G Stillemans (1), R Verbeeck (1), P Wallemacq (2), FT Musuamba (1,2)
(1) Louvain Drug Research Institute, Université Catholique de Louvain, Belgium, (2) Louvain centre for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Belgium
Objectives: To date, several pharmacokinetic (PK) models have been proposed for tacrolimus (TAC) in pediatric and adult patients after liver and kidney transplantation. They are different and lack consistence in their structure and in the covariates included. There is a need for more generic and robust models that could serve for dose individualisation in different settings and situations. This meta-analysis, using data from 9 trials, aimed to propose a library of mechanism-based models for tacrolimus in hepatic and renal transplantation – pediatrics and adults.
Methods: TAC doses and routine TDM trough levels from 289 pediatric and adult liver and kidney allograft recipients during the first year post-transplantation were used to develop a population PK model using NONMEM VII. Patients’ demographic and physiological characteristics, type of transplantation and biochemical test results were tested as covariates to explain interindividual variability in tacrolimus distribution and elimination. The model was then internally and externally validated using graphical and numerical tools.
Results: A two-compartment model with first-order elimination best described the TAC PK. Apparent volumes of distribution of central and peripheral compartments estimates were 30L and 200L, and intercompartmental clearance and blood clearance estimates were 11L/h and 45L/h. The absorption rate was fixed to 4.5h-1. Bodyweight was imputed on apparent clearance and volumes of distribution using an allometric function with the exponent estimated as part of the modelling process. Bias and precision of estimates were within acceptable limits in the model validation. Different models were proposed, using different combinations of covariates.
Conclusions: We developed and validated tacrolimus PK models covering the first year after pediatric and adult liver and kidney transplantation. After implementation in PK software with Bayesian prediction, these models could therefore constitute tools to help clinicians in tacrolimus dose individualisation.
Reference: PAGE 22 (2013) Abstr 2911 [www.page-meeting.org/?abstract=2911]
Poster: Other Drug/Disease Modelling