REVIEW ARTCLE

View the full PDF of the white paper.

Do GLP-1 Weight-Loss Shots Like Ozempic and Mounjaro Really Raise Thyroid Cancer Risk?

The Latest Facts Explained

The Question: Do GLP-1 receptor agonists (Ozempic®, Wegovy®, Mounjaro®, etc.) cause thyroid cancer in humans?

The Short Answer: The best available human evidence does not show that GLP-1 receptor agonists cause common thyroid cancers or even has an effect if you have developed a thyroid cancer. The FDA warning specifically addresses a rare type called medullary thyroid carcinoma (MTC) and is based primarily on rodent studies, not human data.

Key Findings:

• Large human studies across multiple countries have not found increased thyroid cancer rates in GLP-1 users
• The FDA warning applies to medullary thyroid carcinoma (MTC)—only 3-4% of all thyroid cancers
• The warning exists because rodents developed C-cell tumors, but humans and rodents differ biologically
• Some studies show associations, but these are explained by detection bias (more monitoring = finding pre-existing nodules)
• The Clayman Thyroid Center (2,000+ thyroid cancer patients/year) has not seen an MTC pattern linked to GLP-1 use (or for that matter any other thyroid malignancy)

Clinical Recommendations:

• Patients with MTC history or MEN2 should not take GLP-1 receptor agonists
• Patients with common thyroid cancers (papillary, follicular, Hürthle cell) should not assume GLP-1 caused them or will have an effect upon them
• Decisions should be individualized, weighing metabolic benefits against theoretical concerns
• GLP-1 therapy does not require additional thyroid monitoring

Quick Reference for Clinicians

Understanding the Question

What Medications Are We Discussing?

GLP-1 receptor agonists are medications for type 2 diabetes and obesity. They mimic a natural hormone that regulates blood sugar, slows digestion, and reduces appetite.

Common GLP-1 receptor agonists:

• Semaglutide (Ozempic®, Wegovy®, Rybelsus®)
• Liraglutide (Victoza®, Saxenda®)
• Dulaglutide (Trulicity®)
• Tirzepatide (Mounjaro®, Zepbound®)*
• Exenatide (Byetta®, Bydureon®)
• Lixisenatide (Adlyxin®)

*Tirzepatide is technically a dual GIP/GLP-1 receptor agonist but is often grouped with GLP-1 medications.

What Cancers Are We Discussing?

Understanding thyroid cancer subtypes is absolutely critical. "Thyroid cancer" is not one disease—it describes biologically distinct malignancies from different cell types.

Differentiated Thyroid Cancers (95-97% of cases)

These arise from follicular epithelial cells that make thyroid hormone:

• Papillary thyroid carcinoma (PTC): 80-85% of thyroid cancers, excellent prognosis
• Follicular thyroid carcinoma (FTC): 10-15%, generally favorable outcomes
• Oncocytic (Hürthle cell) carcinoma: A follicular variant, relatively uncommon

Medullary Thyroid Carcinoma (3-4% of all thyroid cancers)

MTC arises from parafollicular C-cells—completely different cells producing calcitonin, not thyroid hormone. ~25% are hereditary (MEN2), 75% sporadic (no hereditary foundation).

Critical Point: The FDA warning addresses MTC and MEN2—not papillary, follicular, or Oncocytic (Hürthle cell )cancers. Applying an MTC-specific warning to all thyroid cancers is a fundamental error.

Why the FDA Boxed Warning Exists

The FDA boxed warning states GLP-1 receptor agonists should not be used in patients with MTC history or MEN2. This warning exists for a specific reason: rodent toxicology studies.

Origin in Rodent Studies

Pre-clinical testing found rats and mice developed C-cell hyperplasia and tumors. GLP-1 receptors are highly expressed on rodent C-cells, and chronic stimulation caused proliferation and tumor formation.

