 'DATE', 126, 141, 20, 23), ('24Aug2023', 'DATE', 254, 263, 41, 42)]\n", "[('Data transfer delays to Kayentis portal and the CF version 3.22 embedded improvements to solve this situation was released on ', 'DATE'), ('. When the device was updated with the new version all the data are successfully sent to the portal/database.\\nOn ', 'DATE'), (' Kayentis support team checked all the data flows received to version 3.22 and duplicate questionnaires have been uploaded for some patients/sessions impacted.\\nDuplicates were excluded during the second attempt, however, this second attempt was the one transferred initially, now we transferred the first attempt completed by the patient (and delayed to the portal) with all answers present that why the answers are not the same between the data transfer. may lead to Kayentis UPDD and AAS Diary data has been updated (i.e. new records, deleted records) since the IA concerning data prior to the IA. resulting in The IA outputs including UAS7 and AAS and number of confirmed protocol deviations', '')]\n", "\n", "\n", "\n", "----- [('2nd August 2023', 'DATE', 126, 141, 20, 23), ('24Aug2023', 'DATE', 254, 263, 42, 43)]\n", "~~~~~~~ [('2nd August 2023', 'DATE', 126, 141, 20, 23), ('24Aug2023', 'DATE', 254, 263, 42, 43)]\n", "[('Data transfer delays to Kayentis portal and the CF version 3.22 embedded improvements to solve this situation was released on ', 'DATE'), ('. When the device was updated with the new version all the data are successfully sent to the portal/database. On ', 'DATE'), (' Kayentis support team checked all the data flows received to version 3.22 and duplicate questionnaires have been uploaded for some patients/sessions impacted. Duplicates were excluded during the second attempt, however, this second attempt was the one transferred initially, now we transferred the first attempt completed by the patient (and delayed to the portal) with all answers present that why the answers are not the same between the data transfer. led to Kayentis UPDD and AAS Diary data has been updated (i.e. new records, deleted records) since the IA concerning data prior to the IA. resulting in The IA outputs including UAS7 and AAS and number of confirmed protocol deviations', '')]\n", "\n", "\n", "\n", "----- [('19-Sep-2023', 'DATE', 376, 387, 67, 69), ('28-Oct-2023', 'DATE', 411, 422, 73, 74), ('26-Oct-2023', 'DATE', 472, 483, 82, 83)]\n", "~~~~~~~ [('19-Sep-2023', 'DATE', 376, 387, 67, 69), ('28-Oct-2023', 'DATE', 411, 422, 73, 74), ('26-Oct-2023', 'DATE', 472, 483, 82, 83)]\n", "[('Database updates became required due to Global PPC and new library release in Sep in addition to PPC for PA#02. Combining both global PPC and change requests per PA#02 increased the volume of the change of the database. Also, there is lack of resources for CF programmers due to global PPC which delayed the start of the database updates. may lead to The PA was finalized on ', 'DATE'), (' and FPFV is planned on ', 'DATE'), ('. resulting in Delays in achieving PPC go-live as ', 'DATE'), ('', '')]\n", "\n", "\n", "\n", "----- [('Oct 2022', 'DATE', 38, 46, 7, 9)]\n", "~~~~~~~ [('Oct 2022', 'DATE', 38, 46, 7, 9)]\n", "[('Death of DMC member Joseph Massaro in ', 'DATE'), (', database cleaning activities for transition of EDC from OCR-DC to RAVE EDC (project Rainbow) and delay in DBL for correction of the primary CSR and DMC #9 led to delay in DMC meeting #9 for more than 1 year (per DMC charter meetings should occur every 6 months) resulting in delayed update of the status of the study provided to the DMC members.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay in confirmation of the country list (i.e., breakdown of countries for Phase II and/or Phase III), FPP requirements, and FPP/FSIV priority may lead to inability to accurately rank priorities of translation work resulting in rush timelines fees being applied to countries that will not need them, inability to meet FPP submission, and/or delay in the availability of languages on the eCOA device.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay in fulfilling pre-requisites for Standard requests and complex dynamics and branching of CRF. may lead to delay in timelines resulting in Resulting in overall database go live .', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay in PK reconciliation may lead to PK samples that are not reconciled in PreAdvance cannot be analyzed. Because the reconciliation issues are solved too late, PK samples are reconciled beyond their validated stability period and are therefore out of stability. resulting in Results of some samples are not valid and therefore cannot be delivered', '')]\n", "\n", "\n", "\n", "----- [('Week 1 Day 1/Enrollment date', 'PRECLINICAL STUDY DAY', 18, 46, 3, 8)]\n", "~~~~~~~ [('Week 1 Day 1/Enrollment date', 'PRECLINICAL STUDY DAY', 18, 46, 3, 8)]\n", "[('Delay in updating ', 'PRECLINICAL STUDY DAY'), (' or miscalculation of expected visits by human error may lead to missing or incorrect update of IRT system information resulting in miscalculation of IMP supply for study patients and protocol noncompliance.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay in uploading site documents in CREDI for USA. led to CSR Appendix 16.1.4 cannot be finalize on time. resulting in Dealy in PubCSR.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay of biomarker assay validation (qPCR and ELISA) at Hidalgo may lead to search a new assay. resulting in resulting in delay of data delivery timeline', '')]\n", "\n", "\n", "\n", "----- [('Turkey', 'COUNTRY', 29, 35, 6, 7)]\n", "~~~~~~~ [('Turkey', 'COUNTRY', 29, 35, 6, 7)]\n", "[('Delay of new IMP delivery to ', 'COUNTRY'), (' may lead to late resupply to sites resulting in the additional site not being able to be opened and IMP not available for potential new patients.', '')]\n", "\n", "\n", "\n", "----- [('21-Feb-2024', 'DATE', 130, 141, 24, 25)]\n", "~~~~~~~ [('21-Feb-2024', 'DATE', 130, 141, 24, 25)]\n", "[('Delay of packaging campaign for new IMP kits needed to replace the expiring batch (Shelf life, end of Feb 2024, good to use until ', 'DATE'), ('). may lead to Tight timelines for global replacement of expiring batch. resulting in Drug Supply Continuity at risk if distribution/production is further delayed', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delay of the relabeling of LCZ696 granules and Ora-sweet from E1 label to E2 label may lead to delay of the delivery of LCZ696 granules and Ora-sweet into the sites resulting in delay of the administration of LCZ696 granules and LCZ696 suspension to the specific patients', '')]\n", "\n", "\n", "\n", "----- [('1 month delay', 'DATE', 169, 182, 24, 28)]\n", "~~~~~~~ [('1 month delay', 'DATE', 169, 182, 24, 28)]\n", "[('Delayed delivery of commercial matrices for selectivity analysis supporting PK/PD analysis at SGS led to Rescheduling of analysis slot for the selectivity analysis and ', 'DATE'), (' of the delivery of PK/PD data by SGS. resulting in Risk of delaying CSR timelines beyond the LPLV + 9 mon', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delayed site / country budget (TCF) approval may lead to delay setting up clinical trial agreement with site resulting in delayed SIV and FPFV', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delays in drug supply may lead to IMP not on site in time for planned FPFT resulting in eligible participants unnecessary discontinued, require rescreen', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Delays with bioanalytical analysis led to published CSR exceeding 11-months post LPLV resulting in a deviation from SOP-8109242.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Development of Phase 2 (Open label) and Phase 3 (Double blinded) portions of the study in one database prior to start of Phase 2 study may lead to Phase 3-specific CRF pages being available to sites if a patient is incorrectly documented as a Phase 3 patient on the C1D1 CRF page. resulting in data entry for a Phase 2 patient on Phase 3 CRFs,', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('DM is sending the metrics whereas it has been not followed with PI for their signoff. led to Non-Compliance to the above said CAPA Commitment and possible quality issues during Audit. resulting in Non-Compliance to the above said CAPA Commitment and possible quality issues during Audit', '')]\n", "\n", "\n", "\n", "----- [('Aug 2023', 'DATE', 415, 423, 75, 78)]\n", "~~~~~~~ [('Aug 2023', 'DATE', 415, 423, 75, 78)]\n", "[('DMC requested at DMC #8 to provide follow up information of an AE of \"Acute Myeloid Leukemia\" experienced by a patient that DMC would consider as SAE. Delay in DMC meeting #9 setup, prompt action not taken right after DMC #8 and study handover activities did not involve open/unresolved action items from DMC #8 led to delay in action taken on the open actions from the DMC #8 and closure of the concerned site in ', 'DATE'), (' resulting in querying in EDC not possible anymore, FU information cannot be provided to the DMC.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('DropZone uploaded documents to Subway in incorrect placeholders may lead to Identified misfiled documents in Subway that did not come up as queries in TMF review resulting in Decreased TMF compliance, potentially more issue that did not come up during TMF review may be present in TMF', '')]\n", "\n", "\n", "\n", "----- [('LTT462', 'SUBJID', 24, 30, 3, 4), ('4001', 'Patient Related SiteID', 51, 55, 8, 9)]\n", "~~~~~~~ [('LTT462', 'SUBJID', 24, 30, 3, 4), ('4001', 'Patient Related SiteID', 51, 55, 8, 9)]\n", "[('Drug accountability log ', 'SUBJID'), (' was missing at site ', 'Patient Related SiteID'), (' for one patient led to Unknown drug batch information for this patient resulting in Missing data in the interim CSR in appendix 16.1.6 Listing of subjects receiving test drugs / investigational product from specific batches', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Dynamic competitive landscape for KRAS in 1st line NSCLC may lead to If competitors present positive data for KRAS in 1st line, our control arm for phase III may become obsolete. resulting in ENrollment may become more difficult or drop-out rates may increase.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Dynamic competitive landscape for KRAS in 1st line NSCLC may lead to Our current control arm in phase III may become obsolete if competitors present positive data that changes the standard of care 1st line treatment for KRASmut patients. resulting in We may have increased drop-out rates or have a negative impact on enrollment if our control arm becomes obsolete.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Error in the generation of the RRR form led to From not meeting the requirement and triggering a risk to share unnecessary unblinding data and preventing LSH lock resulting in Form to be update by stat impacting IA delivery timlines', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Error in the generation of the RRR form in the frame of the IA may lead to Unblinded data unnecessary for the IA transferred to an unrestricted domain of LSH resulting in Unblinding team members', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Errors in review of x ray eligibility criteria and QC of MRI imaging (1QEM: DEV-63811) may lead to Higher variability in efficacy measurements resulting in Treatment effects may not be clearly discernible from data', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Extending the shelf life of iptacopan compound and relaxing the storage conditions led to delaying the CMC package readiness resulting in submission delayed by 2-3 months', '')]\n", "\n", "\n", "\n", "----- [('Belgium', 'COUNTRY', 30, 37, 3, 4), ('Belgium', 'COUNTRY', 71, 78, 9, 10)]\n", "~~~~~~~ [('Belgium', 'COUNTRY', 30, 37, 3, 4), ('Belgium', 'COUNTRY', 71, 78, 9, 10)]\n", "[('Following GCO transformation, ', 'COUNTRY'), (' Country resource was reduced (as ', 'COUNTRY'), (' is a satellite country in Project Ambition) led to the departing Country Study Manager did not have resource to address queries reported by TMF QC Manager resulting in 36 unresolved TMF document queries at country (12) and site (24) level at end of trial.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('For the upcoming interim analysis site staff, local teams, vendors and Novartis team members unexpected be out of the office may lead to delays in query responses, data entries, PI signatures and ad-hoc vendor data transfers resulting in delayed iDBL.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Frequent Lu production issue may lead to incapability of providing Lu for the patients resulting in reputation risk of Novartis Japan', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('GCO reorganization and recent team members changes may lead to a delay to follow-up ongoing actions and new one (investigator meeting set-up, date, timelines) resulting in resulting in low reactivity and miss-communication within the team and could have an impact on quality, budget and timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('GCO reorganization and recent team members changes may lead to a delay to follow-up ongoing actions and new one (setting up new CRAs training refresher and investigator meeting for end of recruitment) resulting in low reactivity and miss-communication within the team and could have an impact on quality.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('GCO reorganization and recent team members changes may lead to a delay to follow-up ongoing actions such as LNR and query management, either in clinical or vendor databases resulting in low reactivity and miscommunication within the team and could have an impact on quality, budget and overall primary analysis timelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical instability in Israel and surrounding countries may lead to disruptions in business continuity (e.g. supply chain management, patient safety, patient well-being, data integrity) resulting in potential trial delivery delay.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Geopolitical situation in ISR and neighboring countries. may lead to Challenges for IMP distribution due to ongoing military conflict in Middle East. resulting in Delays in distribution expected for LBN.', '')]\n", "\n", "\n", "\n", "----- [('Guatemala', 'COUNTRY', 0, 9, 0, 1), ('Guatemala', 'COUNTRY', 42, 51, 7, 8)]\n", "~~~~~~~ [('Guatemala', 'COUNTRY', 0, 9, 0, 1), ('Guatemala', 'COUNTRY', 42, 51, 7, 8)]\n", "[('', 'COUNTRY'), (' - Due to political situation in ', 'COUNTRY'), (' the roads are blocked and remote sites remote from the capital cannot ship samples to the airport to go to the central lab which led to 3 patients randomized samples not being shipped to central lab resulting in loss of baseline values for these 3 patients.', '')]\n", "\n", "\n", "\n", "----- [('various', 'COUNTRY', 25, 32, 5, 6), ('various', 'COUNTRY', 111, 118, 20, 22)]\n", "~~~~~~~ [('various', 'COUNTRY', 25, 32, 5, 6), ('various', 'COUNTRY', 111, 118, 20, 22)]\n", "[('High turnover of CRAs in ', 'COUNTRY'), (' countries are resulting in delayed SIVs may lead to High turnover of CRAs in ', 'COUNTRY'), (' countries are resulting in delayed SIVs resulting in Delayed SIVs', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Holiday period in December may lead to delay in LPLV resulting in delay in iDBL timelines', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('If central lab receives only backup PK sample and not the primary, the back-up sample will not be automatically promoted to primary may lead to the back-up sample will not be shipped to BA lab resulting in risk of missing data', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('If the site uses an incorrect kit for B cell subset may lead to the flagging not being properly reported resulting in unblinding a patient', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Implementing ePRO interview mode for the entire study may lead to patient's local language differing from the site staff/interviewer's language resulting in difficulty for the site staff to read the interview assessments aloud to the patient and/or for the patient to understand the question being asked\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('In accordance with SOP-7035752, Non-important PDs are reviewed at a minimum every six month during the regular data review, data quality or CTT (or equivalent) meetings. Additionally, it specifies that at study start-up and as required during study conduct, identify Non-important PDs relevant for the study which will undergo periodic review. Please see the attached SOP\\'s section 3.3.4.2, \"Non-important PD Periodic Review,\" for further information.\\n\\nEven though reviewing the Non-important PDs was a continuing study activity, they had previously been presented in DQT meetings to determine if any trends had been noticed or if any Non-important PDs had met the requirements for an Important PD but not creating the non-important PDs as per the in SOP-7035752 led to not having filed the documentation of the non-important PD listing as well as the review which resulting in a minor deviation. This is going to be added to 1-QEM.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Inadequate informed consent process may lead to subjects not being consented properly and insufficient documentation of informed consent process in source documents resulting in breach of GCP and subject's rights, and potential audit/inspection findings.\", '')]\n", "\n", "\n", "\n", "----- [('7002', 'Patient Related SiteID', 64, 68, 11, 12), ('7002', 'Patient Related SiteID', 101, 105, 19, 21)]\n", "~~~~~~~ [('7002', 'Patient Related SiteID', 64, 68, 11, 12), ('7002', 'Patient Related SiteID', 101, 105, 19, 21)]\n", "[('Inconsistent set up of flex coil or patient positioning by site ', 'Patient Related SiteID'), (' may lead to MRI data from site ', 'Patient Related SiteID'), (' may have inconsistent cartilage measurements resulting in This may lead to higher variability in MRI data', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect conversion factors and unit representation issues have been identified in the Global Reference Metadata led to incorrectly converted International System of Units (SI) values and/or incorrect units for impacted parameters resulting in incorrect values reported in datasets, Tables, Figures, Listings (TFLs) and Clinical Study Reports (CSRs) (Quality Issue Ref #DEV-11040, study-level Impact assessment filed within Quality Issue).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect conversion factors and unit representation issues have been identified in the Global Reference Metadata led to incorrectly converted International System of Units (SI) values and/or incorrect units for impacted parameters resulting in incorrect values reported in datasets, Tables, Figures, Listings (TFLs) and Clinical Study Reports (CSRs) (Quality Issue Ref #DEV-11040, study-level Impact assessment filed within Quality Issue).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect conversion factors and unit representation issues have been identified in the Global Reference Metadata led to incorrectly converted International System of Units (SI) values and/or incorrect units for impacted parameters resulting in incorrect values reported in datasets, Tables, Figures, Listings (TFLs) and Clinical Study Reports (CSRs) (Quality Issue Ref #DEV-11040, study-level Impact assessment filed within Quality Issue).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect conversion factors and unit representation issues have been identified in the Global Reference Metadata led to incorrectly converted International System of Units (SI) values and/or incorrect units for impacted parameters resulting in incorrect values reported in datasets, Tables, Figures, Listings (TFLs) and Clinical Study Reports (CSRs) (Quality Issue Ref #DEV-11040, study-level Impact assessment filed within Quality Issue).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect conversion factors and unit representation issues have been identified in the Global Reference Metadata led to incorrectly converted International System of Units (SI) values and/or incorrect units for impacted parameters resulting in incorrect values reported in datasets, Tables, Figures, Listings (TFLs) and Clinical Study Reports (CSRs) (Quality Issue Ref #DEV-11040, study-level Impact assessment filed within Quality Issue).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Incorrect SDTM annotations are provided for CRF - Provocation test - Fric test, Temp test, Pulse control ergometry test ,Ice cube test, Evidence of Urticaria led to Delay in CRF review and finalization. resulting in A potential delay in screen building, OSRM and screen UAT resulting in database go live timelines.', '')]\n", "\n", "\n", "\n", "----- [('AAA617 production', 'MH', 23, 40, 3, 5)]\n", "~~~~~~~ [('AAA617 production', 'MH', 23, 40, 3, 5)]\n", "[('Inexperience to handle ', 'MH'), (' at scale by Indianapolis Manufacturing site may lead to not being able to supply clinical doses in time and with quality due to production, logistics, distribution issues resulting in patients underdosing', '')]\n", "\n", "\n", "\n", "----- [('22-Feb-2023', 'DATE', 147, 158, 28, 29)]\n", "~~~~~~~ [('22-Feb-2023', 'DATE', 147, 158, 28, 29)]\n", "[('Initially BA lab was planned to be at ICON NL. However, the BA lab was changed to Veeda(India). This was notified to CTT and captured in MoM dated ', 'DATE'), ('. led to The central lab(Q2) sent PK Irinotecan samples to BA lab, ICON NL instead of Veeda(India). resulting in Q2 inadvertently sent Irinotecan PK samples to BA lab ICON NL instead of BA lab Veeda(India).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Initiation of CAAA617D12302 study might be delayed. Ga generator shipment plan might be needed to change because CAAA501A11301 study and CAAA617D12302 study will select same clinical sites and share the Ga generator. Ga generator might be prepared only CAAA501A11301 study. may lead to CAAA501A11301 study will prepare 9 Ga generators because replace of Ga generator will be plannded by CAAA617D12302 study. If CAAA617D12302 study will be delay, AAA501A11301 study needs to prepare up to 18 Ga generators. resulting in Need to update the budget. Approximatory $100,000/Ga ganerator so budget will be increasing up to $900,000.', '')]\n", "\n", "\n", "\n", "----- [('Dr. Choo', 'NAME', 77, 85, 11, 14), ('Dr. Choo', 'NAME', 199, 207, 29, 32)]\n", "~~~~~~~ [('Dr. Choo', 'NAME', 77, 85, 11, 14), ('Dr. Choo', 'NAME', 199, 207, 29, 32)]\n", "[('Insufficient contract oversight led to expiration of contract for DMC Chair ', 'NAME'), (' resulting in performing monthly CMV related data review without valid contract. Based on written confirmation of ', 'NAME'), (' he continued to adhere to the original contract and no confidential data were disclosed.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Insufficient training at the site level and/or lack of attention to instructions provided in the central lab manual. led to The bioanalytical assay to determine the free payload of SDZ475 and RFZ126 is validated in whole blood ONLY. By mistake and lack of attention 45 blood samples (out of 2496 that have been sent to the analysis labs) have been wrongly processed into plasma and therefore not suitable to be analyzed for this PK assay. resulting in As a result, these samples cannot be analyzed which directly impacts the secondary objective for this trial which is to characterize the pharmacokinetics of DYP688.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('IRT Go Live date delayed up to 13NOV23, compare to previous target may lead to Drug supply difficulty to ship treatment in US for FPFV at stretch in December 23 to US (standard of 6 weeks for shipment post IRT Go live) resulting in Delay FPFV in US (target mid DEC 23)', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Kayentis tablet is not synchronizing when patients are entering the study and diary scores are recorded but not available for calculations. led to No UAS7 score could be calculated for these patients resulting in A relapse UAS7 score could not be calculated and patients could be delayed in entering re-treatment.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Lab Manual Not finalized may lead to Delay in DTS Development resulting in Delay in DTS Finalization and ontime Production data in house', '')]\n", "\n", "\n", "\n", "----- [('samples not being able to be analyzed', 'MH', 261, 298, 42, 50)]\n", "~~~~~~~ [('samples not being able to be analyzed', 'MH', 261, 298, 42, 50)]\n", "[('Lack of attention to sample collection instructions in the central lab manual. may lead to Blood samples were wrongly processed into plasma, making them not suitable to analyze for the free payload assay to measure SDZ475 and RFZ126. resulting in Resulting in ', 'MH'), (' to support the secondary objective for this trial.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Lack of compliance with CRO Risk Assessment process may lead to ineffective clinical trial risk management activities resulting in lack of sponsor oversight, non-compliance with ICH GCP E6 R2 guidelines.