{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T14:08:31Z","timestamp":1793801311246,"version":"4.6.1"},"reference-count":24,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Cancer"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Screening for clinical trials is challenging for clinicians due to its time-consuming and repetitive nature. The rise of artificial intelligence (AI) offers an opportunity to improve screening productivity and reproducibility. Pancreatic cancer is characterized by increasing incidence, poor survival outcomes, and an urgent need for improved management strategies.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to assess the performance of AI in evaluating clinical trial inclusion and exclusion criteria, compared to a double-blind human gold standard, using a retrospective cohort.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n In the PANCR-AI (Pancreatic Cancer Retrospective Screening with Artificial Intelligence) pilot study, we retrospectively reviewed cases from our institutional database of patients with advanced pancreatic cancer presented at tumor board meetings between January 2018 and December 2023. Each patient was screened for clinical trials open for inclusion at the time of the multidisciplinary meeting. Manual screening of eligibility criteria for each patient-trial pair was performed by 2 blinded oncologists to determine potential eligibility (gold standard), with a third oncologist resolving discrepancies. Potential eligibility was also assessed using 3 large language models (ie, GPT-4.5, Claude 3.7 Sonnet, and Mistral-7B-Instruct v0.3). Their performance was compared to the human gold standard using standard evaluation metrics (eg, sensitivity, specificity, precision, recall, and\n F<\/jats:italic>\n 1<\/jats:sub>\n -score). Correlations between the risk of failure and the number of words and characters in the criteria were analyzed. The time required to complete the screening was recorded for both human and AI assessments. The number of trials open for enrollment at the time of the tumor board meeting was also recorded as a variable for analysis.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Across 341 patient-trial pairs, the AI models demonstrated high sensitivity, ranging from 83.3% to 92.2%. Analysis of the criteria showed a correlation between the risk of failure and the number of words and the number of characters in the criteria. Overall screening time for manual assessment was significantly longer for the human gold standard (44.70 hours) assessment than for AI (2.53-3.15 hours). Patients were more likely to have been included in a clinical trial if the number of trials open for enrollment was higher at the time of the tumor board meeting (\n P<\/jats:italic>\n =.02).\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Our study highlights the promising performance of AI in clinical trial screening. Future work should explore integration with structured clinical data, such as laboratory values or radiological findings, to improve multimodal comprehension. Expanding the evaluation to a broader range of tumor types and multicenter datasets would improve generalizability. Finally, real-time prospective validation and workflow integration with electronic health records will be critical to assess the feasibility and clinical impact of large language model\u2013assisted screening in daily oncology practice. Addressing these challenges will be essential to move from proof of concept to scalable clinical implementation.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/80268","type":"journal-article","created":{"date-parts":[[2026,2,23]],"date-time":"2026-02-23T21:26:07Z","timestamp":1771881967000},"page":"e80268-e80268","source":"Crossref","is-referenced-by-count":2,"title":["AI-Driven Patient Screening for Clinical Trials in Pancreatic Cancer: The PANCR-AI Pilot Retrospective Comparative Study"],"prefix":"10.2196","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-0683-8313","authenticated-orcid":false,"given":"Arthur","family":"Claessens","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6321-2465","authenticated-orcid":false,"given":"Aliz\u00e9e","family":"Simon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-0804-5323","authenticated-orcid":false,"given":"Agathe","family":"Manchart","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-4102-2714","authenticated-orcid":false,"given":"Laura","family":"Desmonts","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-7736-7923","authenticated-orcid":false,"given":"Cassandre","family":"Leguay","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2617-3562","authenticated-orcid":false,"given":"Jean-Christophe","family":"Faivre","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6876-6473","authenticated-orcid":false,"given":"S\u00e9bastien","family":"Cambier","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7921-4583","authenticated-orcid":false,"given":"Julia","family":"Salleron","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9990-1201","authenticated-orcid":false,"given":"Aur\u00e9lien","family":"Lambert","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,2,23]]},"reference":[{"issue":"1","key":"R1","doi-asserted-by":"publisher","first-page":"10","DOI":"10.3322\/caac.21871","article-title":"Cancer statistics, 2025","volume":"75","author":"Siegel","journal-title":"CA Cancer J Clin"},{"issue":"19","key":"R2","doi-asserted-by":"publisher","first-page":"1817","DOI":"10.1056\/NEJMoa1011923","article-title":"FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer","volume":"364","author":"Conroy","journal-title":"N Engl J Med"},{"issue":"18","key":"R3","doi-asserted-by":"publisher","first-page":"1691","DOI":"10.1056\/NEJMoa1304369","article-title":"Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine","volume":"369","author":"Von Hoff","journal-title":"N Engl J Med"},{"issue":"4","key":"R4","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering11040337","article-title":"The role of AI in hospitals and clinics: