Research Article TACROLIMUS POPULATION PHARMACOKINETICS AND BAYESIAN ESTIMATION IN TUNISIAN RENAL TRANSPLANT RECIPIENTS EMNA GAÏES 1,2 , MOHAMED MONGI BACHA 1,4 , JEAN-BAPTIST WOILLARD 4 , HANENE ELJEBARI 2 , IMED HELAL 3 , EZZEDDINE ABDERRAHIM 3 , NADIA JEBABLI 2 , FRANCK SAINT-MARCOUX 5 , PIERRE MARQUET 5 , TAÏEB BEN ABDALLAH 1,3 , ADEL KHEDER 3 , YOSR GORJI 4 , MOHAMED LAKHAL 1,2 , ANIS KLOUZ 1,2 . 1 Université de Tunis El Manar – Faculté de Médecine de Tunis 1007, 2 Service de Pharmacologie clinique – Centre National de Pharmacovigilance, 3 Service de médicine interne A – Hôpital Charles Nicolle, 4 Laboratoire d’Immunologie de la Transplantation Rénale et d’Immunopathologie (LR03SP01) - Tunis. Tunisie, 5 CHU Limoge, Service de Pharmacologie et toxicologie, Limoge, France. Email: eljebarihanene@yahoo.fr Received: 18 Mar 2013, Revised and Accepted: 25 Jun 2013 ABSTRACT Objective: The inter-dose area-under-the-curve (AUC) has been described as the best marker of exposure to Tacrolimus, suggesting its use for dose adjustment. In a population of Tunisian renal transplant patients, this study aimed (i) at building a population pharmacokinetic (PopPK) model for Tacrolimus, (ii) at identifying factors that explain inter-patient variability, and (iii) at developing a Bayesian estimator (MAP-BE) enabling the estimation of individual AUC. Patients and methods: Full-PK profiles were obtained from 20 stable renal transplant recipients given Prograf® and Tacrolimus blood concentrations were measured by a CMIA technique (ARCHITECT; Abott). PopPK analysis was performed using non linear mixed effects approach (NONMEM program). The following covariates were tested: age, weight, hematocrit, AST, ALT, albumin. PopPK parameters where then used as priors to develop a MAP-BE for the estimation of Tacrolimus AUC using a limited sampling strategy. The predictive performance of the MAP-BE were tested by (i) comparing the estimated AUC to that obtained by the trapezoidal rule; and (ii) its ability to provide similar dose adjustments to those obtained using all the available time-points. Validation was performed by both jackknife and bootstrapping methods. Results: Tacrolimus pharmacokinetics were well described by a two-compartment model combined with an Erlang distribution to describe the absorption phase: residual proportional error was 16% and imprecision parameter estimate was less than 15% (9.3 - 14.4%). Body weight was identified as a covariate influencing the apparent central volume of distribution (inter-patient variability decreased from 28 to 7.7%). MAP-BE based on three blood concentrations measured at 0, 1 and 3 h post-dose provided a good estimation of AUC with a mean bias -1 + 13.2% (-22 to 22%) with 85% of the patients having an AUC bias < 20%. The BE proposed similar doses to those proposed using all concentrations in 19out of 20 patients, with a maximum difference of 0.5 mg. Conclusion: A PopPK model and its associated Bayesian estimator providing good prediction of Tacrolimus exposure have been developed in Tunisian renal transplant recipients. These tools allow us to individualize Tacrolimus dosages based on the AUC using only three concentrations. Keywords: Tacrolimus, Population pharmacokinetics, NONMEM, Renal transplantation, Bayesian estimation. INTRODUCTION Tacrolimus is a calcineurine inhibitor, widely used in renal transplantation. It has a narrow therapeutic index and large inter- and intra individual variability [1] making its drug monitoring necessary. Therapeutic drug monitoring based on Tacrolimus exposure using the inter dose Area under the Curve (AUC) was used to prevent graft rejection and toxic side effect [2-3]. Unlike conventional pharmacokinetic studies, population pharmacokinetic analysis allows an estimate of the mean and variances of PK parameters directly in the population of interest as well as the relationship between these parameters and specific patient covariates using few blood sampling. In fact, many studies investigate population pharmacokinetic and develop bayesian estimator for an individualized Tacrolimus dosage regimen [4-7]. This approach was never established in Tunisian adult renal transplant recipients. In this context, this study aimed (i) at building a population pharmacokinetic (popPK) model for tacrolimus, (ii) at identifying factors that explain inter-patient variability, and (iii) at developing a Bayesian estimator (MAP-BE) enabling the estimation of individual AUC. METHODS Patients and data collection Data were obtained from 20 adult patients who underwent renal transplantation from 2006 to 2010 at the department of Urology in the Charles Nicolle Hospital of Tunis, Tunisia. Data collection was approved by the hospital ethic comity. Patients received Tacrolimus (Prograf®, capsule, 1 mg) twice a day with doses varying between 0.02 and 0.14 mg/kg/d. All patients received concomitantly mycophenolate mofetil. Blood collection Full pharmacokinetics profiles were collected in EDTA tubes at pre- dose, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12 hours post-dose. Blood samples were stored at -20°C until analysis. Tacrolimus assay Tacrolimus blood samples were analyzed at the department of Clinical Pharmacology of Tunis using a chemiluminescent microparticle immunoassay (CMIA) technique (Architect; Abbott). The lower limit of detection was 0.3 ng/ml and it was linear between 2 and 30 ng/ml with a correlation coefficient ≥ 0.90. The average difference bias exhibited by Architect Tacrolimus versus LC/MS/MS in this study was 0.51ng/ml. The 95% confidence interval of the ng/ml difference bias is 0.31ng/ml to 0.71ng/ml. The coefficient of variation of the assay (% CV) was less than 10% and the mean recovery was 102%. Population pharmacokinetic analysis The population pharmacokinetic analysis was conducted by the nonlinear mixed-effects modeling (NONMEM_ version VI) software (GloboMax_ LLC, Ellicott City, MD, USA) using Wings for NONMEM version 614 (developed by N. Holford, available from http://wfn.sourceforge.net/) [8]. All population pharmacokinetic analyses were carried out using the first-order conditional estimation method. One, two and three structural compartment International Journal of Pharmacy and Pharmaceutical Sciences ISSN- 0975-1491 Vol 5, Suppl 3, 2013 A A c c a a d d e e m mi i c c S S c c i i e e n n c c e e s s