Abstract 7238: Quantitative profiling of HLA class I and II antigens and neoantigens in tissue and PBMC samples with an improved mass spectrometry approach

Abstract Human leukocyte antigen (HLA) molecules are critical mediators of immune surveillance, functioning to present antigenic peptides derived from self and non-self proteins to T cells. By facilitating the recognition of pathogenic or abnormal cells, such as cancer cells, HLA molecules play a pivotal role in guiding immune responses and enabling targeted immune interventions. In oncology clinical trials, the use of peripheral blood mononuclear cells (PBMCs) offers significant advantages for immunopeptidomics (IMPX) and potential neoantigen discovery. Unlike solid tissue biopsies, which may be limited in availability and require invasive procedures, PBMCs can be routinely isolated from blood, allowing for serial sampling. PBMCs comprise a diverse population of immune cells, including lymphocytes (T cells, B cells, and NK cells) and monocytes, which collectively serve as a holistic snapshot of the body's immune status, enabling the tracking of immune modulation, and providing pharmacodynamics readout across various disease contexts. We developed and optimized an unbiased IMPX profiling workflow leveraging native lysis and magnetic bead-based immunoprecipitation for isolating both class I and class II immunopeptides. By refining the lysis and pulldown conditions, we ensured effective extraction and enrichment of HLA complexes from minimal starting material, achieving robust profiling across varying sample types, including cell lines, tissues, and PBMCs. To accommodate the often limited PBMC availability in real-world clinical settings, we tested this pipeline with inputs as low as 5 million PBMCs, achieving high-quality, reproducible IMPX data. Further downstream, LC-MS setup and AI-based bioinformatics solutions were optimized to support streamlined data generation and analysis. Overall, the developed workflow demonstrated high reproducibility and specificity across sample types, identifying robust immunopeptide profiles: (1) For class I immunopeptides, >60% of peptides identified were 9-mers, with >80% predicted as strong binders, exhibiting expected amino acid enrichment at anchor residues; (2) For class II immunopeptides, >50% of peptides were 14-to-16-mers, with >50% predicted as strong binders. Input material ramping tests showed that >2, 800 class I immunopeptides could be consistently identified from 2.5 mg of fresh-frozen tissue, and 2, 000-3, 000 class I immunopeptides from as few as 5 million PBMCs. Moreover, we profiled the immunopeptidomes of several representative cancer cell lines (e.g. SHP77, OV56) with known HLA subtype(s) that are commonly used in TCR therapeutics target discovery. In summary, we have developed a high-sensitivity, robust pipeline for class I and II immunopeptidome profiling that is compatible with low-input samples, making it suitable for both target discovery and clinical applications. Citation Format: Arthur Viodé, Anamarija Pfeiffer, Luca Raess, Marco Tognettti, Tikira Temu, Oliver M. Bernhardt, Daniel Redfern, Yuehan Feng, Roland Bruderer, Lukas Reiter. Quantitative profiling of HLA class I and II antigens and neoantigens in tissue and PBMC samples with an improved mass spectrometry approach [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7238.