基于非布司他生理药动学模型预测潜在药物相互作用

Predicting potential drug interactions based on the febuxostat physiological pharmacokinetic model

目的:建立非布司他在健康人体内的生理药动学(PBPK)模型,计算人体内不同部位的药物浓度,预测非布司他对基于CYP2C8和CYP2D6介导代谢的药物的潜在相互作用(DDIs)。方法:通过文献收集和ADMET Predictor^(TM)软件预测获得非布司他建模的理化性质参数、生物药剂学参数、人体生理参数等;同时收集不同给药剂量非布司他在人体内的血药浓度(PK)数据,建立并验证PBPK模型。采用PBPK模型计算非布司他在肠道上皮细胞浓度、血浆达峰浓度、肝脏达峰浓度、肝脏入口游离药物浓度,预测非布司他与基于CYP2C8和CYP2D6代谢酶代谢药物的DDIs。结果:PBPK模型预测非布司他不同剂量下的药-时曲线与实测值拟合良好。预测的非布司他不同剂量下药动学参数(AUC,C_(max),T_(max))与实测值相近,倍数误差<2,模型准确可靠。DDIs预测结果显示采用肝脏达峰药物浓度与肝入口游离药物浓度来代替酶活性位点抑制剂浓度作出的DDIs预测结果更与临床结果相符合,即非布司他在健康人群日服用80 mg剂量下,基本不会对由CYP2C8和CYP2D6酶代谢的药物产生明显的DDIs。结论:所建立的PBPK模型可较好预测非布司他的体内药时曲线与机体不同部位的药物浓度,且能准确预测药物可能的相互作用。

OBJECTIVE To establish a physiological pharmacokinetic(PBPK) model of febuxostat in healthy individuals, and to calculate the concentration of drugs in different sites to predict DDIs mediated by CYP2C8 and CYP2D6. METHODS The characteristics of febuxostat including physicochemical properties, biopharmaceutical parameters, physiological parameters of human body were obtained through literature collection and prediction with ADMET PredictorTM software. Meanwhile, the blood concentration(PK) data with regard to different doses of febuxostat were collected to establish and validate PBPK model. The PBPK model was used to calculate the concentrations of febuxostat in intestinal epithelial cells, peak drug concentrations, liver peak concentrations and liver inlet free drug concentrations to predict the DDIs of febuxostat as well as drugs based on metabolism of CYP2C8 and CYP2D6 metabolic enzymes. RESULTS The drug concentration-time curves of febuxostat at different doses predicted by PBPK model fitted well with measured values. The pharmacokinetic parameters(AUC, Cmax, Tmax) at different doses of febuxostat were similar to the measured values with a fold error of < 2, which showed that the model was accurate and reliable.The results of DDIs prediction showed that compared to DDIs predicted by the enzyme activity site inhibitor concentration, in vitro DDIs predicted by hepatic peak drug concentration and liver inlet free drug concentrations are more consistent with the clinical results,thus, febuxostat does not produce significant DDIs with drugs metabolized by CYP2C8 and CYP2D6 at a dose of 80 mg daily among healthy individuals. CONCLUSION The well-established PBPK model can be a favorable tool to predict in vivo drug concentration-time curve of febuxostat and the drug concentration in different parts of the body as well as drug-drug interactions.