Effective T cell–based cancer immunotherapies depend on receptors that can be engineered to recognize antigens with tumor-restricted expression. Conventional antigen discovery focuses on peptides derived from annotated protein-coding regions, but noncanonical peptides—those generated from transcriptional events outside annotated coding sequences—expand the antigenic landscape and may provide tumor-specific targets. The study aimed to develop a systematic, biology-driven approach to prioritize and functionally validate noncanonical tumor antigens at the receptor level.
The authors combined multiple complementary approaches to identify melanoma-associated noncanonical transcripts. Key elements of the pipeline included de novo transcript analysis to detect transcripts not present in standard annotations, exon-resolved expression quantification to assess tumor enrichment at the exon level, RNA in situ hybridization for spatial validation of tumor expression, and immunopeptidomics to detect peptides presented on major histocompatibility complex molecules. This multimodal strategy was designed to move candidates from computational discovery through molecular and receptor-level validation.
Among three recurrent melanoma-associated noncanonical transcripts identified, the transcript designated EVA003 emerged as the lead candidate. EVA003 was prioritized based on several genomic and expression characteristics: a distinct repeat-enriched genomic architecture, consistent tumor-enriched exon-level expression across independent datasets, and the presence of a genomically unique immunogenic core sequence. These combined features supported biological plausibility for tumor-restricted presentation and immunogenicity.
Using immunopeptidomic methods, the study demonstrated endogenous presentation of peptides derived from EVA003 on HLA-A*03:01 molecules. The authors also detected specific reactivity against EVA003-derived peptides in patient-derived tumor-infiltrating lymphocytes (TILs), linking the putative noncanonical antigen to naturally occurring immune recognition in the tumor microenvironment. Details on the peptide sequences, mass-spectrometry evidence, or the frequency of presentation across patient samples were not included in the abstract and therefore are not reported here.
Single-cell transcriptomic profiling of reactive TILs identified a dominant peptide-reactive clonotype. From this clonotype the investigators isolated a naturally occurring T cell receptor (TCR) specific for the EVA003-derived peptide. This receptor originated from patient TILs and represents a molecularly defined receptor that could be transferred to other T cells for functional testing.
The study transferred the isolated TCR into healthy donor T cells to test functionality. Expression of the receptor conferred antigen-dependent activation and cytotoxicity toward both peptide-pulsed target cells and melanoma cells expressing EVA003 endogenously. These results indicate that a single naturally occurring receptor specific for a noncanonical peptide can mediate recognition and killing of tumor cells presenting that peptide in the context of HLA-A*03:01. The abstract does not provide quantitative measures of activation, killing efficiency, dose–response, or in vivo efficacy.
The work establishes a biologically informed framework for prioritizing noncanonical tumor antigens and advancing them through receptor-level validation. It demonstrates that genomically unique, tumor-enriched noncanonical peptides can be presented to molecularly defined receptors capable of mediating cancer cell killing, supporting the inclusion of prioritized noncanonical antigens in engineered T cell therapeutic strategies. Open questions not addressed in the abstract include population HLA coverage, potential off-target expression, long-term safety, manufacturability of TCR-engineered products, and clinical efficacy.
The conflict of interest statement reports that Enara Bio contributed to antigen discovery and validation experiments. Several authors disclosed relationships with industry or receipt of proprietary data, advisory roles, honoraria, restricted research support, or public grants. The abstract provides no detailed safety, clinical outcome, or large-cohort validation data; such details would need to be consulted in the full text. All factual assertions above are drawn from the article abstract, metadata, and conflict disclosures.