DuoBody®-EpCAMx4-1BB is a novel, clinical stage, bispecific antibody targeting EpCAM and 4-1BB designed to boost antitumor responses conditionally in EpCAM-expressing tumors. By crosslinking EpCAM on tumor cells with 4-1BB on immune cells, DuoBody-EpCAMx4-1BB enhances T-cell activation, proliferation, and antitumor activity in preclinical studies. DuoBody-EpCAMx4-1BB is co-developed by BioNTech and Genmab. The preclinical characterization of DuoBody-EpCAMx4-1BBB will be presented.

Monoclonal antibodies struggle to achieve potent complement activation due to the need for multivalent C1q binding, resulting in the underutilization of complement as a therapeutic mechanism. We have recently described an innovative approach involving bispecific single domain antibodies, BiCE™, which efficiently recruit and activate C1. We now present the 2nd generation BiCE™ IgG molecules that exhibit superior complement-mediated cell killing compared to competing technologies, holding great therapeutic potential.

We previously developed SMART-Ig® technology to efficiently remove soluble antigens from the blood. This time, we aimed for more efficient antigen removal by creating pH-dependent biparatopic antibodies that bind to different epitopes of a soluble monomeric antigen in a pH-dependent manner. These antibodies accelerated cellular uptake by forming larger immune complexes, successfully removing soluble antigens from the blood more efficiently.

The discovery of agonistic antibody drugs has been severely limited by the difficulty of identifying epitopes that support the productive engagement of the signaling complex. Using a combination of experimental and computational approaches, we generated agonist antibodies that activate the ALK1 pathway to treat vasculopathies. The techniques we developed can generate agonist antibodies against any heteromeric receptor complex, opening new opportunities to treat many human diseases with precision biologics drugs.

T cell engaging antibodies (TCEs) are effective therapeutics for patients with diverse malignancies when adequately targeted to tumor biomass. We show that ML methods can support the efficient design of TCEs, including via boolean logic, targeting co-expressed tumor antigens and sparing healthy tissues expressing either antigen, even at high receptor densities. Overall, we demonstrate how AI/ML design with rapid, closed loop wet-lab characterization supports the systematic design of safe and effective TCEs.

Xencor has created a growing set of bispecific antibodies, using principles of avidity-driven selectivity to improve therapeutic index. Building on these modalities are additional efforts to explore T cell costimulation via signal 2 to potentiate anti-tumor activity of T cells.