Development of a Population PK Model of Tacrolimus for Adaptive Dosage Control in Stable Kidney Transplant Patients

Background: Tacrolimus pharmacokinetics (PK) presents a high variability that hampers its therapeutic use. The aims of this study are to: (1) develop a population pharmacokinetic (PPK) model for tacrolimus and to identify the factors that contribute to the variability of tacrolimus PK in renal transplant patients; and (2) to establish a new Bayesian estimator that can easily and routinely be applied in the hospital. A new PPK model may allow efficacy to be optimized, improve dose regimens, minimize side effects, and decrease the cost of extensive area under the curve (AUC) monitoring. Methods: PPK analysis of the full PK profiles of 16 patients on 5 occasions was performed with NONMEM 7.2. Biochemical variables (hematocrit, hemoglobin, aspartate aminotransferase, and others) were analyzed. Results: A 2-open-compartment model with interoccasion variability best described the PK of tacrolimus. Three transit compartments provided the best description of the absorption process. The hematocrit, aspartate aminotransferase, and alanine aminotransferase were not significant in the covariate analysis. External validation with 91 patients proved the good predictability of the model with a bias and precision of 0.37 mcg/L (CI 95%, -0.11 to 1.20 mcg/L) and 0.38 mcg/L (CI 95%, 0.02 to 1.21 mcg/L), respectively. A limited sampling strategy using 1 sampling point at predose (trough concentrations) showed a good performance in AUC(0-12h) estimation with a correlation between AUC(full) and AUC(LSS,) bias and imprecision of r(2) = 0.75, 6.78% (range, -16.26% to 30.06%) and 1.42% (IC 95%, 0.14%-3.61%), respectively. Conclusions: The PPK model developed provides reliable prior information for Bayesian adaptive control of dosage regimens of tacrolimus to achieve the desired AUC goals in stable renal transplant patients.