ABSTRACT Glioblastoma multiforme (GBM) is the most common type of primary brain tumor, with a five-year survival rate of only 5.5%. Chimeric antigen receptor (CAR)-T cell therapy has shown safety but limited efficacy in the treatment of patients with GBM to date. GBM is characterized by dramatic antigen heterogeneity, thus immunotherapy targeting any single antigen is unlikely to achieve complete and durable response. In addition, GBM cells and surrounding tumor stroma overproduce transforming growth factor beta (TGF-β), which not only promotes tumor growth and metastasis, but also actively modulate the immune response by suppressing T-cell function and recruiting suppressive myeloid cells. Our analysis of patient GBM biopsies further confirmed that tumor-infiltrating myeloid cells (TIMs) express both TGF-β and TGF-β receptors, indicating the presence of an immunosuppressive feedback loop in the GBM tumor microenvironment (TME). Here, we aim to develop a multi- pronged immunotherapy featuring bispecific CAR-T cells that can not only attack GBM directly, but also modify the TME to overcome immunosuppression and induce multiclonal immune responses against GBM. Our group has developed a bispecific CAR that can directly target GBM cells through recognition of IL- 13Rα2, a clinically validated GBM-associated surface antigen, while simultaneously converting TGF-β into a stimulant for the engineered T cells. We have demonstrated that bispecific IL-13Rα2/TGF-β CAR-T cells are superior to single-input IL-13Rα2 CAR-T cells in both human GBM xenograft and immunocompetent mouse models. Here, we aim to demonstrate the safety and efficacy of IL-13Rα2/TGF-β CAR-T cells for clinical translation. In addition to demonstrating clearance of IL-13Rα2+ GBM, we will explore the potential for IL- 13Rα2/TGF-β CAR-T cells to induce endogenous immune response against IL-13Rα2– tumor cells using mouse models of heterogenous GBM. We will perform in-depth analyses by flow cytometry, multiplexed immuno- fluorescence, cytometry by time of flight (CyTOF), and single-cell RNA sequencing to understand the impact of bispecific CAR-T cells on the immune composition and function in the GBM TME. We will also perform T-cell receptor beta (TCRβ) sequencing on tumor-infiltrating lymphocytes to quantify the potential for epitope spreading. We will perform rigorous safety evaluations in immunocompetent mouse models, applying worst-case scenario pressure tests to explore the toxicity profile of IL-13Rα2/TGF-β CAR-T cells. Finally, we will verify efficacy of IL-13Rα2/TGF-β CAR-T cell therapy in heterogeneous human gliomas ex vivo & in vivo, using both fresh patient GBM biopsies containing TIMs and GliomaPDOX models—i.e., mice bearing patient-derived GBM tumors that have never been passaged ex vivo and retain the genetic heterogeneity seen in human patients. Successful completion of this project will generate a comprehensive set of preclinical data in support of a phase- 1 clinical trial to...