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Lack of understanding of the Oral Food Challenge requirements of the protocol and first administration of study medication led to Led to treatment being administered to subjects <24 hours after OFC resulting in Led to non compliance of the protocol and therefore a protocol deviation', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Late or incomplete completion of vendor trackers by countries may lead to late activation of vendor portal (i.e. IRT) or late supply shipment or late site qualification resulting in delayed SIVs and recruitment', '')]\n", "\n", "\n", "\n", "----- [('Jordan', 'COUNTRY', 39, 45, 7, 8)]\n", "~~~~~~~ [('Jordan', 'COUNTRY', 39, 45, 7, 8)]\n", "[('Later than expected delivery of IMP to ', 'COUNTRY'), (' may lead to site not resupplied on time resulting in potential unavailability of drug for an enrolled patient', '')]\n", "\n", "\n", "\n", "----- [('30-Apr-2023', 'DATE', 147, 158, 22, 23), ('IMP delivery', 'MH', 82, 94, 11, 14)]\n", "~~~~~~~ [('IMP delivery', 'MH', 82, 94, 11, 14), ('30-Apr-2023', 'DATE', 147, 158, 22, 23)]\n", "[('Lengthy delivery process to Argentina, including custom clearance led to delayed ', 'MH'), (' to local depot, required to replace IMP expiring on ', 'DATE'), (', resulting in late release of IMP for replacement site shipments', '')]\n", "\n", "\n", "\n", "----- [('Sept 2023', 'DATE', 51, 60, 8, 10)]\n", "~~~~~~~ [('Sept 2023', 'DATE', 51, 60, 8, 10)]\n", "[('Limited amount of IMP at study enrollment restart (', 'DATE'), (') may lead to issues if one site is an over-recruiter resulting in in another site not having adequate IMP to screen a patient and may risk diversity of patients coming from one geographical region.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('LNP protocol does allow background and rescue medications for gMG patients. This was reviewed by EU-CTR SME and determined that due to variable background and rescue therapies that could be used in the trial, AXMPs was marked as Not applicable in the protocol may lead to EU CTR comments may requireÂ\\xa0background therapy and/or rescue therapy to be identified asÂ\\xa0AxMPsÂ\\xa0in study-related documentation resulting in May lead to protocol amendment and delay in FPFV timelines in EU region', '')]\n", "\n", "\n", "\n", "----- [('2- 6th Oct', 'DATE', 27, 37, 5, 8)]\n", "~~~~~~~ [('2- 6th Oct', 'DATE', 27, 37, 5, 8)]\n", "[('local holidays in China on ', 'DATE'), (' may delay the resolution of activities (closure of DCRs, actions to close out ePRO devices) may lead to delayed query resolution resulting in delay to DBL', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Logistical challenges at sites led to sites not processing newly obtained tumor biopsies as per ICON Lab Manual resulting in receipt of FFPE tumor blocks instead of fresh tumor samples preserved in ethanol (preferred).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Logistical issues for participants attending site visits may lead to additional costs to the study due to interviews being rescheduled at last minute resulting in additional costs for the study', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('mandatory vaccination against encapsulated bacteria while patients are receiving B cell depeletor agent as SoC RTX may lead to Lack of efficacy of the vaccination resulting in increase cases of infections with encapsulated bacteria', '')]\n", "\n", "\n", "\n", "----- [('MGH/Prof.', 'NAME', 0, 9, 0, 1)]\n", "~~~~~~~ [('MGH/Prof.', 'NAME', 0, 9, 0, 1)]\n", "[('', 'NAME'), (' Kuter (chair of the SSC for this trial): Site is going through an internal review process regarding resources led to they would like to circle back in ~2 months to confirm they have the resources available to take this study and commit on startup timelines resulting in high risk for delaying SIV planned in October 2023 and FPFV by end of 2023, and in worst case, the assessment would confirm that this site does not have the resources to take this additional study', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Misalignment between Study teams and NIBR TMF on study timelines regarding first TMF review led to Delay in performing first TMF review as per timelines described in WP-7006654 resulting in Non compliance with WP but no impact on the TMF quality/completeness as the study is not in an advanced stage. TMF review will be finalized in the next days.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Missing assignment of the RAINBOW training to the sites led to site's staff not being able to access RAVE resulting in delayed PI signature and data entry\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Missing PI signatures which need to be re-sign after transitioned to RAVE is getting delayed. may lead to Trial is a part of Rainbow project. In the process all the pages which were signed in OC database need to be re-sign again after transitioned to Rave. However, some of the sites are not able to provide the signatures in time. resulting in Delayed in Data entries since 3 months after Go -live which will lead to delayed in data review and this would definitely impact the Final DBL', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Missing seasonal Chagas disease due to trial site contract delays may lead to inability to recruit patients resulting in not meeting patient enrollment goals and delay in timelines.', '')]\n", "\n", "\n", "\n", "----- [('1059', 'Patient Related SiteID', 139, 143, 22, 24)]\n", "~~~~~~~ [('1059', 'Patient Related SiteID', 139, 143, 22, 24)]\n", "[('More than 150 Mutiple queries present including 3rd party data queries. may lead to Outstanding queries under site review present at site ', 'Patient Related SiteID'), (' resulting in Final DBL timelines could delay if not resolved.', '')]\n", "\n", "\n", "\n", "----- [('Q4 2023', 'DATE', 60, 67, 9, 12), ('Jan 2024', 'DATE', 120, 128, 19, 20)]\n", "~~~~~~~ [('Q4 2023', 'DATE', 60, 67, 9, 12), ('Jan 2024', 'DATE', 120, 128, 19, 20)]\n", "[('NCV migration timeline may lead to NCV migration occurring ', 'DATE'), (' resulting in May impact RIGHT Choice CSR submission ', 'DATE'), ('', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('NEN patients will not agree additonal PK blood sampling. may lead to PK analysis will not able to assess using NEN patients samples due to PK samples will not collected from 4-6 patinents. resulting in Middle', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('new COVID waves in all regions led to Will lead to reduced screening activities resulting in Possibly will affect the goal of reaching 44 randomized patients by end Q1 2021', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('New EU CTR process with extended TAT of about 8-9 from FPP to FSI may lead to delay the start of the trial in European countries. resulting in Less time for EU countries to enroll patients and hence extend the overall trial duration', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('New ICON PM, ICON RA Lead and CRA are all relatively new to ICON and in their respective roles may lead to Mistakes or overlooked tasks resulting in Timeline delays or compromised quality.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('New indication at NBR with no penetration of Novartis at sites may lead to To extended site feasibility with re-assessing commitment from countries resulting in delaying the country allocation', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('No TMF RACI is available for studies supported by TMF Ops services may lead to Unclear roles and responsibilities of TMF ops services resulting in May result in incomplete activities and overall lack of inspection readiness', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Not following the current guidelines recommendation with fixed schedule of RTX every 6 mo after induction of remission may lead to High drop out rate after remission induction or having PIs not following our protocol and systematically providing RTX at 6mo resulting in non-significant difference between the 2 arms hence not reaching our primary endpoint.', '')]\n", "\n", "\n", "\n", "----- [('January 2024', 'DATE', 119, 131, 20, 22), ('01-Dec-2023', 'DATE', 446, 457, 69, 70), ('08-Dec-2023', 'DATE', 511, 522, 80, 81), ('13-Dec-2023', 'DATE', 567, 578, 88, 89), ('08-Jan-2024', 'DATE', 638, 649, 97, 99)]\n", "~~~~~~~ [('January 2024', 'DATE', 119, 131, 20, 22), ('01-Dec-2023', 'DATE', 446, 457, 69, 70), ('08-Dec-2023', 'DATE', 511, 522, 80, 81), ('13-Dec-2023', 'DATE', 567, 578, 88, 89), ('08-Jan-2024', 'DATE', 638, 649, 97, 99)]\n", "[('Novartis Türkiye is one of the first movers of Lean Digital Core (LDC) SAP-S4 Roll-out Project, which will go-live on ', 'DATE'), ('. led to Specifically for clinical trials supply management, this change will require careful anticipation in order to guarantee supply continuity until the patient. The cut-over dates for the project have been determined as below:\\nLast receipt of all supplies (drugs, lab kits & equipment) at the Turkish customs ', 'DATE'), ('\\nLast day for drug shipment request - both CRAs & IRT ', 'DATE'), ('\\nLast day for drug shipment to clinical site ', 'DATE'), ('\\nStart of deliveries and all payments for customs clearance ', 'DATE'), (' resulting in Risk to not meet FPFV due to customs clearances freeze', '')]\n", "\n", "\n", "\n", "----- [('5001', 'Patient Related SiteID', 39, 43, 6, 7)]\n", "~~~~~~~ [('5001', 'Patient Related SiteID', 39, 43, 6, 7)]\n", "[('Only 2 certified microscopists at site ', 'Patient Related SiteID'), (' may lead to reduced number of microscopists available for readings until retraining in January 2024 resulting in delayed or missed screening and enrollment goals', '')]\n", "\n", "\n", "\n", "----- [('5012', 'Patient Related SiteID', 171, 175, 28, 29)]\n", "~~~~~~~ [('5012', 'Patient Related SiteID', 171, 175, 28, 29)]\n", "[('Optional consent is required for collection of pharmacogenetic sample on the study. Sites must check consent is given prior to obtaining sample. This was not done at site ', 'Patient Related SiteID'), (' for a patient led to patient sample was collected without consent and shipped to central lab resulting in sample requiring destruction.', '')]\n", "\n", "\n", "\n", "----- [('Oct 20', 'DATE', 771, 777, 147, 149)]\n", "~~~~~~~ [('Oct 20', 'DATE', 771, 777, 147, 149)]\n", "[(\"Out of LTS comments added to PK conc after QA released data generated by WUXI for nab-pac and released to IA. This is a process deviation led to 27 PK samples generated by WuXi for nab-pac are out of current established LTS and the conc are kept as the LTS is still on going from Veeda. In order to meet the IA timeline first week of Oct, while WuXi site is closed due to National holiday, the BA monitor added comments to the txt file on Preadvance after the QA'd data uploaded by WUXI and released the pk date for IA. This is a deviation for BA process. Also the data should be deactivated until the established LTS becomes available to make sure the data can be used. resulting in The uploaded data may not be used if the Extended LTS ( expected comes out around \", 'DATE'), (') did not pass.', '')]\n", "\n", "\n", "\n", "----- [('IMP', 'MH', 210, 213, 32, 34), ('IMP', 'MH', 278, 281, 43, 45)]\n", "~~~~~~~ [('IMP', 'MH', 210, 213, 32, 34), ('IMP', 'MH', 278, 281, 43, 45)]\n", "[('Packaging campaign delayed due to shortage in Akylux- pack material. may lead to Packaging campaign planned in October, shifted to end of November, due to shortage in Akylux-packing Material. Current Stock of ', 'MH'), (' at depot is low to support accelerated enrollment. resulting in ', 'MH'), (' at Hub/Depot low to support accelerated enrollment till Dec 2023.', '')]\n", "\n", "\n", "\n", "----- [('6009', 'SUBJID', 16, 20, 3, 4), ('6009', 'SUBJID', 140, 144, 22, 24), ('6009', 'SUBJID', 1028, 1032, 165, 167), ('6009', 'SUBJID', 1152, 1156, 185, 187), ('6009', 'SUBJID', 2153, 2157, 346, 348), ('US ', 'COUNTRY', 24, 27, 5, 7), ('US ', 'COUNTRY', 148, 151, 24, 26), ('US ', 'COUNTRY', 1036, 1039, 168, 169), ('US ', 'COUNTRY', 1160, 1163, 187, 190), ('US ', 'COUNTRY', 2161, 2164, 348, 350), ('26-April-2023', 'DATE', 44, 57, 9, 11), ('26-April-2023', 'DATE', 1056, 1069, 171, 173), ('11-May-2023', 'DATE', 697, 708, 114, 116), ('11-May-2023', 'DATE', 1709, 1720, 276, 278), ('03-August-2023', 'DATE', 988, 1002, 159, 161), ('03-August-2023', 'DATE', 2000, 2014, 321, 323), ('11-May-23', 'DATE', 2118, 2127, 339, 342), ('IMP', 'MH', 2255, 2258, 365, 367), ('Day 90', 'PRECLINICAL STUDY DAY', 2262, 2268, 367, 369)]\n", "~~~~~~~ [('6009', 'SUBJID', 16, 20, 3, 4), ('US ', 'COUNTRY', 24, 27, 5, 7), ('26-April-2023', 'DATE', 44, 57, 9, 11), ('6009', 'SUBJID', 140, 144, 22, 24), ('US ', 'COUNTRY', 148, 151, 24, 26), ('11-May-2023', 'DATE', 697, 708, 114, 116), ('03-August-2023', 'DATE', 988, 1002, 159, 161), ('6009', 'SUBJID', 1028, 1032, 165, 167), ('US ', 'COUNTRY', 1036, 1039, 168, 169), ('26-April-2023', 'DATE', 1056, 1069, 171, 173), ('6009', 'SUBJID', 1152, 1156, 185, 187), ('US ', 'COUNTRY', 1160, 1163, 187, 190), ('11-May-2023', 'DATE', 1709, 1720, 276, 278), ('03-August-2023', 'DATE', 2000, 2014, 321, 323), ('11-May-23', 'DATE', 2118, 2127, 339, 342), ('6009', 'SUBJID', 2153, 2157, 346, 348), ('US ', 'COUNTRY', 2161, 2164, 348, 350), ('IMP', 'MH', 2255, 2258, 365, 367), ('Day 90', 