基于FAERS数据库的索米妥昔单抗不良事件信号挖掘与分析

Signal mining and analysis of mirvetuximab soravtansine based on FAERS database

摘要:

目的 挖掘真实世界中索米妥昔单抗相关药物不良事件（ADE）信号，为临床合理安全用药提供参考。方法 调取美国食品药品监督管理局不良事件报告系统（FAERS）中索米妥昔单抗2004年第三季度至2024 年第三季度的数据，采用报告比值比法和贝叶斯可信区间递进神经网络法进行信号挖掘，分析其ADE发生情况。结果 共得到ADE信号26个，累及9个系统器官分类（SOC），合计报告355份。在已知性别的患者中，女性ADE报告数多于男性；在已知年龄的患者中，45岁以上占比居多；ADE报告主要集中在2023—2024年。累及的SOC主要包括眼器官疾病，各类神经系统疾病，呼吸系统、胸及纵隔疾病，胃肠系统疾病等；报告数较多的ADE信号包括肺部炎症、周围神经病、视物模糊、腹痛、角膜病变等。挖掘到6个药品说明书未记录的新发现可疑不良反应，包括肺纤维化、结肠炎、外渗、血氧饱和度降低等。结论 索米妥昔单抗在真实世界中的常见不良反应与药品说明书有一致性，但存在部分新发现的统计学关联性高的信号，临床用药时应予以重点关注。

Abstract:

Objective To explore the signals of adverse events (ADE) related to mirvetuximab soravtansine in the real world, and to provide reference for clinical medication safety.Methods Data from Q3, 2004 to Q3, 2024, in the US Food and Drug Administration Adverse Event Reporting System (FAERS) database were retrieved for disproportionality and Bayesian analysis. Reporting odds ratio method and Bayesian confidence interval progressive neural network method were used to analyze the occurrence of ADE.Results A total of 26 ADE signals were detected, involving 355 ADE reports, in relation to 9 system organ classes (SOCs). Among patients with documented gender, females reported more adverse drug events (ADEs) than males. In cases with available age data, the majority of ADEs occurred in individuals aged over 45 years. ADE reports were predominantly concentrated between 2023 and 2024. The predominantly involved SOCs included eye disorders, various nervous system disorders, respiratory/thoracic/mediastinal disorders, and gastrointestinal disorders. High-frequency ADE signals comprised pneumonitis, peripheral neuropathy, blurred vision, abdominal pain, and keratopathy. Six new suspected adverse reactions that were not recorded before were found, including pulmonary fibrosis, colitis, extravasation, oxygen saturaion decreased, etc.Conclusion The common adverse reactions of mirvetuximab soravtansine in the real world are consistent with the drug instructions, but there are some newly discovered signals with high statistical correlation. According to the principle of suspicious report, we should pay attention to such signals in clinical practice.