The warning on Ozempic, Wegovy, Mounjaro, and Zepbound is real. It is also based entirely on rodent data, with no confirmed signal in humans. Here is what the evidence actually shows — and what it doesn't.
The pattern shows up with regularity in GLP-1 patient communities, in doctors' waiting rooms, and in telehealth consultations across the country. A 26-year-old at 450 pounds, with type 2 diabetes and elevated cardiovascular risk, has finally gotten a prescription for semaglutide. They fill it. Then they read the label.
The black box warning — the most serious warning the FDA can put on a prescription drug — says the medication causes thyroid C-cell tumors. They stop reading. They do not fill the prescription. They are terrified of cancer.
This is happening regularly. And it reflects a gap between what the warning actually says — and what many patients understandably hear.
The goal of this guide is not to minimize the FDA's warning, dismiss it, or tell anyone whether to take these medications. The goal is to explain what the warning actually says, what evidence it is based on, what it does and does not mean for the general patient population, and what questions are worth discussing with a physician. The FDA's position, the underlying science, and the available post-market data all carry important distinctions that most patients never see.
The black box warning that appears on semaglutide (Ozempic and Wegovy) and tirzepatide (Mounjaro and Zepbound) prescribing information reads, in its essential language:
FDA Black Box Warning — GLP-1 Receptor Agonists (paraphrased from label)
CAUSES THYROID C-CELL TUMORS IN RATS AND MICE AT CLINICALLY RELEVANT EXPOSURES. IT IS UNKNOWN WHETHER [THIS MEDICATION] CAUSES THYROID C-CELL TUMORS, INCLUDING MEDULLARY THYROID CARCINOMA (MTC), IN HUMANS, AS THE HUMAN RELEVANCE OF [DRUG]-INDUCED RODENT THYROID C-CELL TUMORS HAS NOT BEEN DETERMINED.Reading the actual warning language carefully matters. The FDA is making two distinct statements:
The contraindication — the group of patients who should absolutely not take these medications — is specific and limited: patients with a personal or family history of medullary thyroid carcinoma (MTC) or Multiple Endocrine Neoplasia type 2 (MEN2).
Medullary thyroid carcinoma (MTC) is a relatively rare form of thyroid cancer. Unlike the much more common papillary and follicular thyroid cancers — which arise from the thyroid follicular cells that produce thyroid hormone — MTC arises from parafollicular C-cells, which produce a hormone called calcitonin. MTC accounts for roughly 3–4% of all thyroid cancers in the United States. It is the specific type that appeared in GLP-1 rodent studies, and it is the specific type addressed in the contraindication.
Multiple Endocrine Neoplasia type 2 (MEN2) is a hereditary syndrome — caused by mutations in the RET gene — that predisposes carriers to MTC, among other endocrine tumors. It is rare, affecting approximately 1 in 35,000 people. Patients with MEN2, or with a family history suggesting it, should be evaluated by an endocrinologist independent of any GLP-1 medication decision.
Key point: The FDA's warning language itself acknowledges that the human relevance of the rodent findings is unknown. The contraindication applies to a specific and identifiable high-risk population (MTC or MEN2 history), not to the general patient population. These are distinct statements, and the distinction matters for patients trying to understand their individual situation.
The science underlying this warning deserves a direct examination. In animal studies conducted prior to FDA approval, GLP-1 receptor agonists — including semaglutide and tirzepatide — caused C-cell hyperplasia (abnormal cell proliferation) and medullary thyroid carcinoma in rats and mice. These findings were observed at doses that substantially exceeded human therapeutic exposure levels.
Researchers have noted a critical anatomical and physiological difference that is central to understanding whether these rodent findings are relevant to humans: rats and mice express GLP-1 receptors at substantially higher density in their thyroid C-cells than humans do. The GLP-1 receptor is the target through which these medications act, and the relatively sparse GLP-1 receptor expression in human thyroid C-cells is recognized as a plausible biological reason why the rodent signal might not translate to humans.
