Hurwitz Sarah Ruth, Lanes Stephan, Quimbo Tracey, Papazian Anahit, White Jeff, Fisher Vicki, Cziraky Mark J, Crowley Matthew J, Willey Vincent J
Carelon Research, Wilmington, Delaware, USA.
CarelonRx, Morristown, New Jersey, USA.
Pharmacoepidemiol Drug Saf. 2025 Sep;34(9):e70214. doi: 10.1002/pds.70214.
This study assessed serious clinical outcomes comparing glucagon-like peptide 1 receptor agonists (GLP-1-RAs) with sodium glucose co-transporter 2 inhibitors (SGLT2-Is) in patients with type 2 diabetes (T2DM) and patients without diabetes using two chronic weight management (CWM) regimens.
We performed a new user, active comparator cohort study in a large, national U.S. claims database. Adults who initiated GLP-1-RAs, SGLT2-Is, naltrexone/bupropion (NalBup), or phentermine/topiramate (PhenTop) from 1 January 2016 to 31 December 2023 were included. Potential confounding was controlled using propensity score weighting for 82 clinical and demographic covariates, and risk ratios (RRs) were estimated.
This study included 330,684 GLP-1-RA users and 264,277 SGLT2-I users with T2DM. Among CWM patients without diabetes, we studied over 25,000 GLP-1-RA users, 5019 NalBup users, and 3841 PhenTop users. In both indications, GLP-1-RA users had higher rates of hospitalizations for gallbladder and biliary diseases with RRs ranging from 1.14 (95% CI: 1.06-1.22) in T2DM patients to 3.32 (95% CI: 1.44-7.64) in CWM patients. No reduction in the rate of cardiovascular events was observed for GLP-1-RA users with RRs ranging from 0.92 (95% CI: 0.37-2.25) in CWM patients to 1.03 (95% CI: 0.99-1.08) in T2DM patients. In T2DM patients, GLP-1-RA users had a lower rate of acute liver injury (RR: 0.76; 95% CI: 0.64-0.91).
This study corroborates an increased risk of hospitalization for gall bladder and biliary conditions among users of GLP-1-RAs and found similar rates as comparators of MI or stroke when GLP-1-RAs were used for T2DM or CWM. This real-world study complements placebo-controlled trials and can further inform prescribing decisions.
The study protocol was pre-registered at the Center for Open Science's Real-World Evidence Registry and is publicly accessible online (https://doi.org/10.17605/OSF.IO/PSY74).
本研究评估了在2型糖尿病（T2DM）患者和非糖尿病患者中，使用两种慢性体重管理（CWM）方案比较胰高血糖素样肽1受体激动剂（GLP-1-RAs）与钠-葡萄糖协同转运蛋白2抑制剂（SGLT2-Is）后的严重临床结局。
我们在美国一个大型全国性索赔数据库中进行了一项新用户、活性对照队列研究。纳入了2016年1月1日至2023年12月31日开始使用GLP-1-RAs、SGLT2-Is、纳曲酮/安非他酮（NalBup）或苯丁胺/托吡酯（PhenTop）的成年人。使用倾向得分加权法对82个临床和人口统计学协变量进行潜在混杂因素控制，并估计风险比（RRs）。
本研究纳入了330,684名T2DM的GLP-1-RA使用者和264,277名SGLT2-I使用者。在无糖尿病的CWM患者中，我们研究了超过25,000名GLP-1-RA使用者、5019名NalBup使用者和3841名PhenTop使用者。在这两种适应症中，GLP-1-RA使用者因胆囊和胆道疾病住院的发生率较高，RRs范围从T2DM患者中的1.14（95%CI：1.06-1.22）到CWM患者中的3.32（95%CI：1.44-7.64）。未观察到GLP-1-RA使用者心血管事件发生率降低，RRs范围从CWM患者中的0.92（95%CI：0.37-2.25）到T2DM患者中的1.03（95%CI：0.99-1.08）。在T2DM患者中，GLP-1-RA使用者急性肝损伤发生率较低（RR：0.76；95%CI：0.64-0.91）。
本研究证实了GLP-1-RAs使用者胆囊和胆道疾病住院风险增加，并发现当GLP-1-RAs用于T2DM或CWM时，心肌梗死或中风的发生率与对照药物相似。这项真实世界研究补充了安慰剂对照试验，并可为处方决策提供更多信息。
该研究方案已在开放科学中心的真实世界证据注册中心预先注册，可在网上公开获取（https://doi.org/10.17605/OSF.IO/PSY74）。
