| Abstract: |
| Cancer is a leading cause of death worldwide with over 19 million estimated new cases and nearly 10 million deaths in 2020. While the number of clinical trials is increasing over the past decades, the success rate of oncology trials remains the lowest among all therapeutic areas. This challenge necessitates the development of computational tools to predict the effectiveness of drugs of interest and identify predictive biomarkers for various treatment combinations. In model-informed drug discovery (MIDD) and clinical trial design, quantitative systems pharmacology (QSP) models have begun to play crucial roles due to their ability to integrate mechanistic knowledge from cancer biology and pharmacology into a quantitative framework. I will present a modular QSP platform for immuno-oncology (QSP-IO) that describes the cancer-immunity cycle, which allows for varying degrees of complexity based on our research goals. I will introduce the model structure, discuss methods for creating a virtual patient population for in silico clinical trial simulation, and present the results from an in silico clinical trial of PD-L1 inhibition in advanced non-small cell lung cancer. I aim to demonstrate the potential of QSP models to discover predictive biomarkers, make efficacy predictions for drugs of interest, and guide future clinical trials. |
|