Abstract PS4-12-08: Oncotype DX prediction of recurrence by race

Abstract Background: The 21-gene recurrence assay from Oncotype DX is widely used in clinical practice to help predict recurrence risk and the likelihood of benefit of chemotherapy in patients with hormone receptor positive breast cancer. Worse breast cancer outcomes among Black women have been a pernicious and well-documented problem. The role of precision medicine and technical advancements in mitigating or exacerbating these disparities is an important and quickly evolving area of investigation. The TailorRX and RxPonder trials showing the important benefit of Oncotype DX were comprised of 70-85% White patients, with 6-7% Black patients, making real-world outcomes in a diverse population of particular interest. Methods: We conducted a retrospective analysis of patients with hormone receptor positive early breast cancer, who had Oncotype DX testing performed between 2010 and 2021 at a single institution, urban safety-net hospital. Patients with intermediate risk scores, defined as 11-25, were further analyzed for recurrence. Chart review was performed April-July 2025 to assess for disease recurrence. Race was self-reported by patients and extracted from the medical record. Receipt of chemotherapy was defined as chemotherapy given to treat the breast cancer for which Oncotype testing was performed, excluding therapy for subsequent recurrences. Statistical significance was assessed using chi-square testing and logistic regression, with all analyses conducted in Python. Results: Our cohort included 245 patients with intermediate Oncotype scores of 11-25: 72 (29%) Black and 173 (71%) non-Black. Mean and median age at diagnosis were the same for Black and non-Black patients in our cohort, 54 and 55 respectively. Black patients were less likely to have intermediate Oncotype scores (48% vs 62%, p=0.03). In univariate logistic regression, odds ratio for recurrence among those with intermediate Oncotype scores was 2.19 for Black patients compared to non-Black, 95% CI 1.02-4.72, p=0.045. Comparing Black patients to White patients specifically (N=124 White patients), the odds ratio for recurrence was further increased to 2.46, 95% CI 1.05-5.75, p=0.038. In multivariate analysis (N=219 because of data availability), adjusting for age at diagnosis and receipt of chemotherapy, Black patients with intermediate Oncotype scores were still significantly more likely to have recurrence compared to non-Black patients, OR 2.34, 95% CI 1.01-5.42, p=0.047. 6% of Black patients with intermediate Oncotype scores received chemotherapy, while 12% of non-Black patients received chemotherapy, but this was not statistically significant, p=0.31. Conclusions: In our study, Black patients with intermediate Oncotype DX scores were more likely to have disease recurrence compared to their non-Black counterparts. Our findings among a diverse patient cohort at a single urban safety-net hospital suggest that 21-gene recurrence scores may have disparate predictive power for patients of different races. Further research is needed to understand the underlying factors that contribute to these disparities, particularly in settings with comparable access to care. Identifying strategies to enhance validity of predictive tools across diverse patient populations may improve risk stratification and help mitigate racial disparities in breast cancer outcomes. Citation Format: Q. S. Solfisburg, K. P. Verma, R. Soumya, N. Garg, L. Hildebrand, L. J. Oshry, M. Cassidy, R. Dries, N. Y. Ko. Oncotype DX prediction of recurrence by race [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 PS4-12-08.