'PRECLINICAL STUDY DAY', 2262, 2268, 367, 369)]\n", "[('Patient at site ', 'SUBJID'), (' in ', 'COUNTRY'), (' was screened on ', 'DATE'), (' into the CLNP023B12001B study. The study site personnel declared Patient at site ', 'SUBJID'), (' in ', 'COUNTRY'), ('as a screen failure, because the site was not able to complete the screening procedure within the screening visit window. However, the patient was not documented as a screen failure in the Cenduit IRT system. Per protocol, a participant can be re-screened and randomized enrolled into the study if the patient is re-consented and all protocol requirements are satisfied. A new participant ID number should be used for re-screening. The patient was re-screened with a new patient ID and randomized conducted extension day 1 visit on the same day, ', 'DATE'), ('. However, the site personnel, in error, registered the patient as patient with previous ID number the Cenduit IRT system, not the appropriate Patient ID. When the study site personnel identified the error, a Document Change Request Form was completed and forwarded to Cenduit on ', 'DATE'), ('. led to Patient at site ', 'SUBJID'), (' in ', 'COUNTRY'), (' was screened on ', 'DATE'), (' into the CLNP023B12001B study. The study site personnel declared Patient at site ', 'SUBJID'), (' in ', 'COUNTRY'), ('as a screen failure, because the site was not able to complete the screening procedure within the screening visit window. However, the patient was not documented as a screen failure in the Cenduit IRT system. Per protocol, a participant can be re-screened and randomized enrolled into the study if the patient is re-consented and all protocol requirements are satisfied. A new participant ID number should be used for re-screening. The patient was re-screened with a new patient ID and randomized conducted extension day 1 visit on the same day, ', 'DATE'), ('. However, the site personnel, in error, registered the patient as patient with previous ID number the Cenduit IRT system, not the appropriate Patient ID. When the study site personnel identified the error, a Document Change Request Form was completed and forwarded to Cenduit on ', 'DATE'), ('. Study team requested Cenduit IRT to proceed with reversing extension day 1 visit, which took place on ', 'DATE'), (' for that patient at site ', 'SUBJID'), (' in ', 'COUNTRY'), ('in IRT system. resulting in This could have resulted in patient not being able to received ', 'MH'), (' at ', 'PRECLINICAL STUDY DAY'), (' visit and his study treatment could have been disturbed.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Patient transfer from one site to another (due to various reasons) may lead to operational complications resulting in impacting the study data integrity, availability of study documents at site, the data analysis (patients discontinuation)', '')]\n", "\n", "\n", "\n", "----- [(\"02 Nov '23\", 'DATE', 253, 263, 39, 42), ('91 patients', 'SUBJID', 283, 294, 45, 48)]\n", "~~~~~~~ [(\"02 Nov '23\", 'DATE', 253, 263, 39, 42), ('91 patients', 'SUBJID', 283, 294, 45, 48)]\n", "[('PBMC samples collected following the release of an investigator letter which asked sites to stop the collection of PBMC samples with immediate effect led to PBMC samples have been collected and processed following release of the investigator letter on ', 'DATE'), ('. PBMC samples from ', 'SUBJID'), (' have been collected and processed at the central lab after we requested via an investigator letter to stop all collections. This collection was removed from the protocol in PA2 but sites have continued to collect these samples. No data has been transferred. 17 patients who have consented only to PA2 have provided PBMC samples. resulting in Patient PBMC samples were collected after the release of the investigator letter and after the protocol amendment where this collection was removed', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"PI and site staff not knowing business proficient English may lead to inability to enter data correctly and constant queries for the entered data resulting in a decrease in data quality/integrity and a decrease with Novartis' reputation at the site.\", '')]\n", "\n", "\n", "\n", "----- [('1020', 'Patient Related SiteID', 13, 17, 3, 4)]\n", "~~~~~~~ [('1020', 'Patient Related SiteID', 13, 17, 3, 4)]\n", "[('PI from site ', 'Patient Related SiteID'), (' missed redaction of participant personal information in email. led to Confidential information patient information shared with select study team members. resulting in Data privacy issue.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('PI not available within the requested time frame. led to Missing PI signatures at the time of database lock procedure. resulting in Acceptability of data possibly included in a submission.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('PIs stating budgets are low especially against competing trials may lead to Sites not screening / randomizing patients resulting in Trial not enrolling', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Potential delay in data transfer due to technical issue at NGDX for RNA Seq may lead to delay in reconciliation leading to delay in KPI of 6 weeks for planned Biomarker+ IB data cleaning iDBL 6. may lead to Missed KPI of 6 weeks for iDBL for Biomarker+ IB data cleaning due to unavailability of data in due time resulting in KPI of 6 weeks could be missed due to unavailability of data in due time for iDBL', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"PPC can't update the production database correctly for sample numbers may lead to Preadvance reports that are incorrect resulting in PK/PD data can't be QC/QA'd and subsequently PK/PD merged outputs can't be produced.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Programming updates done to resolve the duplicate records issue in PreAdvance by SME resulting in additional errors may lead to re-programming of the file and additional rounds of reconciliation resulting in delay in achieving final database lock.', '')]\n", "\n", "\n", "\n", "----- [('19-Oct-23', 'DATE', 120, 129, 22, 23), ('24-Oct-23', 'DATE', 319, 328, 54, 55)]\n", "~~~~~~~ [('19-Oct-23', 'DATE', 120, 129, 22, 23), ('24-Oct-23', 'DATE', 319, 328, 54, 55)]\n", "[('Protocol Am 1 submitted to EU-CTR in response to Part 1 RFIs with short timelines (12 calendar days from RFI receipt on ', 'DATE'), (') may lead to EU CTR approval of PA1 prior to release of updated IRT/Central Lab Databases + RaveX database resulting in delayed implementation of PA1 due to Vendor/Rave-X readiness. As of ', 'DATE'), (', no site will be activated under PA1 as a PA2 is in preparation, hence the impact on SSU timelines is deemed low.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Protocol Amendment 01 (PA01) was finalized prior to FPFV and oversight by CTT when preparing PA01 led to missed requirement to prepare the Protocol Amendment Impact Assessment Checklist as per the working practice at the time of PA01 finalization resulting in missing Protocol Amendment Impact Assessment Checklist for PA01.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Protocol non-compliance led to incorrect steroid titration (if the patient starts at 9-13 mg of prednisone or a dose equivalent, the next decrease should be between 7-8 mg, then 5-6 mg, finally 2.5 mg) resulting in inconsistent data patterns overall and cannot be compared with other patients (from the protocol deviation minutes).', '')]\n", "\n", "\n", "\n", "----- [('3 timepoints', 'DATE', 298, 310, 56, 59)]\n", "~~~~~~~ [('3 timepoints', 'DATE', 298, 310, 56, 59)]\n", "[('Pulse rate has not been collected in the CRF (Vital signs page) led to protocol required data not collected in the CRF resulting in missing data. Heart rate is recorded on the ECG, so where pulse rate and ECG are collected at the same time this data is coming in house and being reviewed, but for ', 'DATE'), (' there is pulse rate scheduled but no ECG. \\nWill require PPC and retrospective data entry by the sites: sites already unhappy with amount of retrospective data entry as a result of PPCs.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Q2 is responsible for the B and T-cell subsets analysis based on the B-Cell count results (CD19+ events). This is automatically triggered by programming at certain visits where only TBNK is done (no immune cell characterization assessment scheduled). may lead to reflex testing not working properly resulting in missing B and T-cell subsets data when B-cell count >500 events', '')]\n", "\n", "\n", "\n", "----- [('December 2022', 'DATE', 308, 321, 56, 59), ('August 2023', 'DATE', 325, 336, 59, 61)]\n", "~~~~~~~ [('December 2022', 'DATE', 308, 321, 56, 59), ('August 2023', 'DATE', 325, 336, 59, 61)]\n", "[('QTL review was delegated, however delegated person did not perform the task as expected. QTL review was also not a topic in the CTT agenda.\\nDM reminder to send QTLs reports on a monthly basis expired resulting in DM forgetting to provide the reports. led to This led to a lack of oversight of the QTLs from ', 'DATE'), (' to ', 'DATE'), ('. resulting in Non-IEPDs went slightly above the tolerance limit.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Quality issue with the VISION study data may lead to delay of finding a critical issue to the PMDA submission resulting in delay of PMDA submission', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Randomization office did not provide a complete request to NIRT team for randomization data. may lead to Randomization data (Y5) for some patients were not made available to programming team and it was not possible to associate the treatment (QBW/placebo) with specific patients. LSH database has to be unlocked - Root cause and CAPA plan needs to be provided for that. resulting in Delay with data analysis and risk to overall study timelines', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('RAVE EDC system performance issues may lead to delays in timely data entry and query resolution ultimately resulting in delays in iDBL for IA 1', '')]\n", "\n", "\n", "\n", "----- [('2065', 'Patient Related SiteID', 37, 41, 6, 7)]\n", "~~~~~~~ [('2065', 'Patient Related SiteID', 37, 41, 6, 7)]\n", "[('Recent change of CRA personnel (site ', 'Patient Related SiteID'), (\", Mexico) short before database lock (2 working days) may lead to New CRA don't have required trainings and access resulting in Some activities, such as query answers or PI signatures might only be done in a delayed manner, which is jeopardizing database lock timelines.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Recruitment projections numbers provided in the approved FPO plan, are different than the actual recruitment pace, may lead to Lack of study supplies resulting in Patient Randomization & ongoing pts dispensing disruption', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Relabeling of IMP expiring on 30-Apr-2023 at EU Hub led to temporary unavailability of IMP for site shipments resulting in late IMP administration', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Role changes due to project ambition has resulted in CPMs not having a main point of contact for EDC queries/questions may lead to EDC not being updated in a timely manner (queries not being resolved) resulting in delays in data cleaning and inaccurate/incorrect data in EDC.', '')]\n", "\n", "\n", "\n", "----- [('26 Sep 2023', 'DATE', 248, 259, 48, 51), ('Argentina', 'COUNTRY', 295, 304, 57, 59), ('USA', 'COUNTRY', 309, 312, 59, 60)]\n", "~~~~~~~ [('26 Sep 2023', 'DATE', 248, 259, 48, 51), ('Argentina', 'COUNTRY', 295, 304, 57, 59), ('USA', 'COUNTRY', 309, 312, 59, 60)]\n", "[('Root cause is yet to be identified. Cenduit is working on it may lead to There is a stock out at the EU depot for the main study. as IRT orders were triggered and dispatched. Detailed assessment is ongoing and GCS would come back with the data by ', 'DATE'), ('. Assessment would also be done for ', 'COUNTRY'), (' and ', 'COUNTRY'), ('. Detailed assessment would be shared in next couple of days. resulting in Based on risk assessment, decision would be made about the recruitment/enrollment of new patients', '')]\n", "\n", "\n", "\n", "----- [('PBMC biomarker samples', 'MH', 29, 51, 4, 7)]\n", "~~~~~~~ [('PBMC biomarker samples', 'MH', 29, 51, 4, 7)]\n", "[('Several sites not processing ', 'MH'), (' as per ICON Lab Manual led to whole blood, sample quality issues (sample clotting/hemolysis) resulting in samples unable to be used/analyzed as per protocol.', '')]\n", "\n", "\n", "\n", "----- [('Severe or serious', 'AGE', 0, 17, 0, 3)]\n", "~~~~~~~ [('Severe or serious', 'AGE', 0, 17, 0, 3)]\n", "[('', 'AGE'), (' AE/s or worse esp. for infections in CSF-related organs, concretely meningitis or encephalitis, may lead to impact the safety profile of remibrutinib during the ongoing submission process