graph chart displayed on laptop screen

Do Weight Loss Drugs Lower Cancer Risk?

A large observational study found a 41% lower relative risk of obesity-associated cancer diagnoses in adults on GLP-1 drugs. Its design cannot show cause.

A large observational study found about 41% fewer obesity-associated cancer diagnoses in adults prescribed GLP-1 drugs — an association that cannot be called prevention

In a large observational study, obese, nondiabetic US adults prescribed GLP-1 weight-loss drugs were about 41% less likely to be diagnosed with one of 13 obesity-associated cancers over a median two years than matched adults who received diet and exercise counselling. The study cannot show the drugs caused the difference, and randomized trials of these medicines have not found a clear effect on cancer risk.

What this means

In a large observational study, obese, nondiabetic US adults prescribed GLP-1 weight-loss drugs were diagnosed with one of 13 obesity-associated cancers about 41% less often over a median two years than matched adults who received diet and exercise counseling. That is an association, not proof of prevention: who is prescribed a drug differs from who is counseled. A meta-analysis of 48 randomized trials found little or no effect on cancer risk, and the abstract gives no absolute rates, so how many diagnoses were avoided is unknown. Nothing here is a reason to start, stop, or stay on a medication.

What the study measured

The data came from TriNetX, a US electronic health-records database of 113 million patients. Inside a target-trial-emulation framework, researchers identified 229,467 obese (BMI 30 or above), nondiabetic adults with no prior obesity-associated cancer, December 2014 to June 2025, and matched each GLP-1 receptor agonist user one-to-one with a patient given diet or exercise counselling: 161,798 adults in the matched analysis, 80,899 per group, mean age 47.2.

Obesity-associated cancer was diagnosed 41% less often in the GLP-1 group: hazard ratio 0.59 (95% CI 0.53 to 0.67). A weighted analysis confirmed it. The abstract reports lower incidence in every subgroup examined except Black race, without individual effect sizes.

Why 41% less is not prevention

The groups were not assigned at random. Matching balances only what researchers can measure; who gets prescribed a drug versus counseling differs in ways records do not capture. The authors state that prospective trials are needed to confirm causality. The abstract gives no absolute cancer incidence, so the number of diagnoses avoided cannot be calculated, and the authors call the finding short-term.

What randomized trials found

Randomized trials point the other way. A 2025 meta-analysis of 48 placebo-controlled trials, 94,245 participants, concluded that GLP-1 receptor agonists “may have little or no effect” on obesity-related cancer risk, at moderate-to-low certainty. No randomized trial has tested cancer incidence as an endpoint.

Observational studies keep agreeing. A 2026 meta-analysis of 24 studies (about 3.96 million patients) pooled a 30% lower relative risk (RR 0.70, 95% CI 0.54 to 0.89). A 2025 target-trial emulation in 86,632 matched US adults found HR 0.83 (95% CI 0.76 to 0.91), plus a possible increased kidney-cancer signal, HR 1.38 (95% CI 0.99 to 1.93), not significant. A 2024 TriNetX cohort in type 2 diabetes found lower risk than insulin for 10 of 13 obesity-associated cancers. The association reproduces; a cause does not follow from these designs.

Funding and the data source

The authors declared no funding and no competing interests. The data come from TriNetX, a commercial health-data company; no author is disclosed as its employee.

A separate GLP-1 question

Whether the drugs keep weight off after people stop is a separate, live question — see our explainer What Happens When People Stop Weight-Loss Medications. Our companion piece on how to keep weight off after GLP-1 ends covers that question.

Terms explained

  • Target-trial emulation — analysing observational records as if running the randomized trial researchers would design if they could, by first specifying who would be eligible and what would be compared. It narrows some biases but does not remove unmeasured differences the way randomization does.
  • Propensity-score matching — pairing each treated person with a comparison person who had a similar estimated chance of being treated, based on recorded characteristics. It balances only the factors that were recorded.
  • Hazard ratio (HR) — a comparison of how often an outcome occurs over time in two groups. An HR of 0.59 means the outcome occurred about 41% less often in the GLP-1 group.
  • Obesity-associated cancers — the 13 cancers the study grouped together because obesity is linked to them.

Keep reading on Nutritious News: Does WHO Recommend Weight Loss Drugs? covers WHO’s guideline on GLP-1 and related drugs for adult obesity.

Keep reading on Nutritious News: Do Diet Drinks Work as Well as Water for Weight Loss?.

Sources

  1. Primary source: Hsu AH, Ramirez PT, Chang YH, Chiang YC, Santia MC, Meschini T, Mateo-Kubach P, Suri A, Kamat AA. “GLP-1 receptor agonist use and cancer risk in obese nondiabetic adults.” Annals of Oncology 2026;37(8):1178-1185, online 2026-06-07. https://doi.org/10.1016/j.annonc.2026.04.013 (doi:10.1016/j.annonc.2026.04.013; PMID: 42252247). Abstract-level access; the full paper was not available to us at press time.
  2. Context (the randomized counterweight): “Risk for Cancer With Glucagon-Like Peptide-1 Receptor Agonists and Dual Agonists: A Systematic Review and Meta-analysis.” Annals of Internal Medicine 2025. https://doi.org/10.7326/annals-25-02237 (doi:10.7326/annals-25-02237; PMID: 41359966). 48 placebo-controlled trials, 94,245 participants. Abstract-level access.
  3. Context (mixed-signal cohort): “GLP-1 Receptor Agonists and Cancer Risk in Adults With Obesity.” JAMA Oncology 2025. https://doi.org/10.1001/jamaoncol.2025.2681 (doi:10.1001/jamaoncol.2025.2681; PMID: 40839273). Abstract-level access.
  4. Context (pooled observational agreement): “Glucagon-like peptide-1 receptor agonists and the risk of obesity-related cancers: a systematic review and meta-analysis.” Diabetes Research and Clinical Practice 2026;234:113158. https://doi.org/10.1016/j.diabres.2026.113158 (doi:10.1016/j.diabres.2026.113158; PMID: 41722869). 24 studies, about 3.96 million patients. Abstract-level access.
  5. Context (antecedent cohort, type 2 diabetes): “Glucagon-Like Peptide 1 Receptor Agonists and 13 Obesity-Associated Cancers in Patients With Type 2 Diabetes.” JAMA Network Open 2024;7:e2421305. https://doi.org/10.1001/jamanetworkopen.2024.21305 (doi:10.1001/jamanetworkopen.2024.21305; PMID: 38967919). Abstract plus open-access PMC11227080. A journal erratum (JAMA Netw Open. 2025;8(6):e2519224) corrected an error in a figure; the reported results are unchanged.

Related Articles