Principal Scientist Genentech, Inc. South San Francisco, California
Anti-tumor bispecific antibodies, which are designed to target two distinct tumor-associated antigens, have been developed to enhance tumor targeting while reducing systemic toxicity. However, it is still unclear how molecular attributes of bispecific antibody contribute to its in vivo efficacy and toxicity, and how to design bispecific antibody to maximize its therapeutic index during preclinical development. In this presentation, I will describe a bispecific PKPD computational model that can predict PK, minimum effective dose, maximum tolerable dose, and therapeutic index of bispecific antibody based on the experimental measurements of single-arm affinity, cross-arm avidity, and nonspecific clearance. Using a bispecific antibody as an example, I will demonstrate that our in-silico predictions were in line with the in vivo observations in mice. Finally, I will demonstrate that this model could aid the rationale design of bispecific antibodies that aim to improve on-tumor efficacy and reduce off-tumor toxicity.
Learning Objectives:
Utilize the bispecific PKPD model to understand how target expression levels, affinity/avidity of bispecific antibody, and PK properties of bispecific antibody affect its in vivo efficacy and toxicity.
Use the bispecific PKPD model to explore the relationship between dose, anti-tumor efficacy, and off-tumor toxicity, and to predict a dosing regimen that would maximize efficacy and minimize toxicity.
Use the bispecific PKPD model, sensitivity analysis, and parameter scan to suggest a molecular design (i.e. specific affinity/avidity) that could most efficiently enhance efficacy and reduce toxicity.
