• Flatiron’s presence at ESMO AI demonstrates deep methodological expertise in AI and machine learning applications across global markets and showcases the possibility of trustworthy LLM-extracted real-world data at scale.

Flatiron Health today announced its presence at the European Society For Medical Oncology AI & Digital Oncology Congress 2025 (ESMO AI) taking place on November 12-14 in Berlin, Germany. Flatiron’s experts will be presenting five pieces of research, in addition to leading the speaking session, “The VALID Framework for LLM-Extracted Oncology Data Quality: Comprehensive methodology for evaluating LLM-extracted real-world oncology data quality.”

“Our research presented at ESMO AI demonstrates how LLM-extracted real-world data can be rigorously validated across diverse healthcare systems and countries, empowering our partners to generate credible, fit-for-purpose evidence that informs and advances care and research worldwide,” said Nathan Hubbard, Chief Executive Officer at Flatiron Health. “Flatiron’s team of experts continues to demonstrate what responsible AI in oncology looks like—advancing the field through methodological rigor, privacy-preserving approaches, and the frameworks that make AI trustworthy at scale."

Research highlights include:

• The company's first published research exploring the development of digital twin modeling that incorporates LLM-extracted clinical context to enable more personalized survival predictions and treatment planning.

• Multiple studies advancing the use of LLMs in extracting real-world oncology data through validation and quality evaluation across tumor types, healthcare systems, and countries.

• Applications of the company’s Validation of Accuracy for LLM/ML-Extracted Information and Data (VALID) Framework data quality framework, the industry's first comprehensive approach to evaluating AI-extracted real-world data quality.

Schedule a meeting with Flatiron Health at ESMO AI 2025 and follow Flatiron Health on X and LinkedIn for more updates from #ESMOAI25.

Abstracts and Poster Presentations

Structuring GDPR-Compliant Networks to Enable LLM-Extracted Oncology Data on Pseudonymized Patient EHR Data in Europe

Lauren Ellsworth, Lucia Groizard, Florian Stefan, Anna Schwarz, Natalia Viani, Katherine Harrison, Andy Hadjigeorgiou, Blythe Adamson, Igor Serko, Daniel Farrar, Maximilian Hertstein, Yohann Leon, Matt Murchison, Kathi Seidl-Rathkopf

Presentation Number: 412P

Presentation Date: Wednesday, November 12, 2025

Privacy-Preserving Error Analysis Loop For ML-based Extraction of Oncology EHR Data

Lucia Groizard, Lauren Ellsworth, Florian Stefan, Anna Schwarz, Natalia Viani, Katherine Harrison, Andy Hadjigeorgiou, Blythe Adamson, Igor Serko, Kathi Seidl-Rathkopf, Nikola Dolezalova, Serena Dover, Marina Kushnir, Amit Samani

Presentation Number: 393P

Presentation Date: Wednesday, November 12, 2025

A Pan-Tumor and Pan-Country Approach to LLM-Based Extraction of Systemic Therapies From the Electronic Health Record

Natalia Viani, Lucia Groizard, Katherine Harrison, Anna Schwarz, Nikola Dolezalova, Florian Stefan, Andy Hadjigeorgiou. Megan Hildner. Hinrik Gudmundsson, Amit Samani, Serena Dover, Marina Kushnir, Blythe Adamson, Lauren Ellsworth, Kathi Seidl-Rathkopf

Presentation Number: 303P

Presentation Date: Wednesday, November 12, 2025

A Framework for Evaluating Performance of LLM-Based Extraction From the Electronic Health Record Across Different Healthcare Systems

Kathi Seidl-Rathkopf, Anna Schwarz, Natalia Viani, Lucia Groizard, Nikola Dolezalova, Katherine Harrison. Lauren Ellsworth. Jack Burnett, Bryan Bowser, Igor Serko, Nisha Singh, Blythe Adamson, Melissa Estevez, Lauren Dyson

Presentation Number: 389P

Presentation Date: Wednesday, November 12, 2025

Survival Prediction in Advanced NSCLC (aNSCLC) Amid Evolving Standards of Care (SOC): Digital Twin Modeling Incorporating LLM-Extracted Clinical Context

Melissa Estevez, Sandra Griffith, Tori Williams, Larry Bellomo, John Ritten, Patrick Ward, Joe Manfredonia, Marcello Ricottone, Jonathan Kish, Jacqueline Law, Aaron Cohen

Presentation Number: 386P

Presentation Date: Wednesday, November 12, 2025

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/20251106742067/en/

“Flatiron’s team of experts continues to demonstrate what responsible AI in oncology looks like—advancing the field through methodological rigor, privacy-preserving approaches, and the frameworks that make AI trustworthy at scale."