Aya A. Saleh (1), Abdel-Hameed I.M. Ebid (1), Sara M.M. Abdel-Motaleb (1), Mohamed Adel (1), Ahmed F. Mira (2), Mariam A. Ahmed (1)
(1) Department of Pharmacy Practice, Faculty of Pharmacy, Helwan University, Cairo, Egypt. (2) Department of Gastroenterology, International Medical Center, Cairo, Egypt.
Objectives:
The aim of this study is to develop a population pharmacokinetic (PK) model that characterizes the PK of oral tacrolimus in adult living-donor liver transplant recipients and identify the significant predictors of tacrolimus pharmacokinetics.
Methods:
This study was conducted in the Liver and Kidney Transplantation Unit of the International Medical Centre (IMC), Cairo, Egypt. Data were collected from adult de novo living-donor liver transplant recipients who received oral tacrolimus as part of their immunotherapy from March 2015 to December 2018. Eighty-five (85) patients with 2495 tacrolimus trough concentrations were collected over 6 months post-transplantation for model building (Index group). Additionally, data from a further 18 patients with 266 tacrolimus trough concentrations were available for external validation. The pharmacokinetic data were analyzed by nonlinear mixed-effects modelling implemented in Monolix software, version 2018R2. The population pharmacokinetic parameters were estimated using the Stochastic Approximation Expectation-Maximization (SAEM) algorithm, combined with Markov Chain Monte Carlo (MCMC) and simulated annealing implemented in the software.
Results:
A one-compartment with first-order absorption and first-order elimination rate adequately described the data for the index dataset. The residual variability was described with a proportional error model. Post-operative days (POD) had the most significant influence on the CL/F. A linear spline model with 4 different periods was found to be the best model for the POD. These periods were (i) immediate POD: 1-7 days, (ii) early POD: 8-21 days, (iii) intermediate POD: 22-60 days, and (iv) late POD: > 60 days. The typical value of CL/F was 16.2 L/h in the immediate POD (1-7 days POD), which then increased by 22.5%, 35.3%, and 10.8% in the early, intermediate, and late POD, respectively. Additionally, CL/F was lower when the levels of blood urea nitrogen (BUN), international normalization ratio (INR), aspartate aminotransferase (AST) and the total bilirubin (TBIL) were high. The final model for tacrolimus clearance can be described according to the following equation: CL/F(L/h)=16.2× [(1.225,if days 8-21) or (1.353,if days 22-60) or (1.108,if days>60)]+(-0.0322×BUN )+(-1.55×INR)+(-0.00433×AST)+(-0.0708×TBIL). The model predicted the observed data of the external validation reasonably well, with reasonable bias and precision.
Conclusions:
In conclusion, a population PK model was developed for tacrolimus in adult living-donor liver transplant recipients. Tacrolimus CL/F was found to be lower than the reported one in Caucasian adults receiving deceased-donor full liver transplant. Therefore, dosing recommendations should not be directly extrapolated from previous studies that were based on patients receiving full-liver transplant. Our study also indicated that, after taking the days post-transplantation into account, liver function tests are still significant predictors of tacrolimus CL/F and hence dosing requirements. Among different liver function tests, AST, TBIL, BUN, and INR are the most significant predictors of individual tacrolimus dose requirements. The results of the study may be useful for tacrolimus dose optimization in recipients of living-donor liver transplantation.
Reference: PAGE () Abstr 9519 [www.page-meeting.org/?abstract=9519]
Poster: Drug/Disease Modelling - Other Topics