GLP-1 Thyroid Cancer Risk: What the FDA Warning Actually Means in 2026
Published:
The bottom line, up front
The FDA warning on most GLP-1 drugs is real — but it is narrow. The warning is about a rare thyroid cancer called medullary thyroid carcinoma (MTC) and a genetic syndrome called MEN2. MTC is only 1–4% of thyroid cancers in the U.S. For most GLP-1 products, the warning came from animal studies, not human studies, and the same labels say human relevance “has not been determined.”
The biggest human studies through 2026 — including 145,410 GLP-1 users across Scandinavia, 351,000+ adults in a Mayo Clinic / JAMA study, and a 48-trial randomized meta-analysis in 94,000+ people — have not found a real overall increase in thyroid cancer for the general population.
If you have a personal or family history of MTC or MEN2, the warning applies and these drugs are off the table. If you don't, the warning is a precaution from rodent data, not proof of human risk.
At a Glance: Which Situation Are You In?
Jump to the part of the page that fits you.
| Your situation | What it usually means |
|---|---|
| You or a relative had medullary thyroid cancer (MTC) or MEN2 | The warning applies. These drugs are contraindicated. See alternatives below |
| You or a relative had papillary, follicular, or Hürthle-cell thyroid cancer | This is NOT the warning. Tell your doctor, but it's a different cancer type. See section |
| A relative had "thyroid cancer" but you don't know the type | Pause and find out the type before starting. Use the 4-question check |
| You have Hashimoto's, hypothyroidism, or take levothyroxine | Not the warning. Disclose it; standard care continues. See section |
| You have a thyroid nodule | Not the warning by itself. Standard nodule workup applies. See section |
| You've had a thyroidectomy in the past | Depends on why it was removed. See section |
| No personal or family thyroid history | Standard risk-benefit conversation. See evidence section |
| You feel a lump in your neck, persistent hoarseness, trouble swallowing, or shortness of breath | Call your doctor — these are the symptoms the FDA tells GLP-1 users to report. See section |
What the FDA Warning Actually Says (in Plain English)
Quick answer: Most GLP-1 drugs carry a “boxed warning” — the strongest safety alert the FDA uses — for thyroid C-cell tumors, including medullary thyroid carcinoma. The label tells doctors not to prescribe these drugs to anyone with a personal or family history of MTC, or to anyone with MEN2. The same labels say the human relevance of the rodent finding “has not been determined.”
A boxed warning ≠ proven human risk
The FDA places a boxed warning when the evidence suggests serious risk worth flagging at the very top of the prescribing label. It does not mean the drug has been proven to cause that risk in humans. It means the risk was serious enough — and the evidence suggestive enough — that doctors and patients should be told upfront.
Where the warning came from: rodents, not humans
For most GLP-1s, the warning came from studies in rats and mice. When researchers gave rodents long-acting GLP-1 drugs at clinically relevant doses, the rodents grew C-cell tumors in their thyroid glands. Some of those tumors were medullary thyroid cancer. Rodent C-cells have far higher GLP-1 receptor density than human C-cells. Calcitonin (the marker of C-cell activity) does not rise meaningfully in human GLP-1 users in clinical trials.
Why the FDA chose the warning anyway
The FDA had two choices. Approve the drugs without the warning and hope humans were different from rodents. Or approve them with a warning and tell doctors to avoid the high-risk population — people who already have MTC or MEN2 — until human data caught up. They chose the second path. The contraindication is deliberate and narrow.
What Kind of Thyroid Cancer Is the Warning Actually About?
Quick answer: The warning is about medullary thyroid carcinoma (MTC), a rare cancer that starts in the thyroid's C-cells. It is not about the more common types — papillary, follicular, or Hürthle-cell thyroid cancer — which together make up about 95% of thyroid cancers in the U.S. and arise from completely different cells.
Your thyroid has two main kinds of cells:
Follicular cells
Make thyroid hormone (the stuff levothyroxine replaces). Source of papillary, follicular, and Hürthle-cell cancers — about 95% of all thyroid cancers.
