Abstract T-cell based immunotherapies such as CAR-T Cells, BiTEs and an expanding class of next-generation T-cell-engaging modalities, have revolutionized cancer treatment and dramatically improved outcomes for many patients. Yet, the success of these therapies fundamentally relies on identifying safe, tumor-specific targets. However, current practice relies on labor-intensive convoluted bioinformatics efforts that require deep domain expertise and extensive coordination between computational scientists and clinicians. These fragmented workflows slow the pace of discovery and significantly delay the transition of emerging targets into therapeutic development. To overcome these challenges and motivated by recent advances in large language models and agentic AI, we developed ImmunoVerse-Chat, an interactive agentic framework that integrates LLM-driven reasoning with high-performance immunogenomic pipelines to streamline and accelerate tumor-specific antigen discovery. Built upon our previously established ImmunoVerse, the most comprehensive pan-cancer therapeutic T cell targets to date, spanning over 21 tumors, 11 classes of molecular events. Together, these system-level innovations allow ImmunoVerse-Chat to uncover clinically meaningful antigen patterns that are often overlooked or inaccessible to conventional pipelines. ImmunoVerse-Chat streamlines the entire immunopeptidomic workflow from raw multi-omic data to pHLA identification and provides an interactive, reasoning-driven interface that enables rapid comparison of antigen landscapes, real-time target prioritization, and assessment of T-cell therapeutic potential. By leveraging the underlying pan-cancer antigen atlas, the system, through automated visualization modules, further contextualizes these findings across tissue types and molecular aberrations to distinguish shared and tumor-restricted pMHC candidates and uncover recurrent, population-relevant antigens, tumor-resident microbial epitopes, and molecular signatures linked to splicing, immune regulation, and endogenous retrovirus expression, including ERV-derived peptides. These integrated, multi-layered insights directly guide the selection of safe, immunogenic, and clinically meaningful T-cell targets. Overall, ImmunoVerse-Chat combines AI reasoning with multi-omic depth into a unified, interactive, and population-aware engine for T-cell target discovery, and we envision the broad adoption of this platform will democratize antigen discovery across oncology research and accelerate the development of next-generation immunotherapies. Citation Format: Aman Sharma, Guangyuan (Frank) Li, Xinya Liu, Mark Yarmarkovich. ImmunoVerse-Chat: A conversational agentic-AI engine for next-generation immunotherapeutic target discovery abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 23.
Sharma et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: