The Immunogenic Tumor-Specific Neoantigen Database (ITSNdb): A Tool for Comprehensive Performance Evaluation of Neoantigen Immunogenicity Predictors

Abstract

The identification of tumor-specific neoantigen (TSN) immunogenicity is crucial to develop peptide/mRNA based antitumoral vaccines and/or adoptive T cell immunotherapies. In silico immunogenicityprediction of candidate peptides is crucial to speed up the prioritization of such peptides for experimentalvalidation. Up to now, several methods were proposed as TSN immunogenicity predictors, but there arestill several drawbacks in both performance and comprehensive performance evaluation, mainly due to theabsence of well documented and adequate TSN databases.The Immunogenic Tumor-Specific Neoantigen database (ITSNdb) is a tool developed to fairly bench-mark immunogenicity predictors intended to be used over tumoral neopeptides. The proposed ITSNdbenables the analysis of immunogenicity without the interference of other variables such as binding affinity orpeptide processing, as they were considered into the inclusion criteria for the curation of neoantigens.ITSNdb, together with a dataset emulating a true patient neoantigens scenario, as a validation strategy forprioritization, and a list of neopeptides predicted to bind to major histocompatibility complex I (MHC-I)from immune checkpoint blockade immunotherapy (ICB) cohorts, along with their associated patientoutcomes, is available to evaluate tumor neoantigen burden as a biomarker for ICB response.