Discovery and Development of Agonist Antibodies for T cell Receptors

Abstract

Agonist antibodies that activate co-stimulatory immune receptors, such as the tumor necrosis factor (TNF) receptors OX40 and CD137, are an important class of emerging therapeutics due to their ability to regulate immune cell activity. Despite their promise, there are no approved agonist antibodies for treating cancer as demonstrated by previous unsuccessful clinical trials. Although multiple factors are responsible for poor clinical efficacy, one major bottleneck is the reliance on FcγR-mediated crosslinking for sufficient receptor activation. This is inherently problematic because FcγR expression varies greatly on different immune cells, leading to a wide range of receptor agonism. Emerging research suggests that antibodies engaging two different epitopes on the same immune receptor mediate receptor superclustering and enable robust antibody agonism without extrinsic Fc crosslinking. However, there are no systematic methods for identifying such biepitopic (also known as biparatopic) agonist antibodies. Therefore, the objective of this research work is to develop facile methods for reliably identifying biepitopic antibodies to activate immune receptors for immunotherapeutic applications.

Biepitopic antibodies have been shown to mediate potent receptor activation for a variety of immune receptors. Traditionally, the generation of these antibodies requires key steps including animal immunization, epitope binning to identify unique antibody pairs, and combining antibody pairs to engineer biepitopic antibodies. While this approach has been used to successfully discover biepitopic antibodies, it suffers from key limitations. Notably, animal immunization and subsequent antibody isolation is an arduous and unpredictable process. Even when successful clones are discovered from these processes, further epitope binning experiments are needed to select antibody pairs to discover potent immune therapeutics. To overcome these limitations, we developed an antibody screening strategy that greatly simplifies the discovery of biepitopic antibodies. Our approach eliminates the need for animal immunization by using existing, off-the-shelf IgG antibodies specific to the target receptor. Next, we perform in vitro selections by blocking the receptor epitope of the existing antibody and conducting subsequent sorts to identify single-chain antibodies with orthogonal binding sites. Our work has shown that the antibody screening strategy can be used to discover antibodies for a variety of TNF receptors including OX40 and CD137.

Given that receptor clustering of three or more receptors is critical for activating TNF receptors, we first generated biepitopic tetravalent OX40 antibodies by attaching novel single-chain antibodies to the C-termini of the light chain of existing clinical-stage antibodies. These tetravalent biepitopic antibodies showed strong T cell proliferation and cytokine secretion for biepitopic antibodies compared to their monoepitopic counterparts. Next, we sought to improve additional clinical-stage OX40 IgGs which we engineered as biepitopic antibodies to demonstrate the generality of our findings that biepitopic antibodies can mediate superior and FcγR-independent activities. Beyond OX40 IgGs, we also show that biepitopic antibodies can be used to mediate superior T cell proliferation for a second TNF receptor (CD137). Looking forward, we anticipate that these research advancements will accelerate the discovery and development of the next generation of immunotherapeutics.