The aim of this study was, using routine drug monitoring data, to identify patient characteristics that may influence everolimus (EVE) pharmacokinetic parameters and to develop a population pharmacokinetic model to predict EVE whole blood concentrations in cardiac recipients.
Fifty-nine patients were enrolled in the prospective study. Patient’s characteristics were recorded including biological covariates and treatments. CYP3A5 and ABCB1 polymorphisms were determined. Seven hundred seventy-five EVE blood samples were collected for routine drug monitoring. Population pharmacokinetic modeling was carried out using the nonlinear mixed-effects modeling program. Results were analyzed according to a 1-compartment pharmacokinetic model with linear absorption and elimination. The model was evaluated using a bootstrap method and a visual predictive check procedure.
The pharmacokinetic of EVE in cardiac recipients was best described by a 1-compartment model. Interindividual variability was best described by an exponential error model and residual error by a proportional plus additive error model. Estimation of EVE apparent clearance (3.33 ± 0.20 L/h) and apparent volume of distribution (146 ± 33 L) were in accordance with previously published data. Bilirubinemia and cyclosporine significantly influenced EVE clearance. Some covariates that were expected to influence EVE clearance, for example, ABCB1 and CYP3A5 polymorphisms, were not evidenced. No covariates influenced the volume of distribution of EVE.
This study is the first population pharmacokinetic model of EVE in heart transplantation patients. It allows a better description of the pharmacokinetics of EVE. The present population pharmacokinetic model allows estimating a priori and a posteriori EVE concentrations in cardiac recipients and could limit the over and under drug exposure in this population.