Computational structural optimization enhances IL13Rα2 - B7-H3 tandem CAR T cells to overcome antigen-heterogeneity-mediated tumor escape.

Chimeric antigen receptor (CAR) T cell therapy is a highly effective treatment for multiple malignancies. However, one limitation is tumor antigen-heterogeneity and downregulation, which allows tumor cells to evade conventional, monospecific CAR T cells. One approach to overcome this tumor escape is by utilizing a tandem CAR recognizing two antigens. However, tandem CAR constructs often require optimization to achieve cell surface expression and function. Herein, we describe our process of designing an IL-13Rα2-B7-H3 tandem CAR. Interestingly, our original tandem CAR failed to express on the cell surface, leading to a systematic evaluation of 24 tandem constructs varying in their scFv positioning, linkers, and specific amino acids. We identified a "trouble region" in the CAR and optimized it using computational approaches, rescuing surface expression and improving function compared with monospecific CAR T cells. Further, the optimized tandem CAR T cells more effectively eliminated tumors than monospecific CAR T cells in vivo. Our study demonstrates the successful application of structure-guided computational strategies to restore surface expression and antitumor efficacy of an IL13Rα2 - B7-H3 tandem CAR. Our study also highlights the necessity of computational methods to guide the design of synthetic proteins, and that these methods can increase CAR T cell efficacy.