Comparison of predicted benefit using RS clin versus observed benefit in a U.S. registry of stage I ER-positive HER2-negative high oncotype DX RS breast cancer.

Abstract

544 Background: The 21-gene recurrence score assesses the risk of distant breast cancer recurrence and predicts the benefit of adjuvant chemotherapy in ER positive early-stage breast cancer. However, clinicopathologic risk factors continue to impact absolute benefit of adjuvant chemotherapy. Absolute benefit of adjuvant chemotherapy in low clinicopathologic risk Stage I ER positive breast cancer with high Oncotype RS is an important clinical component of shared decision making. The RSClin clinical tool, which integrates the 21-gene recurrence score (RS) and clinicopathologic features, was found to be more prognostic than RS result alone. We assessed RS Clin predicted benefit and compared to absolute benefit in low clinicopathologic risk Stage I ER positive breast cancer with high Oncotype RS as observed in the NCDB. Methods: Using the National Cancer Data Base (NCDB), we identified female patients age 18-75, ER positive and Her 2 negative, 21 gene signature high risk > 25 with negatives surgical margins, <2cm, lymph node negative breast cancers, who had received endocrine therapy with either Tam or AI diagnosed during 2010 - 2017. Clinicopathologic factors and RS were entered into the RSClin calculator, and a predicted benefit of chemotherapy was calculated for each patient. Using NCDB data, Cox proportional hazards regression models were used to project absolute survival benefit at 10 years from diagnosis for those who did vs those who did not receive chemotherapy. NCDB-derived absolute benefit of chemotherapy was compared to estimated absolute benefit as predicted by the RSClin tool. Results: 18,226 patients were identified. Stages T1a = 670(4%), T1b = 4,289(23%), and T1c = 13,267(73%). Median age 59 years (21-75). Race white 84% (15,365), black 10% (1840), and other 1021(6%). AI use 80% (14,711) was greater than Tam use 20% (3515). Chemotherapy was administered in 75% (13,827) of patients. Most patients had high or intermediate grade disease, G3 45% (8225), G2 46% (8328), and G1 9% (1672). Median duration of follow up was 57 months (2-160). Probability of death T1b at 10yrs with chemotherapy was 8.5%, and without chemotherapy was 15.1% with an absolute benefit of 6.6%. Predicted benefit in T1b using RS Clin at 10yrs was 10.8%. Probability of death for T1c at 10yrs with chemotherapy was 15.1%, and without chemotherapy was 23.5% with an absolute benefit of 8.4%. Predicted benefit in T1c using RS Clin at 10yrs was 14%. Conclusions: Patients with stage IB and IC hormone receptor positive HER2 negative breast cancers with high RS had a lower absolute benefit than predicted by RS Clin. The RSClin tool overestimated benefit of therapy in both IB and IC stages requiring caution when using this tool in patients with the lowest clinicopathologic risks.