In lung cancer, PD-1 blocking agents induce prominent and lasting clinical responses in nearly 20% of patients. Increased clinical benefit is associated with elevated tumor PD-L1 protein; increased nonsynonymous mutational load or HLA class-I restricted mutant neoantigens; and cognate CD8+ cytotoxic T cell presence. However, the sensitivity and specificity of these factors to predict clinical benefit to therapy are limited. In most cells, degrading cytosolic proteins including defective (e.g. mutant/truncated proteins) and viral peptides are processed by the ubiquitin-proteasome pathway to generate relatively short 8-15 amino acid long peptides that are presented by HLA class-I molecules to CD8+ T-cells in the plasma membrane. Recognition of these peptides as non-self can trigger an adaptive immune response and activation of effector cytotoxic T-cells to eliminate them. Adequate processing and presentation of peptides requires an intact antigen presenting machinery (APM). Notably, downregulation and somatic mutations of APM components have been found with variable frequency as immune evasion mechanism in diverse human tumors. Surprisingly, there is limited information in lung cancer. In preliminary studies we found prominent association between mutations in APM genes, HLA-I APM protein downregulation, reduced tumor inflammation and resistance to PD-1 axis blockade in lung cancer. We hypothesize that HLA-I APM component defects occur in a proportion of human lung cancers and mediate the anti-tumor immune evasion and resistance to anti-PD-1 therapy through blocking antigen recognition. We propose to evaluate the frequency and role of HLA-I APM defects in primary resistance to PD-1 axis blockers in human lung cancer. We will systematically screen for altered protein levels of key HLA-I APM components in samples from lung cancer patients from 2 large retrospective collections (Aim #1) and in cases treated with PD-1 axis therapies at our institution (Aim #2). The protein levels of the HLA-I APM components will be measured in situ using multiplexed quantitative immunofluorescence (QIF) panels. The HLA-I APM components include i) immunoproteasome subunits LMP2, LMP7 and LMP10, ii) HLA-A, HLA-B and HLA-C heavy chains, and 2 microglobulin; iii) HLA class I-related peptide transporters TAP1 and TAP2; iv) endoplasmic reticulum chaperones calnexin, calreticulin and tapasin; and v) MHC class I chain-related gene MICA. We will also profile tumor-infiltrating lymphocytes (TILs) using QIF and study the association between defective HLA-I APM components, tumor immune infiltration and sensitivity/response to PD-1 agents. The results from these studies could reveal critical immune evasion pathways in lung cancer, develop predictive biomarkers for optimal selection of patients for treatment with immune checkpoint blockers and suggest possible therapeutic strategies for ?non-inflamed? tumors.
Targeting the PD-1 axis induce lasting clinical benefit in ?20% of patients with lung cancer. Resistance to PD-1 axis blockers is not well understood and predictive biomarkers are suboptimal. This project will study the role of the HLA antigen presenting machinery in resistance to PD-1 blocking agents in human lung cancer.
