Errors-in-variables in joint population pharmacokinetic/pharmacodynamic modeling

Pharmacokinetic (PK) models describe the relationship between the administe red dose and the concentration of drug (and/or metabolite) in the blood as a function of time. Pharmacodynamic (PD) models describe the relationship b etween the concentration in the blood (or the dose) and the biologic respon se. Population PK/PD studies aim to determine the sources of variability in the observed concentrations/responses across groups of individuals. In thi s article, we consider the joint modeling of PK/PD data. The natural approa ch is to specify a joint model in which the concentration and response data are simultaneously modeled. Unfortunately, this approach may not Lie optim al if, due to sparsity of concentration data, all overly simple PK model is specified. As all alternative, we propose all errors-in-variables approach in which the observed-concentration data are assumed to be measured with e rror without reference to a specific PK model. We give all example of all a nalysis of PK/PD data obtained following administration of all anticoagulan t drug. The study was originally carried out in order to make dosage recomm endations. The prior for the distribution of the true concentrations, which may incorporate all individual's covariate information, is derived as a pr edictive distribution from all earlier study. The errors-in-variables appro ach is compared with the joint modeling approach and more naive methods in which the observed concentrations, or the separately modeled concentrations , are substituted into the response model. Throughout, a Bayesian approach is taken with implementation via Markov chain Monte Carlo methods.