13种人类恶性肿瘤已被确定为肥胖相关癌症（OAC），即体内脂肪过多与这些特定肿瘤患者患癌症的风险增加和预后较差有关。胰高血糖素样肽受体激动剂 (GLP-1RA) 类药物是治疗 2 型糖尿病 (T2D) 和实现减肥的有效药物，但 GLP-1RA 与 13 种 OAC 事件风险之间的关系尚不清楚比较服用 GLP-1RA 与胰岛素或二甲双胍的 T2D 患者中 13 种 OAC 的事件风险。这项回顾性队列研究基于包含 1.13 亿美国患者的全国多中心电子健康记录 (EHR) 数据库。研究人群包括 1651452 名既往没有 OAC 诊断的 T2D 患者，他们在 2005 年 3 月至 2018 年 11 月期间服用了 GLP-1RA、胰岛素或二甲双胍。数据分析于 2024 年 4 月 26 日进行。GLP-1RA 的处方使用 Cox 比例风险和 Kaplan-Meier 生存分析并应用审查，对暴露后 15 年随访期间发生的 13 个 OAC 中的每一个进行事件（首次）诊断。计算了危险比 (HR)、累积发生率和 95% CI。所有模型均通过倾向评分匹配基线协变量对基线时的混杂因素进行了调整。 在 1651452 名 T2D 患者的研究人群中（平均 [SD] 年龄为 59.8 [15.1] 岁；827873 名 [50.1%] 为男性，775687 名患者为男性） [47.0%] 女性参与者；5780 [0.4%] 美洲印第安人或阿拉斯加原住民，65893 [4.0%] 亚洲人，281242 [17.0%] 黑人，13707 [0.8%] 夏威夷原住民或其他太平洋岛民，以及 1000 780 [60.6%] 白人参与者），与胰岛素相比，GLP-1RA 与 13 种 OAC 中的 10 种风险显着降低相关，包括胆囊癌（HR，0.35；95% CI，0.15-0.83）、脑膜瘤（HR， 0.37；95% CI，0.18-0.74）、胰腺癌（HR，0.41；95% CI，0.33-0.50）、肝细胞癌（HR，0.47；95% CI，0.36-0.61）、卵巢癌（HR，0.52； 95% CI，0.03-0.74）、结直肠癌（HR，0.54；95% CI，0.46-0.64）、多发性骨髓瘤（HR，0.59；95% CI，0.44-0.77）、食道癌（HR，0.60；95% CI，0.42-0.86）、子宫内膜癌（HR，0.74；95% CI，0.60-0.91）和肾癌（HR，0.76；95% CI，0.60-0.91）。 95% CI，0.64-0.91）。尽管没有统计学意义，但与服用胰岛素的患者相比，服用 GLP-1RA 的患者患胃癌的 HR 小于 1（HR，0.73；95% CI，0.51-1.03）。 GLP-1RA 与绝经后乳腺癌或甲状腺癌风险降低无关。在那些显示服用 GLP-1RA 的患者与服用胰岛素的患者相比风险降低的癌症中，服用 GLP-1RA 的患者与服用二甲双胍的结直肠癌和胆囊癌患者的 HR 小于 1，但风险降低并不具有统计学意义。与二甲双胍相比，GLP-1RA 与任何癌症风险降低无关，但与肾癌风险增加相关（HR，1.54；95% CI，1.27-1.87）。在本研究中，GLP-1RA与 T2D 患者中使用胰岛素或二甲双胍相比，特定类型 OAC 的风险较低。这些发现提供了 GLP-1RA 对高危人群癌症预防的潜在益处的初步证据，并支持预防某些 OAC 的进一步临床前和临床研究。

Thirteen human malignant neoplasms have been identified as obesity-associated cancers (OACs), ie, the presence of excess body fat is associated with increased risk of developing cancer and worse prognosis in patients with these specific tumors. The glucagon-like peptide receptor agonist (GLP-1RA) class of pharmaceuticals are effective agents for the treatment of type 2 diabetes (T2D) and for achieving weight loss, but the association of GLP-1RAs with the incident risk of 13 OACs is unclear.To compare the incident risk of each of the 13 OACs in patients with T2D who were prescribed GLP-1RAs vs insulins or metformin.This retrospective cohort study was based on a nationwide multicenter database of electronic health records (EHRs) of 113 million US patients. The study population included 1 651 452 patients with T2D who had no prior diagnosis of OACs and were prescribed GLP-1RAs, insulins, or metformin during March 2005 to November 2018. Data analysis was conducted on April 26, 2024.Prescription of GLP-1RAs, insulins, or metformin.Incident (first-time) diagnosis of each of the 13 OACs occurring during a 15-year follow-up after the exposure was examined using Cox proportional hazard and Kaplan-Meier survival analyses with censoring applied. Hazard ratios (HRs), cumulative incidences, and 95% CIs were calculated. All models were adjusted for confounders at baseline by propensity-score matching baseline covariates.In the study population of 1 651 452 patients with T2D (mean [SD] age, 59.8 [15.1] years; 827 873 [50.1%] male and 775 687 [47.0%] female participants; 5780 [0.4%] American Indian or Alaska Native, 65 893 [4.0%] Asian, 281 242 [17.0%] Black, 13 707 [0.8%] Native Hawaiian or Other Pacific Islander, and 1 000 780 [60.6%] White participants), GLP-1RAs compared with insulin were associated with a significant risk reduction in 10 of 13 OACs, including in gallbladder cancer (HR, 0.35; 95% CI, 0.15-0.83), meningioma (HR, 0.37; 95% CI, 0.18-0.74), pancreatic cancer (HR, 0.41; 95% CI, 0.33-0.50), hepatocellular carcinoma (HR, 0.47; 95% CI, 0.36-0.61), ovarian cancer (HR, 0.52; 95% CI, 0.03-0.74), colorectal cancer (HR, 0.54; 95% CI, 0.46-0.64), multiple myeloma (HR, 0.59; 95% CI, 0.44-0.77), esophageal cancer (HR, 0.60; 95% CI, 0.42-0.86), endometrial cancer (HR, 0.74; 95% CI, 0.60-0.91), and kidney cancer (HR, 0.76; 95% CI, 0.64-0.91). Although not statistically significant, the HR for stomach cancer was less than 1 among patients who took GLP-1RAs compared with those who took insulin (HR, 0.73; 95% CI, 0.51-1.03). GLP-1RAs were not associated with a reduced risk of postmenopausal breast cancer or thyroid cancer. Of those cancers that showed a decreased risk among patients taking GLP-1RAs compared with those taking insulin, HRs for patients taking GLP-1RAs vs those taking metformin for colorectal and gallbladder cancer were less than 1, but the risk reduction was not statistically significant. Compared with metformin, GLP-1RAs were not associated with a decreased risk of any cancers, but were associated with an increased risk of kidney cancer (HR, 1.54; 95% CI, 1.27-1.87).In this study, GLP-1RAs were associated with lower risks of specific types of OACs compared with insulins or metformin in patients with T2D. These findings provide preliminary evidence of the potential benefit of GLP-1RAs for cancer prevention in high-risk populations and support further preclinical and clinical studies for the prevention of certain OACs.