Population Pharmacokinetic Approach with Monte Carlo Integration for Clinical Phase I Trial of Pilocarpine HCl Tablet

Abstract

This paper describes a population pharmacokinetic analysis of a clinical phase I trial of pilocarpine HCl tablet. In the trial, the pharmacokinetics of orally administered pilocarpine HCl were examined in 24 healthy male subjects. A one-compartment model with first-order absorption was adopted and a multivariate log-normal distribution was assumed for the population distribution of the pharmacokinetic parameters. The regression of the distribution volume on the body weight was examined. The maximum likelihood method was employed to estimate the population means, variances and correlation coefficients of the multivariate log-normal population distribution. Instead of the first-order Taylor series approximation, a Monte Carlo integration was used for the multiple integral in the log likelihood. Observations below the detection limit were incorporated via a likelihood term. The population mean estimates for the absorption rate constant k_a, elimination rate constant K and lag time t_lag, were 2.864 (hr^-1), 0.444 (hr^-1) and 0.196 (hr), respectively. The distribution volume V_d was linearly related to the body weight (BW) as V_d (L) =2.959×BW (kg)-23.861. The population standard deviation estimates for k_a, K, V_d and t_lag were 0.764 (log (hr^-1)), 0.197 (log (hr^-1)), 0.169 (log (L)) and 0.346 (log (hr)), respectively.