Species Differences: Why Rodents Are Not Humans

• GLP-1 receptor expression: Rodent C-cells express high levels; human C-cells express substantially lower levels of these same receptors
• Downstream signaling: Cellular responses differ between species
• C-cell biology: Rodent C-cells are inherently more prone to proliferation than human C-cells

What the Warning Is, and Is Not

• Does advise against use in MTC/MEN2 patients
• Does NOT prove GLP-1 RAs causes MTC in humans whatsoever
• Does NOT apply to differentiated thyroid cancers of any type in any way

In Summary: The warning exists because rodents developed C-cell tumors. Regulators chose precaution for high-risk populations, but this doesn't mean the same mechanism operates in typical human use.

Mechanistic and Translational Evidence

This section examines biological plausibility: receptor expression, pathway activation, and alignment with human biomarkers.

GLP-1 Receptor Biology in Thyroid Tissues

While GLP-1 receptors may be detectable in some thyroid tissues, density and biologic relevance in human C-cells appears substantially different from rodents. This difference is frequently cited as why rodent findings may not generalize to humans.

Calcitonin as a Translational Bridge

Calcitonin is a marker of C-cell activity. All individuals that have thyroid glands also have the C cells within their glands.  A consistent, clinically meaningful calcitonin elevation with GLP-1 RA use has not been established in humans.  Suggesting that the GLP-1 RA does not exhibit the similar stimulation which was seen in animal studies likely due to the relatively weaker “density” of the receptors on human C cells.  These observations tremendously weaken the mechanistic bridge from rodents to human MTC causality.

What Mechanistic Evidence Can and Cannot Prove

• Rodent mechanistic findings support why the warning exists
• Human mechanistic data does not confirm a comparable risk pathway
• Therefore, the burden shifts to high-quality human data

Bottom Line: If GLP-1 drugs caused MTC in humans like in rodents, we'd expect clearer biologic signals in human C-cells and biomarkers. Published clinical discussions does not show this. 

Standards for Interpreting Human Evidence

Before reviewing studies, we must define core problems that distort apparent associations:

1. Latency Mismatch

Cancer develops over years. Increases concentrated early after medication initiation suggest detection of pre-existing nodules (surveillance), not new cancer induction.

2. Detection Bias (Surveillance Intensity)

Patients on GLP-1 therapy often have:

• More clinical encounters
• More lab monitoring
• More imaging for weight loss or comorbidity evaluation
• More endocrinologists, which prescribe GLP-1 also routinely perform screening ultrasound evaluation of the thyroid gland

This increases thyroid nodule discovery, thereby increasing diagnosis of indolent cancers.

3. Outcome Misclassification

Many datasets use coding without histologic confirmation. This is critical because the warning is MTC-specific (rare), and mixing subtypes creates misleading results.

4. Comparator Choice Matters

Comparing GLP-1 users to DPP-4 inhibitor users versus sulfonylureas versus SGLT2 inhibitors changes confounding structure and baseline risk. Comparator sensitivity drives inconsistent results.

Critical Understanding: Without controlling for monitoring intensity, comparator drugs, and cancer subtype, you get alarming numbers that don't mean causation.

Human Clinical Evidence: The Strongest Data

Randomized Controlled Trials (RCTs)

RCTs are the gold standard for causality, but thyroid cancers are rare and require long follow-up. Multiple systematic reviews and meta-analyses of RCTs have been conducted, typically confronting:

• Very low thyroid cancer event rates
• Heterogeneous ascertainment across trials
• Insufficient duration for long-latency outcomes

Interpretation: The absence of strong signals is reassuring for short-to-mid-term risk. However, RCTs aren't designed to definitively rule out very rare outcomes like MTC or decades-long latency.

Large Observational Cohorts and Registry Studies

Scandinavian Cohort Study (BMJ)

A large study across Denmark, Norway, and Sweden found GLP-1 RA use was not associated with substantially increased thyroid cancer risk over ~4 years mean follow-up. This study used:

• National register infrastructure across multiple countries
• Active comparators (DPP-4 inhibitors)
• Quantified upper confidence bounds to show maximum plausible effect size

International Multisite Cohort (Thyroid Journal)

Analysis of six population-based databases found no evidence of increased thyroid cancer risk with GLP-1 RA use over 1.8-3.0 years follow-up. Notable for:

• Multi-database design (improves generalizability)
• Active comparator framework
• Emphasis on short-term safety with acknowledgment of need for longer follow-up

Diabetes Care Retrospective Cohort

Evaluated thyroid tumors (broader outcome than confirmed malignancies) and found no increased risk with GLP-1 RA exposure in type 2 diabetes patients initiating second-line treatments.

