有限采样法估算麦考酚钠药-时曲线下面积及其药动学

摘要

Objective To establish a model for estimating area under the concentration -time curve( AUC) of mycophenolic acid(MPA)by a limited sampling strategy. Methods Sixteen renal allograft recipients were treated with enteric coated mycophenolate sodium(EC-MPS)twice a day respectively.The blood samples were collected on postoperative days 4 and 7 before and 0.5,1,1.5,2,3,4,6,8,10,12 h after EC-MPS oral administration.Ultra performance liquid chromatography-UV (UPLC-UV)method was employed to determine the plasma concentration of MPA. Pharmacokinetic parameters of MPA were calculated by DAS 2.0 software. Areas under curve over the period of 0 to 12 h(AUC0-12 h )were calculated by using linear trapezoidal rule.The limited sampling model and relevant parameters were established by multivariate linear stepwise regression analysis with SPSS 17.0. Results The main pharmacokinetic parameters on postoperative days 4 and 7 were as follows:t1/ 2 were(6.93±6.51)and(8.18±5.06)h(P>0.05);Cmax were(10.10±6.44)and(17.26±11.67)mg·L-1(P<0.05);AUC0-12 h were(32.47±15.53)and(41.88±18.30)mg·h·L-1(P<0.05).Some patientsˊ concentration-time curves were bimodal.The second peak appeared at 4~12 h after treatment and the peak obviously decreased compared with the first one.No single plasma concentration can accurately predict AUC0-12 h .After analyzing the correlation between multiple plasma concentrations of MPA and AUC0-12 h by the multivariate linear stepwise regression analysis method,we selected the 0,4,1.5,8 h plasma concentrations of MPA to estimate AUC0-12 h .The computational formula was AUC0-12 h = 6.207+C0 ×6.881+C4 ×2.388+C1.5 ×0.541+C8 ×4.252(r2 =0.888).The predicted value has a good correlation with the measured value(r2 = 0.918)and the relative prediction error was in the acceptable range. Conclusion The multivariate regression model of four points sampling method is a better predictor of MPA- AUC0-12 h .This simple method will be conducive to the further application and popularization in clinic,improve the rational use of mycophenolate sodium,and realize the purpose of curative effect and adverse reaction prediction.%目的：建立有限采样法估算霉酚酸（MPA）药-时曲线下面积的模型。方法16例肾移植患者口服麦考酚钠肠溶片（EC-MPS）720 mg，q12 h，分别于术后第4天、第7天服药前及服药后0.5，1，1.5，2，3，4，6，8，10，12 h 采集静脉血样3 mL，采用超高效液相色谱-紫外（UPLC-UV）分析方法测定 MPA 血浆浓度。以 DAS 2.0版药动学软件进行药动学分析，MPA-AUC0-12 h按梯形法计算。用 SPSS 17.0版软件进行多元线性逐步回归分析方法，建立有限采样法多元回归模型，并计算相关参数。结果术后第4天和术后第7天 MPA 的主要药动学参数 t1/2分别为（6.93±6.51），（8.18±5.06）h（P>0.05）；Cmax分别为（10.10±6.44），（17.26±11.67）mg · L-1（ P <0.05）；AUC0-12 h 分别为（32.47±15.53），（41.88±18.30）mg·h·L-1（P<0.05）。观察到部分患者药-时曲线呈双峰，在服药后4~12 h 出现第二峰，峰值较第一峰明显降低。发现任何单一时间点 MPA 血药浓度不能准确预测 AUC0-12 h。采用多元线性逐步回归分析方法分析多个时间点MPA 血药浓度与 AUC0-12 h的相关性后，综合考虑选择0，4，1.5，8 h 4个时间点的 MPA 血药浓度来估算 AUC0-12 h ，计算公式为 AUC0-12 h =6.207＋ C0×6.881＋ C4×2.388＋C1.5×0.541＋C8×4.252（r2=0.888），其预测值与实测值有很好的相关性（r2=0.918），相对预测误差在可接受范围内。结论该研究建立的4点采样法多元回归模型可较好的预测 MPA-AUC0-12 h ，有利于进一步在临床上推广和应用，从而提高麦考酚钠合理用药水平，达到预测疗效和不良反应的目的。