Abstract
Pharmacometabolomics in early phase clinical trials demonstrate the metabolic profiles of a subject responding to a drug treatment in a controlled environment, whereas pharmacokinetics measure the drug plasma concentration in human circulation. Application of the personalized peak plasma concentration from pharmacokinetics in pharmacometabolomic studies provides insights into drugs' pharmacological effects through dysregulation of metabolic pathways or pharmacodynamic biomarkers. This proof-of-concept study integrates personalized pharmacokinetic and pharmacometabolomic approaches to determine the predictive pharmacodynamic response of human metabolic pathways for type 2 diabetes. In this study, we use metformin as a model drug. Metformin is a first-line glucose-lowering agent; however, the variation of metabolites that potentially affect the efficacy and safety profile remains inconclusive. Seventeen healthy subjects were given a single dose of 1000 mg of metformin under fasting conditions. Fifteen sampling time-points were collected and analyzed using the validated bioanalytical LCMS method for metformin quantification in plasma. The individualized peak-concentration plasma samples determined from the pharmacokinetic parameters calculated using Matlab Simbiology were further analyzed with pre-dose plasma samples using an untargeted metabolomic approach. Pharmacometabolomic data processing and statistical analysis were performed using MetaboAnalyst with a functional meta-analysis peaks-to-pathway approach to identify dysregulated human metabolic pathways. The validated metformin calibration ranged from 80.4 to 2010 ng/mL for accuracy, precision, stability and others. The median and IQR for Cmax was 1248 (849-1391) ng/mL; AUC(0-infinity) was 9510 (7314-10,411) ng·h/mL, and Tmax was 2.5 (2.5-3.0) h. The individualized Cmax pharmacokinetics guided the untargeted pharmacometabolomics of metformin, suggesting a series of provisional predictive human metabolic pathways, which include arginine and proline metabolism, branched-chain amino acid (BCAA) metabolism, glutathione metabolism and others that are associated with metformin's pharmacological effects of increasing insulin sensitivity and lipid metabolism. Integration of pharmacokinetic and pharmacometabolomic approaches in early-phase clinical trials may pave a pathway for developing targeted therapy. This could further reduce variability in a controlled trial environment and aid in identifying surrogates for drug response pathways, increasing the prediction of responders for dose selection in phase II clinical trials.