NONPARAMETRIC DENSITY-ESTIMATION APPLIED TO POPULATION PHARMACOKINETICS

Kinetic parameters are estimated to assess absorption, distribution, m etabolism, and elimination of a drug in a subject. In a group of subje cts, pharmacokinetic population studies are developed to describe the variability and to detect particular subsets by establishing the relat ionships between kinetic parameters and easily measurable subject char acteristics, the covariates (age, body weight, etc.). The usually prop osed methods are based on linear regression equations relating kinetic parameters to the covariates. We propose to measure these dependencie s and describe the interindividual variability through the joint proba bility density function. This function is estimated by a nonparametric method superposing potential functions or kernels over the sample. In this estimation, the Shannon information theory was applied to determ ine the number of individuals needed to describe the variability relia bly and to screen informative covariates with respect to the kinetic p arameters. This approach was used to obtain the nonparametric conditio nal probability density functions of the kinetic parameters, given the covariates. These functions supplied prior information for a Bayesian estimation. The feasibility of the global approach was illustrated by a simulation in which nonlinear relations link covariates and pharmac okinetic parameters. The performance of this new estimator using covar iates was compared with that of the usual Bayesian estimation.