非线性混合效应模型法建立12岁以下癫痫儿童拉莫三嗪的群体药代动力学数学模型  

Establishment of Mathematical Model of Population Pharmacokinetics of Lamotrigine Using NONMEM in Epileptic Children Less than 12 Years Old

作　　者：熊友健[1] 胡榕[2] 卢庆红[1] 涂琼[1] 钟建民[3] XIONG You-jian;HU Rong;LU Qing-hong;TU Qiong;ZHONG Jian-min(Department of Pharmacy,Jiangxi Provincial Children’s Hospital,Nanchang 330006,China;Functional Division,Jiangxi Provincial Children’s Hospital,Nanchang 330006,China;Department of Neurology,Jiangxi Provincial Children’s Hospital,Nanchang 330006,China)

机构地区：[1]江西省儿童医院药学部,南昌330006 [2]江西省儿童医院功能科,南昌330006 [3]江西省儿童医院神经内科,南昌330006

出　　处：《南昌大学学报（医学版）》2020年第6期29-32,59,共5页Journal of Nanchang University：Medical Sciences

基　　金：江西省卫生计生委科技计划(20155590)。

摘　　要：目的采用非线性混合效应模型法(NONMEM法)建立拉莫三嗪(LTG)在12岁以下癫痫儿童中的群体药代动力学(PPK)数学模型,分析其用于预测癫痫患儿LTG血药浓度的可能性。方法收集118例服用LTG的癫痫患儿159份血药浓度监测数据和临床生化指标。通过NONMEM软件建立LTG在癫痫患儿中的PPK数学模型,用自举法(Bootstrap)评估数学模型参数的精确度,并用正态化预测分布误差(NPDE)作进一步评估和验证。结果最终回归数学模型:清除率(CL/F)=1.52×(WT/40)0.87×e-2.63,分布容积(V/F)=290.30×(AMT/100)0.75,CL/F、V/F典型值分别为1.52 L·h^-1和290.30 L。Bootstrap评估显示,最终回归数学模型的稳定性和效能均良好,典型值和固定效应参数的RSE均<16%(6.54%~15.22%)。NPDE分析显示,LTG的PPK最终模型模拟数据为标准正态分布,个体间变异度从72.61%下降到15.23%,表明LTG的PPK模型的变异度良好。结论运用NONMEM法建立的LTG在12岁以下癫痫患儿的PPK数学模型,可用于预测癫痫患儿LTG血药浓度。Objective To establish the mathematical model of population pharmacokinetics(PPK)of lamotrigine(LTG)in epileptic children less than 12 years old using a nonlinear mixed effects modeling(NONMEM)approach,and to analyze the application of the established model in predicting LTG concentrations in children with epilepsy.Methods A total of 159 plasma LTG concentration monitoring data and clinical biochemical indices were collected from 118 epileptic children.The mathematical model of PPK was established using NONMEM software.The precision of model parameters was estimated by Bootstrap sampling method,and further evaluated and validated by normalized prediction distribution errors(NPDE).Results The final regression model for LTG was as follows:clearance(CL/F)=1.52×(WT/40)0.87×e-2.63;volume of distribution(V/F)=290.30×(AMT/100)0.75.The typical values of CL/F and V/F were 1.52 L·h^-1 and 290.30 L,respectively.Bootstrap assessment showed that the final regression model provided good stability and efficiency,and the RSE of both typical values and fixed effect parameters was less than 16%(6.54%-15.22%).NPDE analysis showed that the simulated data of the final PPK model of LTG were consistent with standardized normal distribution,the individual variation decreased from 72.61%to 15.23%,indicating that the PPK model of LTG had a good variation.Conclusion The established PPK model of LTG can be used to predict plasma LTG concentrations in epileptic children less than 12 years old.

关 键 词：拉莫三嗪 非线性混合效应模型法 群体药代动力学 数学模型 12岁以下儿童 癫痫 血药浓度 

分 类 号：R917[医药卫生—药物分析学]