77P BRCA mutations may induce immune-resistance in breast cancer patients treated with IO-only approach: A retrospective analysis

Abstract

Immunotherapy (IO) has been recently introduced in triple-negative breast cancer (TNBC) therapeutic algorithm, however no predictive biomarkers beyond PDL-1 expression demonstrated utility in clinical practice. BRCA mutations account for 15-20% of TNBC, with a positive association with platinum response and an inflammatory background on tumor tissue. Preclinical evidences demonstrated a lack of T-cell activation in BRCA-mutant (BRCAm) tumors due to an HLA machinery defects, but no subgroup effects were observed in IO pivotal trials evaluating chemotherapy-IO combinations. We retrospectively evaluated the effect of germline BRCA mutations in consecutive BC patients underwent any-line IO-only treatment in phase I/II trials at Fondazione IRCCS Istituto Nazionale dei Tumori in Milan. Only mutations evaluated as pathogenic after genetic counselling were considered. IO treatment included PD(L)1, LAG3 and ICOS inhibitors, single agents or in combination. Statistical analysis were performed with RStudio software v2022.07.0. Of 39 pts treated with IO from June 2016 to September 2022, 22 had available germline BRCA status. 4 pts exhibited mutations in BRCA1 gene, 1 in BRCA2. All pts had triple-negative histology, except one (HR+). Median number of previous treatment lines for metastatic disease was 2 (1-8), without significant difference among BRCA groups (p=0.872). None of the other clinical features evaluated significantly differ among BRCA groups. 18 (81.8%) patients received platinum compound as therapy for metastatic disease, while 4 (18.2) received a PARP inhibitor. BRCAm pts demonstrated a significantly shorted PFS compared to wild-type (1.7 vs 3.6 months; HR 4.12, CI 95% 1.32-12.89, p=0.015), whereas a non-significant trend toward a worse OS was observed in BRCAm pts (77.1 vs 42.0 months; HR 2.31, CI 95% 0.65-8.24, p=0.199). A numerically higher primary resistance rate was also observed in BRCAm pts (80% vs 50%). In our small retrospective cohort, the presence of a pathogenic mutation in BRCA1/2 genes resulted in reduced benefit to IO-alone approach. If confirmed in a wider and prospective cohort, these results could guide the decision on first-line treatment of BRCAm TNBC.