目的 挖掘并比较第1代FMS样酪氨酸激酶3（FLT3）抑制剂米哚妥林与第2代FLT3抑制剂吉瑞替尼、奎扎替尼的上市后不良事件信号。方法 基于美国食品药品管理局不良事件报告系统（FAERS）数据库，通过OpenVigil 2.1平台检索2017年第1季度—2025年第3季度以米哚妥林、吉瑞替尼、奎扎替尼为首要怀疑药物的不良事件报告；并从FDA不良事件监测系统（AEMS）下载2025年第4季度—2026年第2季度的原始数据作为增量扩展分析。采用《监管活动医学词典》（MedDRA）28.1版对不良事件术语进行标准化处理，利用比例报告比法（PRR）与报告比值比法（ROR）联合进行信号检测，经混杂因素排除后确认阳性信号，对3种FLT3抑制剂不良事件信号的分布特点及强度差异进行比较分析，并通过增量数据验证主要结论的稳健性。结果 主分析共纳入有效ADE报告6 022份，其中米哚妥林2 005份、吉瑞替尼3 625份、奎扎替尼392份。经ROR-PRR联合检测与混杂因素排除后，最终确认米哚妥林阳性不良事件信号107个、吉瑞替尼141个、奎扎替尼16个。增量分析另行纳入不良事件报告665份（米哚妥林53份、吉瑞替尼396份、奎扎替尼216份），检出阳性信号11、51、25个，关键信号（吉瑞替尼分化综合征、肝功能异常；奎扎替尼QT间期延长；米哚妥林发热性中性粒细胞减少）在增量窗口内均稳健检出且强度排序未变，主要结论稳健。3种FLT3抑制剂检出的不良事件信号均以血液系统及感染类事件为主，但不同药物的不良事件信号谱存在差异：米哚妥林因多靶点抑制KIT/PDGFR激酶使其胃肠道反应类信号更为突出，吉瑞替尼因FLT3/AXL双重抑制作用观察到较突出的肝胆系统毒性及分化综合征信号，奎扎替尼对心肌Ikr电流的作用使其QT间期延长信号的强度相对更高。结论 第1代FLT3抑制剂米哚妥林与第2代FLT3抑制剂吉瑞替尼和奎扎替尼的安全性特征既存在共性也有差异，为FLT3抑制剂上市后安全性监测提供信号层面的参考。

Objective To identify and compare the post-marketing adverse event signals of the first FLT3 inhibitor midostaurin with the second-generation FLT3 inhibitors gilteritinib, and quizartinib. Methods Based on the FAERS, through the OpenVigil 2.1 platform, adverse event reports for midostaurin, gilteritinib, and quizartinib which were the primary suspected drugs, were retrieved from the first quarter of 2017 to the third quarter of 2025. The original data from the fourth quarter of 2025 to the second quarter of 2026 were downloaded from the FDA adverse event monitoring system (AEMS) as an incremental extension analysis. The adverse event terms were standardized using the MedDRA 28.1 version. The PRR and ROR methods were combined for signal detection. Positive signals were confirmed after excluding confounding factors. The distribution characteristics and intensity differences of the adverse event signals of the three FLT3 inhibitors were compared and analyzed. The robustness of the main conclusions was verified through incremental data. Results The main analysis included 6 022 valid adverse drug event reports, among which 2 005 were for midostaurin, 3 625 for gilteritinib, and 392 for quizartinib. After combined ROR-PRR detection and exclusion of confounding factors, it was finally confirmed that there were 107 positive adverse event signals for midostaurin, 141 for gilteritinib, and 16 for quizartinib. The incremental analysis additionally included 665 adverse event reports (53 for midostaurin, 396 for gilteritinib, and 216 for quizartinib), and detected 11, 51, and 25 positive signals respectively. The key signals (differentiation syndrome of gilteritinib, abnormal liver function; QT interval prolongation of quizartinib; fever-induced neutropenia of midostaurin) were all robustly detected within the incremental window and the strength ranking remained unchanged. The adverse event signals detected by the three FLT3 inhibitors were mainly related to the blood system and infections, but the spectra of adverse event signals for different drugs were different: midostaurin, due to its multi-target inhibition of KIT/PDGFR kinases, made the gastrointestinal reaction signals more prominent; gilteritinib, due to its dual inhibition of FLT3/AXL, observed more prominent hepatobiliary system toxicity and differentiation syndrome signals; quizartinib, due to its effect on the Ikr current of the myocardium, made the QT interval prolongation signal relatively stronger. Conclusions Safety profiles of the first-generation FLT3 inhibitor midostaurin, and the second-generation agents gilteritinib and quizartinib share commonalities as well as differences. The findings of this study may serve as a signal-level reference for post-marketing safety surveillance of FLT3 inhibitors.

R973

国家自然科学基金青年项目（82304279）