The FDA's own review process acknowledged this species difference. The agency classified the rodent thyroid C-cell tumor findings as a potential class effect concern requiring a precautionary label warning — while simultaneously acknowledging that human relevance had not been established. This is reflected in the warning language itself.
The large Phase 3 clinical trials that generated the FDA approval for these medications enrolled tens of thousands of patients under carefully monitored conditions:
None of these trials demonstrated a statistically significant signal for medullary thyroid carcinoma or thyroid C-cell tumors in human participants. The available evidence from these controlled clinical programs did not reproduce the rodent carcinogenicity finding in humans.
Important limitation: Clinical trials are not designed to detect rare cancers. MTC is rare even in the general population — a trial would need to be very large and of long duration to reliably detect a meaningful increase in a rare cancer. The absence of a signal in the SUSTAIN, STEP, and SURMOUNT trials is meaningful, but it does not by itself constitute proof that no risk exists. The post-market surveillance data, described below, adds importantly to this picture.
Clinical trials are one piece of the safety evidence picture. Post-market pharmacovigilance — the ongoing monitoring of medications once they are in widespread use by millions of real-world patients — provides a different and complementary dataset.
Semaglutide has been available globally since 2017 (Ozempic for diabetes) and since 2021 (Wegovy for weight management). Tirzepatide has been available since 2022 (Mounjaro for diabetes) and 2023 (Zepbound for weight management). These medications have been used by millions of patients in the United States and globally.
Epidemiological studies and pharmacovigilance analyses examining real-world use of GLP-1 receptor agonists have not identified a confirmed increase in medullary thyroid carcinoma rates in the patient population using these medications. Regulatory agencies monitoring drug safety signals in post-market databases have not issued updated safety communications indicating that a human MTC signal has emerged.
Researchers have noted that the absence of a post-market signal across a large and growing patient population is meaningful evidence — while also acknowledging that it does not constitute absolute proof of safety. MTC has a slow natural history; effects from a medication might not become apparent in the timeframe studied. This uncertainty is worth acknowledging honestly.
The scientific position: Available evidence — from clinical trials enrolling tens of thousands of patients and from post-market surveillance of millions of real-world patients over multiple years — has not demonstrated a confirmed increase in human MTC associated with GLP-1 receptor agonist use. The FDA warning remains in place because the rodent signal exists and because the long-term human data set continues to accumulate. These two facts are both true simultaneously.
The FDA's contraindication is specific. Understanding who it covers — and who it does not — is essential to applying the warning accurately.
Do not use GLP-1 receptor agonists if you have:
For this group, the contraindication is unambiguous and should be observed. An endocrinologist can evaluate thyroid cancer family history in detail.
For everyone else — the broad general patient population without MTC or MEN2 history — the FDA warning is a precautionary label disclosure based on animal data, not a statement that GLP-1 medications cause thyroid cancer in the general population. The FDA's own language acknowledges that the human relevance of the rodent findings is unknown.
A primary care physician or endocrinologist can assess an individual patient's thyroid history, any relevant family history, and any current thyroid symptoms. That clinical evaluation is the appropriate way to determine whether any individual's thyroid history warrants special consideration — not a blanket decision to avoid these medications based on the warning alone.
The FDA's black box warning addresses one category of potential risk associated with GLP-1 receptor agonists. It does not, and by its nature cannot, address the risk calculus that a patient with severe obesity faces when deciding whether to take these medications.
The available evidence on obesity and its health consequences is extensive. Severe obesity (BMI 40 and above) is associated with substantially elevated risk of cardiovascular disease, type 2 diabetes, obstructive sleep apnea, non-alcoholic fatty liver disease, and certain cancers — including endometrial cancer and colorectal cancer. The mortality risk associated with severe, untreated obesity is well-documented across large longitudinal studies.
