AI for evidence-based treatment recommendation in oncology: a blinded evaluation of large language models and agentic workflows
This study demonstrates that while current LLMs show promise in medical decision support, their recommendations require careful clinical supervision to ensure patient safety and optimal care. Further research is needed to improve their clinical use readiness before integration into oncology workflows. These findings provide valuable insights into the capabilities and limitations of LLMs in oncology, guiding future research and development efforts toward integrating AI into clinical workflows.
Authors
- Andrew J. Cowan (ORCID: https://orcid.org/0000-0002-1880-7094)
- Merav Bar (ORCID: https://orcid.org/0000-0001-5334-5158)
- Will Ma
- Samuel M. Rubinstein (ORCID: https://orcid.org/0000-0002-5568-4441)
- Guannan Zhai (ORCID: https://orcid.org/0009-0004-6773-9640)
- Qian Shi (ORCID: https://orcid.org/0000-0002-3777-1170)
- Ningjie Zhang
- En Xie (ORCID: https://orcid.org/0009-0006-2060-6917)
Institutions
- University of North Carolina at Chapel Hill (US)
- Cape Town HVTN Immunology Laboratory / Hutchinson Centre Research Institute of South Africa (ZA)
- Shanghai Jiao Tong University (CN)
- George Washington University (US)
- Princeton University (US)
- University of Washington (US)
- Fred Hutch Cancer Center (US)
- Bristol-Myers Squibb (United States) (US)
- Mayo Clinic in Florida (US)
Publication Details
- Journal
- Frontiers in Artificial Intelligence
- Published
- 2025-12-09
- DOI
- https://doi.org/10.3389/frai.2025.1683322
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00