数学模型预测药源性肝损伤研究进展  

Advances in research on drug-induced liver injury predicted by mathematical models

作　　者：李敏 李思泽 姚莉 相小强[1] LI Min;LI Si-ze;YAO Li;XIANG Xiao-qiang(School of Pharmacy,Fudan University,Shanghai 201203,China)

机构地区：[1]复旦大学药学院,上海201203

出　　处：《中国药理学与毒理学杂志》2021年第5期382-390,共9页Chinese Journal of Pharmacology and Toxicology

基　　金：国家自然科学基金(81473409);上海市2018年度“科技创新行动计划”实验动物研究领域项目(18140900900);复旦大学附属中山医院闵行分院-复旦大学药学院战略合作自主研究课题融合基金(RO-MY201710)。

摘　　要：药源性肝损伤作为引起上市药物退市的主要原因之一,已为新药研发和药物临床使用过程中重点考察的安全性问题。随着信息技术的快速发展,数学模型也开始逐渐用于化合物肝毒性的预测和判断,且取得了较好的效果。药物是否产生肝毒性与药物的化学结构、体内浓度、代谢过程等多个因素有密切关系。根据药物肝毒性原理,多种数学模型被开发和应用于肝毒性预测中,常用方法包括基于定量构效关系(QSAR)模型、毒理基因组学模型和生理药动学(PBPK)模型的方法,该3种模型分别以化合物结构、毒理基因组相关信息和PBPK参数作为基础用于预测药物性肝毒性。本文介绍该3类预测药源性肝损伤方法的构建过程和应用,以期为药源性肝损伤的早期预测和肝毒性药物临床合理应用提供参考。As one of the main causes of drug withdrawal,drug-induced liver injury has become an big safety concern in drug development and clinical use.With the rapid development of information technology,mathematical models have been gradually used to predict and evaluate the hepatotoxicity of compounds,and proven to be effective.The hepatotoxicity of drugs is closely related to their chemical structure,body concentrations and metabolism.Based on the mechanism of drug-induced liver injury,various mathematical models have been developed and applied to the prediction of hepatotoxicity,including the quantitative structure-activity relationship(QSAR)model,computational toxicogenomics model,and physiologically-based pharmacokinetic(PBPK)model.These models make use of compound structures,toxicogenomics data,and PBPK parameters to predict drug hepatotoxicity,respectively.This review introduces these models,in order to provide data for early prediction of drug-induced liver injury and rational clinical application of hepatotoxic drugs.

关 键 词：药源性肝损伤 预测 量化构效关系模型 毒理基因组学 生理药动学模型 DILIsym模型 

分 类 号：R965.1[医药卫生—药理学]