有限采样策略估算肾移植受者霉酚酸钠药-时曲线下面积方法的建立

Use of limited sampling strategy for estimating area under concentration-versus-time curve of mycophenolate sodium in renal allograft recipients

目的建立有限采样策略估算。肾移植受者霉酚酸钠药．时曲线下面积（AUC）的公式。方法2010--2013年南京军区福州总医院接受霉酚酸钠肠溶片（EC-MPS）联合钙调磷酸蛋白酶抑制剂（CNI）及泼尼松三联免疫抑制治疗的35例肾移植受者，血霉酚酸浓度采用均相酶免疫分析（EMIT）法测定，测定时间点为服药前及服药后0．5、1．0、1．5、2．0、3．0、4．0、6．0、8．0、12．0h。采用WINNOLIN软件计算霉酚酸的药代动力学参数。多元逐步回归分析法得出适合中国人群服用不同CNI药物时的简化霉酚酸药-时AUC计算公式。结果35例受者霉酚酸的AUC0-12b为14～67mg·h／L，平均（37±14）mg·h／L；不同肾移植受者之间药代动力学参数个体差异大，霉酚酸的谷浓度（C0）与霉酚酸的AUC0-12b相关性差（r2=0．090）。服用他克莫司的受者采用4个时间点取样的霉酚酸浓度（C4、C6、C8、C12）得出霉酚酸的AUC简化计算公式为AUC=5．678＋1．718×C4＋2．853×C6＋1．812XC8＋3．413×C12；该公式的霉酚酸AUC预测值与AUC0-12h实测值相关性较好（r2=0．890，P〈0．05），绝对预测误差中位数为3．45％（0．41％～24．71％），误差超过15％的比例为11．1％（2／18）。服用环孢素A的患者采用4个时间点取样的霉酚酸浓度（C3、C4、C6、C7）得出霉酚酸AUC简化计算公式为AUC=7．072＋1．525XC3＋1．558×C4＋1．573XC6＋2．285×CE；该公式的霉酚酸AUC预测值与AUC0-12h实测值相关性好（r2=0．952，P〈0．05），绝对预测误差中位数为6．50％（0．02％-12．91％），误差超过15％的比例为0。上述两个公式经Bland-Altman分析，与霉酚酸AUC0-12h一致性良好。结论本研究拟合的4点霉酚酸AUC简化计算公式能较好地反映霉酚酸药．时AUC，适用于肾移植受者服用EC—MPS的治疗药物监测。

Objective To establish a formula for estimating area under the concentration-versus- time curve （AUC） of mycophenolate sodium in Chinese renal allograft recipients with a limited sampling model. Methods A total of 35 renal allograft recipients were recruited from 2010 to 2013 to receive enteric- coated mycophenolate sodium （ EC-MPS）, calcineurin （CNI） and prednisone as immunosuppressive triple therapy. The serum concentration of mycophenolic acid （MPA） was assayed by enzyme multiplied immunoassay technique （EMIT） at pre-dose （ Co ）, 0. 5 （ C0.5 ）, 1.0 （ C1 ）, 1.5 （ C1.5 ）, 2. 0 （ C2 ）, 3.0 （C3 ）, 4. 0 （ C4 ）, 6.0 （ C6 ）, 8.0 （ Cs ） and 12.0 （ C,2 ） h post-dose respectively. Pharmacokinetic parameters of MPA （ Co, CI2, Cm~, T AUC0-12 h） were calculated by software WINNOLIN. Simplified formulae for estimation of MPA-AUC in tacrolimus （ Tac ） group or cyclosporin A （ CsA ） group were established by multiple stepwise regression analysis. Results There were variable MPA AUC0_12 h levels between 14 and 67 mg h/L （mean： 37 ＋ 14）. The MPA trough level （Co） had no correlations with MPA AUC0-12h（r2 =0. 090）. The simplified MPA AUC formula for Tac group was AUC = 5. 678 ＋ 1. 718 x C4 ＋2. 853 x C6 ＋ 1. 812 x Cs ＋ 3. 413 x CIz with four sampling points （ C4, C6 , C8 , C12 ）. Estimated AUC with the formula had eorrelations with AUC0_lz h （ r2 = 0. 890 ）. The mean absolute prediet error （APE） was 3.45% （0. 41% -24. 71% ） and the proportion of APE above 15% stood at 11.1% （2/18）. In CsA group, the simplified MPA AUC formula was AUC = 7. 072 ＋ 1. 525 C3 ＋ 1. 558 x C4 ＋ 1. 573 x C6 ＋ 2. 285 x C8. The correlation was r2 =0. 952, mean APE was 6. 50% （0.02% - 12. 91% ） and proportion of APE above 15% stood at 0. The above formulae were observed to have agreement with AUC0-12h by Bland-Ahman analysis. Conelusion The simplified MPA AUC formulae with 4-point sampling provide an effective approach for estimating full MPA AUC0-12h in Chinese renal recipients on EC-MPS plus tacrolimus or cyclosporin A.