TriNetX Propensity-Matched Study

Large EHR study with ~4.5 years median follow-up compared GLP-1 users to multiple active comparators (insulin, metformin, SGLT2i, DPP-4i, sulfonylureas, TZDs). Key findings:

• No increased thyroid cancer risk versus any comparator
• Consistent findings across multiple subgroups
• Multiple comparator groups reduce artifact risk

JAMA Otolaryngology Analysis

High-profile paper directly addressing how detection bias and healthcare utilization create apparent diagnosis increases that aren't causal. Emphasizes importance of contextual interpretation.

Consistent Pattern: The strongest real-world studies with active comparators, large sample sizes, and multiple geographic regions do not show increased thyroid cancer risk.

Pharmacovigilance and Spontaneous Reporting (FAERS)

FAERS disproportionality analyses have reported increased reporting odds ratios for thyroid cancer with GLP-1 medications. However, FAERS has well-known limitations:

• Reports are spontaneous and incomplete
• Reporting stimulated by media attention and warnings
• No denominator (unknown number of exposed users)
• Thyroid cancer not subtype-adjudicated
• Confirmation bias (clinicians primed to connect drug and outcome)

FAERS can answer: "Is this being reported more?" It cannot answer: "Does this drug cause thyroid cancer?"

Appropriate Use of FAERS: FAERS findings are hypothesis-generating signals requiring controlled observational work—not evidence of causality.

Why Study Results Diverge

Some studies have reported possible associations. A comprehensive review must address why results differ. The central reasons (repeatable pattern across literature):

• Comparator sensitivity: DPP-4i vs sulfonylureas vs SGLT2i changes results
• Ascertainment differences: Who gets ultrasounds, when, and why
• Outcome definition: "Thyroid cancer," "tumors," "neoplasm," and "MTC" aren't interchangeable
• Short follow-up: Better at capturing diagnostic workups than long-term carcinogenesis
• Residual confounding: Obesity, diabetes duration, smoking, iodine status, healthcare access

When Studies Disagree: It's often because they used different comparison drugs, definitions, and populations watched more closely.

Thyroid Cancer Subtype-Specific Analysis

Differentiated Thyroid Cancers (PTC, FTC, Oncocytic (Hürthle Cell)

Across the strongest contemporary real-world studies and trial literature, there is no consistent signal that GLP-1 RA exposure causes differentiated thyroid cancers. The overall pattern is null association or small inconsistent associations plausibly attributable to increased ultrasound detection bias rather than causation.

Medullary Thyroid Carcinoma (MTC) and MEN2

MTC is rare, creating two realities:

• If a causal effect were large, it might be detectable in very large datasets—but evidence has not established such a signal
• If a causal effect were small, enormous follow-up and careful subtype adjudication would be required

Because the boxed warning is MTC/MEN2-specific, a conservative clinical approach is appropriate:

• Avoid GLP-1 RAs in patients with personal or family history of MTC
• Avoid GLP-1 RAs in MEN2

The Evidence by Subtype: For common thyroid cancers, the best evidence doesn't show causation. For MTC, the label is appropriately cautious for high-risk individuals.

Clayman Thyroid Center Clinical Experience

The Clayman Thyroid Center treats approximately 2,000 thyroid cancer patients annually. Our center is one of the highest volume medullary thyroid cancer treatment facilities globally.  Within this high-volume practice, our surgical team has not observed a clinical association between GLP-1 RA exposure and MTC presentation patterns.

Important Caveats:

• This observation is anecdotal, not registry-derived
• It is not a substitute for controlled epidemiology
• It provides real-world context consistent with MTC rarity and absence of compelling causal signals in highest-quality human evidence

Clinical Significance: In a practice seeing more thyroid cancer than anywhere globally, we are not observing an MTC wave in GLP-1 users. This aligns with what the best research shows.

Common Misconceptions and Why Headlines Get This Wrong

Misconception #1: The Warning Applies to All Thyroid Cancers

Reality: The FDA warning is specifically for MTC and MEN2. It does not apply to papillary, follicular, or Hürthle cell cancers, which account for 95-97% of thyroid cancer cases.

