目的：整合定量结构性质关系（QSPR）模型预测化合物在人体的吸收、分布、代谢、排泄 （ADME）性质参数和基于生理药代动力学（PBPK）模型预测人体药代动力学（PK）曲线的方法，并评价该方法的预测能力。方法：以文献报道的具有体外实测理化、生物药剂学性质和临床观测PK性质的 14个化合物作为模型药物。采用ADMET Predictor软件的QSPR模型预测各个化合物的理化与生物药剂学参数，将上述预测的参数加载到GastroPlus软件的PBPK模型中预测各个化合物经口服给药后在人体的PK 曲线以及主要PK参数。对比预测与实测ADME/PK参数间的差异，以评估所用模型的预测效能。结果： QSPR模型预测的理化与生物药剂学性质参数与观测值间的绝对值较为接近，两者具有较好的线性关系（大部分参数的相关系数均接近或超过0.7）；14个化合物中，有6个化合物（43%）的最大血药浓度 （Cmax）预测值落在观测值的2倍误差范围内，9个化合物（64%）的Cmax落在观测值的3倍误差范围内； 有7个化合物（50%）的血药浓度-时间曲线下的面积（AUC）预测值落在观测值的2倍误差范围内，8个化合物（57%）的AUC落在观测值的3倍误差范围内。结论：联合QSPR和PBPK模型可用于评估化合物的ADME性质并进一步预测人体PK特征。经过当前工作的验证，表明该方法具有较高的预测能力。
Abstract
Objective: To integrate the quantitative structural property relationship (QSPR) model to predict the absorption, distribution, metabolism, excretion (ADME) property parameters of compounds in the human body and a physiological pharmacokinetic (PBPK) model to predict human pharmacokinetics (PK) profiles, and then evaluate the predictive performance of the methods. Methods: The 14 compounds with in vitro physicochemical, biopharmaceutical, and clinically observed PK properties reported in the literature were used as model drugs. The QSPR model of ADMET Predictor software was used to predict the physicochemical and biopharmaceutical parameters of each compound, and the above predicted parameters were loaded into the PBPK model of GastroPlus software to predict the PK curve and main PK parameters of each compound after oral administration. The predicted and measured ADME/PK parameters were compared to assess the predictive performance of the current models. Results: The absolute values of the physicochemical and biopharmaceutical parameters predicted by the QSPR models and the observed values were relatively close, and both of them had a good linear relationship (the correlation coeffi cients of most parameters were close to or exceeded 0.7). In these 14 compounds, there were 6 compounds (43%) whose predicted maximum plasma concentrations (Cmax) were within 2-fold error of the observed value, and 9 compounds (64%) whose Cmax were within 3-fold error; there were 7 compounds (50%) whose predicted areas under the concentration-time curve (AUC) were within the 2-fold error of the observed value, and 8 compounds (57%) whose AUC were within 3-fold error. Conclusion: The combined QSPR and PBPK models can be used to evaluate the ADME properties of compounds and further predict human PK profi les. The validation of the current work showed that the method had high predictive ability.
关键词
QSPR模型；PBPK模型；ADME/PK性质；效能评价；新药研发
{{custom_keyword}} /
Key words
quantitative structure property relationship model； physiological based pharmacokinetic model； ADME/PK properties； performance evaluation； new drug research & development
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Mak KK,Epemolu O,Pichika MR.The Role of DMPK Science in Improving Pharmaceutical Research and Development Efficiency [J].Drug Discov Today，2022，27（3）：705-729.
[2] 李敬来，崔孟珣，张振清，等.药物代谢动力学在药物研发中的意义及作用[J].解放军药学学报，2010（3）：4.
[3] Cook D,Brown D,Alexander R,et al.Lessons Learned from the Fate of Astrazeneca's Drug Pipeline:A Five-Dimensional Framework [J].Nat Rev Drug Discov，2014，13（6）：419-431.
[4] 李健，杨进波，王玉珠.模型引导的药物开发在新药研发中的应用[J].中国临床药理学与治疗学，2020，25（1）：8.
[5] Layla Abdel-Ilah,Elma Veljovic,Lejla Gurbeta,et al.Applications of QSAR Study in Drug Design [J].Int J Eng Res & Tech，2017，6（6）：582-587.
[6] 马丽娜.基于QSPR和分子动力学模拟的中药成分肠吸收预测方法研究[D].北京中医药大学，2020.
[7] 李海燕.基于结构预测药物口服吸收/肝清除及生物利用度[D].沈阳药科大学，2009.
[8] Kunal Roy.Advances in QSAR Modeling[M].Springer，2017：339-384.
[9] Naga D,Parrott N,Ecker GF,et al.Evaluation of the Success of High-Throughput Physiologically Based Pharmacokinetic(HT-PBPK)Modeling Predictions to Inform Early Drug Discovery[J].Mol Pharm，2022，19（7）：2203-2216.
[10] Rowland M,Peck C,Tucker G.Physiologically-Based Pharmacokinetics in Drug Development and Regulatory Science[J].Annu Rev Pharmacol Toxicol，2011，51：45-73.
[11] Jones HM,Parrott N,Jorga K,et al.A Novel Strategy for Physiologically Based Predictions of Human Pharmacokinetics[J].Clin Pharmacokinet，2006，45（5）：511-542.
[12] Jones HM,Gardner IB,Collard WT,et al.Simulationof Human Intravenous and Oral Pharmacokinetics of 21 Diverse Compounds Using Physiologically Based Pharmacokinetic Modelling[J].Clin Pharmacokinet，2011，50（5）：331-347.
[13] Grimstein M,Yang Y,Zhang X,et al.Physiologically Based Pharmacokinetic Modeling in Regulatory Science:An Update from the U.S.Food and Drug Administration's Office of Clinical Pharmacology[J].J Pharm Sci，2019，108（1）：21-25.
[14] De Buck SS,Sinha VK,Fenu LA,et al.Prediction of Human Pharmacokinetics Using Physiologically Based Modeling:A Retrospective Analysis of 26 Clinically Tested Drugs[J].Drug Metab Dispos，2007，35（10）：1766-1780.
[15] Zou P,Yu Y,Zheng N,et al.Applications of Human Pharmacokinetic Prediction in First-in-Human Dose Estimation[J].AAPS J，2012，14（2）：262-281.
[16] Jones HM,Gardner IB,Watson KJ.Modelling and PBPK Simulation in Drug Discovery[J].AAPS J，2009，11（1）：155-166.
[17] Ren HC,Sai Y,Chen T.Evaluation of Generic Methods to Predict Human Pharmacokinetics Using Physiologically Based Pharmacokinetic Model for Early Drug Discovery of Tyrosine Kinase Inhibitors[J].Eur J Drug Metab Pharmacokinet，2019，44（1）：121-132.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
