SUMMARY Our long-term objective is to develop highly translatable animal models for testing cancer immunotherapies. Animal models have been essential in cancer research. However, mouse tumor transplantation or tumor genetic models lack critical components for studying human anti-tumor responses such as high mutational load, tumor microenvironment (TME) and tumor heterogeneity or mice have major differences from humans in innate and adaptive immune responses. In response to PAR-20-131, we will use our novel humanized mouse models (HuMice) to test gamma-delta T (γδT) cell- based immunotherapies. Since Vγ9Vδ2+ T cells, a subtype of γδT cells that are most commonly used in adoptive immunotherapy, are unique to primates, traditional mouse models are not ideal to study human γδT cells. This proposal will maximize translational potential of mammalian models by studying Vγ9Vδ2+ T cells-based therapy in HuMice with HLA-matched human melanomas from cell lines or patient-derived xenografts (PDX). Our laboratories have established >500 melanoma PDX and >300 melanoma cell lines, which represent all clinical, genetic and biologic groups of the disease. In Aim 1, we will study adoptive transfer of enhanced γδT cells. We will test a new Vγ9Vδ2+ T cell expansion method in HuMice. We will then equip γδT cells with DR5-CAR that targets both myeloid derived suppressive cells and melanoma cells. We will study alterations in the TME after treatment. To avoid potential toxicity of targeting DR5, we will develop a novel combinatorial CAR that targets PD-L1 and DR5. In Aim 2, we will stimulate endogenous γδT cells for cancer therapy. We will use bromohydrin pyrophosphate (BrHPP) and resiquimod to expand endogenous human γδT cells to treat melanoma bearing HuMice. We will study whether BTN3A1 mAbs expand human γδT cells and govern antitumor functions of both γδ and CD8+ T cells in HuMice. We will then combine expansion of endogenous γδT cells with anti-PD-1 mAbs in melanoma-bearing HuMice. We expect that HuMice will allow us to test expansion of endogenous Vγ9Vδ2+ for cancer therapy for the first time in a model system and our new expansion methods are effective and may be optimized and tested in future clinical trials.