New research showcases the impact of large language models from extracting nuanced clinical information from unstructured data sources to successfully building digital twin models that simulate patient outcomes under different treatment scenarios

NEW YORK--(BUSINESS WIRE)--Apr. 22, 2026-- Flatiron Health today announced its presence at the ISPOR—The Professional Society for Health Economics and Outcomes Research Annual Meeting happening from May 17-20, 2026, in Philadelphia, Pennsylvania. Flatiron's high-quality real-world data and innovative research capabilities are featured across 18+ research acceptances, including seven Flatiron authored research posters as well as a panel presentation “Beyond Black Boxes: Transparent, Validated LLM Workflows for Accelerating Global HTA Submissions and Decisions.”

Flatiron’s presence at ISPOR 2026 closely follows the publication of the Validation of Accuracy for LLM/ML-Extracted Information and Data (VALID) Framework in the Journal of Clinical Oncology Clinical Cancer Informatics.The framework represents the first and most comprehensive, peer-reviewed approach to evaluating the quality and reliability of real-world data extracted by large language models and machine learning—establishing new industry standards for data integrity in oncology research.

"The rapid advancements of large language models and AI-enabled tools have required an incredibly thoughtful approach to data validation and research methodologies," said Emily Castellanos, MD, MPH, Senior Medical Director and Head of Research Oncology at Flatiron Health. "Flatiron’s presence at ISPOR reflects our commitment to advancing the field with high-fidelity, decision-ready real world evidence, reinforcing our position as the gold standard in oncology intelligence and a trusted partner for high-stakes clinical and research decisions.”

Research highlights include:

• Research assessing an LLM-approach that integrates PSA and imaging data to derive real-world progression events in prostate cancer with accuracy and completeness
• A study building and evaluating four different digital-twin modeling approaches to predict survival in patients with advanced non-small cell lung cancer
• An application of the newly published VALID framework to assess quality of an LLM-derived prostate cancer real-world dataset
• A collaborative pilot project led by Friends of Cancer Researchthat demonstrated success of real-world data supported external-control arms depends on the quality of data, consistent methodology, and attentive planning.

Join Flatiron Health at booth #316 Follow Flatiron Health on X and LinkedIn for more updates from #ISPORAnnual.

Speaking Sessions

Beyond Black Boxes: Transparent, Validated LLM Workflows for Accelerating Global HTA Submissions and Decisions
Moderator: Beth Devine
Speakers: Bill Malcolm, Tim Reason, Lockwood Taylor
Speaking Session Date/Time: Wednesday, May 20, 10:00–11:00 AM ET

Abstracts and Poster Presentations

Assessing quality of a LLM-derived prostate cancer (PC) real-world dataset: an application of the validation of accuracy for LLM/ML-extracted information and data (VALID) framework
Patrick J. Ward, Yunzhi Qian, Eunice A. Hankinson, Aaron Dolor, Melissa Estevez
Poster Session: Poster Session 2
Poster Code: MSR65
Poster Session Date/Time: Monday, May 18, 4:00 PM - 7:00 PM

Real-world treatment patterns and outcomes in patients with =2 lines of therapy for recurrent or progressive endometrial cancer
Rachel Bhak, Neeraj N. Iyer, Audrey Hopkins, Murat M. Ikiisik, Edward Kavalerchik, Mala Talekar, Xinye Li, Nada Boualam, Prakirthi Yerram, Fernanda Musa
Author affiliations: Genmab, Providence-Swedish Cancer Institute, Flatiron Health
Poster Session: 2
Poster Code: HSD38
Poster Session Date/Time: Monday, May 18, 4:00 PM - 7:00 PM

Impact of telemedicine encounters on survival outcomes: A time-varying cox analysis using EHR-derived data
Deepika Paratane, Blythe Adamson, Antal T. Zemplenyi
Author affiliations: University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences, Flatiron Health
Poster Session: 3
Poster Code: HSD60
Poster Session Date/Time: Monday, May 18, 10:30 - 1:30 PM

Customization of a large language model approach to capture PSA and imaging derived real-world progression events in prostate cancer
Kelly Magee, Patrick Ward, PhD, Wanjing Chen, Eunice Hankinson, Aaron Dolor
Poster Session: Poster Session 4
Poster Code: RWD123
Poster Session Date/Time: Tuesday, May 19, 4:00 PM - 7:00 PM

Evaluating model performance for between-country survival transportability
Mohamed S. Ali, Harlan Pittell, Elsie Horne, Philani Mpofu, Qianyi Zhang, Blythe Adamson
Poster Session: Poster Session 4
Poster Code: MSR169
Poster Session Date/Time: Tuesday, May 19, 4:00 PM - 7:00 PM

From real-world data (RWD) to digital twins: Building models for patient-level counterfactual prediction in oncology
Sandra Griffith, Joe Manfredonia, Marcello Ricottone, Richard Knoche, Aaron B. Cohen, Jacqueline Law, Melissa Estevez
Poster Session: Poster Session 5
Poster Code: MSR219
Poster Session Date/Time: Wednesday, May 20, 9:00 AM - 11:30 AM

External control arm feasibility across external data sources in oncology: Methodological and regulatory considerations
Bernat Navarro, Kawther Abdilleh, Amy Alabaster, Peter Ansell, Li Chen, Gregory S. Calip, Ruthanna Davi, Janet Espirito, Laura L. Fernandes, Sebastian Zavala Hoffmann, Patricia Luhn, Xinran Ma, Patricia Prince, Mark Riffon, Xiang Yin, Mark Stewart, Hillary Andrews, Jeff Allen
Author Affiliations: Friends of Cancer Research, Pancreatic Cancer Action Network, ConcertAI, AbbVie, Amgen, Medidata, Ontada, COTA Healthcare, iOMEDICO, Genentech, Flatiron Health, Aetion/Datavant, American Society of Clinical Oncology
Poster Session: Poster Session 5
Poster Code: RWD154
Poster Session Date/Time: Wednesday, May 20, 9:00 AM - 11:30 AM

About Flatiron

Flatiron Health is a healthtech company expanding the possibilities for point of care solutions in oncology and using data for good to power smarter care for every person with cancer. Through machine learning and AI, real-world evidence, and breakthroughs in clinical trials, we continue to transform patients’ real-life experiences into knowledge and create a more modern, connected oncology ecosystem. Flatiron Health is an independent affiliate of the Roche Group.

View source version on businesswire.com: https://www.businesswire.com/news/home/20260422858387/en/

Nina Toor
press@flatiron.com

Source: Flatiron Health