Immu-06. Car Design Shapes the Brain Tumor Immune Microenvironment: Impact on Activity and Persistence of Anti-Glioma Adoptive T Cell Immunotherapy

Abstract BACKGROUND B7-H3 CAR T-cells effectively eradicate brain tumors in xenograft models and have proven safe in clinical trials, yet no objective responses were observed in patients. The limited efficacy is partly attributed to the immunosuppressive tumor microenvironment (TME), a characteristic not fully recapitulated in investigations using xenograft models. In this study, we set to optimize B7-H3 CAR structure and examine the effects of CAR design on the brain TME using an immunocompetent glioma model. METHODS We generated a panel of fully murine B7-H3 CARs with variations in transmembrane, co-stimulatory, and activation domains. This enabled us to comprehensively investigate their effector functions within the intact immune system. High-dimensional flow cytometry and single-cell RNA sequencing were then used to investigate changes in the brain TME following CAR T-cell therapy. RESULTS While five out of six B7-H3 CARs with single co-stimulatory domains exhibited potent functionality in vitro, optimizing co-stimulation and signaling failed to enhance their anti-glioma efficacy in vivo. To further enhance therapeutic effectiveness and persistence, we incorporated 4-1BB and CD28 co-stimulation through transgenic expression of 4-1BBL on CD28-based CAR T-cells. Although this design significantly improved anti-glioma efficacy in vitro, it conferred no additional benefits in vivo. Analysis of the TME revealed that CAR T-cell therapy influenced the immune composition of the brain TME. Successful anti-tumor responses were dictated by the recruitment, activation, and spatial localization of unique inflammatory ‘immune hubs’ containing specific subsets of macrophages and endogenous T-cells. Interestingly, depletion of brain macrophages using CSF1R inhibitor markedly inhibited CAR T-cell anti-tumor activity. CONCLUSION Our study highlights the critical role of CAR structural design and its modulation of the TME in mediating the efficacy of CAR T-cell therapy in brain tumors. Yet, more specific targeting and/or reprogramming of the TME is warranted for the successful design of effective adoptive immunotherapies for high-grade glioma.