Ann Intern MedSide Effects & SafetyMeta-AnalysisDecember 9, 2025

Risk for Cancer With Glucagon-Like Peptide-1 Receptor Agonists and Dual Agonists : A Systematic Review and Meta-analysis.

Ko A, Chang YC, Bahar F, Wang TH, Xanthavanij N, Yu CC, Hsieh RJ, See XY, Lo SW, Song J, Hsia YP, Chiang CH, Xu X, Lin S, Chiang CH

View on PubMed DOI: 10.7326/ANNALS-25-02237

Key Finding

A massive analysis of nearly 94,245 people found that GLP-1 medications like Ozempic and Mounjaro don't meaningfully increase or decrease cancer risk, with thyroid cancer odds only 37% higher (but this could easily be due to chance).

What This Study Found

Think of this study as the ultimate safety audit - researchers pooled together 48 different trials involving nearly 100,000 people to see if GLP-1 medications cause cancer. It's like having 48 different security cameras all watching the same intersection to see if accidents really happen more often there. The results are reassuring: these medications appear to be cancer-neutral. For thyroid cancer (the biggest concern), the odds were 37% higher, but the confidence interval was so wide it included 'no effect' - like saying a basketball player's shooting percentage is somewhere between terrible and excellent. The researchers looked at 13 different types of cancer, including the big scary ones like pancreatic, breast, and colorectal cancer. Across the board, the story was the same: no meaningful increase or decrease in risk. It's like checking if a new traffic light causes more accidents and finding that sometimes there are 9 fewer crashes per 10,000 cars, sometimes 6 more - essentially no real change.

Statistics Decoded

OR 1.37 for thyroid cancer means 37% higher odds, but the confidence interval (0.82 to 2.31) crosses 1.0, meaning this could easily be random chance - like flipping a coin and getting heads 6 times out of 10 instead of 5. The '1 fewer to 9 more cases per 10,000 patients' puts this in perspective - even if real, we're talking about tiny absolute numbers. For pancreatic cancer, OR 0.84 actually suggests 16% lower odds, with a range of 9 fewer to 6 more cases per 10,000 people. Moderate certainty means the researchers are reasonably confident but acknowledge some uncertainty remains - like being 75% sure instead of 95% sure.

Why This Matters

This analysis should calm fears about GLP-1 medications causing cancer, addressing one of the biggest patient concerns about these increasingly popular weight loss and diabetes drugs. However, the researchers emphasize that most trials were short-term, so we still need longer studies to be completely sure about cancer risks that might take years to develop.

Original Abstract

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are used for type 2 diabetes mellitus (T2DM) and overweight or obesity, but their association with cancer is unclear. To investigate the risk for obesity-related cancer associated with GLP-1RAs. PubMed, Embase, Web of Science, Scopus, and the Cochrane Central Register of Controlled Trials from inception to August 2025. Randomized placebo-controlled trials reporting any of the following cancer outcomes: thyroid, pancreatic, colorectal, gastric, esophageal, liver, gallbladder, breast, ovarian, endometrial, or kidney cancer; multiple myeloma; or meningioma. Risk of bias was evaluated using the Cochrane Risk of Bias 2 tool, and certainty of evidence was assessed using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach. Odds ratios (ORs) were pooled using random-effects meta-analysis. The review included 48 trials involving 94 245 participants. GLP-1RAs probably have little or no effect on risk for thyroid cancer (OR, 1.37 [95% CI, 0.82 to 2.31]; 1 fewer to 9 more cases per 10 000 patients treated), pancreatic cancer (OR, 0.84 [CI, 0.53 to 1.35]; 9 fewer to 6 more per 10 000), breast cancer (OR, 0.95 [CI, 0.60 to 1.49]; 10 fewer to 12 more per 10 000), or kidney cancer (OR, 1.12 [CI, 0.78 to 1.60]; 5 fewer to 13 more per 10 000) (moderate certainty). GLP-1RAs may have little or no effect on colorectal, esophageal, liver, gallbladder, ovarian, or endometrial cancer; multiple myeloma; or meningioma (low certainty). The effect on gastric cancer is very uncertain. Results were consistent in sensitivity analyses of trials with low risk of bias and studies of semaglutide or tirzepatide and across subgroups stratified by follow-up duration, population, GLP-1RA class, weight loss profile, dose, and duration of action. The included trials were not designed to evaluate cancer outcomes and had short follow-up. GLP-1RAs may have little or no effect on risk for obesity-related cancers. Longer-term studies are needed to clarify potential risks or benefits. None. (PROSPERO: CRD42024608365).