Computational Design of a TCR-Based Bispecific Engager Targeting Cancerous KRAS G12V Mutations

Abstract

Bispecific T-cell engager (BiTE) therapy is a type of immunotherapy that redirects cytotoxic T cells to target tumor cells. Traditionally, a BiTE structure dual binds to the CD3 receptor on the T-cell surface and to the mutated antigen presented on the surface of the cancer cell as an extracellular protein. An example of this is blinatumomab, an FDA-approved BiTE structure for the treatment of B-cell acute lymphoblastic leukemia (B-ALL). However, traditional BiTE complexes are at risk of off-tumor effects due to their targeting of cell surface antigens, which are not specific to certain cancer variants. In this study, we utilized a computational approach, an area that remains underdeveloped in current immunotherapy design, to design the initial structures of two TCR-BiTEs that consist of a single-chain variable fragment (scFv) in conjunction with a T-cell receptor (TCR) that binds specifically to the KRAS G12V neoantigen, which currently does not have any FDA-approved targeted therapy treatments. Ultimately, we present two promising TCR-BiTEs that are specific to two different HLAs (HLA:03:01 and HLA:11:01) targeting KRAS G12V, serving as valuable starting points for further evaluation and design in vitro. We validated the design of our TCR-BiTE structures through AlphaFold tools, free energy estimation methods, and molecular dynamics analysis, thereby also providing a potential computational pipeline that can be applied in the design of TCR-BiTE structures targeting other mutations in addition to KRAS G12V.