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

Paper Details

Abstracts, authors, affiliations, and disclosures.

SESSION

702. CAR-T Cell Therapies: Basic and Translational: Poster I

Program: Poster Abstract Session
Saturday, December 12, 2026
05:30 p.m. - 07:30 p.m.
MCCNO - Hall F
Technology and Procedures, Assays, Imaging, Molecular Testing, Omics Technologies, Bioinformatics, Cell Expansion, Computational Biology, Emerging Technologies, Gene Editing, Profiling, Treatment Considerations, Combination Therapy, Drug Development, Biological Therapies, Natural Killer (NK) Cell Therapies, Transplantation, Bispecific Antibody Therapy, Chimeric Antigen Receptor, (CAR)-T Cell Therapies, Gene Therapy, Immunotherapy, Non-Biological Therapies, Pharmacology, Adverse Events, Artificial Intelligence (AI), Deep Learning, Generative AI, Machine Learning, Biological Processes, Pathogenesis, Immune Mechanism, Metabolism, Molecular Biology, Multi-Systemic Interactions, Clinical Practice (Health Services and Quality), Clinical Guidelines, Quality Improvement, Diseases, Immunology, Autoimmune Disorders, GVHD, Infectious Diseases, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, ALL, Non-Hodgkin Lymphoma, B Cell Lymphoma, Aggressive Lymphoma, Research, Fundamental Science, AML, Translational Research, Clinical Research, Clinical Trials, Epidemiology, Real-World Evidence, Study Population, Human, Adult, Pediatric, Animal Model
PAPER

2346 Discovery and identification of signal optimized second-generation human CAR T cells

Andrea Garmilla, Marcela V. Maus, MD, PhD2, Michael Birnbaum, PhD3
1 Harvard Medical School, Boston, Massachusetts, United States, Koch Institute For Integrative Cancer Research at MIT, Cambridge, Massachusetts, United States, Mass General Brigham Cancer Institute, Boston, Massachusetts, United States
2 Cellular Immunotherapy Program, Boston, Massachusetts, United States, Harvard Medical School, Boston, Massachusetts, United States, Mass General Brigham Cancer Institute, Boston, Massachusetts, United States
3 Koch Institute For Integrative Cancer Research at MIT, Cambridge, Massachusetts, United States, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States

Introduction Immunotherapies have revolutionized cancer treatment over the last decade. Of these, chimeric antigen receptor (CAR) T cell therapies have led to durable complete responses in patients with liquid tumors such as leukemia and multiple myeloma. CAR T cell therapy redirects patient-derived T cells to target cancer cells by expressing a tumor-specific synthetic receptor. The success of this therapy has led to the FDA approval of seven different CAR T cell products for the treatment of liquid tumors. However, despite the therapies approved for multiple myeloma, a subset of patients do not benefit due to poor persistence and efficacy of the CAR T cell product.
It is known that the CAR intracellular signaling domains (ICDs) lead to different T cell phenotypes. Until recently, CAR design has relied on construction and validation of individual receptors, making it difficult to study new ICD combinations. Thus, researchers have largely relied on precedent in choosing ICD combinations to test in CAR T cell trials. To address this, our lab has developed a library-based selection approach, termed CARPOOL, that can identify CARs with novel or improved function by assembling and assaying hundreds of thousands of CARs in T cells. While previous CARPOOL selections were performed in vitro, we have moved to an in vivo screening approach to allow for the selection of CARs that are relevant in a more physiological context to more directly assess clinically relevant CAR functions.

Methods This screen was conducted in a model of multiple myeloma using an ICD library with an scFv targeting BCMA. This model provides adequate CAR T cell recovery post treatment in vivo. We adapted library-based screening approaches for use in vivo to functionally identify second-generation CAR T cells that have improved function. We designed and generated a 2nd-generation CAR ICD library that consists of two position-constrained ICDs and an 18 nucleotide barcode. The library contains 18 co-stimulatory and 24 ITAM containing activating domains from immune cells including T, B, and NK cells, many of which have not previously been evaluated in the clinic. To simplify downstream analysis, each ICD combination corresponds to a barcode. These 432 combinations were pooled and can be used in a modular format such that changing the extracellular portions of the CAR requires a single cloning step. In vivo screens were conducted, and after CAR T cell recovery short-read next-generation sequencing was used to quantitate the frequency of each CAR variant via unique 18 nucleotide barcodes. Nanopore Sequencing was used as long-read sequencing to map ICD combinations to each unique barcode.

Results Our in vivo screens identified several enriched ICD combination with 29/412 identified combinations with log2(fold change) > 0.5. Of these, the FDA approved ICD combination 4-1BB CD3ζ  used as a positive control in our screen was highly enriched. A non-signaling CAR containing no ICDs was depleted across mice in the study. The enriched combinations are comprised of ICDs naturally occurring in T cells as well as mutated ITAM containing domains and costimulatory domains that are found in other immune cells. We found that relationships between enrichment of ITAM-containing and costimulatory domains were not necessarily related to the overall enrichment of each ICD domain individually. In vitro characterization of these combinations showed CAR T cell activation and killing capacity comparable to the clinical control, 4-1BB CD3ζ CAR. In vivo efficacy of these candidates in ongoing. Thus, we show that in vivo selections can identify functional CARs containing ICD combinations that have not yet been tested in preclinical or clinical spaces.

Conclusion
CAR T cells are a promising cellular therapy against multiple myeloma, but further optimization of the CAR itself could help overcome barriers to treatment efficacy. Current pre-clinical and clinical approaches largely rely on previously approved CAR architectures. However, it is not known whether these architectures present the most optimal phenotypes. By leveraging the ability of CARPOOL to assay a diverse library of CARs for optimized function in vivo, these findings have the potential to offer a more comprehensive understanding of the ICD combinations, and thus CAR T cell phenotypes, that are critical in treating multiple myeloma.

Andrea Garmilla: No relevant conflicts of interest to declare.