The new semaglutide pancreatic cancer study found no increased risk compared with other diabetes medicines. That's reassuring, but it doesn't establish zero risk over a lifetime. The finding concerns adults with type 2 diabetes, and the time researchers could follow them was limited.
For someone worried by an Ozempic headline, the useful answer is narrower than “safe forever.” This study adds evidence to a prescribing conversation. It doesn't replace a review of someone's own pancreatic history.
What the semaglutide pancreatic cancer study asked
As of October 1, 2026, the day this article was written, the European Association for the Study of Diabetes's September 28 release reported new findings for its Milan meeting. Anton Pottegård of the University of Southern Denmark and colleagues investigated pancreatic cancer after people started semaglutide.
This was a post-authorization safety study, meaning safety research required after a medicine reaches the market. The EMA study catalogue, NN9535-4447, confirms regulators required it and Novo Nordisk, semaglutide's manufacturer, funded it. Both facts matter when judging the findings.
Researchers used national prescription, patient and cancer records. They matched people starting semaglutide with people starting other diabetes treatments at a similar treatment stage.
The ClinicalTrials.gov record, NCT04572165, names Ozempic and Rybelsus. Comparisons included sulfonylureas, pills that stimulate insulin release; SGLT-2 inhibitors, medicines that help the kidneys remove sugar; and insulin, the hormone that lowers blood sugar. Semaglutide acts like GLP-1, a gut hormone involved in appetite and blood sugar.
This was observational research: scientists watched existing treatment choices rather than assigning medicines by chance. Matching makes the groups more comparable, but cannot erase every difference between them.
The result: similar risk in all three countries
Among 97,464 semaglutide users, 131 cancers occurred versus 123 with comparators, across 167,399 versus 140,306 person-years. Person-years means the combined years people were followed.
A hazard ratio compares how quickly diagnoses occur between groups; 1 means no difference. A 95% confidence interval gives the estimate's uncertainty range.
| Country | Hazard ratio | 95% confidence interval |
|---|---|---|
| Denmark | 1.00 | 0.66–1.51 |
| Sweden | 0.82 | 0.56–1.22 |
| Norway | 0.91 | 0.55–1.51 |
| Combined | 0.91 | 0.71–1.16 |
The combined range crosses 1, allowing a decrease or a modest increase. Higher cumulative doses or longer treatment showed no increasing-risk pattern.
What this means for someone prescribed semaglutide
The useful message is that this large comparison did not uncover the feared increase. The raw case counts alone cannot answer the question, because the groups contributed different amounts of tracking time. The adjusted comparison is more informative.
It also fits earlier evidence. Dankner and colleagues, JAMA Network Open, 2024, studied a population of 543,595 adults with type 2 diabetes. They found no support for increased pancreatic cancer incidence during seven years after starting GLP-1 medicines compared with basal insulin, a long-acting insulin treatment.
That earlier study covered the medicine family, rather than semaglutide alone. Its authors still called for monitoring beyond seven years. Reassuring findings and a need for longer observation can both be true.
The broader question of whether semaglutide causes cancer includes other cancer types and the separate thyroid warning. This week's pancreatic result cannot settle those questions.
What the Nordic findings cannot answer
The release reports mean follow-up of 1.40–1.85 years after a one-year delay before cancer tracking began. Mean means average. That window is short for judging cancers that may take years to develop.
The registered study population also matters. It concerned type 2 diabetes treatment, rather than Wegovy users taking semaglutide solely for weight management. The same molecule does not make those populations identical.
The registry's exclusions included prior acute or chronic pancreatitis and a history of any cancer other than non-melanoma skin cancer. Pancreatitis means inflammation of the pancreas. Findings from this selected group cannot establish the risk for someone excluded from it.
Nor can matching turn a medical-record study into proof of cause. Differences in diabetes severity, testing and treatment choices can still affect a comparison. Manufacturer funding deserves transparency; a regulator's requirement doesn't remove the need to examine the methods.
Separate trial evidence adds context
As of October 1, 2026, the day this article was written, a Diabetes, Obesity and Metabolism analysis published online June 15, 2026 also addressed this question. It appears in the September issue. Chiang and colleagues combined six randomized trials, studies assigning treatment by chance, with 38,057 participants.
There were 32 pancreatic cancers with semaglutide and 41 with placebo, an inactive comparison treatment. The overall odds ratio, a comparison of the odds of a diagnosis, was 0.78, with a 95% confidence interval of 0.49–1.24. It did not show a statistically clear difference.
These trials included SUSTAIN-6, SELECT, FLOW, STRIDE, PIONEER-6 and SOUL. Cancer counts came from side-effect reports, rather than a dedicated cancer assessment designed in advance. Few cases and limited follow-up leave uncertainty. The analysis supports continued study, rather than prescribing semaglutide to prevent cancer.
The pancreatitis warning still applies
Cancer and inflammation are different outcomes. A reassuring cancer result doesn't cancel a pancreatitis warning.
As of October 1, 2026, the day this article was written, the FDA-posted Ozempic injection label prints “Revised: 05/2026.” Section 5.2 retains the acute-pancreatitis warning. It describes persistent or severe abdominal pain, sometimes reaching the back, with or without nausea or vomiting.
That pattern warrants prompt medical assessment. The label directs clinicians to discontinue Ozempic and begin appropriate management if pancreatitis is suspected. The separate semaglutide and pancreatitis guide explains that clinical concern.
Does the study favor switching to tirzepatide?
This study didn't compare semaglutide with tirzepatide. It cannot show that switching between them lowers pancreatic cancer risk. The tirzepatide versus semaglutide comparison covers the trials used to compare other treatment outcomes.
Through Promise, semaglutide and tirzepatide are dispensed as compounded medications, prepared by a pharmacy for individual prescriptions. This is different from an FDA-approved product: the formulations offered here are not FDA-approved. The Nordic findings do not establish identical outcomes for a compounded preparation.
As of October 1, 2026, the day this article was written, the FDA's compounding page is marked “Content current as of: 06/22/2026.” It explains that compounded drugs do not undergo its review for safety, effectiveness or quality before marketing.
A licensed provider may still prescribe a compounded formulation when medically appropriate; that decision is between the patient and the doctor.
What to watch next
A full report can help readers examine the matching methods, exclusions and additional analyses behind the conference findings. Longer follow-up would address the biggest remaining question: whether the reassuring pattern holds many years after treatment starts.
For an individual visit, prior pancreatic illness and the exact cancer history still matter more than a headline alone. At Promise, a licensed provider reviews every request, and not everyone qualifies. A useful review keeps both the reassuring evidence and its limits in view.
