基于药物基因组学的丙戊酸钠治疗儿童癫痫的群体PPK/PPD模型研究

Sodium valproate is a antiepileptic drug which widely used in children with epilepsy. One of the hotspots of the project based on the treatment of drug monitoring study population pharmacokinetic/pharmacodynamic binding model theory, nonlinear mixed effect model mainstream methods. First retrospectively collected data of 200 cases of valproate treatment of epilepsy in children and the genotypes of cytochrome P450 enzyme and UGT using the illumina genechip initially establish population pharmacokinetic model of epilepsy in children; further 200 cases were collected prospectively sodium valproate treatment of 1-year follow-up data, monthly monitoring of steady-state concentration to assess the clinical effect, check EEG-related receptor genotype determination of metabolic pathways and therapeutic targets, establish and perfect the sodium valproate population pharmacokinetic/model pharmacodynamics binding. The project can quantitatively describe the metabolic enzyme and the target receptor genotype valproate predict the efficacy and adverse effects of sodium valproate treatment of epilepsy in children,which it is the progress of individualized treatment，and understanding the mechanism of sodium valproate pharmacological effects, to promote new the development of anti-epileptic drugs.

丙戊酸钠为一线的广谱抗癫痫药，广泛应用于儿童癫痫。丙戊酸钠药代动力学及药效学个体差异较大, 其剂量与血药浓度之间的相关性较差，其疗效也存在很大差异。目前，丙戊酸钠相关代谢酶和受体的基因多态性与丙戊酸钠代谢和临床效应的相关性尚不明确，与其他因素的交互作用对其代谢和疗效的影响更是缺乏研究。进行儿童癫痫的丙戊酸钠个体化治疗研究势在必行。本研究以药物基因组学为基础，应用自定义芯片的方法测定细胞色素P450酶、UGT等基因型，收集患者资料，初步建立儿童癫痫的群体药动学模型；进而应用该模型，前瞻性收集200例丙戊酸钠治疗及随访数据，建立和完善丙戊酸钠的群体药动学/药效学结合模型。旨在定量描述代谢酶和靶受体基因型对丙戊酸钠治疗的影响，预测丙戊酸钠治疗儿童癫痫的疗效和不良反应，将个体化治疗模式应用于临床。

项目背景：丙戊酸(VPA)为广谱抗癫痫药,在人体内的代谢途径主要有3条:葡萄糖醛酸代谢途径(UGT, 50%),线粒体内的β-氧化途径(40%)和细胞色素酶CYP450介导的氧化途径(10%)。本项目测定了丙戊酸代谢、转运和效应通路中各相关基因的多态性及其对儿童癫痫患者的稳态血药浓度及疗效的影响，并建立群体药动学/药效学模型，为个体化治疗奠定基础。.项目内容：本研究运用MassARRAy Sequenom时间飞行质谱生物芯片对384例中国汉族癫痫儿童丙戊酸代谢、转运和效应通路上的29个基因的90个位点进行基因分型，并分析他们与丙戊酸稳态血药浓度及疗效的相关性，建立丙戊酸群体药动学/药效学模型。.重要结果及关键数据：UGT1A6 rs28898617与丙戊酸口服液稳态浓度升高相关(Beta = 5.31, 95%CI = 0.78-9.83, P = 0.024)。rs2279020 (GABRA1, G>A)会降低丙戊酸口服液的耐药风险(OR = 0.42, 95% CI = 0.21–0.84, P = 0.014)。UGT1A6 rs28898617与丙戊酸缓释片稳态浓度升高相关(BETA = 0.32, 95% CI = 0.082–0.55, P = 0.0089)。在特发性全面性癫痫中，三个位点与疗效相关：UGT2B7 rs7668282 (OR = 2.66, 95% CI = 1.19–5.91, P = 0.017)，CYP3A4 rs2242480 (OR = 0.27, 95% CI = 0.095–0.79, P = 0.017)和SCN1A rs10188577 (OR = 0.40, 95% CI = 0.17–0.94, P = 0.035)。在部分性癫痫中，一个位点与疗效相关：CYP2A6 rs28399433 (OR = 5.06, 95% CI = 1.17–22.01, P = 0.031)。.课题组已应用NONMEM法初步建立PPK/PPD模型，并验证该模型具有较好的拟合效果。PPK/PPD模型研究中发现ABAT、FABP2和UGT1A9可能影响VPA的药物清除率，但仍需进一步扩大数据进行验证。.科学意义：丙戊酸血药浓度及疗效受到基因多态性的影响，课题组已应用NONMEM法初步建立并验证PPK/PPD模型，为个体化治疗提供依据。