Abstract PS3-04-14: Artificial intelligence (AI) as a decision support tool for practicing oncologists: breast cancer cases

Abstract

Abstract Background: Second opinions and multidisciplinary review (MDR) can inform and impact the management of cancer patients, yet these resources are not universally accessible. We previously presented over 400 anonymized complex cancer scenarios (cases) via video conference to MDR panel experts—including specialists in medical oncology, surgical oncology, radiation oncology, radiology, and a moderator—across 5 major tumor types and tracked their treatment recommendations. In this study, we focus on breast cancer cases to compare how recommendations from 3 leading AI models compare to those of MDR panel experts. Methods: We selected 50 complex breast cancer scenarios previously adjudicated by MDR panels from a larger database of over 400 anonymized cancer cases collected between 2020 and 2021. These cases were analyzed by 3 different foundational AI models (OpenAI’s ChatGPT 4.5, Anthropic’s Claude Opus 4 and Google’s Gemini Ultra) using PrecisCa’s proprietary method for prompts. The AI recommendations from each system were scored on a scale of 1-5 (5 being the highest) for completeness, reasoning, clarity, menu of options, recency, and relevance as compared to the MDR panel recommendations. The maximum possible competence score was 30 points per scenario and an aggregate score of 1500. Final AI recommendations were also compared with current National Comprehensive Cancer Network (NCCN) guidelines for serious omissions. Comparison in reverse (additional AI options that the experts missed) was not performed due to interval changes in treatment recommendations over the past 4 years. Results: Patient characteristics and aggregate concordance and competence scores are listed in Table 1 below. While there was some variation in the concordance and competence rates of the 3 AI models, there was excellent concordance with the expert opinion for all of them. Discordant cases were reviewed and primarily involved minor differences that would not have altered the patients’ management significantly. Relatively speaking within the 6 categories listed above, AI systems excelled in clarity and reasoning and less so in relevance and menu of options. Conclusions: This study demonstrates a high degree of concordance between 3 leading AI models and expert MDR panels for common yet complex breast cancer clinical scenarios. These findings suggest that AI tools have matured sufficiently to serve as valuable decision support aids in oncology practice, particularly where expert review may be limited or unavailable. Careful human oversight remains essential to ensure safe and personalized cancer care. Citation Format: M. Jahanzeb, K. Haines, T. Buchholz, R. Butler, W. J. Gradishar, S. A. Hurvitz, T. A. King, R. L. Mahtani, T. Mamounas, H. McArthur, M. Morrow, A. P. O'Dea, J. O'Shaughnessy, M. D. Pegram, H. S. Rugo, C. Shah, F. Vicini, N. Wolmark. Artificial intelligence (AI) as a decision support tool for practicing oncologists: breast cancer cases [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS3-04-14.

Ähnliche Arbeiten