A bispecific T-cell engager (BiTE) enhances the antitumor capabilities of T-cells by directing them to recognize a tumor-specific antigen. Several BiTEs are currently in clinical trials, and there is one FDA-approved BiTE for cancer therapy at the moment. However, one of the biggest challenges BiTE is currently facing is cytokine release syndrome (CRS). While most patients have mild flu-like symptoms, some patients experience a severe inflammatory syndrome, which could ultimately cause multi-system organ failure and death. Thus, the key is to accurately predict the risk-benefit ratio of each individual before starting the treatment. Unfortunately, there is no reliable in vitro or in vivo model to predict the toxicity and efficacy of BiTE at the moment. Additionally, one of the most challenging factors for pre-clinical evaluation of BiTEs for toxicity and efficacy is the inherent differences among patients’ immune systems. Here, we developed an in vivo PBMC-humanized mouse model that can evaluate both toxicity and efficacy of BiTE and other immune-oncology therapeutics in each individual mouse. The platform was validated using an anti-CD28 superagonist (TGN1412), which none of the pre-clinical in vitro assays and in vivo studies, including non-human primates, executed before clinical trials were able to predict the observed clinical toxicities. This model uses luciferase-tagged human B-cell lymphoma Raji tumor cells to allow the measurement of tumor burden and response to treatment, along with toxicity simultaneously. In addition, we showed that this platform can determine individual PBMC donor differences, and potential adverse drug combinations and drug dose-response with efficacy and safety on individual PBMC donors. The model we developed can potentially be used as a predictive and reproducible platform to identify patient, cancer, and therapy combinations at risk for developing CRS. This model will also help to optimize the therapeutic drug dose range to improve the safety profile of BiTE and other immune-oncology drugs. We believe this platform will greatly benefit not only the scientific community but potentially cancer patients as well.

Citation Format: Jiwon Yang, Jing Jiao, Kyle Draheim, Danying Cai, Michael A. Brehm, Leonard D. Shultz, Dale L. Greiner, Mingshan Cheng, James G. Keck. A novel, rapid, sensitive, and reproducible in vivo platform to assess efficacy and toxicity of bispecific antibodies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2049.