Socioeconomic and racial disparities in receipt of treatment among early-stage ER+/PR+ HER2- breast cancer: Insights from invasive breast cancer OncoDX recurrence score database.

Abstract

520 Background: Oncotype DX Recurrence Score (RS) predicts recurrence risk and potential benefit from chemotherapy (CT) in early-stage ER+ and HER2- breast cancer (BC). However, real-world CT decision patterns remain poorly understood. This work evaluated how socioeconomic, racial and clinical factors influence CT decisions beyond RS-guided recommendations. Methods: We analyzed patients with early-stage ER+/PR+, HER2- BC in SEER OncoDX RS Database (2004-2019). Variables included age, rurality, poverty status, socioeconomic status (SES), RS, histology, tumor grade, nodal status, PR status, and CT receipt. Treatment disparities among patients with similar RS were evaluated using chi-square tests. Multivariable logistic regression model (MVA) identified independent predictors of CT utilization among patients with similar RS. Results: A total of 208,674 patients were included (median age 60). Median follow up was 4.9 years. Among postmenopausal (post-M, ≥50yrs) women with RS≥26 (CT indicated), lowest SES (60.6% vs 66.5% [highest], P<0.001) and Hispanic ethnicity (63.2% vs 63.3% [White], P=0.016) were associated with lower rate of CT. When CT was not indicated in post-M women (RS<26), Black patients were more likely to receive treatment (9.0% vs 7.9% [White], P<0.001). Among premenopausal (pre-M, <50yrs) women with N0 disease and RS>15, highest SES (41.0% vs 44.2% [lowest], P=0.032) and White race (41.2% vs 46.8 [Black], P<0.001) received less CT. Among pre-M women with N1-3 disease, rural residence (42.9% vs 47.1 [urban], P = 0.004) and White race (44.6% vs 50.1% [Black], P = 0.042) were associated with less CT. MVA showed that invasive ductal carcinoma, grade III tumor, PR-negative status and Black race increased the likelihood of CT irrespective of RS (Table 1). Conclusions: CT decisions were frequently influenced by socioeconomic, racial and clinicopathological factors beyond genomic guided recommendations. Our data highlights a gap between clinical guidelines and real-world practice. Future research should focus on understanding the underlying reasons for deviations in treatment selection and examining changes in practice patterns in the post-RxPONDER era. MVA for CT decision by RS (OR, 95%CI). Pre-M, N0, ≤15 Pre-M, N0, >15 Pre-M, N1-3 Post-M, <26 Post-M, ≥26 SES (highest vs lowest) - 0.88 (0.78-0.99) - - 1.34 (1.22-1.49) Race (B vs W) - - - 1.10 (1.01-1.19) 1.14 (1.03-1.26) Histology (IDC vs ILC) - 1.22 (1.11-1.33) 1.16 (1.01-1.32) 0.84 (0.80-0.88) 1.19 (1.09-1.28) Grade (III vs I) 4.89 (3.72-6.42) 7.13 (6.42-7.93) 5.91 (4.96-7.04) 4.07 (3.80-4.36) 2.44 (2.21-2.71) PR (pos vs neg) - 0.49 (0.43-0.56) 0.44 (0.32-0.60) 0.53 (0.50-0.57) 0.82 (0.77-0.87) Not statistically significant.