Abstract 4096: Assessing individual variability in efficacy and toxicity of autologous and allogeneic chimeric antigen receptor T-cell immunotherapy using a PBMC-humanized mouse model

Abstract

Abstract Chimeric antigen receptor T-cell (CAR-T) therapy has emerged as a revolutionary therapeutic to fight cancer. However, side effects such as cytokine-release syndrome and neurotoxicity often accompany and remain unpredictable. While autologous CAR-Ts are FDA-approved, allogeneic CAR-Ts are in active development in pursuit of “off-the-shelf” availability and reduced manufacturing costs. However, the same concern for adverse events applies. Importantly, how the immune characteristics of the patient may predispose towards toxicity is not well modeled. Further, it is unclear how autologous versus allogeneic products interact with the patient immune system. Toxicity does not correlate with in vitro characterization of the CAR-Ts, adding unpredictability to the toxicity and efficacy. These highlight the need for in vivo mouse models to explore biological variation, predictive biomarkers, and therapeutic intervention. Therefore, we developed a humanized mouse model using NSG™-MHC Class I/II double knock-out (DKO) mice engrafted with human PBMCs to access individual variability in CAR-T-induced toxicity and efficacy. DKO mice are optimal for assessing human cytokine induction from CAR-Ts as they show delayed onset of GvHD and low baseline cytokine levels. To test donor variation in autologous CAR-T toxicity, we humanized mice and treated with autologous CD19 CAR-Ts containing the CD28 costimulatory domain and CD3z chain from two human PBMC donors (Donor A and B). CAR-T successfully eliminated human B cells from both PBMC donors in vivo. We observed drastic body weight loss and 83% of Donor A mice (engrafted with 17M PBMCs) reached the humane endpoint. In contrast, Donor B mice did not experience body weight loss. CAR-T induced human cytokines such as IFN-ꝩ, IL-2, IL-4, IL-6, IL-10, TNFα, MIG, MIP-1a, and IP-10, recapturing clinical data. To test donor variation in allogeneic CAR-T toxicity, we engrafted mice with 10M PBMCs from 6 healthy donors and treated with allogeneic CAR-Ts from Donor A or B. Allogenic efficacy was greater in Donor B CAR- than Donor A CAR-Ts. Donor-specific variability was observed; mice humanized with one of the six donors, experienced significant body weight loss only from the Donor B CAR-Ts but not from Donor A CAR-Ts. Principal component analysis of cytokine data revealed that Donor B CAR-Ts induced more distinguished cytokine responses than Donor A CAR-Ts, driven by IFN-ꝩ and IL-10. Lastly, the data indicates that the CAR-T cells show poor expansion in allogeneic treatment, corresponding to the current literature. PBMC-humanized mice provide an in vivo platform to assess the toxicity and efficacy of autologous and allogeneic CAR-T treatment. This platform can be used to preclinically assess cytokine induction from the interaction between the patient immune system and allogeneic CAR-Ts from healthy donors. Citation Format: Won Lee, Destanie Rose, Katelyn Burleigh, Blair Armstrong, Heather Gustafson, Rebecca Gardner, James G. Keck, Jiwon Yang. Assessing individual variability in efficacy and toxicity of autologous and allogeneic chimeric antigen receptor T-cell immunotherapy using a PBMC-humanized mouse model. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4096.