C-cells (parafollicular cells)
Make calcitonin. Source of medullary thyroid carcinoma (MTC) — the cancer named in the GLP-1 warning. Only 1–4% of all thyroid cancers.
| Type | Origin cell | Share of U.S. thyroid cancers | In GLP-1 warning? |
|---|---|---|---|
| Papillary | Follicular cell | ~85% | No |
| Follicular | Follicular cell | ~10% | No |
| Hürthle/oncocytic | Follicular cell | ~2% | No |
| Medullary (MTC) | C-cell | 1–4% | YES — this is the warning |
| Anaplastic | Undifferentiated | <1% | Not specifically named |
GLP-1 Thyroid Cancer Risk in Humans: What Does the Research Actually Show?
Quick answer: The biggest active-comparator cohorts and the largest randomized-trial meta-analysis do not show a real overall increase in thyroid cancer risk for GLP-1 users. Two smaller randomized-trial meta-analyses found small fragile signals on rare events — neither found a signal for MTC or papillary cancer specifically. The early French case-control signal is best explained by detection bias.
| Study | Design | Participants | Finding | Limitation |
|---|---|---|---|---|
| Bezin et al., 2023, Diabetes Care | French nested case-control, national insurance database | T2D patients on second-line drugs, 2006–2018 | Adjusted HR 1.58 for all thyroid cancer; 1.78 for MTC at 1–3 years cumulative use | Case-control vulnerable to detection bias; Diabetes Care commentary flagged statistical and design concerns |
| Pasternak et al., 2024, BMJ | Active-comparator new-user cohort; Denmark, Norway, Sweden | 145,410 GLP-1 starters vs 291,667 DPP-4 inhibitor starters | HR 0.93 (95% CI 0.66–1.31) for thyroid cancer; HR 1.19 (0.37–3.86) for MTC; rules out >31% relative increase | MTC subgroup small; mean follow-up 3.9 years |
| Brito / Mayo Clinic et al., 2025, JAMA Otolaryngology–H&N Surgery | Target-trial emulation, U.S. claims database | 351,000+ adults with T2D; 41,112 GLP-1 starters | Main mITT HR 1.24 (0.88–1.76, not significant). First-year HR 1.85 (1.11–3.08), then signal disappears. GLP-1 users had more thyroid ultrasound at 12 months: 2.1% vs 1.5% | Diabetes population; rare MTC hard to detect; first-year signal explained by detection bias |
| Baxter et al., 2025, Thyroid | International multisite cohort, 6 countries | 92,497 GLP-1 users vs 2,484,408 DPP-4 inhibitor users | Adjusted HR 0.81 (95% CI 0.59–1.12); no dose-response trend | Median follow-up 1.8–3 years per cohort |
| Silverii et al., 2024, Diabetes Obes Metab | Systematic review + meta-analysis of RCTs | 64 trials; 26 reported ≥1 thyroid cancer event | Fixed-effect MH-OR 1.52 (1.01–2.29) for overall thyroid cancer — fragility index 1; NNH 1,349. No significant signal for papillary cancer or MTC specifically | Signal is fragile (single event flip would erase it); RCTs not powered for cancer |
| Duchemin et al., 2025, Diabetes Obes Metab | Systematic review + meta-analysis of RCT harms data | Pooled randomized-trial harms data | Peto OR 1.62 (95% CI 1.08–2.44, I² 11%) — small statistically significant rare-event thyroid cancer signal | Rare-event harms meta-analysis; small event counts; trial follow-up still short for cancer latency |
| Annals of Internal Medicine, December 2025 | Systematic review + meta-analysis of placebo-controlled RCTs | 48 RCTs, 94,245 participants (semaglutide in 42% of trials) | "Little or no effect" on thyroid cancer (moderate certainty) | Median trial follow-up only 70 weeks. Citation to be confirmed against primary source on next refresh |
| TriNetX retrospective cohort, 2025 (PMC) | Propensity-matched real-world EHR | 89,646 long-term GLP-1 users vs multiple comparators | No increased thyroid cancer risk vs insulin, metformin, SGLT2, DPP-4, sulfonylureas, or thiazolidinediones | EHR coding accuracy; residual confounding |
| Clayman Thyroid Center White Paper, Feb 2026 | Clinical synthesis from highest-volume U.S. thyroid center (~2,000 surgeries/year) | Institutional case review + literature | No observed pattern of MTC tied to GLP-1 use; no convincing evidence of papillary, follicular, or Hürthle-cell causation | Narrative review, not RCT |
