Background Improvement of current prognosis biomarkers will enhance our ability to identify cancer patients at higher risk of recurrence and will further advance the personalization of patient monitoring and treatment. We hypothesized that the presence of a mutation alone is not sufficient to generate an immunogenic neoepitope, but that significant differences must exist between the Human Leukocyte Antigen (HLA)- and/or T Cell Receptor (TCR)-interfaces of the neoepitope and its non-mutated form, or with other self-epitopes, in order to be recognized as non-self by the immune system. As such, cancer patient clinical outcomes may be better understood by neoepitope analyses that integrate these considerations.
Methods We analyzed large scale (n=412) bladder cancer genomic data from The Cancer Genome Atlas (TCGA) using Ancer, an automated machine-learning-based pipeline we designed for neoantigen screening and vaccine design. Ancer shares components with other commercial-grade screening platforms used routinely in immunogenicity assessments of biologics and infectious disease antigens, such as the EpiMatrix algorithm for HLA-I and HLA-II neoepitope identification, and the JanusMatrix algorithm for tolerated, tolerogenic, and cross-reactive T cell epitope identification. Evaluation of patient survival with Ancer was compared to other analyses employing tumor mutational burden (TMB) or neoepitopes identified with the commonly used NetMHCpan-4.0 and NetMHCIIpan-3.1 T cell epitope prediction tools.
Results We stratified bladder patients based on their Ancer HLA-I and HLA-II neoepitope burdens and observed significantly prolonged disease free and overall survival in patients whose tumor contained both high numbers of HLA-I and HLA-II neoepitopes compared to other individuals. Stratifications performed with Ancer were superior to comparative analyses performed with TMB or with NetMHCpan and NetMHCIIpan. In addition, we showed that Ancer’s precise filtering and characterization of mutated epitopes contributed to this increased association with survival, as showcased by gradual improvements in survival analyses performed after each of its filtering step. Multivariate survival analyses indicated that Ancer neoepitope content remained a significant factor in patient overall survival even when adjusted for TMB, and other clinical covariates such as age at diagnosis and disease stage, unlike analyses involving NetMHCpan and NetMHCIIpan neoepitopes.
Conclusions Our analysis suggests that enhanced presence of CD8, CD4 T cell epitopes, and limited inclusion of Treg epitopes in the tumor genome plays a key role in cancer survival. Ancer scoring provides a predictive method for predicting patient outcomes, by defining the number of true neoepitopes and by identifying Treg epitopes that would interfere with T cell-based immune activation and response to the tumor.
This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.