Population Pharmacokinetics Of Tacrolimus(FK506) In Chinese Liver Transplant Patients

FK506 is a macrolide lactone immunosuppressant that is used clinically for the prophylaxis or reversal of organ rejection after liver and kidney transplantation.The administration of FK506 is complicated by a narrow theraputic index,significant inter- and intrapatient variabilities in its pharmacokinetics,and the potential for serious drug interactions and toxicities.Also we have established a mutiple-center study protocol according to population PK study guide formulated by FDA combined with the PK characteristic and clinical usage of FK506.Steady state serum concentrations data (n=702) were collected from 67 patients of Shanghai First People's Hospital and Changzheng Hospital during their routine clinical care. The dosage history and sampling time were recorded in detail and the serum concentration of FK506 was determined by the IMx method.After fixed the values of the zbsorption rate,bioavailability and volumn of distribution,Population CL was calculated by using NOMMEM,with a one-compartment model of first-order absorption and elimination.The final regression model for FK506 clearance was: CL=2.37·Dose^0.542·WT^0.732·Age·1.23^MYC·1.67^LAM·1.46^INS CL:clearance(Unit:L/hr) Dose:daily dose(Unit:g/kg) WT:weight(Unit:kg) Age=0.836 for age>45 years old Age=1.49 for people is between 4 and 10 years old Age=1 for otherwise MYC=1 for comedication with MYC MYC=0 for otherwise LAM=1 for comedication with LAM LAM=0 for otherwise INS=1 for comedication with INS INS=0 for otherwiseGender and comedication with gliquidone do not alter the FK506 clearance.The clearance of FK506 remain unchanged with weight and dose.The comedication with MYC,LAM and INS do not alter FK506 clearance.The results also showed that the inter-individual variability in FK506 clearance, described by a proportional error model, had a variation coefficient of 15% and the residual variability, described using an additive model, was 0.829μg/ml.The final regression model for Clwas validated in a different group of 22 patients and the external validation revealed an important improvement in the predictive performance of FK506 concentrations in comparison with the simplest model which did not include any covariates.The mean and the variance of standardized prediction error was close to 0 and 1 respectively, indicating that the model was stable and applicable.The model was applied to estimate the individual CL for patients receiving FK506 in terms of patients'dose, age, weight and comedication with other antiepileptic drugs, in order to establish individualized dosing regimens, which can reduce the risk of toxity and improve the clinical effect.