(251c) Engineered Multispecific Antibodies to Interrogate and Manipulate Immune Checkpoint Protein Trafficking

Groundbreaking advances in immunotherapy, the mobilization of a patient’s own immune system against disease, have revolutionized the cancer therapy landscape. A particularly transformative advance is the development of antibody drugs that block immunosuppressive networks known as immune checkpoint pathways, thereby unleashing the power of immune cells to destroy tumors. These drugs have been remarkably successful in multiple cancer types. However, the majority of patients (>70%) do not respond to treatment, and most tumor regressions are only partial. The determinants of clinical response are not well understood, and certain cancer types resist immune checkpoint blockade altogether. Moreover, the role of immune checkpoint protein trafficking in immune regulation and disease pathogenesis remains unexplored. There is thus an urgent demand for innovative tools to interrogate immune checkpoint pathways, and for new interventions to therapeutically modulate their activity.

OBJECTIVE

We addressed current limitations by engineering novel multispecific antibodies targeting immune checkpoint proteins that elucidate immune biology and also form the basis for a new class of cancer therapeutics. Whereas all current clinical antibodies targeting immune checkpoint proteins act through a common mechanism of inhibiting pathway activation by blocking ligand/receptor interaction, we pursued an alternative strategy designed to down-regulate the surface presentation of immune checkpoint proteins and thereby prevent the possibility of pathway activation. This approach employs a concept known as multiparatopic antibody-mediated down-regulation, wherein a multispecific antibody targets two or more epitopes on a single protein to bias protein trafficking and drastically down-regulate surface presentation. We hypothesized that multiparatopic antibodies that engage multiple epitopes on the immune checkpoint protein programmed death-ligand 1 (PD-L1) would abolish immunosuppressive signaling to stimulate robust anti-cancer immunity.

RESULTS

We used a combination of immunization and directed evolution techniques to discover a panel of anti-PD-L1 antibodies that are non-competitive with the FDA-approved PD-L1-targeted neutralizing antibody atezolizumab. We then incorporated the variable domains of these novel antibodies together with atezolizumab into engineered biparatopic and triparatopic antibodies in a range of orientations. Trafficking studies revealed that these multiparatopic antibodies induced rapid and robust clustering, internalization, and lysosomal degradation of PD-L1 in an epitope- and topology-dependent manner. T cell activation assays demonstrated that the complementary mechanisms of ligand blockade and receptor downregulation employed by multiparatopic antibodies led to more durable immune stimulation, and in vivo studies in tumor-bearing mice showed that our lead multiparatopic antibody dramatically reduced availability of PD-L1 at the site of disease.

CONCLUSIONS AND IMPACT

We established a panel of multiparatopic antibodies against PD-L1 that offer new insight into immune checkpoint pathways and demonstrate how we may manipulate these pathways for targeted immune activation. The universality of our immunostimulatory approach enables efficacy across cancer types, and our unique trafficking-focused strategy allows for synergy with other immunotherapeutic modalities, such as vaccines, cytokine therapy, and adoptive cell transfer. Moreover, our versatile multiparatopic antibody designs can be extended to other classes of transmembrane proteins to interrogate and manipulate molecular trafficking for a broad range of research and medical applications.