Abstract 3850: Modeling individual variability in CD19 CAR-T therapy: An in vivo study with PBMC-humanized mice

Abstract Chimeric antigen receptor T-cell (CAR-T) therapy targeting CD19 has emerged as a transformative treatment for hematological malignancies. Despite its efficacy, significant inter-patient variability in response and associated toxicities pose challenges for its broader application. Understanding these variations is critical to optimize therapy and predict outcomes. This study aims to capture the individual variability observed in CD19 CAR-T cell therapy responses using PBMC-humanized mouse model, with a focus on efficacy, expansion, and cytokine induction profiles. Three human PBMCs (Donors A, B, C) used throughout the studies were pre-screened for in vivo cytokine induction and humanization levels. Autologous CD19 CAR-Ts with a CD28 costimulatory domain and CD3z chain were generated with varying transduction rates: 92%, 52%, and 95% for Donors A, B, and C, respectively. In the first experiment, irradiated female NSG™-MHC I/II double knock-out (DKO) mice were reconstituted with human PBMCs. Six days post humanization, the mice were treated with equal total number of autologous CAR-Ts. All three CAR-Ts showed efficacy, eradicating CD19 cells in peripheral blood. We observed a moderate but significant body weight loss only in mice treated with Donor B autologous CAR-Ts. Notably, CAR-T expansion varied across three donors and did not correspond with the purity (% transduction) of CAR-Ts. Donor B CAR-Ts with the lowest purity showed the highest CAR-T expansion. Donor B PBMCs resulted in the highest humanization level of CD19+ cells and we hypothesize that more CD19 targets led to higher CAR-T expansion. By measuring 33 cytokines from sera collected 2 days post treatment, we explored if human cytokine induction corresponded to the CAR-T expansion. Contrary to initial expectations, cytokines typically associated with robust CAR-T expansion, such as IL-2, IL-6, IL-10, IFNγ, and TNFα, did not peak in the Donor B cohort. To see if individual variability in efficacy and CAR-T cell expansion shows similar variation with increased targets of CD19 CAR-Ts, we engrafted DKO mice with B cell lymphoma (Raji-luc). Seven days after tumor engraftment, we humanized mice with Donor A, B, or C PBMCs and subsequently treated mice with autologous CAR-Ts. Donor A showed the fastest clearance of Raji-luc measured by IVIS bioluminescence imaging. In parallel with the previous experiment, we observed that Donor B CAR-Ts showed the highest expansion level among the three CAR-Ts. The study illustrates how a PBMC-humanized mouse model can reflect the variability in patient responses to CD19 CAR-T therapy. This model serves as a valuable tool for further investigation into the complex dynamics of CAR-T efficacy and toxicity, potentially guiding more tailored approaches in treatment design. Citation Format: Won Lee, Destanie Rose, James G. Keck, Jiwon Yang. Modeling individual variability in CD19 CAR-T therapy: An in vivo study with PBMC-humanized mice abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 3850.