2024 January 17; 16(730): eade2886. doi:10.1126/scitranslmed.aade2886. | Guangyuan Li, Shweta Mahajan, Siyuan Ma, Erin D. Jeffery, Xuan Zhang, Anukana Bhattacharjee, Meenakshi Venkatasubramanian, Matthew T. Weirauch, Emily R. Miraldi, H. Leighton Grimes, Gloria M. Sheynkman, Tamara Tilburgs, Nathan Salomonis
The study introduces Splicing Neo Antigen Finder (SNAF), a computational workflow designed to identify and prioritize tumor-specific and immunogenic neoantigens derived from splicing events. SNAF employs deep-learning algorithms for immunogenicity prediction and probabilistic models to assess tumor specificity. The workflow is validated using mass spectrometry (MS) and other experimental methods, demonstrating its ability to predict MHC-bound peptides and induce T-cell reactivity. The analysis reveals that splicing neoantigens are frequently shared among patients with melanoma, correlate with overall survival, and can predict response to immunotherapy. Additionally, SNAF identifies a new class of tumor-specific extracellular neo-epitopes termed ExNeoEpitopes, which can be recognized by CAR-T therapies or monoclonal antibodies. The study highlights the potential of splicing neoantigens as shared targets for targeted cancer immunotherapy and provides a comprehensive pipeline for their identification and validation.The study introduces Splicing Neo Antigen Finder (SNAF), a computational workflow designed to identify and prioritize tumor-specific and immunogenic neoantigens derived from splicing events. SNAF employs deep-learning algorithms for immunogenicity prediction and probabilistic models to assess tumor specificity. The workflow is validated using mass spectrometry (MS) and other experimental methods, demonstrating its ability to predict MHC-bound peptides and induce T-cell reactivity. The analysis reveals that splicing neoantigens are frequently shared among patients with melanoma, correlate with overall survival, and can predict response to immunotherapy. Additionally, SNAF identifies a new class of tumor-specific extracellular neo-epitopes termed ExNeoEpitopes, which can be recognized by CAR-T therapies or monoclonal antibodies. The study highlights the potential of splicing neoantigens as shared targets for targeted cancer immunotherapy and provides a comprehensive pipeline for their identification and validation.