An integrative computational framework for personalized detection of tumor epitopes in melanoma immunotherapy

Abstract

In aggressive solid tumors like melanoma, a strategy for therapy personalization can be achieved by combining high-throughput data on the patient's specific tumor mutation and expression profiles. A remarkable case is dendritic cell-based immunotherapy, where tumor epitopes identified from the patient's specific mutation profiles are loaded on patient-derived mature dendritic cells to stimulate cytotoxic T cell mediated anticancer immunity. Here we present a personalized computational pipeline for the selection of tumor-specific epitopes based on 1) patient specific haplotype; 2) cancer associated mutations; and 3) expression profiles of mutation carrying genes. We applied our workflow to one melanoma patient. Specifically, we analyzed tumor whole exome sequencing and RNA sequencing data to first detect tumor-specific mutations followed by epitope prediction based on the patient's HLA haplotype and filtering of epitopes using expression profile and binding affinity. We performed docking studies to predict the best set of epitopes targeting the patient's alleles. The proposed workflow enables us to find personalized tumor-specific epitopes for stimulating cytotoxic T-cell responses.