MR1 in combination with tumor mutational burden and PD-1/PD-L1 expression as a potentially novel clinical predictor for T cell exhaustion and immune checkpoint inhibitor response.

Abstract

3109 Background: Immune checkpoint inhibitors (ICIs) restore T cell function by reversing T cell exhaustion. Variable response to ICIs warrants the development of precise predictive biomarkers, which is challenging due to difficulty in capturing the interplay of factors involved with tumor cell immune recognition. High intratumoral expression of MR1, the MHC-I related protein basally expressed on cancer cells, may drive T cell exhaustion through presentation of cancer-specific antigens. Here, we construct a database to study the relationship between MR1, tumor mutational burden (TMB), the PD-1/PD-L1 axis and T cell exhaustion across 8,975 sequenced tumors and 27 cancer types. Methods: RNA Seq by expectation maximization (RSEM) values from the TCGA were collected and normalized along with expression data for markers of interest (Table). TMB was defined as the number of non-synonymous somatic mutations per sample. For each cancer, 5 cohorts were created based on ascending mean expression levels of MR1, PD-1, PD-L1, and increasing TMB. For each cancer, an “immunogenicity score” for these factors was computed, and its relationship with T cell exhaustion signatures was assessed via linear regression. Data is presented as adjusted R2 and p-value. Results: While PD-1 and T cell exhaustion marker expression were correlated across cancers, the “immunogenicity score” (IS) correlated with exhaustion markers specifically in cancers with FDA-approved ICIs. Excluding MR1 from the score weakened the correlation with EOMES and TBET expression (Table). Each component of the score analyzed independently failed to show a statistically significant correlation for both EOMES and TBET expression. Conclusions: In this cross-cancer analysis, we support the hypothesis that presentation of metabolic intermediates in cancer cells via MR1 may drive T cell exhaustion. Also, the novel “immunogenicity score”, which incorporates MR1 into standard biomarkers for response to ICIs may convey the global picture of cancer cell recognition by the immune system and warrants further investigation as a tool for predicting clinical response. [Table: see text]