for marketing authorization and review by HA with major HA Q&A resulting in resulting in reduced or delayed marketing authorization', '')]\n", "\n", "\n", "\n", "----- [('29August2023', 'DATE', 229, 241, 34, 35), ('12Oct2023', 'DATE', 265, 274, 39, 41)]\n", "~~~~~~~ [('29August2023', 'DATE', 229, 241, 34, 35), ('12Oct2023', 'DATE', 265, 274, 39, 41)]\n", "[('Significant delay in processing requests, multiple issues with assigned trainings, and OC-RDC system issues, all associated with the Data Management Vendor Syneos. may lead to New PI access request to OC-RDC system requested on ', 'DATE'), (', and is still pending ', 'DATE'), ('. Numerous follow-up, escalations, and troubleshooting directly with Syneos at the site level has not resulted in system access. Additional issues regarding connectivity and access requests in the past have been previously escalated. resulting in The PI has conveyed frustration and unwillingness to continue on the trial with use of this system and associated vendor Syneos. If actions or solutions cannot be presented to the PI, there is risk of site withdrawal from the trial. This trial is a single centered study, and it is vital action is taken to demonstrate trust between the site and NVS team.', '')]\n", "\n", "\n", "\n", "----- [('1001', 'Patient Related SiteID', 5, 9, 1, 2)]\n", "~~~~~~~ [('1001', 'Patient Related SiteID', 5, 9, 1, 2)]\n", "[('Site ', 'Patient Related SiteID'), (' not being able to enroll patients (contract and unexperienced site) may lead to inability to meet study power and meet seasonal disease in Belem resulting in elongated timelines and potential need for TTG amendment.', '')]\n", "\n", "\n", "\n", "----- [('4000', 'Patient Related SiteID', 5, 9, 1, 2), ('6-Oct', 'DATE', 25, 30, 4, 5), ('6-10 Oct', 'DATE', 156, 164, 21, 23), ('24-Oct', 'DATE', 230, 236, 34, 35), ('24-Oct', 'DATE', 326, 332, 48, 50)]\n", "~~~~~~~ [('4000', 'Patient Related SiteID', 5, 9, 1, 2), ('6-Oct', 'DATE', 25, 30, 4, 5), ('6-10 Oct', 'DATE', 156, 164, 21, 23), ('24-Oct', 'DATE', 230, 236, 34, 35), ('24-Oct', 'DATE', 326, 332, 48, 50)]\n", "[('Site ', 'Patient Related SiteID'), (' was initiated (', 'DATE'), (') with partial imaging qualification (CT/MRI and PET qualified/SPECT not qualified), as screening/imaging visit is planned on ', 'DATE'), (' (requires only CT/MRI and PET) and treatment is planned later on ', 'DATE'), (' (requires SPECT) may lead to possibility that SPECT qualification not obtained prior to ', 'DATE'), (' resulting in postponement of planned patient treatment', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Site error when inputting randomisation cohort details in IRT led to Sites have made mistakes in choosing if subjects are part of the MRI cohort and their MS subtype within IRT. This has resulted in patients who are not in the MRI cohort but for whom MRIs have been performed, patients randomised to the MRI cohort for whom no MRIs have been performed and patients reported with the incorrect MS subtype per IRT. resulting in This causes confusion regarding the correct cohort for patients and may have downstream issues to ensure the correct patients are present in the analysis sets.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Site no longer using original experienced study coordinator and is now searching for a new study coordinator may lead to delay in site initiation, decreased patient recruitment, and inability to perform study conduct if unexperienced resulting in delay of timelines such as SIV/FPFV/LPLV and potential protocol deviations.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Site submitting an accession for B Cell panel 2.4; SBAFF; TBNK panel (which are blinded params) under the incorrect visit where the blinding is not the same as in the correct visit may lead to analyte(s) not properly blinded or reported resulting in patient unblinded', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"Site using local lab to assess patient safety in addition to central lab required per protocol led to patients having more blood drawn during the study conduct resulting in additional patient's burden and PIs not willing to assess central lab value\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Sites are not following the sample processing instructions in the Lab Manual. Instead of aliquoting whole blood samples, sites are centrifuging them to produce plasma. led to Received samples at the testing lab that were plasma instead of whole blood. resulting in Unable to use samples for planned analysis.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Skin Prick Test not done as per protocol led to Not doing dilutions correctly, not following the protocol requirements for the testing, , incorrect milk/egg dilution resulting in Potential affect on data quality', '')]\n", "\n", "\n", "\n", "----- [('1Sep2023', 'DATE', 62, 70, 12, 13), ('1Sep2023', 'DATE', 563, 571, 109, 110), ('4Oct2023', 'DATE', 100, 108, 18, 19), ('881', 'Patient Related SiteID', 484, 487, 91, 92)]\n", "~~~~~~~ [('1Sep2023', 'DATE', 62, 70, 12, 13), ('4Oct2023', 'DATE', 100, 108, 18, 19), ('881', 'Patient Related SiteID', 484, 487, 91, 92), ('1Sep2023', 'DATE', 563, 571, 109, 110)]\n", "[('Snapshot is planned for DEM for ALL treatment arm with DCO as ', 'DATE'), (' and planned Snapshot Date is ', 'DATE'), ('.\\nHowever there are lot of missing data into the study and huge count of open queries for site action and most of the count are from ESP and US sites.\\n\\nBelow count from latest Metrics report\\nMissing Visits = 20 (out of which 16 is for ALL arm)\\nMissing Pages = 548 (Out of which 223 is for ALL arm)\\nMissing Lab data = 52 (Out of which 32 is for ALL arm)\\nOpen queries to site = ', 'Patient Related SiteID'), (' (Out of which 379 is for ALL arm) may lead to All Data in is expected on 1', 'DATE'), (\", Considering pending missing data there's a risk that site will not be able to update all expected data before ADI.\\nDelay Data entry will impact ongoing Data review tasks resulting in Pending Data entry and query resolution is impacting the Monthly Data revie task for all line functions (DM, Clinical and Medical review), further delay in Data entry will have impact on Snapshot date. \\nDelay in acheiving ADI and resolving queries in a timely manner will have an impact on quality of data review and will also delay the Data review rounds planned for the DEM snapshot.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Study design showing a maintenance period without SoC recommendation as per guidelines triggered concerns from key opinion leaders in France and Germany may lead to decrease the commitment from sites who will either not participate or not recruit, or not follow the tailored approach for RTX in the remission maintenance period resulting in Not reaching our primary endpoint', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Study specific CRF pages requiring Standards approval, may lead to long discussions, resulting in DB go-live on a critical path for FPFV', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Study uses an old process with less PreAdvance involved led to confusion on reports and process to be followed for sample reconciliation resulting in delays in sample reconciliation and potentially data extractions (for PK, PD and IG samples).', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('StudyÂ’s use of multiple capsule strengths that look similar or identical may lead to misidentification of capsules by patients if more than one strength is prescribed resulting in dosing errors', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The challenging study design means countries/sites are making conservative enrollment projections which may lead to an underallocation of patients resulting in the approved baseline FPO plan only projecting enrollment for 100 (and not 108) patients.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The CSR signature page was sent by PI by mail without a tracking number. led to The CSR signature page signed by the PI was not received (hard copy with wet-ink signature). resulting in The PI CSR signature page was lost and not received.', '')]\n", "\n", "\n", "\n", "----- [('21-DEC-2023', 'DATE', 107, 118, 17, 19)]\n", "~~~~~~~ [('21-DEC-2023', 'DATE', 107, 118, 17, 19)]\n", "[('The CVAY736Y2102 study was endorsed by ICDB for early termination. may lead to database lock planned for ', 'DATE'), ('. resulting in very tight timelines and may lead to delays in locking the trial.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The IMP Expiry Date of November 30, 2023 may lead to the IRT not able to dispense medication beyond November 28, 2023 resulting in no medication available for the last dispensation visit.', '')]\n", "\n", "\n", "\n", "----- [('February 2023', 'DATE', 23, 36, 4, 6)]\n", "~~~~~~~ [('February 2023', 'DATE', 23, 36, 4, 6)]\n", "[('The issues faced since ', 'DATE'), (' for converting the Word to PDF in Credi may lead to mismatched tables, incorrect bookmarks and/or duplicated tables resulting in errors in the published interim CSR', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The lack of protocol specific language, ambiguity of the monitoring plan wording and absence of a dedicated eCRF field for record of meal times/intake led to lack of monitoring oversight and incapacity to verify at data review steps that the IMP intake followed protocol indications resulting in protocol deviations not being identified before DBL and in uncertainty regarding the food intake impact on PK data (and therefore in drug effect).', '')]\n", "\n", "\n", "\n", "----- [('India', 'COUNTRY', 79, 84, 11, 12)]\n", "~~~~~~~ [('India', 'COUNTRY', 79, 84, 11, 12)]\n", "[('The location of the bioanalytical laboratory (Veeda Clinical Research Ltd.) in ', 'COUNTRY'), (' may lead to delayed turnaround times for PK analysis resulting in delayed trial timelines', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The main ICF v01.02 was developed to incorporate the changes of the study protocol amendment 01. After release of the main ICF, 2 discrepancies was identified at the Summary of Changes and not at the actual document content. According to the document development practice, the track changes do not apply to Summary of Changes and a new table (SoC) is needed every time and ICF is developed, for this reason a new ICF was not developed at that time led to incorrect information recorded at the ICF v01.02 Summary of Changes of the document resulting in mismatch between the SOC and the actual content of the the main ICF.', '')]\n", "\n", "\n", "\n", "----- [('1461 (Netherlands)', 'Patient Related SiteID', 18, 36, 4, 6)]\n", "~~~~~~~ [('1461 (Netherlands)', 'Patient Related SiteID', 18, 36, 4, 6)]\n", "[('the need for site ', 'Patient Related SiteID'), (' to hold a scientific board to approve the commercial supply (mandatory as per local regulation) may lead to lead to delay of the LPLT and LPLV resulting in a potential delay of DBL', '')]\n", "\n", "\n", "\n", "----- [('30-06-21', 'DATE', 150, 158, 24, 25), ('10-02-22', 'DATE', 242, 250, 41, 42)]\n", "~~~~~~~ [('30-06-21', 'DATE', 150, 158, 24, 25), ('10-02-22', 'DATE', 242, 250, 41, 42)]\n", "[('The off-line finalization of the RIS checklist by Parexel, led to an inconsistency between the actual RIS date i.e. date when Parexel finalized RIS (', 'DATE'), (') and the RIS date which appears in Impact (date when the RIS is uploaded to Subway ', 'DATE'), (')” resulting in low or no impact to the conduct of the clinical study.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The QMF149G2301 study protocol specifies in Section 8.4.3 that ECGs must be recorded in triplicate. However, the Masterscope is currently not programmed to collect ECGs in triplicate by default. Instead the Masterscope has been programmed to collect single ECGs with the provision for optional ECGs to be performed at the investigatorÂ’s discretion, which led to some ECGs being performed as a single measure for ten patients currently on treatment period and one in run-in, resulting in lesser accuracy and potentially missing an abnormal finding and for the investigators to take appropriate action to ensure patient safety. The investigators, however, can perform optional ECGs or repeat ECGs if clinically required.', '')]\n", "\n", "\n", "\n", "----- [('6005', 'Patient Related SiteID', 38, 42, 7, 8), ('6007', 'Patient Related SiteID', 44, 48, 8, 9)]\n", "~~~~~~~ [('6005', 'Patient Related SiteID', 38, 42, 7, 8), ('6007', 'Patient Related SiteID', 44, 48, 8, 9)]\n", "[('The RIS checklists of US sites (6004, ', 'Patient Related SiteID'), (', ', 'Patient Related SiteID'), (') were approved in the DMS after the SIV visits which led to non-chronological order of events in CTMS (SIV prior to RIS) resulting in system discrepancy. CSM confirmed that everything from the RIS checklists was in place prior to the SIV, only the system approvals were late hence the impact is minor.