| FAERS pharmacovigilance, 2022–2023 | FDA Adverse Event Reporting System | Spontaneous reports | Reporting odds ratios 4.7× and ~8× higher for GLP-1s vs comparator drugs | FAERS measures reporting volume, not incidence. A submitted report does not mean the drug caused the event |
The honest takeaway
The large active-comparator cohorts (Pasternak, Baxter, Mayo/JAMA) and the largest randomized-trial meta-analysis (Annals 2025) all converge on no real overall increase. Two smaller randomized-trial meta-analyses (Silverii 2024 and Duchemin 2025) found small fragile signals driven by rare events — neither found a signal for MTC or papillary cancer specifically. The early French case-control signal stands almost alone, and the Mayo Clinic / JAMA target-trial emulation showed why: GLP-1 users get more thyroid imaging, so more incidental thyroid findings get diagnosed in the first year. The label warning remains specific to MTC and MEN2 — populations every one of these studies excluded.
Why the early French signal probably wasn't what it looked like
The 2023 Bezin study found that 1–3 years of GLP-1 use was linked to higher thyroid cancer rates. That's a real signal. But case-control studies have a known weakness called detection bias: if a drug carries a thyroid warning, doctors order more thyroid imaging. More imaging finds more thyroid problems — including small cancers that were already there and would have been found later anyway.
The Mayo Clinic / JAMA 2025 target-trial emulation showed this directly. GLP-1 users got thyroid ultrasounds more often than people on other diabetes drugs — 2.1% vs. 1.5% within 12 months. And the entire excess thyroid cancer signal was concentrated in the first year, with the hazard ratio dropping to non-significance after that. That's the fingerprint of detection bias: a burst of early diagnoses, then nothing.
Why the warning still exists despite the reassuring data
- One.The rodent finding for most of the class is real. Rats and mice on long-acting GLP-1s grow C-cell tumors in a dose-dependent way. That's not a fluke.
- Two.Rodent thyroids and human thyroids are biologically different. Rodent C-cells have far higher GLP-1 receptor density than human C-cells. Calcitonin does not rise meaningfully in human GLP-1 users in clinical trials. So the bridge from rodent biology to human biology is weak — but not zero.
- Three.MTC is rare and serious. Even a small absolute risk in a high-risk genetic subgroup like MEN2 is worth keeping the warning for. The FDA chose to keep the contraindication narrow (MTC and MEN2) rather than apply it to the whole population.
The 4-Question Family History Check
Quick answer: Most people who searched this page after hearing “thyroid cancer runs in my family” are not actually contraindicated. The contraindication is for medullary thyroid cancer (MTC) or MEN2 — not for the more common papillary, follicular, or Hürthle-cell types. Use the four questions below to figure out which group you're in.
Question 1: Has any blood relative been diagnosed with thyroid cancer?
Question 2: Do you know what kind of thyroid cancer it was?
Question 3: Is there a known diagnosis of MEN2 in the family — or a known RET gene mutation?
Question 4: Have you personally ever had MTC or been diagnosed with MEN2?
Your path
Path A — No personal or family thyroid cancer history
The MTC/MEN2 thyroid contraindication does not appear to apply based on your answers. The general human evidence applies — discuss benefits and risks with your prescriber and learn the symptoms to watch for.
Path B — Personal or family MTC, or MEN2
Every GLP-1 with the boxed warning is contraindicated for you. That includes Ozempic, Wegovy, Rybelsus, Mounjaro, Zepbound, Saxenda, Victoza, Trulicity, Bydureon BCise, Foundayo, and Xultophy. Talk to your endocrinologist about non-GLP-1 options. See the alternatives section below.
