目的 评估目前已发表的中国成人患者丙戊酸群体药动学(population pharmacokinetic, PPK)模型的预测性能,选择最适合双相情感障碍患者的模型。方法 在PubMed和CNKI数据库中系统检索已发表的丙戊酸PPK模型,将收集的双相情感障碍患者丙戊酸治疗药物监测数据代入模型中进行评估。评价指标包括:相对预测误差(PE%)、平均预测误差(ME)、平均相对预测误差(ME%)、平均绝对误差(MAE%)、均方根误差(RMSE)和实测浓度-预测浓度图。结果 共纳入9个丙戊酸PPK模型。外部验证数据集共纳入48例患者(48个浓度点作为已知浓度,157个浓度点作为验证浓度)。在无浓度反馈的情况下,模型9(徐静静)的ME%<10%;模型1(Lin等)和9的PE%在±30%以内的分布比例超过了70%,ME趋近于0;模型1、5(焦正等)和9的PE%在±20%以内的分布比例超过了50%,MAE<20 μg·mL-1,RMSE<25 μg·mL-1。使用Bayesian反馈进行模型评估后发现,模型1、5、6(黄春新等)、7(黄春新等)、9的预测性能均有显著的提高,PE%在±30%以内的分布比例超过了80%,PE%在±20%以内的分布比例超过了60%。结论 目前已发表的丙戊酸PPK模型中,模型1和9显示出了较好的预测能力,即使无Bayesian反馈,也可为初步剂量方案的制订提供参考。另外,模型1、5、6、7和9可结合Bayesian反馈,为双相情感障碍患者临床实践中剂量方案的调整提供参考。
Abstract
OBJECTIVE To evaluate population pharmacokinetic (PPK) model external predictability of valproate and choose the most suitable model for Chinese adult patients with bipolar disorder. METHODS Published models were screened from the database of PubMed and CNKI, and were validated using an external dataset collected from therapeutic drug monitoring. They were evaluated by predictive error% (PE%), mean error (ME), ME%, mean absolute error (MAE), root mean squared error (RMSE) and observed-predicted concentrations fitting plots. RESULTS Nine published models were evaluated with 48 patients (48 samples as prior concentrations, and 157 samples as posteriori Bayesian forecasting concentrations). ME% was less than 10% of Model 9 (Xu) without a prior concentration. ME were close to 0 and the percentages of PE%<±30% were more than 70% of Model 1 (Lin, et al) and 9 without a prior concentration. MAE were less than 20 μg·mL-1, RMSE were less than 25 μg·mL-1 and the percentages of PE%<±20% were more than 50% of Model 1, 5 (Jiao, et al) and 9 without a prior concentration. The posteriori Bayesian forecasting approach improved the predictive performance of the model 1, 5, 6 (Huang, et al), 7 (Huang, et al), and 9. The percentages of PE%<±20% and <±30% were increased to 60% and 80%, respectively. CONCLUSION The published models of 1 and 9 show good predictive ability, and may help developing initial dosing regimens. Model 1, 5, 6, 7 and 9 combined with posteriori Bayesian forecasting approach may be a useful tool for guiding dose adjustments in patients with bipolar disorder in clinical practice.
关键词
丙戊酸 /
双相情感障碍 /
模型验证 /
群体药动学 /
贝叶斯
{{custom_keyword}} /
Key words
valproate /
bipolar disorder /
model validation /
population pharmacokinetic /
Bayesian
{{custom_keyword}} /
中图分类号：
R969.1
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] CIPRIANI A, REID K, YOUNG A H, et al. Valproic acid, valproate and divalproex in the maintenance treatment of bipolar disorder[J]. Cochrane Database Syst Rev, 2013,(10): CD003196.
[2] HIEMKE C, BERGEMANN N, CLEMENT H W, et al. Consensus guidelines for therapeutic drug monitoring in neuropsychopharmacology: update 2017[J]. Pharmacopsychiatry, 2018, 51(1-2): 9-62.
[3] METHANEETHORN J. Population pharmacokinetics of valproic acid in patients with mania: implication for individualized dosing regimens[J]. Clin Ther, 2017, 39(6): 1171-1181.
[4] ZHAO S M, JIANG D C, WANG Y W, et al. Evaluation of the prediction performance of population pharmacokinetic models of sodium valproate in epileptic children in China[J]. Chin Pharm J (中国药学杂志), 2020, 55(3): 233-238.
[5] TAUZIN M, TRELUYER J M, NABBOUT R, et al. Simulations of valproate doses based on an external evaluation of pediatric population pharmacokinetic models[J]. J Clin Pharmacol, 2019, 59(3): 406-417.
[6] ZHONG D F, LI G, LIU C X. Guidance on bioanalysis: method validation and analysis of study samples (draft)[J]. Drug Eval Res(药物评价研究), 2011, 34(6): 409-415.
[7] LIN W W, JIAO Z, WANG C L, et al. Population pharmacokinetics of valproic acid in adult Chinese epileptic patients and its application in an individualized dosage regimen[J]. Ther Drug Monit, 2015, 37(1): 76-83.
