| OBJECTIVE Using a population pharmacokinetic?PPK?model in kidney transplant recipients to quantitatively investigate the effects of demographic characteristics,physiological and biochemical indicators,genetic polymorphisms,and co-medications on tacrolimus pharmacokinetics,so as to provide a safe and effective theoretical basis for the individualized administration of TAC in kidney transplant recipients.METHODS The steady state trough concentration data,physiological and biochemical indicators,gene polymorphisms and co-medications information of kidney transplant recipients during outpatient or hospitalization from June 2015 to March 2019 were collected.The PPK model of TAC was established by the nonlinear mixed effect model?NONMEM?method,and the effects of covariates including gender,age,weight,blood routine,liver function,renal function,combination medications,and gene polymorphisms on the pharmacokinetic parameters of TAC were investigated by forward inclusion and backward exclusion.The internal verification methods include the goodness of fit plots?GOF?,normalized prediction distribution errors method?NPDE?,and Bootstrap.The root mean square error?RMSE?and mean prediction error?MPE?were used as evaluation indicators for external verification.Bayesian method on the basis of maximum posterior?MAPB?was used to evaluate the predictive performance of the model when it has 0 to 4 trough concentration points,separately,and the individual prediction error?IPE%?,median individual prediction error?MDIPE?,median absolute individual prediction error?MAIPE?,F20?|IPE|%within 20%?and F30?|IPE|%in 30%?as the evaluation indicators.The initial dose for different typical patients was selected from the final model by Monte Carlo simulation and summarized into a table,so as to improve the clinical application efficiency.RESULTS 1.In this study,218 patients and 2236 steady-state trough concentrations were included,among which 162 patients with 1634 trough concentrations were classified into the modeling group and 56 patients with 602 trough concentrations were divided into the model validation group.The clinical data of the patients including gender,age,weight,blood,liver function,kidney function,concomitant medications?erythromycin,wuzhi capsules,voriconazole,etc?and 12 gene polymorphisms that may be related to TAC transport or metabolism.2.In this study,a PPK model of TAC was established.The covariates of CYP3A5*3gene polymorphism,hematocrit?HCT?,erythromycin,voriconazole,and wuzhi capsules were included in the model.The final model apparent clearance rate?CL/F?and apparent distribution volume?V/F?were 24.6 L·h-1and 502 L,respectively,and the objective function value?OFV?was 5323.3.3.The internal verification through GOF,NPDE and Bootstrap method proved that the final model had higher stability and practicability,and the external verification results showed that the RMSE value between the tacrolimus observed concentration and the individual predicted concentration was 4.63 and the MPE was 0.1,indicating that the model had low prediction deviation and good prediction performance.4.The Bayesian prediction results indicated that the prediction performance of the model was related to the number of trough concentrations.Trough concentration point can significantly improve model prediction performance,even if there was only one.When two trough concentrations were obtained,the prediction performance of the final model was higher?MDIPE<10%,MAIPE<25%,IF20>50%,IF30>60%?and remained basically stable thereafter.More trough concentrations information does not represent a further reduction in prediction errors.Finally,the initial doses of tacrolimus were determined according to Monte Carlo simulations for various types of patients.The dose requirements vary widely among typical patients?from 1 to 14 mg?.The results were compiled into a reference table of initial doses of tacrolimus for clinical application.CONCLUSION A population pharmacokinetic model of tacrolimus was established in Chinese adult kidney transplant recipients using the NONMEM method.CYP3A5*3gene polymorphism,HCT,erythromycin,voriconazole,and wuzhi capsules had significant effects on the pharmacokinetics of tacrolimus.Internal and external verification results showed that the model was stable and reliable,indicating that the model could be served as an effective tool to guide tacrolimus in individualized treatment of kidney transplant recipients. |
PDF Full Text Request