Defining Heterogeneity of CAR T cell populations to treat B16F10 melanoma 2310016

Abstract Introduction Chimeric antigen receptors (CAR) T cells recognize tumor associated antigen and have been approved to treat relapsed/refractory B cell lymphomas but not yet for solid tumors. In patients, the pre-transfer CAR product is heterogenous, containing T cells from distinct differentiation stages which can influence proliferation and persistence in vivo. These subsets may also adopt distinct effector, memory or exhausted states correlated with the antigen load and tissue environment. Understanding this heterogeneity is critical for optimizing CAR T cell therapy. Methods To investigate how CAR T cell subsets contribute individually and collectively, we are using DARLIN in-vivo lineage tracing and scRNA-seq to determine clonal relationships between the pre-transfer product and CAR T cells found at Day 7 and Day 40 after transfer. This will reveal the differentiation capacity and spatial localization of CAR T populations to provide further insights that may improve CAR-T therapy. Results We have demonstrated in a B16F10 melanoma model, TA99-CAR T cells (TRP1-specific), infiltrating the tumor are characterized by a CD44+ Ly6c+ PD-1+ population, indicating an activated and potentially exhausted phenotype. In contrast, CAR T cells localized in secondary lymphoid organs predominantly express CD62L and lack Ly6c expression, suggesting a central memory-like state with greater proliferative potential. Furthermore, when armoring CAR T cells with superkine IL-2 and IL-33, these cells demonstrated enhanced infiltration into the tumor microenvironment and adopted a terminal effector phenotype, characterized by high expression of terminal differentiation marker, KLRG1. Conclusion These data suggest that CAR T cell intrinsic and tissue-specific cues could shape CAR T cell differentiation potential. Funding Source NIH R01CA297297-01 Topic Categories Tumor Immunology: Checkpoints, Prevention, and Treatment (TIPT)