Determination of the most influential sources of variability of tacrolimus trough blood levels in adult liver transplant patients: a bottom-up approach
C. Gérard, J. Stocco, A. Hulin, B. Blanchet, M. Tod
EMR3738, ciblage thérapeutique en oncologie, faculté de médecine Lyon Sud, France
Objectives: Tacrolimus (TAC), an immusuppressive drug used for the prevention of graft rejection after liver transplantation, presents a large pharmacokinetic variability. Blood concentration of TAC are routinely measured to achieve and maintain target trough blood concentration (TBC). Most part of the interindividual variability remains unexplained. The objective was to identify predictive factors influencing TBC of TAC using a bottom-up approach.
Methods: A physiologically-based pharmacokinetic (PBPK) model of TAC was proposed, taking into account the body weight (BW), the proportion of adipose tissue, hematocrit, lipid fraction of organs, liver function, CYP3A5 genotype of patient and concomitant drugs (CYP3A4 inhibitors). TAC concentration profiles were simulated in a virtual population defined by a set of covariate values similar to those from a real population of 66 transplanted patients included in a multicentric PK study. Correlations between covariates were accounted for. For the validation of the PBPK model, TBC were compared with those observed in the real population. Then, the impact of each covariate has been tested on TBC of TAC in order to identify the most influential ones.
Results: For CYP3A5 *3/*3 genotype (non-expressor) of liver donor and three levels of drug-drug interaction (without inhibition, low and moderate inhibition), means were 7.84 vs 7.80, 8.22 vs 8.25 and 10.30 vs 9.97 ng/ml for observed vs simulated TBC of TAC.
With a dosage regimen of 0.04 mg/kg every 12 hours, significant increase of the TBC of TAC (6.61 to 17.1 ng/mL) has been found when the BW increase (50 to 110 kg). For the liver function (degrees of cirrhosis with Child Pugh score of A to C), hematocrit (0.19 to 0.43), proportion of adipose tissue (0.10 to 0.30), the same trend in TBC was found: means of 9.48 to 14.39, 4.92 to 11.8 and 6.00 to 8.83 ng/mL, respectively. Without drug-drug interaction, mean TBC of TAC were 3.68, 5.21 and 7.80 when CYP3A5 genotypes of the liver donor were *1/*1, *1/*3 and *3/*3, respectively.
Conclusions: Bottom-up approach allowed taking into account the influence of relevant covariates on TBC of TAC. The most influential covariates were BW, hematocrit, liver function and CYP3A5 genotype of the liver donor. These covariates should therefore be taken into account in therapeutic drug monitoring to adjust dosage regimen for each transplant recipient patient.
