Abstract
PURPOSE: The purpose of this study was to assess prognosis of estrogen receptor (ER)-low expression and its dynamics in HER2- metastatic breast cancer and to compare sensitivity to nivolumab between ER-low and triple-negative breast cancer (TNBC). EXPERIMENTAL DESIGN: Two cohorts were analyzed: a multicenter cohort of 982 patients with HER2- metastatic breast cancer and one prospective cohort of 110 patients with ER <10%/HER2- metastatic breast cancer enrolled in the TONIC trial (testing nivolumab). Endpoints were overall survival (OS) and post-relapse survival (PRS) in the retrospective cohort and progression-free survival, OS, and clinical benefit rate in the TONIC trial. RESULTS: A total of 7.3% of retrospective cases had ER-low breast cancer, 15 of 110 of the TONIC trial. In the retrospective cohort, patients with ER-low breast cancer had significantly poorer OS (P < 0.001) and numerically shorter PRS (P = 0.230) compared with ER+/HER2- breast cancer and numerically longer OS (P = 0.098) and significantly longer PRS (P = 0.017) compared with TNBC. In the TONIC trial, patients with ER-low breast cancer, compared with TNBC, showed similar response to nivolumab (clinical benefit rate: 20.0% vs. 22.1%; P = 1), progression-free survival (median, 1.7 vs. 2.0 months; P = 0.5), and OS (median, 5.3 vs. 8.6 months; P = 0.3). Among patients with primary ER+/HER2- breast cancer (n = 565), the conversion toward ER-low breast cancer or TNBC at metastasis conferred independent negative prognostic impact both for OS (P = 0.002 and P = 0.001, respectively) and PRS (P = 0.018 and P = 0.001, respectively). CONCLUSIONS: We provided evidence of the prognostic role of ER-low expression and its dynamics in patients with HER2- metastatic breast cancer. We offered insights into sensitivity to anti-PD1 in metastatic breast cancer, showing that patients with ER-low breast cancer have comparable likelihood of responding to nivolumab as those with TNBC.