The available evidence on semaglutide's cardiovascular effects is also substantial. The SELECT trial — a large randomized controlled trial of over 17,000 patients with overweight or obesity and established cardiovascular disease but without diabetes — demonstrated a 20% reduction in major adverse cardiovascular events (heart attack, stroke, and cardiovascular death) in the semaglutide group compared to placebo over approximately 3.3 years. This is a meaningful, pre-specified, clinically significant result.
The point here is not to tell any patient what to do. A 26-year-old at 450 pounds facing a decision about semaglutide is facing a risk comparison that involves a theoretical, rodent-based thyroid risk on one side, and documented, epidemiologically-established cardiovascular and metabolic mortality risk on the other. The FDA warning is required to focus on the potential drug risk. Framing that risk against the risks of untreated disease is a question for a physician, not a label.
Researchers and clinicians who specialize in obesity medicine have noted this asymmetry: patients who decline GLP-1 therapy because of the thyroid warning may be accepting substantially larger documented risks in its place. That is not a reason to dismiss the warning — but it is a reason to have the conversation with a qualified physician rather than making a decision in isolation.
If you are considering a GLP-1 receptor agonist and are concerned about the thyroid cancer warning, the following are the clinically relevant questions — the ones that will actually inform whether the warning has direct relevance to your situation.
If your answers to the contraindication-specific questions (MTC history, MEN2) are negative, and you have no current thyroid symptoms, your doctor will likely assess the warning as applying to the general precautionary category — a disclosure based on animal data, not a confirmed personal risk factor.
Attorney's framing: The FDA warning is written for a population of millions of patients. The relevant clinical analysis is individual. The questions above are what a physician uses to filter general precautionary language down to individual clinical relevance. They are worth having explicitly with your prescriber — not because the answers are likely to reveal a problem, but because they are the right questions to ask.
For patients without MTC or MEN2 history who begin a GLP-1 receptor agonist, the FDA's recommended monitoring approach is symptom-based, not surveillance-based.
Any new or enlarging mass in the neck, particularly in the midline or lower anterior neck region where the thyroid sits.
A persistent change in voice quality not explained by an upper respiratory illness or other clear cause.
New dysphagia — particularly difficulty swallowing solid food — that is unexplained and persists.
New or unexplained shortness of breath, particularly if associated with any neck symptoms.
The FDA does not recommend routine thyroid ultrasound for patients without a personal or family history of MTC or MEN2 prior to starting or during GLP-1 therapy. Baseline or interval ultrasound is not part of the standard monitoring protocol for the general patient population.
Routine thyroid function tests — specifically TSH (thyroid-stimulating hormone) and free T4 — are also not MTC screening tools. Medullary thyroid carcinoma arises from the C-cells, which are separate from the follicular cells that govern TSH levels. A normal TSH does not exclude MTC. For patients concerned specifically about MTC, calcitonin levels are a more specific biomarker — but ordering calcitonin as a screening test in asymptomatic, low-risk patients is not current FDA guidance and is a clinical decision for a physician to make in context.
Practical note: If you develop any of the four symptoms above while on a GLP-1 medication, contact your prescriber — do not simply stop the medication without medical guidance. A thorough clinical evaluation will determine whether further workup is warranted. The symptoms listed are common causes of many non-serious conditions; prompt reporting allows for appropriate clinical assessment.
| Claim or Question | What the Evidence Shows | Source |
|---|---|---|
| GLP-1 RAs cause C-cell tumors in rodents | Yes — confirmed in animal studies at supratherapeutic doses | FDA prescribing information; pre-approval animal data |
| Rodents have higher GLP-1 receptor density in thyroid C-cells than humans | Yes — recognized species difference acknowledged in FDA review | FDA scientific review; published receptor biology research |
| Human clinical trials show MTC signal | No confirmed MTC signal in SUSTAIN, STEP, or SURMOUNT trials | Published trial results; FDA drug approval reviews |
| Post-market surveillance confirms MTC increase in humans | No confirmed increase identified in pharmacovigilance to date | Regulatory pharmacovigilance; epidemiological studies |
| GLP-1 RAs are contraindicated for MTC/MEN2 patients | Yes — clear, unambiguous FDA contraindication | FDA black box warning; prescribing information |
| General patients (no MTC/MEN2 history) face confirmed thyroid cancer risk | Not established — warning is precautionary, based on animal data | FDA warning language; available human data to date |
| Semaglutide reduces cardiovascular events in obese patients | Yes — SELECT trial showed 20% reduction in major CV events | SELECT trial (NEJM, 2023); FDA label update |
This table summarizes the state of available evidence as of May 2026. The evidence base for long-term GLP-1 safety continues to accumulate. Consult a physician for individual clinical guidance.
