Abstract B089: Application of precision cancer immunotherapy design tools to bladder cancer: Non-self-like neoepitopes as a prognostic biomarker

Abstract

Precision cancer immunotherapy targeting mutations expressed by cancer cells has proven to effectively control the tumor of patients in multiple clinical trials (Sahin et al., Nature 2017; Ott et al., Nature 2017). However, the selection of immunogenic T-cell neo-epitopes remains challenging and many epitopes selected using traditional methodologies fail to induce effector T-cell responses. Poor performance may partially be due to inclusion of mutated epitopes cross-conserved with self-epitopes recognized by regulatory (Treg), anergic, or deleted T-cells. Vaccination with self-epitopes can lead to weak effector responses, active immune suppression, and toxicity due to immune-mediated adverse effects. In addition, most cancer vaccine studies focus on the selection of CD8 T-cell neo-epitopes due to an apparent lack of robust and accurate CD4 T-cell epitope prediction tools. We have developed Ancer, an integrated and streamlined neo-epitope selection pipeline, that accelerates the selection of both CD4 and CD8 T-cell neo-epitopes from next-generation sequencing (NGS) data. Ancer leverages EpiMatrix and JanusMatrix, predictive algorithms that have been extensively validated in prospective vaccine studies for infectious diseases (Moise et al., Hum Vaccines Immunother 2015; Wada et al., Sci Rep 2017). Distinctive features of Ancer are its ability to accurately predict Class II HLA ligands, or CD4 epitopes, with EpiMatrix, and to identify tolerated or Treg epitopes with JanusMatrix. In addition, screening candidate sequences with JanusMatrix enables to the removal of neo-epitopes that may trigger off-target events, which have in some cases abruptly halted the development of promising cancer therapies. Ancer was applied to NGS data derived from the BLCA bladder cancer cohort from The Cancer Genome Atlas (TCGA) database. On average, 55 out of 204 missense mutations in bladder cancer patients’ tumors met Ancer’s quality control standards, in an initial analysis carried out for a representative set of 11 patients. This subset of high-quality missense variants was then screened using Ancer settings defined by the unique HLA of each patient, to derive the best vaccine candidate sequences encompassing these mutations. A median number of 24 (interquartile range: 15-64) candidate sequences were generated for each patient under study. The time required to select sequences for all of the patients in this study was less than two days. This initial analysis of eleven BLCA bladder cancer cohort patients demonstrates the capacity of Ancer to define a sufficient number of candidate sequences for vaccinating bladder cancer patients in a precision immunotherapy setting. We also assessed Ancer’s ability to predict patient outcomes on a larger subset of 58 individuals. While the disease-free status of BLCA patients could not be explained by their tumor mutational burden (AUC = 0.55, p-value = 0.1328), nor by their load of missense mutations (AUC = 0.54, p-value = 0.1740), the number of neoepitopes highly different from self, as defined by Ancer, significantly segregated disease-free patients from patients who recurred or progressed (AUC = 0.68, p-value = 0.0214). These results suggest that defining the number of true neoepitopes using Ancer may represent a novel biomarker for more robust antitumor immune response and higher likelihood of disease-free survival.Our analysis of the BLCA cohort from the TCGA database showcases the value of Ancer in clinical settings. Ancer can be used to identify high-value candidate sequences for inclusion in personalized therapies while removing potentially tolerated or tolerogenic self-epitopes from consideration. Our next step will be to investigate whether Ancer-defined neoepitope load will serve as a biomarker for prognosis and response to therapy in the full BLCA cohort. Citation Format: Guilhem Richard, Randy F. Sweis, Leonard Moise, Matthew Ardito, William A. Martin, Gad Berdugo, Gary D. Steinberg, Anne S. De Groot. Application of precision cancer immunotherapy design tools to bladder cancer: Non-self-like neoepitopes as a prognostic biomarker [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr B089.