Dataset Information

Discovery of prevalent, clinically actionable tumor neoepitopes via integrated biochemical and cell-based platforms

ABSTRACT: Strategies for maximizing the potency and specificity of cancer immunotherapies have sparked efforts to identify recurrent epitopes presented in the context of defined tumor-associated neoantigens. Discovering these neoepitopes can be difficult owing to the limited number of peptides that arise from a single point mutation, a low number of copies presented on the cell surface, and variable binding specificity of the human leukocyte antigen (HLA) class I complex. Due to these limitations, many discovery efforts focus on identifying neoepitopes from a small number of cancer neoantigens in the context of few HLA alleles. Here we describe a systematic workflow to characterize binding and presentation of neoepitopes derived from 47 shared cancer neoantigens in the context of 15 HLA alleles. Through the development of a high-throughput neoepitope-HLA binding assay, we surveyed 24,149 candidate neoepitope-HLA combinations resulting in 587 stable complexes. These data were supplemented by computational prediction that identified an additional 257 neoepitope-HLA pairs, resulting in a total of 844 unique combinations. We used these results to build sensitive targeted mass spectrometry assays to validate neoepitope presentation on a panel of HLA-I monoallelic cell lines engineered to express neoantigens of interest as a single polypeptide. Altogether, our analyses detected 84 unique neoepitope-HLA pairs derived from 37 shared cancer neoantigens and presented across 12 HLA alleles. We subsequently identified multiple TCRs which specifically recognized two of these neoantigen-HLA combinations. Finally, these novel TCRs were utilized to elicit a T cell response suggesting that these neoepitopes are likely to be immunogenic. Together these data represent a validated, extensive resource of therapeutically relevant neoepitopes and the HLA context in which they can be targeted.

INSTRUMENT(S): Orbitrap Eclipse, Orbitrap Fusion Lumos

ORGANISM(S): Homo Sapiens (ncbitaxon:9606)

SUBMITTER: Christopher Rose

PROVIDER: MSV000090323 | MassIVE |

REPOSITORIES: MassIVE

ACCESS DATA

Similar Datasets

Project description:Background Clinical success of T cell receptor (TCR) gene therapy has previously been demonstrated for NY-ESO-1 TCR gene therapy. To increase numbers of cancer patients that can be treated with TCR gene therapy we aimed to identify a novel set of high-affinity cancer specific TCRs targeting different cancer testis (CT) antigens in prevalent HLA class I alleles. Methods In this study, we selected based on publicly available gene expression databases the most promising CT genes to target. From these selected genes we identified by HLA peptidomics the naturally processed and presented HLA class I peptides. With these peptide-HLA tetramers were generated, and by single cell sorting CT specific CD8+ T cells were selected, and expanded from the allo-HLA repertoire of healthy donors. By several functional assays high avidity CT-specific T cell clones with safe recognition pattern were selected. To evaluate the potential for clinical application in TCR gene therapy, TCRs were sequenced and transferred into peripheral blood derived CD8+ T cells. Results In total we identified, 7 novel CT-specific TCRs that effectively target MAGE-A1, MAGE-A3, MAGE-A6 and MAGE-A9 in the context of human leukocyte antigen(HLA) -A1, -A2, -A3, -B7, -B35 and -C7. TCR gene transfer into CD8⁺ T cells resulted in efficient cytokine production and cytotoxicity of variety of different tumor types without detectable cross-reactivity. In addition, major in vivo antitumor effects of MAGE-A1 specific TCR engineered CD8⁺ T cells was observed in an orthotopic xenograft model for established multiple myeloma, in which bone marrow located tumor cells were completely eradicated after T cell injection. Conclusion The identification of 7 novel CT-specific TCRs, reactive against CT antigens presented in a variety of different HLA class I alleles, allows selection of therapeutic TCRs for an increased number of cancer patients, and will improve development of personalized TCR gene therapy.