Does semaglutide cause thyroid cancer in humans?
Available evidence has not established a causal link between semaglutide and thyroid cancer in humans. The FDA black box warning is based on animal studies — specifically rodent data showing C-cell tumors at doses far exceeding human therapeutic levels. Rodents have substantially higher GLP-1 receptor density in thyroid C-cells than humans do, which the FDA acknowledges as a meaningful species difference.
No large human clinical trial, including the SUSTAIN or STEP series enrolling tens of thousands of patients, has demonstrated a signal for medullary thyroid carcinoma (MTC). Epidemiological and pharmacovigilance data from millions of real-world patients have not confirmed an increase in MTC rates. The warning is a precautionary disclosure required when animal data raises a concern — not a statement that the risk has been confirmed in humans.
Who should definitely not take GLP-1 medications because of thyroid risk?
The FDA has established a clear contraindication for patients with a personal or family history of medullary thyroid carcinoma (MTC) or Multiple Endocrine Neoplasia type 2 (MEN2). This contraindication appears in the black box warning on semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound) and should be taken seriously.
For everyone else, the warning is a precautionary disclosure based on animal data. An endocrinologist or primary care physician can evaluate whether an individual's thyroid history warrants special consideration before initiating therapy.
Why did the FDA put a black box warning if there's no human evidence?
Black box warnings — the most prominent label warnings the FDA can require — are triggered under a regulatory standard that does not require confirmed human harm. When animal studies show a serious potential risk, FDA regulations can require a black box warning even when the relevance to humans is uncertain. The FDA's own label language reflects this: the warning states that the relevance to humans is "unknown."
The FDA is disclosing a signal from animal data and imposing a precautionary contraindication for high-risk populations — standard regulatory practice for serious potential hazards identified in animal studies. It does not mean the FDA has found that GLP-1 medications cause thyroid cancer in people. It means the agency found animal data that required disclosure under its labeling rules.
Should I get a thyroid ultrasound before starting semaglutide?
FDA guidelines do not recommend routine thyroid ultrasound before starting a GLP-1 receptor agonist for patients without a personal or family history of MTC or MEN2. The clinical questions your doctor will want answered are: Do you have any history of MTC or MEN2? Do you have any neck masses, thyroid nodules, hoarseness, or difficulty swallowing? Has your thyroid been evaluated before?
For patients without thyroid-specific history or symptoms, a baseline thyroid ultrasound is not the current standard of care before initiating GLP-1 therapy. Note also that routine thyroid function tests (TSH, free T4) are not MTC screening tools — MTC arises from C-cells that produce calcitonin, not from the follicular cells that regulate TSH.
What thyroid symptoms should I report while on a GLP-1 medication?
The FDA recommends reporting the following symptoms to your healthcare provider promptly while on a GLP-1 receptor agonist: a lump or swelling in the neck; hoarseness or a change in voice quality; difficulty swallowing (dysphagia); or shortness of breath. These symptoms could indicate a thyroid mass and should be evaluated — though they are also common symptoms of many non-serious conditions.
The FDA does not recommend routine monitoring for asymptomatic patients without prior thyroid history. If you experience any of these symptoms, contact your prescriber rather than stopping your medication unilaterally — the clinical evaluation will determine whether the medication is a factor.