Abstract
Over the past 2 decades, tumor immunotherapies have witnessed remarkable advancements, especially with the emergence of immune checkpoint-targeting bispecific antibodies. However, a quantitative understanding of the dynamic cross-talking mechanisms underlying different immune checkpoints as well as the optimal dosing and target design of checkpoint-targeting bispecific antibodies still remain challenging to researchers. To address this challenge, we have here developed a multi-scale quantitative systems pharmacology (QSP) model platform that integrates a diverse array of immune checkpoints and their interactive functions. The model has been calibrated and validated against an extensive collection of multiscale experimental datasets covering 20+ different monoclonal and bispecific antibody treatments at over 60 administered dose levels. Based on high-throughput simulations, the QSP model platform comprehensively screened and characterized the potential efficacy of different bispecific antibody target combination designs, and model-based preclinical population-level simulations revealed target-specific dose-response relationships as well as alternative dosing strategies that can maintain anti-tumor treatment efficacy while reducing dosing frequencies. Model simulations also pointed out that combining checkpoint-targeting bispecific antibodies with monoclonal antibodies can lead to significantly enhanced anti-tumor efficacy. Our mechanistic QSP model can serve as an integrated precision medicine simulation platform to guide the translational research and clinical development of checkpoint-targeting immuno-modulatory bispecific antibodies.