Design of a humanized CD40 agonist antibody with specific properties using AlphaFold2 and development of an anti-PD-L1/CD40 bispecific antibody for cancer immunotherapy.

Abstract

Bispecific antibodies (BsAbs) represent a promising strategy for cancer immunotherapy. Challenges in immunotherapy include inefficient early events in the immune response cycle, such as antigen presentation and T cell priming. Background stimulation of CD40 with agonistic antibodies is a promising strategy to enhance the therapeutic efficacy of immune checkpoint inhibitors (ICIs). Assisted by Alphafold2(AlphaFold-Multimer), we developed a humanized CD40 agonistic antibody that exhibits activation only in the presence of cross-linking. It also demonstrates that the current AlphaFold2(AlphaFold2-Multimer) can predict antibody-antigen complexes. Due to the unique epitope, it demonstrates superior activation compared to APX005M (S267E). Building upon this, we created a novel bispecific antibody (anti-PD-L1/CD40 bispecific antibody, referred to as "BA4415") designed to activate CD40 signaling specifically in the context of PD-L1 while simultaneously blocking PD-1/PD-L1 signaling. Results from functional evaluations using effector cells revealed the superior biological activity of BA4415 compared to the combination of each monoclonal antibody. BA4415 demonstrated the ability to enhance T-cell cytokine release in vitro assays, exhibiting superior functional attributes compared to the anti-PD-L1 antibody. Furthermore, in humanized transgenic mice challenged with huPD-L1-expressing tumor cells, BA4415 induced superior anti-tumor activity. This novel anti-PD-L1/CD40 bispecific antibody holds potential for strong anti-tumor therapeutic efficacy by selectively restricting CD40 stimulation in tumors.