Proteomic determinants of response to checkpoint blockade in malignant pleuralmesothelioma

PROJECT SUMMARY / ABSTRACT Malignant pleural mesothelioma (MPM) is a fatal cancer of the lining of the lungs that has defeated standard therapies for decades. Immune checkpoint inhibitors (ICIs) are revolutionizing cancer treatment, and tumor- specific neoantigens are critical components of the vigorous anti-tumor T cell responses possible from these

agents. Emerging clinical data show that ICIs result in meaningful extension of life in half of patients with MPM but are associated with immune-related side effects. There is no reliable biomarker for identifying MPM patients who will respond to ICIs. Such a test would avoid unnecessary toxicity, triage non-responders to potentially more

effective treatment, and could even extend long-term survival. Our preliminary data show that MPM tumors from patients that respond favorably to ICIs are rich in 1) distinct cellular and tissue architectural features of MPM's unique immune contexture, and 2) abundant tumor neoantigens that are detected at the peptide level and

expressed concomitantly with the specific HLA proteins that are required for their presentation to T cells. Our central hypothesis is that response to PD-1 blockade in MPM can be predicted by a clinically-applicable biomarker of its immunologic organization and requires neoantigen:MHC concordance. In Aim 1, we will apply

the high dimensional platforms of time-of-flight mass cytometry (CyTOF) and imaging mass cytometry (IMC) to dissect the cellular networks and immuno-architectural features of MPM that govern response to nivolumab. Based on these features, we derived a novel score that predicts response to nivolumab in MPM and we have

developed an innovative bioinformatics platform to abstract this score from standard formalin-fixed paraffin- embedded (FFPE) clinical tissue sections. In Aim 1, we will optimize and prospectively validate this score in patients with MPM. In Aim 2, we will perform mass spectrometry (MS) on HLA-typed tumors to investigate

neoantigen biology and challenge the prevailing biomarker of neoantigen burden that is used clinically to predict response to ICIs in other tumors, but with low accuracy. This biomarker has relied exclusively on neoantigens predicted in silico and does not directly measure neoantigens in tumors. We found that high quantities of MHC-

I and MHC-II neoantigens detected by MS in MPM tumors (termed neoantigen abundance) correlated with sensitivity to nivolumab. More interestingly, we found that the MPM tumors most likely to respond had high peptide level expression of neoantigens concordant with high protein level expression of the specific HLA

proteins required for their presentation (termed neoantigen:MHC concordance). These novel metrics will be tested prospectively in Aim 2 where we will also validate the immunogenicity of MS-detected neoantigens and determine the phenotype of neoantigen-reactive T cells. Our results will define core elements of the

immunoproteomic structure of MPM and result in the clinical translation of innovative biomarkers expected to directly improve the care of MPM patients. More broadly, completion of this project will advance our understanding of neoantigen biology and of mechanisms of response and resistance to ICIs in human cancer.