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('The study TMF was transferred from the MDCO Veeva to NVS Subway during the study, may lead to discrepancies in the study documents filed in Subway with potential missing docs, TMF not kept up to date with filing in real time as documented were generated, resulting in the study TMF not being GCP compliant and inspection ready.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[(\"The study's UK SSO SSU Manager currently handling more than 50 studies ,out of which 15 are in start up may lead to to delays in UK sites start-up resulting in the trial being behind in enrollment and not meeting the TM LPFV recruitment plans\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('There might be delay in 1st production transfer for MRI (Imaging) for this study. The study team is aware of this risk. The initial specifications were delayed by Vendor which resulted in delay in DTS writing. Post DTS approval, Vendor informed that their production set-up is not ready. They need at least 60 days for the set-up. VPM and VPG are currently involved in timeline negotiation with Vendor. may lead to This may lead to delay in Production transfer, delay data review (reconciliation data). MRI data review is necessary to define eligibility. resulting in The overall delay will impact the identification of potential data issues/protocol deviations.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('There might be delay in 1st production transfer for MRI (Imaging) for this study. The study team is aware of this risk. The initial specifications were delayed by Vendor which resulted in delay in DTS writing. Post DTS approval, Vendor informed that their production set-up is not ready. They need at least 60 days for the set-up. VPM and VPG are currently involved in timeline negotiation with Vendor. may lead to This may lead to delay in Production transfer, delay data review (reconciliation data). MRI data review is necessary to define eligibility. resulting in The overall delay will impact the identification of potential data issues/protocol deviations.', '')]\n", "\n", "\n", "\n", "----- [('9th Oct 2023', 'DATE', 111, 123, 20, 23)]\n", "~~~~~~~ [('9th Oct 2023', 'DATE', 111, 123, 20, 23)]\n", "[('Third Party Data from Science 37 Vendor for QS domain- Emphasis and Sympact was received for the first time on ', 'DATE'), (\". The data needs to be reviewed and may have an impact on cleaning for the IA scheduled in December. may lead to This may impact cleaning of data for the upcoming IA timeline in December. resulting in As ePRO's are part of upcoming IA, the data review and cleaning activities for the QS domain may impact the downstream data analysis.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('TMF Ops services support the upload of EU CTR documents to Subway may lead to EU CTR documents are not uploaded on time resulting in delayed submission and unmet milestone', '')]\n", "\n", "\n", "\n", "----- [('ed.9', 'DATE', 10, 14, 2, 3)]\n", "~~~~~~~ [('ed.9', 'DATE', 10, 14, 2, 3)]\n", "[('TNO155 IB ', 'DATE'), (\" includes a new requirement for sexually active male patients to use a condom during sexual intercourse was extended from 10 to 12 days after the last dose of study treatment. led to since the study doesn't enroll additional patients, a new protocol amendment will not be done to reflect this change resulting in HA/IRB and sites will know about the requirements throught the IB and the ICF but not the protocol.\", '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('To ensure stability of samples, especially for proteomics. It would be ideal to have sites who have -80 degree freezers. led to This may be a challenge and it is possible to store samples at minus 20 degrees but for no longer than one or two weeks. resulting in This will impact the shipments to the central lab and site would require dry ice on site for freezing samples after collection', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Transferring confidential information with external DMC members may lead to compromised data security resulting in unauthorized access to study data', '')]\n", "\n", "\n", "\n", "----- [('01-Dec-2023', 'DATE', 44, 55, 9, 11), ('08-Jan-2024', 'DATE', 59, 70, 11, 13)]\n", "~~~~~~~ [('01-Dec-2023', 'DATE', 44, 55, 9, 11), ('08-Jan-2024', 'DATE', 59, 70, 11, 13)]\n", "[('TUR supply cut off period may lead to From ', 'DATE'), (' to ', 'DATE'), (' IMP distribution to TUR will be blocked due to switch of SAP governance system resulting in Due to delay in replacement of the expiring batch of IMP (expiry date: 31-Jan) the timelines for \\ndistribution of the new batch to TUR are even more challenging (see Risk R0XX). Potential stock out for TUR.', '')]\n", "\n", "\n", "\n", "----- [('#6004', 'Patient Related SiteID', 10, 15, 2, 3), ('9001', 'Patient Related SiteID', 20, 24, 4, 5)]\n", "~~~~~~~ [('#6004', 'Patient Related SiteID', 10, 15, 2, 3), ('9001', 'Patient Related SiteID', 20, 24, 4, 5)]\n", "[('Two sites ', 'Patient Related SiteID'), (' and ', 'Patient Related SiteID'), (' with screen failures only have been included into CSR, however screen failures are not included into data analysis. may lead to CSR publishing resulting in No serious impact on CSR publishing timelines and data disclosure', '')]\n", "\n", "\n", "\n", "----- [('27th Sep 2023', 'DATE', 38, 51, 5, 8), ('27th Sep 2023', 'DATE', 166, 179, 28, 31)]\n", "~~~~~~~ [('27th Sep 2023', 'DATE', 38, 51, 5, 8), ('27th Sep 2023', 'DATE', 166, 179, 28, 31)]\n", "[('Unavailability of updated database on ', 'DATE'), (' as per Protocol Amendment 1 may lead to Unavailability of Local Lab Cardiac enzyme CRF for completion by site on ', 'DATE'), ('. resulting in Delay in data entry by site', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Unclear roles and responsibilities of TMF Ops Services team may lead to Identified issues that did not come up in the TMF review and could not find clear process to resolve; Missing signposts in Subway (unclear who is responsible for completing and follow up with relevant line function) resulting in Decreased TMF compliance', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Unclear, overlooked or misunderstood use of the ICF checklist in fully outsourced studies led to Retrospective completion of the ICF checklist and the Novartis ICF based on the company-internal template resulting in A risk of missing essential and mandatory ICF elements.', '')]\n", "\n", "\n", "\n", "----- [('patient reimbursement', 'MH', 30, 51, 4, 6)]\n", "~~~~~~~ [('patient reimbursement', 'MH', 30, 51, 4, 6)]\n", "[('Unexperienced site not having ', 'MH'), (' properly set-up may lead to decreased patient enrollment and patients not completing study resulting in a delay in trial milestones and not meeting necessary amount of patients to complete the study.', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('Unresponsive procurement group may lead to delay in study activities resulting in delay in recruitment', '')]\n", "\n", "\n", "\n", "----- [('29Sep2023', 'DATE', 50, 59, 7, 8)]\n", "~~~~~~~ [('29Sep2023', 'DATE', 50, 59, 7, 8)]\n", "[('Up-versioning of coding dictionary at RAVEX since ', 'DATE'), (' may lead to resulting in slowness of RAVEX and performance issue leading to delay in database developer and DM activities like Screen UAT resulting in Leading to delay in database Go live timelines', '')]\n", "\n", "\n", "\n", "----- [('2545', 'Patient Related SiteID', 9, 13, 2, 3)]\n", "~~~~~~~ [('2545', 'Patient Related SiteID', 9, 13, 2, 3)]\n", "[('USA Site ', 'Patient Related SiteID'), (' was closed with IRB prior to green light from Global team/CRA did not close out as per Novartis SOP led to 3 TMF queries (affected docs are FDA 1572, DSL & FD for one Sub-I) not being resolved for the site resulting in not having confirmation needed for 3 Sub IÂ’s at the site who need to be listed in Final CSR appendix 16.1.4 as well as receiving the updated affected documents from the site', '')]\n", "\n", "\n", "\n", "----- []\n", "~~~~~~~ []\n", "[('War conflict in Israel may lead to disruption of study conduct resulting in resulting in impact on upcoming patient´s visits and other study related activities.', '')]\n", "\n", "\n", "\n", "----- [('Israel', 'COUNTRY', 32, 38, 5, 6), ('Israel', 'COUNTRY', 1991, 1997, 330, 332), ('Jordan', 'COUNTRY', 86, 92, 15, 16), ('Jordan', 'COUNTRY', 204, 210, 34, 35), ('Jordan', 'COUNTRY', 1134, 1140, 193, 194), ('Jordan', 'COUNTRY', 1286, 1292, 219, 221), ('Jordan', 'COUNTRY', 1947, 1953, 323, 325), ('November 2023', 'DATE', 1253, 1266, 213, 216)]\n", "~~~~~~~ [('Israel', 'COUNTRY', 32, 38, 5, 6), ('Jordan', 'COUNTRY', 86, 92, 15, 16), ('Jordan', 'COUNTRY', 204, 210, 34, 35), ('Jordan', 'COUNTRY', 1134, 1140, 193, 194), ('November 2023', 'DATE', 1253, 1266, 213, 216), ('Jordan', 'COUNTRY', 1286, 1292, 219, 221), ('Jordan', 'COUNTRY', 1947, 1953, 323, 325), ('Israel', 'COUNTRY', 1991, 1997, 330, 332)]\n", "[('With the political situation in ', 'COUNTRY'), (' and knowing that we have 1 patient included in ', 'COUNTRY'), (', CTT prepared a risk assessment to document the current situation in this country.\\nNo impact for this study in ', 'COUNTRY'), (' so far and the patient can remain in the study. may lead to Summary of risk assessment: status of the site to perform protocol visits / performed safety exams locally (if require) / lab. Kits / IMP and on-site monitoring visits.\\n1)\\tCountry /Site level:\\n•\\tDoes the site have sufficient kits of all kinds? CRA: Yes a new shipment was placed last week to make sure we have everything. \\n•\\tWhat is the expiration? re-supply needed? CRA no re supply is needed at the moment. \\n•\\tAny issue with sending the samples outside to LabCorp? CRA: No issues. \\n•\\tSafety of the patients: Does any local lab perform the lab tests in case it any difficult to send the samples? CRA: yes, the local lab has all the needed tests that needs to be done to the patient. \\n•\\tMonitoring plan: any issue with scheduling the MV (on site?) CRA: no problem with scheduling onsite visit to the site.\\n2)\\tIMP:\\nIMP shipment that was recently released in ', 'COUNTRY'), (', thereÂ’s enough drug at the depot for visits through June 2024, with some extra. \\n•\\tRemaining IMP that has a ', 'DATE'), (' expiry date at the ', 'COUNTRY'), (' depot will be extended, a relabeling activity would be required.\\no\\tRelease of CoA is still pending, estimated to be generated next month and then we can plan the relabeling dates. \\n•\\tGCS is also shipping syringes and the moment, the site has enough on shelf for the next 6 visits, plus thereÂ’s an extra box at the depot.\\no\\tGiven the current situation though, best if I send a few more boxes to the depot\\n•\\tShould another patient enroll, then we can recalculate if additional material is still needed, depending also on the potential relabeling. \\n3)\\tCentral lab.\\nCurrently, there is no disruption to shipping/transportation of kits/samples to and from ', 'COUNTRY'), (' due to the geopolitical situation in ', 'COUNTRY'), ('. Central Lab. is monitoring the situation. resulting in Minor', '')]\n", "\n", "\n", "\n", "----- [('Feb 2024', 'DATE', 78, 86, 13, 15), ('15-Nov-2024', 'DATE', 249, 260, 39, 40)]\n", "~~~~~~~ [('Feb 2024', 'DATE', 78, 86, 13, 15), ('15-Nov-2024', 'DATE', 249, 260, 39, 40)]\n", "[('Wrong DP allocated for the most current packaging campaign with shelf life of ', 'DATE'), ('.\\nUrgent new packaging campaign requested by CTSM with expedited timelines. The current projections is that kits with Expiry date of Jun/2025 will be available by ', 'DATE'), ('. may lead to Delay of packaging campaign with suitable expiry date. resulting in Global replacement of expiring batch at risk for countries with long distribution lead time.