Abstract LB300: Recapitulating the immune dynamics of cytokine release syndrome in humanized immune system mouse models

Abstract Cytokine release syndrome (CRS) is a severe and potentially fatal immune reaction that can be triggered by various immunotherapies including CAR-Ts, T-cell engagers, and immune checkpoint inhibitors. The preclinical assessment of such therapies therefore relies on a deeper understanding of human immune dynamics (e.g. T and myeloid cell interplay) and the availability of immunocompetent mouse models amenable to human tumor engraftment such that lead candidates can be assessed without needing to rely on murine analogs. Here, we induced CRS with the anti-human CD3 monoclonal antibody OKT3 in two mouse models that have human immune elements: PBMC humanized NCG mice (PBMC mice), where the only persisting human immune population is T lymphocytes, and CD34+ HSC humanized NCG mice (CD34 mice), which reconstitute a complete human immune system with B, T, NK, and myeloid cells albeit. Our findings reveal a dynamic cascade of common and model-specific physiologic and immune events at 1, 3, 6, and 24 hours post-OKT3 treatment. Indeed, within 6 hours of injection, OKT3 induced breathing difficulty, oedema, diarrhea, and hyperventilation in both models, but incidence and severity of these symptoms were greater, reaching loss of consciousness, in the PBMC model. Similarly, as the PBMC model harbors a superior circulating T cell count, it produced higher levels of circulating IFNγ (and TNF-α) compared to the CD34 model. However, CD34 mice had more circulating IL-2, likely explaining the increase in T-cell proliferation, which was not observed in PBMC mice where T cells retained an exhausted and predominantly apoptotic phenotype. Finally, as PBMC mice do not retain myeloid populations, we could only detect myeloid-produced cytokines CXCL10 and CCL12 in CD34 mice commensurate with marked expansion of circulating CD14+ monocytes and macrophage liver infiltration, underscoring the systemic and organ-specific dimensions of CRS pathology. Our comparative results validate feasibility and characterize CD34 humanized mice as a CRS model for tailored testing of specific therapeutic mechanisms of actions where a complete immune system is required. As CD34 mice offer significantly wider treatment windows due to longer, GvHD-free life spans, and produce less allogeneic response, they can be simultaneously leveraged to assess safety and efficacy of cell and antibody therapies. Citation Format: Audrey Wetzel, Joséphine Zangari, Pierre Wallet, Anaïs Meynet-Cordonnier, Sébastien Tabruyn, Emilie Bayon, Dan Georgess. Recapitulating the immune dynamics of cytokine release syndrome in humanized immune system mouse models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB300.