Does semaglutide cause thyroid cancer? Human studies have not established that it does. The warning comes from tumors in rats and mice. A small or delayed risk in people remains unresolved.
The firm restriction concerns a particular cancer history. Wegovy and Ozempic must not be used with a personal or family history of medullary thyroid carcinoma (MTC), a rare thyroid cancer. They also must not be used in people with multiple endocrine neoplasia syndrome type 2 (MEN 2), an inherited tumor condition.
Does semaglutide cause thyroid cancer? What the warning says
As of September 30, 2026, the day this article was written, both labels retain their thyroid warning. The FDA PDFs print “Revised: 06/2026” for Wegovy and “Revised: 05/2026” for Ozempic injection.
Both full boxed warnings contain this sentence, verbatim:
In rodents, semaglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures.
That means more tumors with higher doses and longer exposure, at blood levels relevant to treatment.
A boxed warning is the most prominent safety warning on a prescription label. Here, the labels say the relevance of those animal tumors to humans is unknown. They also specify the MTC and MEN 2 exclusions above. A contraindication means the medicine must not be used in that situation.
Why the warning began with rats and mice
C cells are thyroid cells that make calcitonin, a hormone involved in calcium balance. MTC starts in those cells. It differs from papillary and follicular cancers, the more common thyroid cancers arising from other thyroid cells.
The label describes lifetime exposure in mice and rats. The animal finding matters even though its meaning for people remains uncertain.
What the human thyroid studies found
Semaglutide belongs to the GLP-1 family, medicines that copy a gut hormone involved in appetite and blood sugar. Much of the thyroid evidence combines that family. It cannot give a precise risk for semaglutide alone.
The French study found a concerning association. Bezin and colleagues, Diabetes Care, 2023, compared 2,562 thyroid-cancer cases with 45,184 people without that diagnosis. One to three years of GLP-1 use was associated with higher thyroid-cancer rates.
A hazard ratio (HR) compares rates during follow-up; 1 means no difference. A 95% confidence interval (CI) is the range showing how uncertain the estimate is. The French estimates were HR 1.58 (95% CI 1.27–1.95) for all thyroid cancers and 1.78 (1.04–3.05) for MTC.
For all thyroid cancer, that means a rate about 58% higher in this comparison. It does not show that the medicine caused those cancers.
The Scandinavian study found a different result. Pasternak and colleagues, BMJ, 2024, compared 145,410 GLP-1 users with 291,667 people starting another diabetes medicine. Thyroid cancer occurred at about 1.3 versus 1.5 cases per 10,000 people followed for a year. The overall estimate was HR 0.93 (95% CI 0.66–1.31); for MTC, 1.19 (0.37–3.86).
Average GLP-1 follow-up was 3.9 years, but semaglutide users averaged only 1.1 years. That shorter window matters for a cancer question. Both studies observed treatment rather than assigning it by chance, so differences in testing or patient histories could influence their findings.
What newer reviews add
A meta-analysis is a review that combines study results. Reviews of randomized trials, which assign treatment by chance, have also disagreed. Odds ratios (ORs) compare the odds of a diagnosis between groups; 1 again means no difference.
Silverii and colleagues, Diabetes, Obesity and Metabolism, 2025, found a thyroid-cancer signal: OR 1.55 (95% CI 1.05–2.27). Their review included trials lasting at least 52 weeks.
As of September 30, 2026, the day this article was written, a newer trial review offered an inconclusive estimate. Eisa and Barood, AACE Endocrinology and Diabetes, published online March 12, 2026, included 15 trials with 84,237 participants. There were only 43 thyroid cancers, and the estimate was OR 1.52 (95% CI 0.86–2.68). The authors rated certainty very low; the results could not exclude a meaningful increase.
As of September 30, 2026, the day this article was written, a September 1 review in Cancer Control added another comparison. A risk ratio (RR) compares risks between groups. Across 17 observational thyroid studies, it found RR 1.09 (95% CI 0.98–1.21), an inconclusive overall result.
Against metformin specifically, the estimate was higher: 1.31 (1.02–1.67). Metformin is another diabetes medicine, and these comparisons cannot establish cause.
The evidence supports continued attention to the warning. It cannot settle risk over decades. The broader question about semaglutide and cancer also involves cancers outside the thyroid.
What a prescriber needs to know
The useful family-history detail is the exact diagnosis. “My relative had thyroid cancer” leaves an important question unanswered: was it MTC, papillary cancer, or another type? Previous records can help a clinician clarify that history.
A personal or family MTC history rules out Wegovy and Ozempic. MEN 2 in the patient also rules them out. A family MEN 2 diagnosis belongs in the review because it can signal an inherited risk requiring evaluation. Watching for symptoms does not replace these restrictions.
Hypothyroidism, meaning an underactive thyroid, is a different diagnosis. So are common thyroid nodules, lumps within the gland. Neither is automatically the MTC or MEN 2 exclusion, but symptoms, prior surgery and any unfinished thyroid evaluation still matter.
Neck symptoms and screening are separate questions
Both labels identify a neck lump, persistent hoarseness, trouble swallowing and trouble breathing as symptoms to report to a physician. New or unexplained symptoms need medical evaluation. They cannot diagnose cancer on their own.
Routine calcitonin blood tests or thyroid ultrasound, a scan using sound waves, have uncertain value for detecting MTC early during treatment. That is different from investigating a known nodule or an abnormal test. The labels call for further evaluation of those findings.
Would switching to tirzepatide avoid the warning?
Tirzepatide carries the same thyroid-history restrictions, so switching does not remove them. As of September 30, 2026, the day this article was written, Zepbound's label retained its thyroid box. It prints “Revised: 8/2026.” The guide to tirzepatide and thyroid cancer explains its own rat findings and the human evidence.
What this means for a compounded prescription
A compounded medication is prepared by a pharmacy for an individual prescription. Through Promise, semaglutide is dispensed as a compounded medication, which is different from an FDA-approved product: the formulation offered here is not FDA-approved. The FDA 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. Compounding does not remove the thyroid concern or make an MTC or MEN 2 history acceptable.
At Promise, a licensed provider reviews every request and not everyone qualifies. The review needs the actual diagnosis and family history, with the warning and the limits of the human evidence both clear.
