瑞戈非尼的安全警戒信号检测与评价

目的:运用数据挖掘的方法检测瑞戈非尼的不良反应信号,为临床提供用药参考。方法:采用比例报告比值法(PRR)、报告比值比法(ROR)、英国药品和保健产品管理局的综合标准法(MHRA)和贝叶斯可信区间递进神经网络法(BCPNN)4种数据挖掘方法,对美国食品药品监督管理局(FDA)的不良事件报告系统(FAERS)中2012年第四季度-2018年第三季度共24个季度的数据进行关于瑞戈非尼不良反应的信号检测。结果:4种检测方法检测出重叠的关于瑞戈非尼的不良反应信号136个,存在药品说明书未涵盖的新不良反应信号102个。应特别关注涉及新ADR信号较多且信号中等的肝胆系统。另外,说明书未提及其有生殖系统相关的不良反应,此次检测中有4个信号。结论:通过数据挖掘的信号检测方法,发现瑞戈非尼药品说明书未涉及的新的不良反应信号102个,应引起临床重视。

Signals detection and evaluation of adverse reaction induced by regorafenib

OBJECTIVE To detect the adverse drug reaction (ADR) of regorafenib by data mining and provide reference for clinical medication. METHODS Signals related to adverse effects of regorafenib were detected in 24 quarters of data from the Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) from the fourth quarter of 2012 to the third quarter of 2018 using four data mining methods:proportional reporting ratio (PRR), reporting odds ratio (ROR), United Kingdom Medicines and Healthcare Products Regulatory Agency omnibus standard method (MHRA), and the Bayesian Confidence Propagation Neural Network (BCPNN). RESULTS The four detection methods detected 136 overlapping adverse reaction signals concerning regorafenib and 102 new adverse reaction signals were not covered in the package insert. Special attention should be paid to the hepatobiliary system, which is involved in more and moderate signals of new ADRs. In addition, there were four new ADR signals of reproductive system which were not listed in the package insert. CONCLUSION 102 new adverse reaction signals which were not listed in the package insert of regorafenib were found by the signal detection method of data mining, which should be paid attention to in clinical practice.