Path C — Family history of papillary, follicular, or Hürthle-cell thyroid cancer (no MEN2)
The MTC/MEN2 thyroid contraindication does not appear to apply by current U.S. label criteria. The Clayman Thyroid Center white paper (Feb 2026) and the University of Colorado Anschutz endocrinology team (2025) both frame GLP-1s as reasonable in this category, with normal monitoring. Confirm with your prescriber.
Path D — Family history of thyroid cancer of unknown type
Pause before starting. Most thyroid cancers in the U.S. are papillary, so the odds favor "you're fine." But MTC is the one that matters here, and it's worth a phone call. Ask your relative — or ask their doctor for the pathology report — to confirm the cancer type. Look for the word "medullary." If MEN2 is mentioned anywhere in the family, ask your doctor about RET genetic testing before starting.
Path E — You personally had a non-MTC thyroid cancer (papillary, follicular, or Hürthle-cell), in remission
The MTC/MEN2 thyroid contraindication does not appear to apply by current U.S. label criteria. The Clayman Thyroid Center and CU Anschutz teams consider GLP-1 therapy reasonable in this group with monitoring. Decisions should be individualized with your endocrinology and oncology teams.
Other Thyroid Conditions That Are Not the FDA Warning
Quick answer: Hashimoto's thyroiditis, hypothyroidism, hyperthyroidism, Graves' disease, benign thyroid nodules, and prior non-medullary thyroid cancer are not the FDA contraindication. Tell your prescriber, but don't assume any of these disqualify you from a GLP-1.
Hashimoto's thyroiditis and hypothyroidism
Hashimoto's is an autoimmune attack on the thyroid that often leads to hypothyroidism. It is not MTC, not MEN2, and not a thyroid C-cell tumor — different cells, different category, different warning. If you take levothyroxine (Synthroid, Levoxyl, Tirosint), there's one practical wrinkle: GLP-1s can change how your stomach empties, and some people need their levothyroxine dose adjusted. Get your TSH checked a few months after starting and again if your weight changes a lot. That's a normal endocrinology adjustment, not a thyroid cancer concern.
Benign thyroid nodules
Thyroid nodules are common — more than half of adults over 60 have at least one if you scan them. Most are benign. They are not the FDA contraindication. If you already have a known nodule, your prescriber may want to make sure it's been worked up under standard guidelines before starting a GLP-1. That's not because the GLP-1 will affect the nodule. It's because the FDA labels say nodules found on physical exam or imaging should be evaluated normally, regardless of GLP-1 status.
Hyperthyroidism and Graves' disease
Not the FDA contraindication. These are also follicular-cell conditions and have nothing to do with the C-cell warning. Standard endocrinology care applies independently.
Prior papillary, follicular, or Hürthle-cell thyroid cancer
Not the FDA contraindication. These cancers come from the follicular cells, not the C-cells. Bryan Haugen MD, an endocrinologist at the University of Colorado Anschutz Medical Campus, has been particularly clear in 2025: for patients with a history of papillary thyroid cancer, GLP-1s are not contraindicated by label, and his clinical view is that they're medically safe to use with standard monitoring. The Clayman Thyroid Center reached the same conclusion in their February 2026 white paper. For confirmed MTC patients — the ones the warning is actually about — the same physicians say no, until the data is better.
What if I've already had a thyroidectomy?
Thyroidectomy by itself doesn't answer the GLP-1 question. The reason for the surgery is what matters. If your thyroid was removed because of MTC or MEN2, the warning still applies — your medical history doesn't change because the gland is gone. If it was removed for papillary cancer, follicular cancer, benign nodules, Graves' disease, or goiter, that's a different category and not the FDA contraindication. Bring the operative note and pathology report to your prescriber. The pathology report is the document that answers the question.
One thing TSH and free T4 don't do
Standard thyroid function tests (TSH and free T4) check whether your thyroid hormone levels are normal. They do not screen for medullary thyroid cancer. MTC can occur with completely normal TSH. So if your doctor said "your thyroid labs are fine, you're safe to start," that's true for hormone function — but it isn't the same thing as ruling out MTC. Calcitonin is the C-cell marker, not TSH.
