Racial differences in predictive value of the 21-gene recurrence score assay: a population-based study using the SEER database.

Purpose: The 21-gene recurrence score (RS) assay is currently used for predicting chemotherapeutic benefits for hormone receptor-positive (HR +) early-stage breast cancer patients without consideration regarding racial differences in that predictive value. This study aimed at demonstrating racial differences in the predictive values of the 21-gene RS assay.

Methods: The study cohort was selected from the Surveillance, Epidemiology, and End Results (SEER) database. Breast cancer-specific mortality (BCSM) was compared between patients who received chemotherapy (the "CTx group") and those who did not (the "no CTx group") to estimate the predictive value of the assay. This comparison was repeated for each racial group.

Results: Among 88,498 T1 - 2N0 HR + breast cancer patients who had results of 21-gene RS, 13,123 patients had RS > 25, which included 10,697 Whites, 1282 Blacks, and 1,144 Asian Americans/Pacific Islanders (AAPIs). Chemotherapy was administered to 8364 patients (63.4%). The adjusted hazard ratio for BCSM in the CTx group (vs. no CTx group) was 0.734 (95% confidence interval [CI] 0.588-0.917) in Whites, 0.748 (95% CI 0.428-1.307) in Blacks, and 1.343 (95% CI 0.558-3.233) in AAPIs. No subgroup within patients with RS > 25 among non-White women showed a significant predictive value of the 21-gene RS assay, except for Black women with grade 3 tumors.

Conclusion: The predictive value of the 21-gene RS assay for assessing chemotherapy benefit was validated in White women based on the SEER database, although the predictive value was not warranted in non-White women.