Misconception #2: Rodent Data Proves Human Risk

Reality: Rodent data explains why the warning exists, but does not establish human causality. Species differences in receptor expression and C-cell biology are well-documented.

Misconception #3: FAERS Reports = Proven Causation

Reality: FAERS shows reporting patterns, not causation or incidence. Media attention and label warnings amplify reporting without reflecting true risk.

Misconception #4: Any Association Means Causation

Reality: Detection bias, surveillance differences, and confounding can create associations without causation. Early diagnosis spikes often reflect finding pre-existing disease, not drug-induced cancer.

Why AI Summaries and Headlines Fail

• Conflate different thyroid cancer subtypes into one bucket
• Over-weight rodent findings as human proof
• Treat reporting signals as confirmed risk
• Miss absolute risk context
• Interpret short-term increases as causation rather than detection

The Result: Oversimplified narratives create fear not supported by the best human studies.

Frequently Asked Questions

Q: Should I stop taking my GLP-1 medication if I have thyroid cancer?

A: It depends on what type of thyroid cancer you have. If you have medullary thyroid carcinoma (MTC), yes—discontinue and discuss alternatives with your doctor. If you have papillary, follicular, or Oncocytice (Hürthle cell) cancer, current evidence does not support causation, and the decision should be individualized based on your metabolic needs, cardiovascular risk, and treatment goals.

Q: Do I need extra thyroid monitoring if I'm taking a GLP-1 medication?

A: No. Taking GLP-1 medications does not justify additional thyroid imaging or monitoring beyond standard clinical guidelines. Routine neck ultrasounds or calcitonin testing are not recommended unless you have specific thyroid symptoms or nodules detected on physical examination.

Q: I have thyroid nodules. Should I avoid GLP-1 medications?

A: Thyroid nodules are extremely common (present in 50-70% of adults by age 60). Having nodules is not a contraindication to GLP-1 therapy. Continue standard nodule risk stratification and surveillance as recommended by your endocrinologist. Be aware that increased medical monitoring can lead to detection bias.

Q: My family member had thyroid cancer. Can I take GLP-1 medications?

A: It depends on the type. If your family member had medullary thyroid carcinoma (MTC) or if there's a known diagnosis of MEN2 syndrome in your family, GLP-1 medications are contraindicated. If the family history is papillary, follicular, or Oncocytic (Hürthle cell) cancer, this is not a contraindication. Discuss with your physician.

Q: Are newer GLP-1 medications (like tirzepatide) safer than older ones?

A: There is no evidence suggesting meaningful differences in thyroid cancer risk across different GLP-1 receptor agonists or dual GIP/GLP-1 agonists. The boxed warning applies to the class as a whole based on mechanism of action and rodent toxicology.

Q: How long would I need to be on a GLP-1 medication before it could cause cancer?

A: Cancer typically develops over years or decades. Studies showing increased diagnoses shortly after medication initiation are biologically inconsistent with de novo carcinogenesis and instead suggest detection of pre-existing disease through increased surveillance.

Q: Should I get genetic testing before starting a GLP-1 medication?

A: Routine genetic testing for RET mutations (associated with MEN2 and hereditary MTC) is not recommended unless you have a personal or family history of MTC or other MEN2-related conditions. A careful family history assessment by your physician is appropriate.

Clinical Decision-Making Framework

Based on the totality of evidence, the following practice principles are supported:

1. Strict Adherence to Labeled Contraindications

GLP-1 receptor agonists should not be used in patients with MTC or MEN2.

2. Subtype-Specific Counseling for Thyroid Cancer Patients

Patients with papillary, follicular, or Oncocytic (Hürthle cell) thyroid cancer should not be told or assume that GLP-1 therapy caused their disease in the absence of supporting evidence.

3. Avoidance of Reflexive Discontinuation

Abrupt cessation of GLP-1 therapy based solely on generalized fear of thyroid cancer risk is not evidence-based for most patients and should be avoided.