Project description:Background: Patient derived organoids (PDOs) can be established from colorectal cancers as in vitro models to interrogate cancer biology and its clinical relevance. We applied mass spectrometry (MS) immunopeptidomics to investigate neoantigen presentation and whether this can be augmented through interferon gamma (IFN) or MEK-inhibitors. Methods: Four PDOs from chemotherapy refractory and one from a treatment naïve CRC were expanded to replicates with 100 million cells each, and HLA class I and class II peptide ligands were analysed by MS. Results: We identified an average of 9,936 unique peptides per PDO which compares favourably against published immunopeptidomics studies, suggesting high sensitivity. Loss of heterozygosity of the HLA locus was associated with low peptide diversity in one PDO. Peptides from genes without detectable expression by RNA-sequencing were rarely identified by MS. Only 3 out of 612 non-silent mutations encoded for neoantigens that were detected by MS. Treatment of four PDOs with IFN upregulated HLA class I expression and qualitatively changed the immunopeptidome, with increased presentation of IFN-inducible genes. HLA class II presented peptides increased on average 16-fold with IFN treatment. MEK-inhibitor treatment showed no consistent effect on class I or II HLA expression or the peptidome. Importantly, no additional HLA class I or II presented neoantigens became detectable with any treatment. Conclusions: Only 3 out of 612 non-synonymous mutations encoded for neoantigens that were detectable by MS. Although MS has sensitivity limits and biases, and likely underestimated the true neoantigen burden, this established a lower bound of the percentage of non-silent mutations that encode for presented neoantigens, which may be as low as 0.5%. This could be a reason for the poor responses of non-hypermutated CRCs to immune checkpoint inhibitors. MEK-inhibitors recently failed to improve checkpoint-inhibitor efficacy in CRC and the observed lack of HLA upregulation or improved peptide presentation may explain this.

Project description:Tumor antigens are central to antitumor immunity. Recent evidence suggests that peptides from non-canonical (nonC) aberrantly translated proteins can be presented on HLA-I by tumor cells. Here, we investigated the presentation and immunogenicity of nonC antigens across different cancer types to better understand their contribution to cancer immunosurveillance and to address whether they could be exploited therapeutically. To this end, we employed a proteogenomics pipeline to identify nonC HLA-I ligands derived from off-frame translation of coding sequences and non-coding regions (UTR, ncRNA, intronic and intergenic) in patient-derived tumor cell lines (TCL) of different histological types (4 gynecological cancer, 3 melanoma and 2 head and neck cancer patients). First, peptides bound to HLA-I were isolated and analyzed by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) using state-of-the-art procedures. Amino acid (Aa) sequences were identified through the previously described pipeline Peptide-PRISM, with some modifications. Briefly, for each MS spectrum, the top 10 candidates were first identified by de novo sequencing and later mapped to a database including the 3-frame transcriptome and 6-frame genome. Additionally, whole-exome sequencing (WES) information of each TCL was included to interrogate the presentation of mutated peptides derived cancer-specific NSM. The false-discovery rate (FDR) was calculated independently for each category considering the search space and peptide length in a stratified mixture model as previously described. Next, in order to select nonC peptides preferentially presented by tumor cells, immunopeptidomics data from several healthy tissues and samples available from the HLA ligand atlas was used to filter out peptides presented in healthy tissues. To overcome a potential bias toward frequent alleles, the peptides were excluded at the ORF level rather than the Aa sequence. As a result, we found that from a total of 839 unique nonC peptides detected in our tumor samples, 38.5% were predicted to derive from ORFs also present in healthy tissue (nonC-HL). Hence, 61.6% (n=517) were considered preferentially presented on tumor HLA-I, referred to as non-canonical tumor ligands (nonC-TL). Out results showed that, although nonC-TL constitued the most abundant source of candidate tumor antigens, as compared to neoantigens, cancer-germline or melanoma-associated antigens, pre-existing antitumor T cells in cancer patients preferentially recognized neoantigens rather than nonC-TL. Nonetehless, nonC-TL elicited de novo T-cell responses via in vitro sensitization of donor lymphocytes. We identified TCRs specific to three nonC-TL, two of which mapped to the 5’ UTR regions of HOXC13 and ZKSCAN1, and one mapping to a non-coding spliced variant of the C5orf22C. These immunogenic nonC-TL were expressed across tumor types but were barely or not detected in healthy cells. Our findings predict a limited contribution of nonC-TL to cancer immunosurveillance but demonstrate they are attractive novel targets for widely applicable immunotherapies.

Dataset Information

Discovery of prevalent, clinically actionable tumor neoepitopes via integrated biochemical and cell-based platforms

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure