目的 基于3大数据库，通过深度挖掘大规模真实世界上市后监测20年的数据，剖析胰高血糖素样肽-1受体激动剂（GLP-1RAs）中7种不同药物与骨骼及关节不良事件之间的相关性，为临床用药安全、药物监管决策及后续研究提供高效、可靠的初步证据。方法 挖掘和提取截至2025年第2季度的美国食品药品监督管理局（FDA）不良事件报告系统（FAERS）、日本药物不良事件报告数据库（JADER）和加拿大警戒不良反应数据库（CVARD）中与GLP-1RAs相关骨骼与关节不良事件报告。应用报告比值比（ROR）、比例报告比值比法（PRR）和信息成分法（IC）检测骨骼与关节不良事件阳性信号，并分析GLP-1RAs不同药物对其信号的影响。结果 共提取到GLP-1RAs相关骨骼与关节不良事件报告5 511例（FAERS数据库5 475例，JADER数据库7例，CVARD数据库29例），按照国际医学用语词典（MedDRA）共提取出65个骨骼与关节相关首选术语（PT）。检测报告显示，在FAERS数据库中，GLP-1RAs类药物中的艾塞那肽、利拉鲁肽、利司那肽、度拉糖肽、司美格鲁肽、替尔泊肽与阿必鲁肽7类药物，大多表现出无显著相关性，仅有利拉鲁肽与神经病性关节病、司美格鲁肽与肩骨折及椎间盘损伤存在风险信号。在JADER和CVARD数据库中，这7类药物信号皆为阴性，仅是司美格鲁肽与骨折、利拉鲁肽与肢体疼痛也呈现出阳性信号。但由于这些阳性信号上报病例较少，难以评估GLP-1RAs与骨骼及关节不良事件之间是否存在关联。结合火山图的结果分析表明，与GLP-1RAs相关联的PT虽包括关节痛、类风湿关节炎、关节肿胀和骨坏死，但与信号检测的PT不一致，证实无统计学显著性的风险信号。结论 从整体分析层面看，GLP-1RAs与骨骼及关节不良事件无显著关联性，但需重点关注的是，应对高风险人群（如高龄、绝经女性，以及有骨折或骨质疏松症病史）开展个体化评估，定期监测其骨密度及关节症状，平衡GLP-1RAs的降糖、减重方面的获益与骨骼关节安全风险，为患者的用药决策提供参考依据。

Objective Based on three major databases, this study aims to analyze the associations between seven different glucagonlike peptide-1 receptor agonists(GLP-1 RAs) bone and joint adverse events by in-depth mining of large-scale real-world post-marketing surveillance data spanning 20 years. It is intended to provide efficient and reliable preliminary evidence for clinical medication safety, drug regulatory decisions, and subsequent research. Methods Reports of GLP-1 RAs related bone and joint adverse events were extracted from the U.S. Food and Drug Administration(FDA) Adverse Event Reporting System(FAERS), Japanese Adverse Drug Event Report Database(JADER), and Canadian Vigilance Adverse Reaction Database(CVARD) up to the second quarter of 2025. Positive signals of bone and joint adverse events were detected using the reporting odds ratio(ROR), proportional reporting ratio(PRR), and Information Component(IC) methods, and the effects of different GLP-1 RAs on these signals were analyzed. Results A total of 5 511 reports of GLP-1 RA-related musculoskeletal and joint adverse events were retrieved(5 475 from FAERS, seven from JADER, and 29 from CVARD), corresponding to 65 preferred terms(PTs) related to bones and joints according to the Medical Dictionary for Regulatory Activities(MedDRA). Disproportionality analysis revealed that among the seven GLP-1 RAs evaluated(exenatide, liraglutide, lixisenatide, dulaglutide, semaglutide, tirzepatide, and albiglutide), most did not exhibit significant associations in the FAERS database. Significant risk signals were only detected for Liraglutide associated with neuropathic arthropathy, and for Semaglutide associated with shoulder fracture and intervertebral disc injury. In the JADER and CVARD databases, no positive signals were identified for the seven drugs, with the exception of Semaglutide associated with fracture and Liraglutide associated with limb pain. However, due to the limited number of reported cases for these positive signals, it is difficult to definitively assess the potential association between GLP-1 RAs and musculoskeletal adverse events. Supplementary analysis of the volcano plots indicated that while GLP-1 RAs associated PTs included joint pain, rheumatoid arthritis, joint swelling, and osteonecrosis, these specific PTs did not coincide with those identified by the disproportionality analysis, confirming the absence of statistically significant risk signals for these conditions. Conclusion Overall, our analysis suggests that GLP-1 RAs are not significantly associated with musculoskeletal and joint adverse events. However, it is crucial to emphasize the need for individualized risk assessment in high-risk populations, such as the elderly, postmenopausal women, and patients with a history of fractures or osteoporosis. Regular monitoring of bone mineral density and joint symptoms is recommended to balance the therapeutic benefits of GLP-1 RAs(e.g., glycemic control and weight reduction) against potential musculoskeletal risks, thereby providing a solid evidence base for clinical decision-making.

R977

国家自然科学基金青年科学基金项目(82000842); 广东省医学科研基金项目(A2024138,B2025580)