TCASIA: an integrative atlas linking alternative splicing landscapes for cancer immunotherapy

Abstract

Abstract Background Identifying robust predictive biomarkers for immune checkpoint blockade (ICB) remains a challenge. While alternative splicing (AS) shapes tumor immunity, the lack of large-scale, consensus-based datasets constrains the characterization of ICB-specific splicing landscapes. Methods We developed TCASIA (the Cancer Alternative Splicing Immunotherapy Atlas), harmonizing >4000 samples across 29 cohorts. A rigorous “2-out-of-4” multi-algorithm consensus framework (rMATS, MAJIQ, SUPPA2, and SplAdder) with a fuzzy matching strategy was used to establish a high-confidence AS landscape. Results TCASIA provides a robust toolkit linking splicing perturbations with clinical phenotypes (response, survival, and immune infiltration). Using this platform, we recapitulated the established FAS exon 6 skipping mechanism and identified IL7R exon 6 skipping as a functionally supported candidate associated with reduced CD8+ T-cell infiltration and poor survival in melanoma. Conclusions TCASIA provides an analytical atlas for investigating splicing-mediated immune evasion and prioritizing isoform-specific candidates for further functional and clinical validation in the context of ICB treatment.