Associations of immune checkpoint predictive biomarkers MHC-I and MHC-II with clinical and molecular features in a diverse breast cancer cohort.

Abstract

Immunotherapy (IO) in triple negative breast cancer (TNBC) has improved survival outcomes, with promising improvements in pCR rates among early high-risk HR+/HER2- breast cancers. However, biomarkers are needed to select patients likely to benefit from IO. MHC-I and tumor-specific MHC-II (tsMHC-II) expression are candidate biomarkers for PD-(L)1 checkpoint inhibition, but existing data from clinical trials included limited racial/ethnic diversity. We performed multiplexed immunofluorescence assays in the Carolina Breast Cancer Study (CBCS, n=1628, 48% Black, 52% non-Black). Intrinsic subtype and P53 mutant-like status were identified using RNA-based multigene assays. We ranked participants based on tumoral MHC-I intensity (top 33% categorized as "MHC-IHigh") and MHC-II+ (≥5% of tumor cells as tsMHC-II+). MHC-I/II were evaluated in association with clinicopathological features by race. Black participants had higher frequency of TNBC (25% vs. 12.5%, p = < 0.001) and Basal-like (30% vs. 14%, p = < 0.001) tumors overall, and higher frequency of Basal-like (11% vs. 5.5%, p = 0.002) and TP53 mutant tumors (26% vs. 17%, p = 0.002) among HR+/HER2-. The frequency of tsMHC-II+ was higher in HR+/HER2- Black participants (7.9% vs. 4.9%, p = 0.04). Black participants also had higher frequency of MHC-Ihigh (38.7% vs. 28.2%, p <0.001), which was significant among HR+/HER2- (28.2% vs. 22.1%, p = 0.02). In this diverse study population, MHC-I and MHC-II tumor cell expression were more highly expressed in HR+/HER2- tumors from Black women, underscoring the importance of diverse and equitable enrollment in future IO trials.