AI Model Enhances Recurrence Risk Assessment in HR-Positive, HER2-Negative Breast Cancer
A groundbreaking AI model, developed through the fusion of clinical, molecular, and histopathological data, has significantly improved recurrence risk stratification in hormone receptor (HR)-positive, HER2-negative breast cancer, as presented at the San Antonio Breast Cancer Symposium (SABCS) in December 2025. This subtype of breast cancer is the most prevalent, with at least 50% of recurrences occurring post-five years, as explained by Joseph A. Sparano, MD, chief of the Division of Hematology and Oncology at the Mount Sinai Tisch Cancer Center.
The Oncotype DX (ODX) 21-gene recurrence score, a widely used prognostic and predictive tool, has limitations in forecasting recurrence beyond the five-year mark. Sparano's team aimed to develop a more comprehensive diagnostic test by studying tumor specimens from the TAILORx trial. Their AI model, ICM+, evaluates both digitized slide images and molecular, clinical characteristics, offering a more accurate prognosis for cancer recurrence up to 15 years, including early and late recurrence.
ICM+, a multimodal model integrating pathomic imaging, clinical, and expanded molecular data, outperformed ODX in predicting overall distant recurrence at 15 years (C-index 0.705 vs. 0.617) and late recurrence post-five years (C-index 0.656 vs. 0.518) in a training set of 2,806 patients. This superior performance was also evident in a holdout validation set of 1,621 patients, with C-indices of 0.733 and 0.705, respectively.
This research, a collaboration between the ECOG-ACRIN Cancer Research Group and Caris Life Sciences, has the potential to lead to a new diagnostic test for HR-positive, HER2-negative, node-negative breast cancer, which constitutes half of all breast cancers in the U.S. Sparano emphasized the AI's ability to provide more accurate recurrence risk estimates and personalized treatment decisions.
However, Sparano noted that the study's scope did not include predicting chemotherapy benefit or the efficacy of long-term endocrine therapy beyond five years. The research was funded by the Breast Cancer Research Foundation, the National Cancer Institute, and the U.S. Postal Service Breast Cancer Research Stamp Fund. Sparano's consulting roles and institutional research support were also disclosed.
This study highlights the potential of AI in developing more precise diagnostic tools, offering hope for improved treatment outcomes for breast cancer patients.
