504 CS2009, a novel first-in-class trispecific antibody targeting PD-1, CTLA-4, and VEGFa, demonstrates promising anti-tumor activity in preclinical evaluation

Background

Immune checkpoint blockage (ICB) has become a standard-of-care treatment for various tumor types. While the clinical benefits of PD-1 inhibition are well established, combining it with CTLA-4 blockage has demonstrated enhanced efficacy. However, anti-CTLA-4 agents also induces significant immune-related adverse events (irAEs), thus restricting its anti-tumor effect. One strategy to overcome this problem is to build bispecific molecules aiming to 1) attenuate anti-CTLA-4 arm to balance the immune-toxicity and immune-efficacy (e.g. cadonilimab and KN046); 2) preferentially bind to PD-1/CTLA-4 double positive tumor infiltrating T cells (TILs) (e.g. volrustomig and vudalimab). To improve the effect of ICB, VEGF inhibitors including bevacizumab, ramucirumab, lenvatinib, etc. were applied to interrupt tumor angiogenesis, hence promote vascular normalization in tumor immune microenvironment, and improve T lymphocyte infiltration. We therefore developed a trispecific molecule, CS2009, which adds an anti-VEGF arm to further increase the anti-tumor activity of ICB.

Methods

The lead molecule, J30 was screened out from 100+ multispecific molecules, and constructed with a bivalent anti-VEGF arm, a monovalent anti-PD-1 arm and a monovalent anti-CTLA4 arm.

Results

J30 can concurrently bind to three targets, which was validated by protein-based and cell-based binding assays. In addition, J30 preferentially targets PD-1/CTLA4 double positive cells because of avidity-driven binding triggered by the two functional arms with balanced affinity. The anti-VEGF arm of J30 is fully functional assessed with both reporter assay and HUVEC proliferation assay. In the studies using immune competent humanized mice, J30 showed superior anti-tumor activities to its major competitors such as cadonilimab, AK112 and volrustomig. J30 has also demonstrated favourable stability and developability, with a yield level similar to monoclonal antibodies (mAbs). PK/toxicity studies in cynomolgus monkeys indicated that J30 had a mAb-like PK profile and was well tolerated. Ki67 as a T cell proliferation marker and a reliable PD marker also demonstrated that J30 could induce the activation of T cells remarkably.

Conclusions

J30 is a promising trispecific molecule for the treatment of various advanced solid tumors as a next-generation ICB. Current data supports further clinical research on CS2009.