Should You Get a Calcitonin Test or Thyroid Ultrasound Before Starting?
Quick answer: No, not as a routine pre-flight check. The FDA labels themselves say routine calcitonin or thyroid ultrasound monitoring is “of uncertain value” for early MTC detection in GLP-1 users and may lead to unnecessary procedures. Targeted testing is reasonable when family history is unclear, when a nodule is already present, when calcitonin was previously elevated, or when symptoms call for it.
When NOT to get baseline screening
If you have no personal or family thyroid cancer history, no nodule, no symptoms, and no MEN2 risk — the labels do not recommend baseline calcitonin or thyroid ultrasound just because you're starting a GLP-1. Calcitonin has low specificity (lots of false positives), thyroid disease is common in the population, and routine monitoring tends to generate worry and procedures without catching meaningful disease earlier.
When baseline testing does enter the conversation
- Confirmed or possible MTC/MEN2 family history that needs clarification
- An existing palpable thyroid nodule, or one already seen on prior imaging
- Previously elevated calcitonin from another workup
- Symptoms from the FDA-required list (lump, hoarseness, trouble swallowing, shortness of breath)
- Another thyroid workup your clinician can document a specific reason for
How to read a calcitonin result if you do get one
Normal serum calcitonin is typically below 10–20 pg/mL, depending on sex and the lab's assay. Mild elevations have many non-cancer causes:
- Smoking
- Kidney disease
- Proton pump inhibitors (omeprazole, pantoprazole, etc.)
- Some autoimmune conditions
- Lab-to-lab variation
The Ozempic prescribing information notes that significantly elevated calcitonin — typically above 50 ng/L on confirmatory testing — warrants further evaluation. A single mildly high reading is not a diagnosis. It's a trigger for the next step, which is usually a stimulated calcitonin test or a referral.
The three questions to actually ask your prescriber
- 1Based on my history, should I get a baseline calcitonin?
- 2Is there a thyroid finding I should follow up on with my primary care doctor before we start?
- 3If my family history is unclear, what's our plan to clarify it?
Symptoms to Watch for While on a GLP-1
Quick answer: The FDA-required Medication Guide for every boxed-warning GLP-1 names four symptoms to report to your prescriber: a lump or swelling in the neck, persistent hoarseness, trouble swallowing, or shortness of breath. These are general thyroid mass symptoms and are not specific to GLP-1 use — but any of them deserves prompt evaluation.
A lump or swelling in your neck
Front and middle, not the side jaw area.
Hoarseness that doesn't go away
A few days of a scratchy voice from a cold doesn't count. We're talking weeks of it without explanation.
Trouble swallowing
Food or pills feel stuck. Worsening over time.
Shortness of breath
Especially when lying flat or when it's getting worse.
What these symptoms usually turn out to be
Most neck lumps are benign. They're often enlarged lymph nodes from a recent infection, sebaceous cysts, or benign thyroid nodules. Most hoarseness is viral, allergic, reflux-related, or vocal strain. Most swallowing trouble is reflux or muscle-related. Most shortness of breath is cardio, asthma, allergies, or anxiety.
The point of the symptom list isn't to make you panic-check your neck every morning. It's to tell you what's worth a phone call if it shows up and won't go away.
When to call your prescriber
Persistent (more than 2 weeks) or progressively worsening symptoms.
When to go to urgent care or ER
Acute breathing difficulty, sudden severe swelling, or rapid worsening.
Which GLP-1s Carry the Warning — and What About the New Pills?