\\nThis includes LBN, OMN, TUR, COL, BRA. If replacement of expiring batch of IMP is not performed on time there is a risk for failed dispensations.', '')]\n" ] } ], "source": [ "input_excel['RISK_ISSUE_STATEMENT_REDACTED'] = input_excel['RISK_ISSUE_STATEMENT'].apply(run_model_on_excel)" ] }, { "cell_type": "code", "execution_count": 71, "id": "5b7a685f-ca5e-4caf-828d-964ed8967f23", "metadata": {}, "outputs": [], "source": [ "input_excel.to_excel(r'Risk_Data_Output_v4.xlsx')" ] }, { "cell_type": "markdown", "id": "79aaf022-fd2a-499f-b47c-290ae23f967e", "metadata": {}, "source": [ "### Parse json and evaluate" ] }, { "cell_type": "code", "execution_count": 23, "id": "0c13d33a-ea12-42c7-bcf0-14562411432b", "metadata": {}, "outputs": [], "source": [ "import gc\n", "import torch\n", "\n", "def evaluate_json_paragraph(input_paragraph):\n", " if len(input_paragraph.strip()) < 30:\n", " sentences = [input_paragraph]\n", " else:\n", " sentences = seg.segment(input_paragraph)\n", " # print(sentences)\n", " para_value_list = []\n", " for sentence in sentences:\n", " sent_token_len = get_token_len(sentence)\n", " final_prompt_to_model = final_prompt+'\\n'+whole_task+'\\n'+sentence+\"\\nAnswer:\"\n", " input_ids = tokenizer(final_prompt_to_model, return_tensors='pt').input_ids.cuda()\n", " output = model.generate(inputs=input_ids, temperature=1, do_sample=False, max_new_tokens=sent_token_len*3, eos_token_id=16981)\n", " model_generated_output = tokenizer.decode(output[0][input_ids[0].shape[0]:])\n", " if debug:\n", " print(model_generated_output)\n", " sent_value_list = extract_values(model_generated_output, input_paragraph)\n", " para_value_list.extend(sent_value_list)\n", " del output\n", " del input_ids\n", " torch.cuda.empty_cache()\n", " gc.collect()\n", " para_value_list = remove_overlapping_entities(para_value_list)\n", " para_value_list = basic_postprocessing(para_value_list)\n", " print('\\n\\n')\n", " print('-----', para_value_list)\n", " return para_value_list" ] }, { "cell_type": "code", "execution_count": 24, "id": "f16757b6-bbea-4bb5-afb9-cf7ed5305d3a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Tue Dec 26 08:17:10 2023 \n", "+---------------------------------------------------------------------------------------+\n", "| NVIDIA-SMI 545.23.08 Driver Version: 545.23.08 CUDA Version: 12.3 |\n", "|-----------------------------------------+----------------------+----------------------+\n", "| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n", "| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n", "| | | MIG M. |\n", "|=========================================+======================+======================|\n", "| 0 Tesla V100-SXM2-16GB Off | 00000000:00:1E.0 Off | 0 |\n", "| N/A 31C P0 37W / 300W | 10636MiB / 16384MiB | 0% Default |\n", "| | | N/A |\n", "+-----------------------------------------+----------------------+----------------------+\n", " \n", "+---------------------------------------------------------------------------------------+\n", "| Processes: |\n", "| GPU GI CI PID Type Process name GPU Memory |\n", "| ID ID Usage |\n", "|=======================================================================================|\n", "| 0 N/A N/A 1461406 C .../miniconda3/envs/llm_env/bin/python 5264MiB |\n", "| 0 N/A N/A 3976095 C .../miniconda3/envs/llm_env/bin/python 5370MiB |\n", "+---------------------------------------------------------------------------------------+\n" ] } ], "source": [ "!nvidia-smi" ] }, { "cell_type": "code", "execution_count": 25, "id": "f2a62b30-4e8b-413e-8001-01e41aebd224", "metadata": {}, "outputs": [], "source": [ "from difflib import SequenceMatcher\n", "\n", "def calculate_overlap(start1, end1, start2, end2):\n", " # Calculate the overlapping range\n", " overlap = max(0, min(end1, end2) - max(start1, start2))\n", " # Calculate the total range covered by both spans\n", " total_range = max(end1, end2) - min(start1, start2)\n", " # Calculate the percentage overlap\n", " return (overlap / total_range) * 100\n", "\n", "def compare_entities(ground_truth, model_output):\n", " metrics = {'TP': [], 'FP': [], 'TN': [], 'FN': []}\n", " model_output_checked = []\n", "\n", " for gt_entity in ground_truth:\n", " match_found = False\n", " for mo_entity in model_output:\n", " # if gt_entity[0] == mo_entity[0] and calculate_overlap(gt_entity[2], gt_entity[3], mo_entity[2], mo_entity[3]) >= 90:\n", " percentage_match_string = SequenceMatcher(None, gt_entity[0], mo_entity[0]).ratio()\n", " if ((gt_entity[0] in mo_entity[0]) or (mo_entity[0] in gt_entity[0]) or percentage_match_string > 0.8) and (gt_entity[1] == mo_entity[1]) and calculate_overlap(gt_entity[2], gt_entity[3], mo_entity[2], mo_entity[3]) >= 50:\n", " metrics['TP'].append(gt_entity)\n", " model_output_checked.append(mo_entity)\n", " match_found = True\n", " break\n", " if not match_found:\n", " metrics['FN'].append(gt_entity)\n", "\n", " for mo_entity in model_output:\n", " if mo_entity not in model_output_checked:\n", " metrics['FP'].append(mo_entity)\n", "\n", " # Assuming True Negatives are not directly identifiable in this context\n", " return metrics\n", "\n", "def update_master_metrics(master_metrics, new_metrics):\n", " for metric in ['TP', 'FP', 'FN']:\n", " for entity in new_metrics[metric]:\n", " entity_type = entity[1]\n", " try:\n", " master_metrics[entity_type].setdefault(metric, 0)\n", " master_metrics[entity_type][metric] += 1\n", " except KeyError:\n", " continue\n", " return master_metrics\n", "\n", "# # Example Usage\n", "# ground_truth_list = [('Rama', 'NAME', 3, 8), ('Raipur', 'CITY', 9, 15)]\n", "# model_output_list = [('Rama', 'NAME', 3, 8), ('Raipur', 'CITY', 9, 15), ('Hyderabad', 'CITY', 12, 21)]\n", "\n", "# # Initialize master metrics dictionary\n", "# master_metrics = {'NAME': {}, 'CITY': {}}\n", "\n", "# # Compare entities\n", "# metrics = compare_entities(ground_truth_list, model_output_list)\n", "# print(metrics)\n", "# # Update master metrics\n", "# master_metrics = update_master_metrics(master_metrics, metrics)\n", "# print(master_metrics)" ] }, { "cell_type": "code", "execution_count": 48, "id": "2b026d0c-fd19-4f22-a228-ed2405d9d89b", "metadata": {}, "outputs": [], "source": [ "import os\n", "from tqdm import tqdm\n", "folder_name = r\"./test_files/kisqali/\"\n", "# file_name = \"11-2-pos_m5331-gp15104-p-csr-body.pdf.json\"\n", "file_name = \"11-2-pos_m5351-lee011a2301p01--legacy-clinical-study-report.pdf.json\"\n", "debug = True\n", "file_path = os.path.join(folder_name, file_name)\n", "with open(file_path, 'r') as f:\n", " gt_json = json.load(f)" ] }, { "cell_type": "code", "execution_count": 49, "id": "e0e48487-3a7e-40cd-8024-fb2717416142", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Tue Dec 26 10:55:59 2023 \n", "+---------------------------------------------------------------------------------------+\n", "| NVIDIA-SMI 545.23.08 Driver Version: 545.23.08 CUDA Version: 12.3 |\n", "|-----------------------------------------+----------------------+----------------------+\n", "| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n", "| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n", "| | | MIG M. |\n", "|=========================================+======================+======================|\n", "| 0 Tesla V100-SXM2-16GB Off | 00000000:00:1E.0 Off | 0 |\n", "| N/A 30C P0 37W / 300W | 10536MiB / 16384MiB | 0% Default |\n", "| | | N/A |\n", "+-----------------------------------------+----------------------+----------------------+\n", " \n", "+---------------------------------------------------------------------------------------+\n", "| Processes: |\n", "| GPU GI CI PID Type Process name GPU Memory |\n", "| ID ID Usage |\n", "|=======================================================================================|\n", "| 0 N/A N/A 1461406 C .../miniconda3/envs/llm_env/bin/python 5264MiB |\n", "| 0 N/A N/A 3976095 C .../miniconda3/envs/llm_env/bin/python 5270MiB |\n", "+---------------------------------------------------------------------------------------+\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n", "To disable this warning, you can either:\n", "\t- Avoid using `tokenizers` before the fork if possible\n", "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n" ] } ], "source": [ "!nvidia-smi" ] }, { "cell_type": "code", "execution_count": 50, "id": "db001fff-0906-4572-ae65-4d73c2d7bb40", "metadata": {}, "outputs": [], "source": [ "# gt_json" ] }, { "cell_type": "code", "execution_count": null, "id": "6b279b7a-5701-4897-9364-642a8894e2dd", "metadata": { "scrolled": true }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ " 0%| | 0/17737 [00:00'):\n", " if '

' in paragraph:\n", " splitter = '

'\n", " if '

' in paragraph:\n", " splitter = '

'\n", " if '

' in paragraph:\n", " splitter = '

'\n", " paragraph = paragraph.split(splitter)[1]\n", "\n", " \n", " for para_entity in para_dict['entities']:\n", " start_ = para_entity['position'][0]\n", " end_ = para_entity['position'][1]\n", " # print(label['text'], label['labels'][0])\n", " gt_list.append((para_entity['entity'], para_entity['tag'], start_, end_))\n", "\n", " # Getting model list\n", " model_output = evaluate_json_paragraph(paragraph)\n", " model_output = [(i[:4]) for i in model_output]\n", "\n", " para_metrics = compare_entities(gt_list, model_output)\n", " fp_list.extend(para_metrics['FP'])\n", " fn_list.extend(para_metrics['FN'])\n", " master_metrics = update_master_metrics(master_metrics, para_metrics)\n", " if debug:\n", " print('Para', para_no)\n", " print('GT values', gt_list)\n", " print('Model values', model_output)\n", " print('\\n')\n", " print(para_metrics)\n", " if debug:\n", " print('---------------------para-end------------------------')\n", " \n", " # if para_no == 50:\n", " # break\n", "# ['data'][5]['entities']" ] }, { "cell_type": "code", "execution_count": null, "id": "368c3e7b-72de-4a7c-b31c-8e69e4d4f040", "metadata": {}, "outputs": [], "source": [ "fn_list" ] }, { "cell_type": "code", "execution_count": null, "id": "1cd73710-896b-43df-af75-188e3ae91197", "metadata": {}, "outputs": [], "source": [ "master_metrics" ] }, { "cell_type": "code", "execution_count": null, "id": "edfa43d4-8cf3-47f2-b357-7de47cab96f8", "metadata": {}, "outputs": [], "source": [ "with open(\"./11-2-pos_m5351-lee011a2301p01--legacy-clinical-study-report_output_complete.pdf.json\", \"w\") as f:\n", " json.dump(master_metrics, f)" ] }, { "cell_type": "code", "execution_count": 37, "id": "a1fa43cc-7ea2-44be-bb20-a81a25d15ae9", "metadata": {}, "outputs": [], "source": [ "# gt_list" ] }, { "cell_type": "markdown", "id": "d3535590-cab2-48d1-81ef-1b49edd4a02b", "metadata": {}, "source": [ "### For labels of annotated format. 'label' key" ] }, { "cell_type": "code", "execution_count": null, "id": "7810a22a-137f-4a33-b34c-70b4133ceb7c", "metadata": {}, "outputs": [], "source": [ "master_metrics = {i: {} for i in ui_mapping.values()}\n", "for para_no, para_dict in tqdm(enumerate(gt_json), total=len(gt_json)):\n", " # if para_no != 33:\n", " # continue\n", " paragraph = para_dict['ner']\n", " # assert paragraph == input_paragraph\n", " if debug:\n", " print(\"Input paragraph:\", len(paragraph))\n", " print(\"Copie paragraph:\", len(input_paragraph))\n", " try:\n", " labels = para_dict['label']\n", " model_output = evaluate_json_paragraph(paragraph)\n", " model_output = [(i[:4]) for i in model_output]\n", " if debug:\n", " print('The model output is:', model_output)\n", " gt_list = []\n", " for label in labels:\n", " start_ = label['start']\n", " end_ = label['end']\n", " # print(label['text'], label['labels'][0])\n", " gt_list.append((label['text'], label['labels'][0], start_, end_))\n", " if debug:\n", " print('The ground-truth is:', gt_list)\n", " \n", " para_metrics = compare_entities(gt_list, model_output)\n", " # final_metrics['TP']\n", " if debug:\n", " print('Paragraph level metrics are:', para_metrics)\n", " master_metrics = update_master_metrics(master_metrics, para_metrics)\n", " # print(para_metrics)\n", " if debug:\n", " print('---------------------para-end------------------------')\n", " # if para_no == 10:\n", " # break\n", " except:\n", " pass" ] }, { "cell_type": "code", "execution_count": null, "id": "f2d621e8-84da-48ae-96a7-a5ecbd1bb0db", "metadata": {}, "outputs": [], "source": [ "master_metrics" ] }, { "cell_type": "code", "execution_count": null, "id": "bcc6cfc7-20e3-41b0-8d37-cb083b0ce673", "metadata": {}, "outputs": [], "source": [ "!nvidia-smi" ] }, { "cell_type": "code", "execution_count": null, "id": "a7cc9f90-c4f6-4627-8069-c34d22ceaed0", "metadata": {}, "outputs": [], "source": [ "para_dict" ] }, { "cell_type": "code", "execution_count": null, "id": "bf2efc6c-70fd-4803-91ae-e8e0f9c93c86", "metadata": {}, "outputs": [], "source": [ "len('Subject No.: 21092/12253')" ] }, { "cell_type": "code", "execution_count": null, "id": "2d9d90c9-ef31-4387-816b-3f7452221498", "metadata": {}, "outputs": [], "source": [ "result = []\n", "for value_tuple in para_value_list:\n", " value = value_tuple[0]\n", " value_entity = value_tuple[1]\n", " value_positions = get_entity_position(value, input_paragraph)\n", " for entity_position in value_positions:\n", " word_positions = get_word_from_character((entity_position[0], entity_position[1]), input_paragraph)\n", " result.append((value.strip(), value_entity, entity_position[0], entity_position[1], word_positions[0], word_positions[1]))" ] }, { "cell_type": "code", "execution_count": null, "id": "c5d822b1-5b4b-47a9-9e5a-70683a7c0670", "metadata": {}, "outputs": [], "source": [ "result" ] }, { "cell_type": "code", "execution_count": null, "id": "3a53eb23-df2b-468b-9d34-f967a5800e99", "metadata": {}, "outputs": [], "source": [ "result = []\n", "for value_tuple in para_value_list:\n", " value = value_tuple[0]\n", " value_entity = value_tuple[1]\n", " print('---', value)\n", " value_positions = get_entity_position(value, input_paragraph)\n", " print('~~~', value_positions)\n", " for entity_position in value_positions:\n", " word_positions = get_word_from_character((entity_position[0], entity_position[1]), input_paragraph)\n", " result.append((value.strip(), value_entity, entity_position[0], entity_position[1], word_positions[0], word_positions[1]))\n", "\n", "print(result)" ] }, { "cell_type": "code", "execution_count": null, "id": "ccc95d9c-7251-4113-9cfc-78305739300b", "metadata": {}, "outputs": [], "source": [ "def remove_overlapping_entities(entities):\n", " \"\"\"\n", " Remove overlapping entities from a list, giving priority to entities that appear earlier in the list.