A Personalized Multiscale Modeling Framework for Dose Selection in Precision Medicine

Abstract

Precision medicine (PM) refers to the use of available genomic information from an individual patient to select the most appropriate therapy for a disease. In this paper, we have developed a mechanistic multiscale modeling framework for the whole human body integrated with human hepatocyte genomic data. The model is validated by estimating the concentrations of several biomarkers in different amino acid inborn errors of metabolism (IEMs). To demonstrate the potential application of our multiscale modeling framework to precision medicine, we present a computational study of a specific disease. In this study, a genetic deficiency called Kelley–Seegmiller syndrome (KSS) is simulated for eight adult patients. Using our approach, we estimate the proper dosage of the drug for each subject needed to prevent hyperuricemia. These in silico results demonstrate that there is a significant difference in the optimal dose of the drug among individuals. In addition, the required dosages for the second and third days are different from those for the first day of treatment in each patient. Both results have important implications in terms of drug efficacy, drug side effects, and cost of the treatment. We also compare the pharmacological effect of available commercial drug tablets with these optimal values on disease progression. The results in this paper highlight the potential of the proposed modeling framework in opening up new opportunities in systems pharmacology and personalized medicine.