Abstract PS3-08-08: Anthracycline vs. Non-Anthracycline Chemotherapy Regimens for ER+/HER2− Early-Stage Breast Cancer Based on Oncotype DX Recurrence Score: A Retrospective Study

Abstract Introduction: Choosing between anthracycline (A)-based versus non-A-based chemotherapy (chemo) regimens among patients with estrogen receptor-positive (ER+), HER2-negative (HER2-) early-stage breast cancer (BC) remains challenging due to the lack of data to guide this decision. In this study, we utilized the Flatiron Health Research Database to examine demographic, clinical, and pathologic factors associated with administration of A versus non-A-based chemo regimens for patients with stage I-II, ER+/HER2- BC. Methods: Patients age 18+, who were diagnosed with stage I-II, N0-1, ER+/HER2- BC, who initiated chemo within 180 days of BC diagnosis, and who had an available Oncotype DX recurrence score (RS) were included in this cohort. Patients who received at least one dose of doxorubicin were designated as having received A-based chemo. Logistic regression was performed to determine associations between the A-based chemo decision and each clinical factor. Cox regression was performed to compare survival probabilities between patients who received A- versus non-A-based chemo, using propensity score matching to balance baseline clinical characteristics and reduce confounding due to treatment selection. Results: Of the 512 patients included in the analysis, 406 (79.3%) received non-A-based chemo. Factors significantly associated with receiving A-based chemo included higher stage, larger tumor size, lymph node involvement, younger age, and higher RS (Table 1). Tumor grade was not significantly associated. In a multivariable analysis, factors that remained significant included stage, lymph node, age, and RS. We matched patients 1:1 on these factors, resulting in 190 participants for survival analysis. Patients who received A-based regimens had numerically better, though not statistically significant, recurrence-free survival (RFS) compared to those who received non-A-based regimens (hazard ratio [HR]=0.64 [0.24-1.70], p=0.37). A-based regimen had a lower HR in the subset of patients with RS>30 (HR=0.57 [0.12-2.64], p=0.90) than those with RS<=30 (HR=0.71 [0.20-2.52], p=0.60), although the difference was not significant. Conclusion: Our study suggests that oncologists use treatment with anthracycline-containing chemotherapy regimens for patients who are younger and those with higher-risk BC. In this relatively small, retrospective cohort, there was a suggestion of improved RFS with anthracycline, but confidence intervals were wide. Larger randomized studies or meta-analyses are needed to determine whether anthracyclines truly improve outcomes in these patients. Citation Format: M. Wei, C. Carroll, S. Kimani, H. N. Lynn, F. Bingjian. Anthracycline vs. Non-Anthracycline Chemotherapy Regimens for ER+/HER2− Early-Stage Breast Cancer Based on Oncotype DX Recurrence Score: A Retrospective Study [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 PS3-08-08.