\n", "\n", " :param entities: List of entities, where each entity is represented as a tuple (start_pos, end_pos).\n", " :return: A new list of entities with no overlaps.\n", " \"\"\"\n", " non_overlapping = []\n", "\n", " for current_entity in entities:\n", " overlap = False\n", " for added_entity in non_overlapping:\n", " # Check if the current entity overlaps with any already added entity\n", " if not (current_entity[3] <= added_entity[2] or current_entity[2] >= added_entity[3]):\n", " overlap = True\n", " break\n", " \n", " if not overlap:\n", " non_overlapping.append(current_entity)\n", "\n", " return non_overlapping\n", "\n", "# Example usage\n", "# entities = [('', 0, 10), ('', 5, 15), (16, 20), (18, 25)]\n", "non_overlapping_entities = remove_overlapping_entities(result)\n", "print(non_overlapping_entities)" ] }, { "cell_type": "code", "execution_count": null, "id": "fd4ca473-c733-4bc0-84ac-1f03ad26f556", "metadata": {}, "outputs": [], "source": [ "def get_word_from_character(span, text): \n", " # Returns the word based on character position\n", "\n", " text = text.strip()\n", " words_text = text.split(\" \")\n", " words_text = [w for w in words_text if w != \"\"]\n", " wrd_index = []\n", " char_counter = 0\n", " for idx, word in enumerate(words_text):\n", " wrd_index.append((char_counter, char_counter + len(word)))\n", " char_counter += len(word) + 1\n", "\n", " left_indices = [wrd_ind[0] for wrd_ind in wrd_index]\n", " right_indices = [wrd_ind[1] for wrd_ind in wrd_index]\n", "\n", " start_char, end_char = span\n", " assert not start_char > end_char\n", "\n", " nearest_start = left_indices[0]\n", " nearest_end = right_indices[0]\n", " nearest_start_diff = 989989\n", " nearest_end_diff = 989989\n", " start_wordloc = -1\n", " end_wordloc = -1\n", "\n", " for wordloc, indx in enumerate(left_indices):\n", " diff = abs(indx - start_char)\n", " if diff <= nearest_start_diff and start_char >= indx:\n", " nearest_start_diff = diff\n", " nearest_start = indx\n", " start_wordloc = wordloc\n", "\n", " for wordloc, indx in enumerate(right_indices):\n", " diff = abs(indx - end_char)\n", "\n", " if diff <= nearest_end_diff and end_char <= indx:\n", " nearest_end_diff = diff\n", " nearest_end = indx\n", " end_wordloc = wordloc\n", "\n", " if end_wordloc == -1 or end_wordloc == start_wordloc:\n", " end_wordloc = start_wordloc\n", " return (start_wordloc, end_wordloc + 1)" ] }, { "cell_type": "code", "execution_count": null, "id": "00aecf1f-fab0-419d-8df4-ed0c95aa7338", "metadata": {}, "outputs": [], "source": [ "import re\n", "def get_gender_reg(para_data):\n", " gender_candidates = ['Male', 'Female', 'male', 'female', 'He', 'Him', 'His', 'he', 'him', 'his', 'She', 'Her', 'she', 'her']\n", " gender_identifier_string = '|'.join(gender_candidates)\n", " \n", " of_interest = []\n", " \n", " pattern = gender_identifier_string\n", " pattern_occurences = re.findall(pattern, para_data, re.IGNORECASE)\n", " # try:\n", " if pattern_occurences:\n", " for pattern in pattern_occurences:\n", " phrase_spans = re.finditer(r'\\b{0}\\b'.format(pattern), para_data)\n", " for word in phrase_spans:\n", " word_index = get_word_from_character((word.start(), word.end()), para_data)\n", " of_interest.append((para_data[word.start():word.end()], \"GENDER\", word.start(), word.end(), word_index[0], word_index[1]))\n", "\n", " of_interest = sorted(list(set(of_interest)))\n", " # except:\n", " # pass\n", "\n", " return of_interest" ] }, { "cell_type": "code", "execution_count": null, "id": "a59e751f-26d5-419f-ba87-45d4b6ae1872", "metadata": {}, "outputs": [], "source": [ "input_text = \"25291/2006 46/M 10SEP2005 66 kg 159 cm HEADACHE \"\n", "get_gender_reg(input_text)" ] }, { "cell_type": "code", "execution_count": null, "id": "39859a64-4871-4ecf-96b7-e35f2a8392e0", "metadata": {}, "outputs": [], "source": [ "input_text[37: 40]" ] }, { "cell_type": "code", "execution_count": null, "id": "e4ea194b-bec2-43b3-890e-a946b8780961", "metadata": {}, "outputs": [], "source": [ "import re\n", "def get_circumstances_reg(para_data):\n", " '''\n", " Rules for Circumstances entity identification\n", " '''\n", "\n", " with open('circumstances.txt') as f:\n", " words = f.readlines()\n", " words = [sent.strip('\\n').strip(' ') for sent in words]\n", " words = '|'.join(words)\n", "\n", " pattern = r'(\\s|^)('+words+'\\b)'\n", " prog = re.compile(pattern)\n", " pattern_occurences = prog.findall(para_data)\n", " pattern_occurences = [patt[1].strip() for patt in pattern_occurences]\n", "\n", "\n", " of_interest = []\n", "\n", " try:\n", " if pattern_occurences:\n", " for pattern in pattern_occurences:\n", " phrase_spans = re.finditer(r'\\b{0}\\b'.format(pattern), para_data)\n", " for word in phrase_spans:\n", " word_index = get_word_from_character((word.start(), word.end()), para_data)\n", " of_interest.append((para_data[word.start():word.end()], \"CIRCUMSTANCES\", word.start(), word.end(), word_index[0], word_index[1]))\n", " \n", " of_interest = sorted(list(set(of_interest)))\n", " except:\n", " pass\n", "\n", " return of_interest" ] }, { "cell_type": "code", "execution_count": null, "id": "61976447-a84c-41d8-a6f2-10c1d1b9531e", "metadata": {}, "outputs": [], "source": [ "para_data = \"There was a delivery.\"\n", "get_circumstances_reg(para_data)" ] }, { "cell_type": "code", "execution_count": null, "id": "daf3d6e1-0938-459b-ae25-f8e54fff997e", "metadata": {}, "outputs": [], "source": [ "def get_entity_position(phrase, paragraph):\n", " # Search for the phrase in the paragraph\n", " if len(phrase) <= 4:\n", " phrase = ' '+phrase\n", " positions = search_phrase_in_paragraph(phrase, paragraph)\n", " print(positions)\n", " # if phrase.startswith(' '):\n", " \n", " # Check for a 90% match of the phrase in the paragraph\n", " if len(positions) == 0:\n", " if len(phrase) > 6:\n", " positions = check_90_percent_match(phrase, paragraph)\n", " if positions:\n", " positions = merge_overlapping_tuples(positions)\n", " return positions" ] }, { "cell_type": "code", "execution_count": null, "id": "41596514-841e-4526-9558-c6c9889dfb66", "metadata": {}, "outputs": [], "source": [ "paragraph = \"This is her third visit to the place.\"\n", "value_positions = get_entity_position(\"her\", paragraph)" ] }, { "cell_type": "code", "execution_count": null, "id": "da0d2d99-8ddf-4a94-90fc-a7ab362bc92b", "metadata": {}, "outputs": [], "source": [ "result = []\n", "for entity_position in value_positions:\n", " word_positions = get_word_from_character((entity_position[0], entity_position[1]), paragraph)\n", " result.append((value.strip(), \"GENDER\", entity_position[0], entity_position[1], word_positions[0], word_positions[1])) # Append tuples to the result list" ] }, { "cell_type": "code", "execution_count": null, "id": "6206a326-d6fa-4c81-bc7b-354cc9bc4182", "metadata": {}, "outputs": [], "source": [ "result" ] }, { "cell_type": "code", "execution_count": null, "id": "4d78ff67-b541-441a-a240-259cf12290b5", "metadata": {}, "outputs": [], "source": [ "value_positions" ] }, { "cell_type": "code", "execution_count": null, "id": "0b4b86d7-4faf-4e77-8bde-369b65e151cb", "metadata": {}, "outputs": [], "source": [ "paragraph[8:11]" ] }, { "cell_type": "code", "execution_count": null, "id": "3495f86d-be95-47cc-a19d-a88c174c8f3a", "metadata": {}, "outputs": [], "source": [ "def basic_postprocessing(entity_list):\n", " filtered_list = []\n", " for entity in entity_list:\n", " if entity[1] == \"SUBJID\" and not any(char.isdigit() for char in entity[0]):\n", " continue\n", " if entity[1] == \"DATE\" and not any(char.isdigit() for char in entity[0]):\n", " continue\n", " elif entity[1] == \"DATE\" and 'day' in entity[0].lower():\n", " temp_entity = (entity[0], 'PRECLINICAL STUDY DAY', entity[2], entity[3], entity[4], entity[5])\n", " filtered_list.append(temp_entity)\n", " continue\n", " filtered_list.append(entity)\n", " return filtered_list" ] }, { "cell_type": "code", "execution_count": null, "id": "9acbda2c-f895-4400-be52-c33df9ce063b", "metadata": {}, "outputs": [], "source": [ "output_list = [('9006001', 'SUBJID', 11, 18, 2, 3), ('Day 5', 'DATE', 11, 18, 2, 3), ('This subject', 'SUBJID', 113, 125, 14, 16), ('this subject', 'SUBJID', 378, 390, 54, 56), ('this subject', 'SUBJID', 439, 451, 66, 68), ('9006001', 'SUBJID', 11, 18, 2, 3), ('This subject', 'SUBJID', 113, 125, 14, 16), ('this subject', 'SUBJID', 378, 390, 54, 56), ('this subject', 'SUBJID', 439, 451, 66, 68)]\n", "basic_postprocessing(output_list)" ] }, { "cell_type": "code", "execution_count": null, "id": "3a39dfab-46d2-42e4-9f71-10292e59ec86", "metadata": {}, "outputs": [], "source": [ "def remove_overlapping_entities(entities):\n", " \"\"\"\n", " Remove overlapping entities from a list, giving priority to entities that appear earlier in the list.\n", "\n", " :param entities: List of entities, where each entity is represented as a tuple (start_pos, end_pos).\n", " :return: A new list of entities with no overlaps.\n", " \"\"\"\n", " non_overlapping = []\n", "\n", " for current_entity in entities:\n", " overlap = False\n", " for added_entity in non_overlapping:\n", " # Check if the current entity overlaps with any already added entity\n", " if (not (current_entity[3] <= added_entity[2] or current_entity[2] >= added_entity[3])) or ((current_entity[4] == added_entity[4] or current_entity[5] == added_entity[5])):\n", " overlap = True\n", " break\n", " \n", " if not overlap:\n", " non_overlapping.append(current_entity)\n", "\n", " return non_overlapping" ] }, { "cell_type": "code", "execution_count": null, "id": "787c3d9a-f0e3-47a0-9e38-63d35d088efa", "metadata": {}, "outputs": [], "source": [ "temp_ent = [('https://www.antennahouse.com/', 'ADDRESS', 22, 51, 4, 5), ('https://www.antennahouse.com/', 'ADDRESS', 75, 104, 4, 5), ('https://www.antennahouse.com/', 'ADDRESS', 22, 51, 4, 5), ('https://www.antennahouse.com/', 'ADDRESS', 75, 104, 4, 5), ('https://www.antennahouse.com/', 'Website URL', 22, 51, 4, 5), ('https://www.antennahouse.com/', 'Website URL', 75, 104, 4, 5)]\n", "remove_overlapping_entities(temp_ent)" ] }, { "cell_type": "code", "execution_count": null, "id": "5402c531-abba-4f63-9352-fad587653a18", "metadata": {}, "outputs": [], "source": [ "Linking to an external file (attachment-sample-1.pdf).：external-destination=\"./attachment-sample-1.pdf\"\n", "Linking to a website (https://www.antennahouse.com/)：external-destination=\"https://www.antennahouse.com/\"" ] }, { "cell_type": "code", "execution_count": 60, "id": "2ff361d3-8cbb-498d-a246-b27b0df3ace7", "metadata": {}, "outputs": [], "source": [ "import boto3\n", "import json\n", " \n", "secret_client = boto3.client(service_name='secretsmanager', region_name='eu-west-1')\n", "secrets = json.loads(secret_client.get_secret_value(SecretId='SCRT-DATAANONYMIZATION-RDS-DEV-001')['SecretString'])" ] }, { "cell_type": "code", "execution_count": 61, "id": "7b409243-e6f3-405b-b8dd-92c31b58177c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'username': 'dataanonymizationdbuser',\n", " 'password': 'Gramex@123',\n", " 'engine': 'sqlserver',\n", " 'host': 'dataanonymizationdev01.cx4gszcuntlv.eu-west-1.rds.amazonaws.com',\n", " 'port': 1433,\n", " 'dbInstanceIdentifier': 'dataanonymizationdev01',\n", " 'Cipher Key': 'privacyai@2021$',\n", " 'File Cipher Key': '0123456789abcdef',\n", " 'client_id': 'd3d579b6-2596-4ac4-80df-1d6a38aeacee',\n", " 'client_secret': '80a04479-ca4e-4d5b-83c2-c8d80221d7c8',\n", " 'oauth_username': 'clinicaltrial_@pidev',\n", " 'oauth_password': 'clinicaltrial_@pidev2021',\n", " 'grant_type': 'password',\n", " 'elastic_token': '2_4+0&JjFt;aw;hK',\n", " 's3_bucket': 'novartisrsbgbiedataanonymization001',\n", " 'model_api_key': 'IMRhW6t3Uq5EJ1MxXfXo6w'}" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "secrets" ] }, { "cell_type": "code", "execution_count": 63, "id": "d6ba81e2-09a2-4c3b-a80e-e863aba2a607", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'privacyai@2021$'" ] }, "execution_count": 63, "metadata": {}, "output_type": "execute_result" } ], "source": [ "secrets['Cipher Key']" ] }, { "cell_type": "code", "execution_count": null, "id": "3e8600bf-f99d-432d-a9d6-793f4743f7d7", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.18" } }, "nbformat": 4, "nbformat_minor": 5 }