2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Assays, Imaging, Molecular Testing, Omics Technologies, Bioinformatics, Cell Expansion, Gene Editing, Profiling, Treatment Considerations, Combination Therapy, Biological Therapies, Bispecific Antibody Therapy, Immunotherapy, Artificial Intelligence (AI), Deep Learning, Biological Processes, Apoptosis, Pathogenesis, Genomics, Immune Mechanism, Metabolism, Microbiome, Molecular Biology, Multi-Systemic Interactions, Clinical Practice (Health Services and Quality), Diseases, Immunology, Neutropenia, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Research, Fundamental Science, Translational Research, Clinical Research, Clinical Trials, Study Population, Human, Adult, Elderly
PAPER

3928 Leveraging an immunoproteasome-specific molecular glue with deep learning tools to identify highly immunogenic myeloma-specific neoantigens

James J Ignatz-Hoover, MD PhD, Dahlia Messeloff2, Kyle Hoffman, PhD3, James Driscoll II, MD, PhD4
1 Case Western Reserve University, Cleveland, Ohio, United States, UH Seidman Cancer Center, Cleveland, Ohio, United States
2 Fuchs Mizrachi School, Beachwood, Ohio, United States
3 Bioinformatics Solutions, Waterloo, Ontario, Canada
4 Case Western University, Cleveland, Ohio, United States, UH Seidman Cancer Center, Cleveland, Ohio, United States

Multiple myeloma (MM) is a mostly incurable plasma cell dyscrasia characterized by immune dysfunction. Chronic stimulation of T-cells that recognize MM-specific antigens leads to decreased anti-MM cytotoxicity and subsequent disease progression. Immunoproteasomes (IP) are highly specialized holoenzymes that degrade misfolded proteins for presentation in MHC-I complexes, increasing the diversity of antigens recognized for T-cell surveillance. We have previously reported a novel drug that stimulates IP activity. This increased IP activity promotes antigen processing, modulating antigen abundance and diversity subsequently promoting autologous, allogenic and antigen restricted MM cell killing. This increased killing suggests that the upregulated antigens are immunogenic and may be actionable targets for T-cell directed therapies. However, tumor and antigen heterogeneity limit antigen identification, immunogenicity characterization and subsequent targeting in a clinical setting. AI deep learning platforms can integrate extensive input data to not only improve antigen identification but also to predict immunogenicity and HLA binding, expediting the analysis, interpretation and targeting of the MM immunopeptidome by orders of magnitude using only a small amount of tissue.

Here, we utilize our IP activator to modulate antigen presentation on CD138+ patient MM Cells. We then analyze the patient immunopeptidome using LC-MS/MS. Resulting data is analyzed using AI deep learning methods to identify antigens, map antigens to source proteins if applicable, predict HLA binding and immunogenicity. Using this unique pipeline we are able to identify 4,572 antigens present in CD138+ tumor cells but not in CD138- control cells. Of these 4,572 antigens, 251 are upregulated by ~2.5-fold on IP treatment, and 1,676 new antigens are identified after IP treatment not present before treatment. Of these MM specific antigens upregulated on IP activation, 29 map to no known protein, suggesting they are neoantigens. 34 antigens are predicted to be immunogenic, and 32 are predicted to be strong binders to HLA*A02.1. The top strong binders were modeled in Alphafold-3 in an HLA*A02.1 complex revealing high overall model certainty. Here, combining experimental and AI Tools, we retool the MM immunoproteasome and identify novel antigenic peptides. This pipeline represents a novel framework for identification and validation of new targets for personalized medicine in MM patients.

James J Ignatz-Hoover, MD PhD: Regeneron - Honoraria; Sanofi - Honoraria