Abstract
586 Background: A recurrence score (RS) derived from a 21-gene RT-PCR assay (Oncotype DX) stratifies patients (pts) with early estrogen receptor (ER)-positive, HER2-normal breast cancer (BC) into 3 groups based on risk of 10-year distant metastasis rate: low, intermediate, and high. High RS predicts benefit from chemotherapy follow-by hormonal therapy (CH-T); whether intermediate RS predicts benefit from CH-T is unknown. For Stage I, NCCN Clinical Practice Guidelines recommend use of RS for T1b BC with unfavorable features (angiolymphatic invasion, high histologic or nuclear grade) and T1c BC. However, data are limited on how RS influences treatment of stage I BC. We sought to define the impact of RS on treatment in pts with stage I BC. Methods: In 1057 pts with ER-positive, HER2-normal stage I BC with RS available, we assessed the interaction between RS and treatment by tumor category. Kaplan-Meier survival curves and log-ranktests were used to compare outcomes. Results: Pts with T1a, T1b, and T1c BC did not differ in the distribution of low, intermediate and high RS (P=0.53) (Table). In the T1b and T1c groups, 57% and 52%, respectively, of pts with intermediate RS did not receive chemotherapy. Median follow-up is 25.7 months. Among T1b pts with intermediate RS, disease-free survival did not differ between pts who received HT and those who received CH-T (P=0.84). Among T1b pts with intermediate RS, pts who received CH-T had higher RS than pts who received HT (median 23 vs 21, P<0.001). Conclusions: Less than 10% of patients with stage I had high RS.More than half of pts with stage I BC and intermediate RS received HT alone. Pts with higher RS were more likely to receive CH-T. Among pts with T1b BC and intermediate RS, pts who received HT alone had the similar DFS as those who received CH-T.