transforming healthcare in the 21st century","volume":"11","author":"Maleki Varnosfaderani","journal-title":"Bioengineering (Basel)"},{"issue":"9","key":"R5","doi-asserted-by":"publisher","DOI":"10.1093\/jnci\/dju229","article-title":"Adult cancer clinical trials that fail to complete: an epidemic?","volume":"106","author":"Stensland","journal-title":"J Natl Cancer Inst"},{"issue":"4","key":"R6","doi-asserted-by":"publisher","first-page":"e174","DOI":"10.1200\/JOP.2012.000763","article-title":"Screening intervention to identify eligible patients and improve accrual to phase II-IV oncology clinical trials","volume":"9","author":"Chen","journal-title":"J Oncol Pract"},{"issue":"2","key":"R7","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1007\/s10115-022-01779-1","article-title":"Information extraction from electronic medical documents: state of the art and future research directions","volume":"65","author":"Landolsi","journal-title":"Knowl Inf Syst"},{"issue":"4","key":"R8","doi-asserted-by":"publisher","first-page":"286","DOI":"10.4258\/hir.2023.29.4.286","article-title":"Named entity recognition in electronic health records: a methodological review","volume":"29","author":"Durango","journal-title":"Healthc Inform Res"},{"issue":"Suppl 5","key":"R9","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-019-0931-8","article-title":"Natural language processing for populating lung cancer clinical research data","volume":"19","author":"Wang","journal-title":"BMC Med Inform Decis Mak"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2024.104720","article-title":"Extracting lung cancer staging descriptors from pathology reports: a generative language model approach","volume":"157","author":"Cho","journal-title":"J Biomed Inform"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-015-0149-3","article-title":"Increasing the efficiency of trial-patient matching: automated clinical trial eligibility pre-screening for pediatric oncology patients","volume":"15","author":"Ni","journal-title":"BMC Med Inform Decis Mak"},{"key":"R12","unstructured":"ChatGPT (version 45). OpenAI. URL: https:\/\/chat.openai.com\/ [Accessed 27-05-2025]"},{"key":"R13","unstructured":"Claude 37 sonnet. Anthropic. 2025. URL: https:\/\/www.anthropic.com\/news\/claude-3-7-sonnet [Accessed 27-05-2025]"},{"key":"R14","unstructured":"Jiang AQ Sablayrolles A Mensch A et al. Mistral 7B. arXiv. Preprint posted online on Oct 10, 2023. [doi: 10.48550\/arXiv.2310.06825]"},{"key":"R15","unstructured":"Guiding principles on the use of large language models in regulatory science and for medicines regulatory activities. European Medicines Agency. 2024. URL: https:\/\/www.ema.europa.eu\/en\/documents\/other\/guiding-principles-use-large-language-models-regulatory-science-medicines-regulatory-activities_en.pdf [Accessed 21-10-2025]"},{"issue":"12","key":"R16","doi-asserted-by":"publisher","first-page":"2629","DOI":"10.1007\/s10439-023-03272-4","article-title":"Prompt engineering with ChatGPT: a guide for academic writers","volume":"51","author":"Giray","journal-title":"Ann Biomed Eng"},{"issue":"3","key":"R17","doi-asserted-by":"publisher","DOI":"10.2196\/27767","article-title":"Accuracy of an artificial intelligence system for cancer clinical trial eligibility screening: retrospective pilot study","volume":"9","author":"Haddad","journal-title":"JMIR Med Inform"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1200\/CCI.19.00079","article-title":"Artificial intelligence tool for optimizing eligibility screening for clinical trials in a large community cancer center","volume":"4","author":"Beck","journal-title":"JCO Clin Cancer Inform"},{"issue":"5","key":"R19","doi-asserted-by":"publisher","first-page":"473","DOI":"10.1016\/j.bulcan.2024.01.010","article-title":"D\u00e9veloppement d\u2019une solution d\u2019intelligence artificielle pour am\u00e9liorer le screening en recherche clinique","volume":"111","author":"G\u00e9dor","journal-title":"Bull Cancer"},{"issue":"2","key":"R20","doi-asserted-by":"publisher","first-page":"209","DOI":"10.1093\/jamiaopen\/ooaa002","article-title":"Evaluation of an artificial intelligence clinical trial matching system in Australian lung cancer patients","volume":"3","author":"Alexander","journal-title":"JAMIA Open"},{"issue":"1","key":"R21","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1007\/s43441-019-00030-4","article-title":"Improving clinical trial participant prescreening with artificial intelligence (AI): a comparison of the results of AI-assisted vs standard methods in 3 oncology trials","volume":"54","author":"Calaprice-Whitty","journal-title":"Ther Innov Regul Sci"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-53081-z","article-title":"Matching patients to clinical trials with large language models","volume":"15","author":"Jin","journal-title":"Nat Commun"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","DOI":"10.1186\/s12885-024-11959-7","article-title":"Evaluation of an artificial intelligence-based clinical trial matching system in Chinese patients with hepatocellular carcinoma: a retrospective study","volume":"24","author":"Wang","journal-title":"BMC Cancer"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.1016\/j.annonc.2025.12.006","article-title":"Next-generation multicenter studies: using artificial intelligence to automatically process unstructured health records of patients with lung cancer across multiple institutions","author":"Aldea","journal-title":"Ann Oncol"}],"container-title":["JMIR Cancer"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/cancer.jmir.org\/2026\/1\/e80268","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:13:39Z","timestamp":1782843219000},"score":1,"resource":{"primary":{"URL":"https:\/\/cancer.jmir.org\/2026\/1\/e80268"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,23]]},"references-count":24,"URL":"https:\/\/doi.org\/10.2196\/80268","relation":{},"ISSN":["2369-1999"],"issn-type":[{"value":"2369-1999","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,23]]},"article-number":"v12i1e80268"}}