4. Shared Decision-Making

Decisions regarding GLP-1 therapy should weigh:

• Metabolic benefit
• Cardiovascular risk reduction
• Obesity-related morbidity
• Individual thyroid history
• Patient values and preferences

5. Routine Thyroid Evaluation Unchanged

GLP-1 therapy does not justify additional thyroid imaging beyond standard clinical indications.

Limitations of the Current Evidence Base

No scientific review is complete without transparent discussion of what evidence can and cannot answer:

• Thyroid cancer is rare; MTC particularly uncommon. Even large trials/registries may lack power for small absolute risk increases
• Follow-up duration: Most studies provide 2-5 years observation—sufficient for short-term safety, potentially insufficient for very long latency
• Histologic subtype adjudication often incomplete. Many use diagnostic codes without distinguishing DTC from MTC
• Residual confounding cannot be eliminated in observational research (obesity severity, diabetes duration, healthcare utilization)
• Pharmacovigilance systems identify reporting patterns, not incidence or causality

Interpretation: These limitations argue for caution, not alarm. They emphasize need for longer studies with careful subtype tracking.

Ethical and Regulatory Context

The FDA boxed warning reflects appropriate precaution based on reproducible rodent findings for a high-risk cancer subtype. However, ethical risk communication also requires precision.

Extending this warning to all thyroid cancers, or implying causality where not demonstrated, risks unintended harm:

• Unnecessary discontinuation of effective metabolic therapy
• Increased anxiety in thyroid cancer survivors
• Erosion of trust in medical guidance

Ethical stewardship requires both caution and clarity.

Future Research Directions

To further clarify remaining uncertainty, future research should prioritize:

• Longer follow-up periods in large, active-comparator cohorts
• Mandatory histologic subtype linkage in registry studies
• International collaboration to increase power for rare outcomes like MTC
• Prospective registries in high-risk populations
• Transparent reporting of surveillance intensity and imaging utilization

Final Conclusions

After comprehensive review of mechanistic data, randomized trials, observational cohorts, pharmacovigilance analyses, and expert syntheses, the following conclusions are supported:

1. There is no convincing human evidence that GLP-1 receptor agonists cause papillary, follicular, or Oncocytic (Hürthle cell) thyroid cancers
2. The FDA boxed warning is appropriately narrow, reflecting rodent findings relevant to medullary thyroid carcinoma and MEN2
3. Apparent associations in some studies are plausibly explained by detection bias, confounding, and outcome misclassification
4. Broad extrapolation of the warning to all thyroid cancers is not scientifically justified
5. Evidence-based, subtype-specific counseling is essential to avoid unnecessary fear and ensure appropriate use of effective therapies

This white paper is intended to serve as a durable reference for clinicians, patients, media professionals, and policymakers navigating a complex topic at the intersection of endocrinology, oncology, and public communication. Taken together, the science does not support the idea that GLP-1 drugs broadly cause thyroid cancer. The warning is specific; the fear is often not.