[8] HU M, ZHANG J H, SUN H, et al. Estimation of relative clearance of valproic acid with nonlinear mixed effect model in epileptic out-patients[J]. Chin J Clin Pharm (中国临床药学杂志), 2000, 9(1): 33-36.
[9] HU M, ZHANG J H, SUN H, et al. Construction and clinical application of the population pharmacokinetic model of valproic acid[J]. Eval Anal Drug-Use Hosp China (中国医院用药评价与分析), 2001, 1(2): 95-97.
[10] HU M, ZHANG J H, SUN H, et al. Estimation of the influence of three anti-epileptic drugs on relative clearance of valproic acid with NONMEM[J]. Chin Pharm J (中国药学杂志), 2002, 37(3): 201-205.
[11] JIAO Z, ZHONG M K, HU M, et al. Population pharmacokinetic modeling of valproic acid clearance[J]. Chin J Hosp Pharm (中国医院药学杂志), 2004, 24(9): 515-517.
[12] JIAO Z, ZHONG M K, SHI X J, et al. Investigation of valproic acid-carbamazepine interaction by mixed effect modeling in Chinese epilepsy patients[J]. Chin Pharm J (中国药学杂志), 2004, 39(2): 130-132.
[13] HUANG C X, ZHOU G Y, HUANG Y L, et al. Estimation of pharmacokintic parameters of valproic acid by Bayesian approach in epileptic patients[J]. Chin J Clin Pharm (中国临床药学杂志), 2005, 14(5): 278-281.
[14] HUANG C X, LIN H, HAN F X, et al. Estimation of pharmacokinetics parameter of valproic acid and administration with individuation by Bayesian[J]. Chin J Hosp Pharm (中国医院药学杂志), 2005, 25(4): 324-329.
[15] ZHANG Z C, LI X X, FENG W Y. Population pharmacokinetics of sodium valproate in epileptic patients[J]. J Shenyang Pharm Univ (沈阳药科大学学报), 2009, 26(12): 1001-1003.
[16] XU J J. A pharmaceutical intervention: controlling the epilepsy seizure of postoperative brain injury based on the population pharmacokinetics of valproic acid [D]. Shanxi Medical University, 2016.
[17] METHANEETHORN J. A systematic review of population pharmacokinetics of valproic acid[J]. Br J Clin Pharmacol, 2018, 84(5): 816-834.
[18] WHITING B, KELMAN A W, GREVEL J. Population pharmacokinetics. Theory and clinical application[J]. Clin Pharmacokinet, 1986, 11(5): 387-401.
[19] VASUDEV K, DAS S, GOSWAMI U, et al. Pharmacokinetics of valproic acid in patients with bipolar disorder[J]. J Psychopharmacol, 2001, 15(3): 187-190.
[20] CLOYD J C, FISCHER J H, KRIEL R L, et al. Valproic acid pharmacokinetics in children. IV. Effects of age and antiepileptic drugs on protein binding and intrinsic clearance[J]. Clin Pharmacol Ther, 1993, 53(1): 22-29.
[21] ZHAO M M, SUN Y X, QIU F, et al. Comparison of high performance liquid chromatography, fluorescence polarization immunoassay and mass spectrometer method in determination of valproic acid in serum[J]. Chin J Clin Pharmacol (中国临床药理学杂志), 2015, 31(5): 370-374.
[22] OWEN J S, FIEDLER-KELLY J, LU W, et al. Introduction to Population Pharmacokinetic/Pharmacodynamic Analysis with Nonlinear Mixed Effects Models (PK/PD建模实践NONMEM软件入门)[M]. Beijing: Chemical Industry Press, 2020: 178.
[23] DESOKY E, MEINSHAUSEN J, BUHL K, et al. Generation of pharmacokinetic data during routine therapeutic drug monitoring: Bayesian approach vs. pharmacokinetic studies[J]. Ther Drug Monit, 1993, 15(4): 281-288.
[24] PROOST J H. Adaptive control of drug dosage regimens using maximum a posteriori probability Bayesian fitting[J]. Int J Clin Pharmacol Ther, 1995, 33(10): 531-536.
[25] ZHAO C Y, JIAO Z, MAO J J, et al. External evaluation of published population pharmacokinetic models of tacrolimus in adult renal transplant recipients[J]. Br J Clin Pharmacol, 2016, 81(5): 891-907.
[26] LIN W W, WANG C L, HUANG P F, et al. Timing and quantitative administration of carbamazepine and valproic acid based on population pharmacokinetics models and its clinical application[J]. Chin J Hosp Pharm (中国医院药学杂志), 2015, 35(12): 1108-1112.
[27] GUO Q, LIU T T, JING L, et al. Predictive performance and analysis of a vancomycin population pharmacokinetic model in Chinese pediatric patients[J]. Acta Pharm Sin(药学学报), 2019, 54(3): 528-532.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
首都卫生发展科研专项资助(首发2018-4-2124);北京市属医院科研培育计划资助(PZ2020031)
{{custom_fund}}
PDF(1853 KB)
