An Optimal Sampling Strategy for Tacrolimus in Pediatric Liver Transplant Recipients Based on a Population Pharmacokinetic Model
Nastya Kassir (1,2,3), Jean-Romain Delaloye (1,4), Mohamad-Samer Mouksassi (2,3), Anne-Laure Lapeyraque (1,5), Line Labbé (2), Fernando Alvarez (5), Michel Lallier (4), Mona Beaunoyer (4), Yves Théorêt (1,6,7), Catherine Litalien (1,5,7)
(1) Clinical Pharmacology Unit, CHU Ste-Justine, Montreal, Quebec, Canada; (2) Faculty of Pharmacy, Université de Montréal, Montreal, Quebec, Canada; (3) Pharsight - A Certara Company, Montreal, Quebec, Canada; Departments of (4) Surgery, (5) Pediatrics, and (6) Biochemistry, CHU Ste-Justine, Montreal, Quebec, Canada; (7) Department of Pharmacology, Université de Montréal, Montreal, Quebec, Canada.
Objectives: Trough concentration is commonly used for tacrolimus (TAC) dosing optimization despite its inadequacy in predicting correct drug exposure. Although tacrolimus systemic exposure may be best reflected by AUC, extensive blood sampling is not suitable for outpatient monitoring. Instead, a model based AUC is being considered. Objectives were to 1) develop a population pharmacokinetic (PK) model for TAC in pediatric liver transplant patients; and 2) determine a limited sampling strategy that accurately predicts TAC exposure.
Methods: Study included 28 patients receiving TAC orally twice daily. Median patient age and weight were 7.3 years old, and 20.4 kg, respectively. A population PK model was fitted to whole blood concentration data (PhoenixTM NLME). Covariate analysis was performed using a stepwise forward additive and a backward elimination approach. Influence of the following covariates was explored: body weight, body mass index, age, gender, type of transplant (full or cut-down liver), liver function tests, hematocrit, hemoglobin, drug interactions, and time post-transplantation. Using the final model, a practical optimal sampling strategy was developed using WinPOPT software. The best limited sampling strategy among combinations of a maximum of four sampling time-points was tested. Precision of individual parameter estimates was obtained using simulation and re-estimation using a candidate design.
Results: Concentration-time profiles of TAC were adequately fitted by a 2-compartment model with first order absorption. Weight was found to be significant on both oral clearance (CL/F) and central volume of distribution (Vc/F). Although not statistically significant, CL/F had a trend to be higher in patients transplanted with full liver, as compared to those who received a cut-down liver. Estimates of CL/F and Vc/F for a patient weighing 20 kg were 16.9 L/h and 47.5 L, respectively. Based on the optimized sampling strategy, the expected standard error on population CL/F and AUC0-12 was very low (1.9 %). This design also enabled the estimation of empirical Bayesian estimates of interest (CL/F and AUC0-12) with good precision.
Conclusions: The population PK model of tacrolimus, and empirical Bayesian estimates based on three or four blood concentration measurements, represent an accurate and convenient method to predict tacrolimus AUC0-12 in pediatric liver transplant recipients, despite high inter-individual variability in PK and patient demographics.