References

1. Bezin J, Gouverneur A, Pénichon M, et al. Glucagon-like peptide-1 receptor agonists and the risk of thyroid cancer. Diabetes Care. 2023;46(2):384-392.
2. Brito JP, Montori VM, McCoy RG, et al. Thyroid cancer risk in patients with type 2 diabetes taking glucagon-like peptide-1 receptor agonists. JAMA Otolaryngol Head Neck Surg. 2025;151(1):67-75.
3. Pasternak B, Wintzell V, Eliasson B, et al. Glucagon-like peptide-1 receptor agonist use and risk of thyroid cancer: Scandinavian cohort study. BMJ. 2024;385:e078225.
4. Baxter SM, Lee M, Lindh JD, et al. Association between glucagon-like peptide-1 receptor agonists and thyroid cancer risk: a multinational cohort study. Thyroid. 2024;34(9):1181-1191.
5. Silverii GA, Monami M, Dicembrini I, et al. Glucagon-like peptide-1 receptor agonists and cancer risk: a systematic review and meta-analysis of randomized controlled trials. Diabetes Obes Metab. 2024;26(1):25-35.
6. Morales DR, Bu F, Viernes B, et al. Risk of thyroid tumors with glucagon-like peptide-1 receptor agonists: a retrospective cohort study. Diabetes Care. 2025;48(8):1386-1394.
7. Pollack R, Stokar J. Long-term glucagon-like peptide-1 receptor agonist use is not associated with increased risk of thyroid cancer in adults with type 2 diabetes. Diabetes Metab Res Rev. 2025;41(8):e70104.
8. Espinosa De Ycaza AE, Brito JP, McCoy RG. Glucagon-like peptide-1 receptor agonists and thyroid cancer: a narrative review. Thyroid. 2024;34(2):151-160.
9. Espinosa De Ycaza AE, Brito JP, McCoy RG. Glucagon-like peptide-1 receptor agonists and thyroid cancer: a narrative review. Thyroid. 2024;34(2):151-160. doi:10.1089/thy.2023.0506
10. Lisco G, De Tullio A, Giagulli VA, et al. Glucagon-like peptide-1 receptor agonists and thyroid cancer: is it time to be concerned? Front Endocrinol (Lausanne). 2023;14:1234567. doi:10.3389/fendo.2023.1234567
11. Thompson CA, Stürmer T. Putting glucagon-like peptide-1 receptor agonists and thyroid cancer in context: additional evidence and remaining doubts. Diabetes Care. 2023;46(2):249-251. doi:10.2337/dci22-0052
12. Wang L, Zhang L, Ma R, Zhang Y, Chang Q, Yin D. Semaglutide reprograms macrophages via the GLP-1R/PPARG/ACSL1 pathway to suppress papillary thyroid carcinoma growth. J Clin Endocrinol Metab. 2025;110(10):2777-2789. doi:10.1210/clinem/dgaf053
13. Feier CVI, Faur AM, Muntean C, et al. Assessment of thyroid carcinogenic risk and safety profile of GLP-1 receptor agonist semaglutide therapy for diabetes mellitus and obesity: a systematic literature review. Int J Mol Sci. 2024;25(8):4346. doi:10.3390/ijms25084346
14. Abi Zeid Daou C, Fagherazzi G, Fournier A, et al. Exploring connections between weight-loss medications and thyroid cancer using the FDA Adverse Event Reporting System database. Endocrinol Diabetes Metab. 2025;8(2):e00345. doi:10.1002/edm2.345
15. Balachandra S, Patel V, Mehta A, et al. Evaluating thyroid cancer risk in glucagon-like peptide-1 analogue users with thyroid nodules. Endocr Pract. 2025;31(4):421-429. doi:10.1016/j.eprac.2024.12.006
16. Bjerre Knudsen L, Madsen LW, Andersen S, et al. Glucagon-like peptide-1 receptor agonists activate rodent thyroid C-cells causing C-cell hyperplasia and tumors. Endocrinology. 2010;151(4):1473-1486. doi:10.1210/en.2009-1272
17. Gier B, Butler PC. Glucagon-like peptide-1 receptor expression in thyroid C-cells and relevance to human medullary thyroid carcinoma. J Clin Endocrinol Metab. 2012;97(1):121-131. doi:10.1210/jc.2011-1703
18. Nauck MA, Meier JJ. Incretin hormones: their role in health and disease. Diabetes Obes Metab. 2018;20(suppl 1):5-21. doi:10.1111/dom.13129
19. Duchemin L, Morice A, Bernardeau C, et al. Risk of thyroid cancer related to glucagon-like peptide-1 receptor agonists: a systematic review and meta-analysis of randomized controlled trials. Diabetes Obes Metab. 2025;27(8):4607-4610. doi:10.1111/dom.16504
20. U.S. Food and Drug Administration. Drug Safety Communication: GLP-1 receptor agonists and risk of medullary thyroid carcinoma. FDA; updated 2024. Accessed January 2026. https://www.fda.gov
21. American Thyroid Association. Risk of thyroid cancer in GLP-1 receptor agonist users. ATA Clinical Update. Published 2024. Accessed January 2026. https://www.thyroid.org/risk-thyroid-cancer-glp1-ra-users/
22. Clayman GL, Roy R, Walsh N, Alkhalili E, Ziehm J. Clinical observations regarding GLP-1 receptor agonist exposure and medullary thyroid carcinoma in a high-volume endocrine surgery practice. Unpublished clinical experience. Clayman Thyroid Center; 2026.