Balachandra Sanjana, Syed Rohma, Song Zhixing, Kasmirski Julia, Gillis Andrea, Fazendin Jessica, Lindeman Brenessa, Chen Herbert
Section of Endocrine Surgery, Department of Surgery, University of Alabama at Birmingham, Birmingham, Alabama.
Section of Endocrine Surgery, Department of Surgery, University of Alabama at Birmingham, Birmingham, Alabama.
J Surg Res. 2025 Aug;312:104-110. doi: 10.1016/j.jss.2025.05.016. Epub 2025 Jun 27.
Thyroid nodules are common and require vigilant monitoring due to malignancy potential. They become particularly concerning in patients using glucagon-like peptide (GLP-1) analogs, which are associated with a proposed increased risk of medullary thyroid cancer. This study aims to evaluate the incidence of thyroid cancer in patients with thyroid nodules treated with GLP-1 analogues.
We conducted a retrospective cohort study using the TriNetX database looking at pediatric and adult patients diagnosed with thyroid nodules (ICD-10-CM E04) and treated with GLP-1 analogues (ATC A10BJ) from 1995 to 2024. In addition, we compared the GLP-1 cohort to a cohort of patients who developed thyroid nodules after starting metformin. Demographic data and biochemical markers were assessed. The primary outcome was the incidence of thyroid cancer, identified by the first occurrence of ICD-10-CM C73 following the diagnosis of thyroid nodules and the start of GLP-1 analogues. Descriptive statistics summarized baseline characteristics, and Kaplan-Meier survival analysis estimated the cumulative incidence of thyroid cancer. Analyses were performed using the TriNetX Analytics platform, with statistical significance defined as P < 0.05.
We identified 1,401,568 patients using GLP-1 analogues and 2,779,340 patients with thyroid nodules in our database. Among these, 171,460 patients had both conditions, with 98,142 (57%) developing thyroid nodules after starting GLP-1 analogues. The average age of the GLP-1 cohort was 60 ± 13 years, with 72% (n = 66,195) being female and 66% (n = 60,855) identifying as White. Kaplan-Meier analysis indicated that the survival probability, or the likelihood of not developing thyroid cancer by the end of the study, was 91.042% for the GLP-1 cohort, with 4687 cases of thyroid cancer observed. In comparison, the metformin cohort (n = 306,114) had a higher survival probability of 94%, with 11,898 cases of thyroid cancer observed. The risk ratio of 0.99 (95% confidence interval: 0.96-1.03) between the cohorts indicates no significant difference in the risk of developing thyroid cancer for patients on GLP-1 analogues compared to those on metformin.
Our study indicates a relatively low incidence of thyroid cancer among patients with thyroid nodules treated with GLP-1 analogues.
甲状腺结节很常见，由于存在恶变可能，需要进行密切监测。对于使用胰高血糖素样肽（GLP-1）类似物的患者，甲状腺结节尤其令人担忧，因为这类药物被认为会增加甲状腺髓样癌的发病风险。本研究旨在评估接受GLP-1类似物治疗的甲状腺结节患者中甲状腺癌的发病率。
我们使用TriNetX数据库进行了一项回顾性队列研究，观察1995年至2024年期间诊断为甲状腺结节（ICD-10-CM E04）并接受GLP-1类似物（ATC A10BJ）治疗的儿童和成人患者。此外，我们将GLP-1队列与开始使用二甲双胍后出现甲状腺结节的患者队列进行了比较。评估了人口统计学数据和生化指标。主要结局是甲状腺癌的发病率，通过甲状腺结节诊断后首次出现ICD-10-CM C73以及开始使用GLP-1类似物来确定。描述性统计总结了基线特征，Kaplan-Meier生存分析估计了甲状腺癌的累积发病率。使用TriNetX分析平台进行分析，统计学显著性定义为P < 0.05。
我们在数据库中识别出1,401,568名使用GLP-1类似物的患者和2,779,340名甲状腺结节患者。其中，171,460名患者同时患有这两种疾病，98,142名（57%）在开始使用GLP-1类似物后出现甲状腺结节。GLP-1队列的平均年龄为60±13岁，72%（n = 66,195）为女性，66%（n = 从开始使用GLP-1类似物到研究结束时未发生甲状腺癌的概率，即生存概率，GLP-1队列为91.042%，观察到4687例甲状腺癌病例。相比之下，二甲双胍队列（n = 306,114）的生存概率更高，为94%，观察到11,898例甲状腺癌病例。队列之间的风险比为0.99（95%置信区间：0.96 - 1.03），表明与使用二甲双胍的患者相比，使用GLP-1类似物的患者发生甲状腺癌的风险没有显著差异。
我们的研究表明，接受GLP-1类似物治疗的甲状腺结节患者中甲状腺癌的发病率相对较低
