Abstract 273: A mass spectrometry survey of frequent HLA alleles successfully presenting common tumor specific mutations for immune recognition

Abstract

Abstract Tumor-specific mutated proteins can create non-self, mutation-containing ‘neoepitopes' that are immunogenic and are attractive targets for adoptive T-cell therapies. To avoid the complexity of patient-specific, private neoepitopes and their T-cell responses, there has been major interest in targeting common shared mutations in driver genes using off-the-shelf T-cell receptors engineered into autologous lymphocytes. The diversity of MHC alleles which could present these select neoantigens is an obstacle to this strategy. Most neoepitopes will not be successfully presented by most HLA alleles, and algorithms predicting epitope-HLA binding will either miss valid epitopes or identify too many candidate alleles to investigate thoroughly. One method to definitively demonstrate that an MHC allele can successfully present an epitope is to elute peptides bound to that MHC allele on the cell surface and analyze them by mass spectrometry (MS). It has the additional advantages of defining the precise epitope processed from an antigen and estimating its abundance. We undertook a comprehensive analysis by MS of several common shared mutations in driver oncogenes and their presentation by common Class I HLA alleles. To overcome the low sensitivity of MS (compared to T-cells), we created mono-allelic cell lines expressing one Class I HLA allele and one common mutated oncogene. Peptides bound to MHC on the surface of these cell lines were immunoprecipitated, eluted, and analyzed using liquid chromatography-tandem mass spectrometry (LC-MS/MS). As expected, most HLA alleles could not present a detectable neoepitope from the candidate antigens. Our system was able to identify precise neoepitopes derived from four commonly mutated oncogenes (KRAS, EGFR, BRAF, and PIK3CA) presented by the HLA-A*03 HLA superfamily. The majority of mutated peptides were consistently found in biological replicates, demonstrating the reproducibility of our system. By evaluating intensities and fragmentation data for mutated peptides, we were able to improve the reliability of detection of neoepitopes at the cell surface. Although most of the identified epitopes were predicted to bind using the canonical NetMHCpan algorithm, many epitopes predicted to bind with high affinity were not observed using our pipeline. HLA-A*0201 has been intensely studied because of its high frequency in the U.S. population but we have not found it to consistently present mutated peptides from our candidate oncogenes. Our data show that we could utilize a MS approach to demonstrate which shared oncogene-derived neoepitopes are unequivocally processed and presented by common HLA alleles, which will allow greater focused efforts on developing TCRs against these promising HLA-neoantigen pairs. Citation Format: Catherine M. Ade, Yue A. Qi, Sudipto Das, Ken-ichi Hanada, Tapan Maity, Xu Zhang, Thorkell Andresson, Udayan Guha, James C. Yang. A mass spectrometry survey of frequent HLA alleles successfully presenting common tumor specific mutations for immune recognition [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 273.