Interrogating T cells specific to mutated and shared tumor epitopes in mouse and man

Abstract

It is by now acknowledged that immunological recognition of cancer is a many-sided interplay, able to both promote and reject tumor growth. With cancer immunotherapy, the ambition is to harness immunity towards long lasting tumor elimination, for which T cells are important mediators. Tumor cells differ genetically from healthy tissue, which will be visible to surveilling T cells via aberrant peptide-MHC presentation. Hence, a large effort is being put into understanding T cells and their epitope targets in cancer, via techniques to monitor and augment tumor specific T cells. Little is known about the characteristics that govern immunogenicity of T cell epitopes, and current strategies depend on in silico prediction of peptide-MHC binding affinity. The outlined thesis contains three research papers that investigate T cells and their associated epitopes in mice and humans, with emphasis on cancer. In the first study, we developed novel conditional ligands for murine H-2 alleles H-2Dd and H-2Kd , that enable rapid generation of multiple pH-2 multimers via UV-mediated peptide exchange, without the need for individual peptide-H-2 refolding. pH-2d multimers were successfully used for fluorescent tetramer staining and large-scale DNA barcode labelled multimer libraries. This provides the first description of H-2Dd and H-2Kd conditional ligands, and proof-of-concept of DNA barcode labelled multimers for murine H-2d haplotype screening. This technology will enable epitope mapping in a wide variety of disease models on e.g. BALB/c background, including syngeneic models of cancer. The second study investigated neo-epitope vaccination in the murine CT26 syngeneic tumor model. We explored DNA and peptide-based delivery of neo-epitopes and found only DNA vaccination to induce protective tumor immunity. Correspondingly, DNA based vaccination preferentially induced neo-epitope specific CD8+ T cells, whereas peptide vaccination induced neo-epitope specific CD4+ T cells. Our DNA based vaccination strategy thus represents an interesting framework for future neo-epitope discovery, from which putative epitope libraries are often large and will benefit from the technology outlined in the first study. In the third study we show T cell recognition of epitopes from previously undescribed shared tumor associated antigens in breast cancer patients. Similar to the first study, this study employs high-throughput DNA barcode labelled pHLA multimers, and the investigated T cell epitopes do not give rise to recognition in HLA-matched healthy donors. Breast cancer remains a major cause of female mortality worldwide and is not thoroughly researched in the field of immunotherapy. Breast cancer harbors low mutational burden compared to other cancer types, thus investigations on shared tumor epitopes are of importance compared to mutation-derived neo-epitopes. As such, our findings represent novel T cell targets in breast cancer, that might be of relevance in patient immune monitoring. Collectively, these studies represent tools for, and investigations of, T cell epitopes in cancer. With an increased understanding of what the T cells “see” and how to enhance them, we can better steer them towards tumor eradication via immunotherapy.