Abstract

Abstract Background: Neuroblastoma is a childhood tumor characterized by relatively few somatic mutations and low MHC expression, which has thus far largely precluded it from investigation using adoptive immunotherapy. Methods: We characterized MHC-presented antigens in 8 patient derived xenograft (PDX) tumors using LC/MS/MS immunopeptidomics. We developed a method to identify antigens derived from differentially expressed proteins by combining RNA-seq data from 153 neuroblastoma and 1641 healthy tissues, and ligandomic data from 190 healthy tissues. We also performed functional characterization on the ability HLA-A2 neuroblastoma lines to elicit a T-cell response using CEF1 antigen-specific T-cell hyrbidoma when challenged with flu virus, and performed bioinformatic analysis of the TME. Results: From 8 PDX tumors, we identified a total of 14119 MHC-presented antigens. Interestingly, we observed antigens from all tumors which stained negative for MHC by IHC, suggesting that antigens can be identified from tumors below the detection limits of the standard staining protocols. We first searched the ligandome dataset for all possible neoantigens of 8-14aa arising from known mutations. Not having found any mutated neoantigens, we developed a method to search for tumor antigens derived neuroblastoma-specific proteins. We discovered 83 MHC ligands that derive from neuroblastoma-specific proteins, which we expect to be promising targets for adoptive T-cell therapy. We also found a number of recurrent antigens across tumor samples, suggesting that the proteins from which these antigens derive could be used in HLA-agnostic tumors vaccines. To test whether the MHC expression in neuroblastoma is sufficient to induce a T-cell response, we characterized the ability of neuroblastoma cells to elicit a CD8 response to CEF1 antigen. We demonstrate that despite low MHC expression in all lines tested, 4/7 lines are able to induce a robust T-cell response to flu antigen greater than HLA-A2 melanoma cells (>20pg/mL IL-2 release in all responsive lines). We also characterized immune activity in patient tumors using Granzyme A (GZMA) and Perforin (PRF1) as surrogates for T-cell activity. We observed a strong correlation between known T-cell-recruiting cytokines (particularly CXCL5/9/10) and GZMA/PRF1 (p=2.51x10-35), providing further evidence that T-cells are active in the tumor microenvironment despite low MHC expression on tumor cells. Conclusions: We identify neuroblastoma-specific MHC class I antigens in each of the 8 PDX tumors we tested, yielding a total of 83 novel antigens. We conclude that neuroblastoma tumors harbor promising tumor-specific targets for immunotherapy despite low rates of somatic mutation and low MHC expression, and that at least a subset of these tumors are able to induce a robust response using antigen-specific T-cell hybridomas. We also describe methods of identifying tumor-specific MHC antigens in other tumors with low mutational burden. Note: This abstract was not presented at the meeting. Citation Format: Mark Yarmarkovich, Moreno Di Marco, Olivia Padovan, Jenna Lobby, Laurence Eisenlohr, Dimitrios Monos, Stefan Stevanovic, John M. Maris. MHC class I immunogenicity and novel tumor antigen discovery in neuroblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5824. doi:10.1158/1538-7445.AM2017-5824

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call