Quick answer: Every commonly prescribed GLP-1 and GIP/GLP-1 drug in 2026 carries the boxed warning, including Ozempic, Wegovy (injection and oral pill), Rybelsus, Mounjaro, Zepbound, Saxenda, Victoza, Trulicity, Bydureon BCise, the just-approved oral Foundayo, and the combination drug Xultophy. Two older agents — Byetta (exenatide IR) and Adlyxin (lixisenatide) — historically have not carried the boxed warning. Last verified .
| Brand | Active ingredient | Use | Boxed warning? | MTC/MEN2 contraindication? |
|---|---|---|---|---|
| Ozempic | semaglutide (injection) | Type 2 diabetes / CV risk | Yes | Yes |
| Wegovy (injection) | semaglutide | Weight management | Yes | Yes |
| Wegovy oral (tablet) | semaglutide | Weight management (Dec 2025 approval) | Yes | Yes |
| Rybelsus | semaglutide (oral, T2D) | Type 2 diabetes | Yes | Yes |
| Mounjaro | tirzepatide | Type 2 diabetes | Yes | Yes |
| Zepbound | tirzepatide | Weight management / OSA | Yes | Yes |
| Saxenda | liraglutide | Weight management | Yes | Yes |
| Victoza | liraglutide | Type 2 diabetes | Yes | Yes |
| Trulicity | dulaglutide | Type 2 diabetes | Yes | Yes |
| Bydureon BCise | exenatide ER | Type 2 diabetes | Yes | Yes |
| Foundayo | orforglipron (oral pill) | Weight management (April 1, 2026 approval) | Yes (different rodent profile — see below) | Yes |
| Xultophy 100/3.6 | insulin degludec + liraglutide | Type 2 diabetes | Yes (via liraglutide) | Yes |
| Byetta | exenatide IR | Type 2 diabetes | Historically no (verify current label) | Historically no |
| Adlyxin | lixisenatide | Type 2 diabetes | Historically no (verify current label) | Historically no |
Sources: each row reflects current DailyMed prescribing information as of May 8, 2026. Verify current Byetta and Adlyxin labels on DailyMed before relying on historical notes.
What's new with Foundayo (the oral pill from Lilly)
Foundayo (orforglipron) was approved by the FDA on — the first oral, non-peptide GLP-1 that doesn't require any food or water timing rules. Same MTC/MEN2 contraindication. But the rodent story is genuinely different: the Foundayo label specifically says orforglipron is not pharmacologically active in rats or mice and did not produce thyroid tumors in those rodent studies. The FDA still required the boxed warning and the MTC/MEN2 contraindication as a class precaution.
The FDA approval letter also requires Lilly to run a medullary thyroid carcinoma registry-based case series for at least 15 years, with study completion scheduled for August 2042 and the final report due August 2043. That's the longest mandated thyroid follow-up of any GLP-1 in the class.
Alternatives If You're Contraindicated
Quick answer: If you have a personal or family history of MTC or MEN2, every boxed-warning GLP-1 is off the table. Alternatives by indication: for type 2 diabetes, options include metformin, SGLT2 inhibitors, DPP-4 inhibitors, and possibly Byetta or Adlyxin (verify current labels). For weight management, options include phentermine/topiramate (Qsymia), naltrexone/bupropion (Contrave), and bariatric procedures.
For type 2 diabetes
- Metformin: First-line, cheap, decades of safety data.
- SGLT2 inhibitors: Empagliflozin, dapagliflozin, canagliflozin. Cardiovascular and kidney benefits in T2D with heart failure or CKD. No thyroid warning.
- DPP-4 inhibitors: Sitagliptin, linagliptin, saxagliptin. Work on the same biological pathway as GLP-1s but don't carry the thyroid C-cell warning.
- Insulin: When appropriate.
- Byetta or Adlyxin: Historically, when no other option works — verify current label.
For weight management
- Phentermine/topiramate (Qsymia): Older, oral, effective; different mechanism.
- Naltrexone/bupropion (Contrave): Works on appetite differently.
- Bariatric surgery: Sleeve gastrectomy, Roux-en-Y gastric bypass. Durable weight loss for eligible patients.
- Endoscopic procedures: Intragastric balloon, endoscopic sleeve gastroplasty. Less invasive, shorter results.
