Abstract PS3-01-21: Prediction of Breast Cancer Risk and Tumor Aggressiveness in the Athena Breast Health Network

Abstract Background: The Athena Breast Health Network (Athena) is a University of California (UC) and Sanford Health initiative integrating clinical care and research to drive improvements in breast screening. Here, we evaluated cancer occurrence and tumor type in our UC San Francisco (UCSF) Athena cohort. We assessed the association with 5-year invasive cancer risk level by the Breast Cancer Surveillance Consortium (BCSC) version 3 (BCSCv3) model and the development of aggressive or non-aggressive cancer. Methods: BCSCv3 risk scores were calculated for 10,031 consented UCSF Athena participants enrolled 2011-2022 with a median follow-up of 6.9 years. The BCSCv3 model includes age, race/ethnicity, prior breast biopsy results, breast density, first and second-degree family history, body mass index, menopausal status, and age at first live birth. Risk information was gathered prior to breast cancer diagnosis. The ‘high’ risk threshold was previously set at >1.67% 5-year risk by the BCSCv3 model (PMID37976443). All others were classified as ‘low’ risk.To identify those who developed invasive breast cancer after intake, we performed a cancer linkage with the San Francisco Mammography Registry and the UCSF cancer registry (complete through Dec. 2018 and Jan. 2022 respectively).Tumor aggressiveness was defined in two ways. Advanced or non-advanced cancer was classified by the AJCC prognostic pathologic stage (>stage II is advanced, <stage II is non-advanced). Faster and slower growing cancer considered tumor biology only, where one or more of the following criteria defined faster: HR-negative, HER2-positive, and/or grade 3, all other were slower growing cancers.We used logistic regressions to examine associations between BCSCv3 5-year risk score and breast cancer incidence and tumor aggressiveness. Results: Of the 10,031 participants, 2838 (28%) participants were designated as ‘high’ risk, and 7193 (72%) participants were ‘low’ risk.Overall, 213 (2%) developed breast cancer. 34 (16%) developed advanced cancer and 179 (84%) developed non-advanced cancer. Among women with breast cancer, 60 (28%) developed faster-growing cancers and 153 (72%) developed slower-growing cancers.The median 5-year BCSCv3 risk score for women with and without breast cancer was 1.73% and 1.38% respectively (p<0.001). The average 5-year BCSCv3 risk score for women with advanced and non-advanced cancers was 1.67% and 1.74%, respectively. The average 5-year BCSCv3 risk score for women with faster and slower growing cancers was 1.65% and 1.76%, respectively. In logistic regression analysis adjusted for age, women in the high-risk group had significantly higher odds of developing non-advanced cancer compared to healthy controls (OR = 3.08, 95% CI 1.87-5.08, p<0.001), whereas the association for advanced cancer was not statistically significant (OR = 1.24, 95% CI 0.56-2.77, p=0.59).Similarly, when analyzing slow vs. fast growing cancer, in logistic regression analysis adjusted for age, women in the high-risk group had significantly higher odds of developing slower-growing cancer compared to healthy controls (OR = 2.02, 95% CI 1.39-2.92, p<0.0001), whereas the association for faster-growing cancer was not statistically significant (OR = 1.62, 95% CI 0.88-2.99, p=0.12). Conclusions: This study confirms that higher 5-year BCSCv3 risk scores are associated with higher overall breast cancer development. Participants with the highest 5-year BCSCv3 risk scores are more likely to develop cancers with less aggressive features, suggesting the BCSCv3 model may preferentially predict less aggressive tumors. Those with the highest 5-year BCSCv3 risk may be more likely to benefit from endocrine risk reduction therapy. The Analysis using the recently developed BCSC advanced cancer model is ongoing and may offer better discrimination for those at risk for advanced cancers. Citation Format: K. Leggat-Barr, J. Tice, K. Badal, R. Sayaman, B. Parker, S. Hartman, H. Anton-Culver, H. Park, A. Ziogas, A. Petruse, A. Naiem, K. Malvin, L. Sabacan, T. Lewis, T. Glatt, S. Kapoor, A. Verma, P. Warner, E. Tsopurashvili, A. Griffin, V. Lee, J. McGuire, S. Borowsky, WISDOM Study and Athena Breast Health NetworkInvestigators and Advocate Partners, C. Kaplan, R. Hiatt, A. Stover-Fiscalini, K. Kerlikowske, Y. Shieh, L. Esserman, L. van ‘t Veer. Prediction of Breast Cancer Risk and Tumor Aggressiveness in the Athena Breast Health Network [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 PS3-01-21.