Assessment of thyroid cancer risk associated with glucagon-like peptide 1 receptor agonist use.
Vilsbøll T, Stellfeld M, Aroda VR, Dandanell S, David JP, Kristiansen CTP, Rasmussen S, Roberts FL, Hegedüs L
Key Finding
Massive analysis of 101,732 people taking GLP-1 drugs like Ozempic and Victoza found no increased thyroid cancer risk, with only 1 case per 10,000 patient-years reported in real-world surveillance.
What This Study Found
Statistics Decoded
Why This Matters
This should reassure the millions of people using GLP-1 drugs for diabetes and weight loss that thyroid cancer isn't a realistic concern, with massive real-world evidence showing no increased risk despite early theoretical worries from animal studies.
Original Abstract
Published literature has raised concerns regarding a causal association between glucagon-like peptide 1 receptor agonist (GLP-1RA) use and thyroid cancer risk in adults. In this analysis, we evaluated the association between thyroid cancer risk and GLP-1RA use. Our evaluation included data from (i) 93 liraglutide or semaglutide phase 2 and 3 clinical trials, of which six were cardiovascular outcome trials (CVOTs); (ii) post-marketing surveillance from the Novo Nordisk safety database (evaluating data for liraglutide and semaglutide); and (iii) the US Merative™ MarketScan® Commercial Database. Hazard ratios (HRs) were estimated for overall thyroid cancer risk for treatment groups in the clinical trials and for individuals with type 2 diabetes treated with any GLP-1RAs versus sodium-glucose cotransporter-2 inhibitors (SGLT2is) for the real-world data. Data from all clinical trials across 101 732 participants (totalling 206 950 patient-years of exposure [PYE]) reported low numbers of thyroid cancer events; across all trials, HRs were 1.70 (95% confidence interval [CI] 0.99, 3.03) and 1.83 (0.70, 6.71) for pooled GLP-1RA versus pooled placebo and active comparator, respectively. For CVOTs, the HR (95% CI) for pooled GLP-1RA versus pooled placebo was 1.41 (0.72, 2.81). Post-marketing surveillance data showed a low thyroid cancer reporting rate of 0.001 cases/100 PYE and did not support an association between liraglutide or semaglutide exposure and the number of thyroid cancer events. Analysis of the US Merative™ MarketScan® Commercial Database reported an HR of 0.87 (95% CI 0.58, 1.29) for the occurrence of thyroid cancer in individuals using any GLP-1RAs versus SGLT2is. The totality of data analysed did not suggest an association between liraglutide or semaglutide use and thyroid cancer risk in adults.