What We Actually Verified for This Page
Primary sources we read directly:
- FDA-approved Medication Guides via DailyMed for: Ozempic, Wegovy (injection and oral), Rybelsus, Mounjaro, Zepbound, Saxenda, Victoza, Trulicity, Bydureon BCise, Xultophy 100/3.6, Foundayo. Last checked May 8, 2026.
- FDA news release announcing Foundayo (orforglipron) approval on April 1, 2026.
- FDA approval letter for Foundayo, including the 15-year MTC registry-based case series requirement (completion August 2042; final report due August 2043).
- Pasternak et al., BMJ 2024 — full PMC text (DOI 10.1136/bmj-2023-078225).
- Bezin et al., Diabetes Care 2023 — full PMC text (DOI 10.2337/dc22-1148) and the Thompson/Stürmer commentary.
- Brito / Mayo Clinic et al., 2025, JAMA Otolaryngology—Head & Neck Surgery — mITT HR 1.24, first-year HR 1.85, 2.1% vs. 1.5% ultrasound rates.
- Baxter et al., Thyroid 2025 — abstract (DOI 10.1089/thy.2024.0387).
- Silverii et al., Diabetes Obes Metab 2024 — fixed-effect MH-OR 1.52 (1.01–2.29), fragility index 1, NNH 1,349; no MTC or papillary signal.
- Duchemin et al., Diabetes Obes Metab 2025 — Peto OR 1.62 (1.08–2.44), I² 11%.
- TriNetX retrospective cohort, 2025 — PMC12582397.
- Clayman Thyroid Center White Paper, February 2026.
- American Thyroid Association MTC patient page (1–2% prevalence figure).
- National Cancer Institute SEER thyroid subtype distribution (3–4% prevalence figure).
Sources cited from secondary references (flagged for direct verification before re-publication):
Annals of Internal Medicine December 2025 systematic review and meta-analysis (48 RCTs, 94,245 participants). Top-line conclusion drawn from secondary summaries; we will replace with the primary citation on next quarterly refresh.
What we did not verify:
- Non-U.S. labels (UK, EU, Australia, Canada). Drug labels differ across jurisdictions; this page covers U.S. labeling only.
- Whether Byetta's and Adlyxin's current 2026 labels include any newer C-cell language than their original approvals indicated. Verify on DailyMed.
Frequently Asked Questions
- Does Ozempic cause thyroid cancer?
- The biggest human studies through 2026 have not shown that Ozempic (semaglutide) causes thyroid cancer in the general population. The drug carries an FDA boxed warning for thyroid C-cell tumors based on rodent studies, and it is contraindicated for people with a personal or family history of medullary thyroid cancer (MTC) or MEN2. The Pasternak BMJ 2024 study of 145,410 GLP-1 users found no real increase, and the Mayo Clinic / JAMA 2025 target-trial emulation in 351,000 adults found that the early signal was best explained by detection bias.
- Does Wegovy cause thyroid cancer?
- Wegovy carries the same FDA boxed warning as Ozempic because both contain semaglutide. The warning is mostly about a rare cancer called medullary thyroid carcinoma. The general human evidence does not show a real increase in thyroid cancer for Wegovy users without an MTC or MEN2 history. If you have either, Wegovy is contraindicated.
- Does Mounjaro or Zepbound have a thyroid cancer warning?
- Yes. Both Mounjaro and Zepbound contain tirzepatide and both carry an FDA boxed warning for thyroid C-cell tumors. Both are contraindicated for people with a personal or family history of medullary thyroid carcinoma or MEN2. Human evidence for tirzepatide specifically is more limited than for semaglutide because tirzepatide has been on the market for less time, but the trial data so far is consistent with the broader GLP-1 evidence.
- Is the GLP-1 thyroid cancer warning about all thyroid cancers?
- No. The warning is mostly about medullary thyroid carcinoma (MTC) and MEN2. MTC is a rare cancer that comes from the thyroid’s C-cells and accounts for only 1–4% of all thyroid cancers. The more common types — papillary, follicular, and Hürthle-cell — come from different cells and are not what the warning is about.
- Can I take a GLP-1 if my mom had thyroid cancer?
