NovumRNA: Accurate prediction of non-canonical tumor antigens from RNA sequencing data

Abstract

Non-canonical tumor-specific antigens (ncTSAs) can expand the pool of targets for cancer immunotherapy, but require robust and comprehensive computational pipelines for their prediction. Here, we present NovumRNA, a fully automated Nextflow pipeline for predicting different classes of ncTSAs from patients’ RNA sequencing data. We extensively benchmarked NovumRNA using publicly available and newly generated datasets, demonstrating the robustness of its analytical modules and predictions. NovumRNA analysis of RNA-seq data from colorectal cancer organoids and patients’ tumor samples revealed comparable ncTSA potential for microsatellite stable and unstable tumors and candidate therapeutic targets for patients with low tumor mutational burden. Finally, our investigation of glioblastoma cell lines demonstrated increased ncTSAs burden upon indisulam treatment, and detection by NovumRNA of therapy-induced ncTSAs, which we could validate experimentally. These findings underscore the potential of NovumRNA for identifying synergistic drugs and novel therapeutic targets for immunotherapy, which could ultimately extend its benefit to a broader patient population.