While immunotherapy is rapidly growing as an excellent treatment for many cancer types, most immunotherapies, particularly in combination can induce immune related adverse events (irAEs) which are generally caused by immune cell activation and expansion. Although most irAEs are low grade (1-2), higher grade events (3-4) can prevent patients from completing therapy and can also be life threatening. Thus, there is a strong need to develop pre-clinical mouse models to identify which immunotherapeutic combinations induce the best anti-tumor responses without inducing severe irAEs. To replicate irAEs associated with immunotherapies in humans, we utilized the FoxP3-DTR mouse model where regulatory T-cell (Treg) depletion can be achieved by injecting diphtheria toxin (DT), releasing Treg associated immune suppression. Transient Treg depletion by a single dose of DT induced mild T-cell activation and increased T-cell immune checkpoint expression, making mice more sensitive to irAEs development from clinically relevant immunotherapies. In three tumor models, we demonstrated that transient Treg depletion with anti-PD-1 or TIM-3 had a good anti-tumor efficacy and mild irAEs. Conversely, combining transient Treg depletion with an antibody targeting the co-stimulatory receptor CD137 had a lower therapeutic window due to increased irAEs. This combination was effective in suppressing established tumors, but induced severe irAEs including elevated serum cytokines and acute liver inflammation. These irAEs were similar to those seen in clinical trials of anti-CD137 antibodies where liver toxicity was a major concern and were driven by a rapid systemic CD8+ T-cell expansion. We also demonstrated that irAEs from anti-CD137 therapy could be mitigated by neutralising TNF without impacting on anti-tumor efficacy. We believe this model can provide valuable data on the efficacy and irAEs associated with novel combination immunotherapies which may aid clinicians and pharmaceutical companies in clinical trial design. We are currently utilizing this model to predict the therapeutic irAEs and efficacy of other novel immunotherapies that are in pre-clinical and early stage clinical trials.

This abstract is also being presented as Poster B58.

Citation Format: Stephen J. Blake, Jing Liu, Harjunpaa Heidi, Mark J. Smyth, Michele W.L. Teng. Use of a novel mouse model to investigate immune related adverse events arising from immunotherapies. [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2016 Oct 20-23; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2017;5(3 Suppl):Abstract nr PR07.