Abstract

We constructed population pharmacokinetic (PK) models for the five constituents of daikenchuto (DKT), a traditional Japanese herbal medicine. Data were collected from two randomized PK studies conducted in Japan and the United States. Participants received single oral doses of 2.5 g, 5 g, and 10 g of DKT. The plasma concentrations of five DKT constituents—hydroxy-α-sanshool (HAS), hydroxyl-β-sanshool (HBS), 6-shogaol (6S), 10-shogaol (10S), and ginsenoside Rb₁ (GRB1)—were determined by liquid chromatography-tandem mass spectrometry. A total of 1859 samples from 55 participants (US, n = 36; Japanese, n = 19) were included in the analysis. Population PK models of HAS, HBS, 6S, and 10S were best described by a one or two-compartment model with a bolus input. On the other hand, the model of GRB1 was best described by a one-compartment model with nonlinear extravascular input. Among the covariates evaluated, body mass index (BMI) and age were found to influence oral clearance (CL/F) and volume of distribution (Vd/F) for HAS and HBS, respectively. The influence of body weight on CL/F and Vd/F for 6S was demonstrated. Marked differences were observed in mean plasma concentrations of HAS and HBS between Japanese and US participants. However, the simulation results indicated that the difference in plasma concentrations may be attributed to the difference in demographic factors such as BMI, body weight, and age, whereas ethnic difference between the Japanese and US participants was considered minimal.