- Probably yes, depending on the type. Most thyroid cancers in the U.S. are papillary, which is not the FDA contraindication. The contraindication is specifically for medullary thyroid carcinoma or MEN2. Find out the exact pathology subtype your mom had. Look for the word “medullary” — that’s the one that matters. If it was papillary, follicular, or Hürthle-cell, the MTC/MEN2 contraindication does not appear to apply by current U.S. label criteria.
- What thyroid cancer symptoms should I watch for on Ozempic, Wegovy, Mounjaro, or Zepbound?
- The FDA-required Medication Guide for every boxed-warning GLP-1 names four symptoms: a lump or swelling in the neck, persistent hoarseness, trouble swallowing, or shortness of breath. These aren’t specific to GLP-1 use — they’re general thyroid mass symptoms. But any of them deserves a phone call to your prescriber if it lasts more than two weeks or gets worse.
- Should I get a thyroid ultrasound or calcitonin test before starting a GLP-1?
- Not as a routine pre-flight check. The FDA labels themselves say routine calcitonin or thyroid ultrasound is “of uncertain value” and may lead to unnecessary procedures. Targeted testing makes sense if your family history is unclear, you have an existing nodule, you’ve previously had an elevated calcitonin, or your prescriber has a specific reason. For most people with no thyroid history and no symptoms, baseline screening isn’t recommended.
- Does Hashimoto’s or hypothyroidism disqualify me from a GLP-1?
- No. Hashimoto’s and hypothyroidism are not the FDA contraindication. The contraindication is for medullary thyroid carcinoma or MEN2 — different cells, different category. You should disclose your thyroid history to your prescriber, and if you take levothyroxine, your TSH may need re-checking once your weight stabilizes. But neither condition disqualifies you.
- Does the new oral Foundayo pill have the same thyroid warning?
- Yes, and no. Foundayo carries the boxed warning for thyroid C-cell tumors and the MTC/MEN2 contraindication. But its rodent data is different from the rest of the class — the Foundayo label specifically says orforglipron did not produce tumors in rodents. The FDA still required the warning as a class precaution, and the agency required Lilly to run a 15-year MTC registry-based case series (final report due August 2043).
- Is the GLP-1 thyroid cancer warning based on humans or animals?
- For most products in the class, animals. The boxed warning was placed because rats and mice given long-acting GLP-1 drugs developed thyroid C-cell tumors at clinically relevant doses. The same labels say the human relevance “has not been determined.” Foundayo (orforglipron) is the exception — its label says it did not produce tumors in rodents but still carries the contraindication as a class precaution. Human studies through 2026 — totaling over half a million GLP-1 users across multiple countries — have generally not found a real increase in thyroid cancer for the general population.
- Does compounded semaglutide or tirzepatide have the same thyroid warning?
- The thyroid concern travels with the active ingredient. Compounded semaglutide is the same molecule as Ozempic and Wegovy; compounded tirzepatide is the same molecule as Mounjaro and Zepbound. A compounding pharmacy mixing the drug doesn’t remove the C-cell signal. Bring the same MTC/MEN2 family and personal history to your prescriber regardless of whether the product is brand-name, compounded, or an unapproved version.
The Honest Summary
The FDA boxed warning is real. It's specific to medullary thyroid carcinoma and MEN2. For most GLP-1s, it came from rodent data, not human data — and Foundayo's label says even the rodent data was clean for that drug.
Human studies through 2026 — including the Pasternak Scandinavian cohort, the Baxter international cohort, the Mayo Clinic / JAMA target-trial emulation, and the largest randomized-trial meta-analysis — do not show a real overall increase in thyroid cancer for the general population. Two smaller randomized-trial meta-analyses found small fragile rare-event signals; neither found a signal for MTC or papillary cancer specifically. The early French case-control signal is best explained by detection bias.
If you have a personal or family history of MTC or MEN2, every boxed-warning GLP-1 is off the table. Talk to your endocrinologist about alternatives.
Most people who searched this page do not need to be afraid. They need to find out what kind of thyroid cancer was actually in their family. That single question — was it medullary? — ends most of the uncertainty on this topic.