The FDA includes a Boxed Warning — its strongest required disclosure — on long-acting GLP-1 receptor agonists because rodent studies found C-cell tumors in the thyroid gland. Whether this finding applies to people has not been established. "Not established" is not the same as "definitively safe," and that distinction is what this article is here to clarify.
This article explains what the FDA Boxed Warning is and what the available evidence shows. It is not a basis for starting, stopping, or changing any medication. Those decisions belong to you and your prescribing provider.
What the FDA Boxed Warning Actually Says
A Boxed Warning is the FDA's most serious required disclosure — reserved for risks identified in preclinical or clinical data that warrant explicit communication before prescribing. Every long-acting GLP-1 receptor agonist currently approved in the United States carries a Boxed Warning about thyroid C-cell tumors.
The warning discloses that in animal studies, thyroid C-cell tumors — including both benign adenomas and malignant carcinomas — were found in animals given GLP-1 receptor agonists at doses substantially higher than standard therapeutic levels over long durations. The FDA label states that the relevance of these findings to humans has not been determined. Healthcare providers are required to discuss this risk with patients before prescribing.
The warning also establishes specific contraindications: GLP-1 receptor agonists are contraindicated in patients who have a personal or family history of medullary thyroid carcinoma (MTC), or in patients who have Multiple Endocrine Neoplasia type 2 (MEN2). For people with these specific histories, the combination of elevated baseline risk and uncertain rodent signal places these medications outside current FDA labeling.
A Boxed Warning does not mean a drug is categorically unsafe for all patients. It means a specific identified risk requires explicit disclosure, and that certain patient histories change the prescribing calculus entirely.
The FDA Boxed Warning requires disclosure and establishes contraindications based on rodent data. It does not state that GLP-1 medications have been shown to produce thyroid cancer in people. Understanding the difference between "not established in humans" and "confirmed safe" is what this article is for.
C-Cells and Why Medullary Thyroid Cancer Is Different
The thyroid gland contains two functionally distinct cell populations. Follicular cells — the large majority — produce thyroxine and triiodothyronine, the thyroid hormones most people associate with metabolism and energy regulation. C cells, also called parafollicular cells, are a smaller population with a different job: they produce calcitonin, a hormone involved in calcium homeostasis.
Medullary thyroid cancer (MTC) originates specifically from C cells. This makes it biologically distinct from the more common thyroid cancers, which arise from follicular cells. According to the American Cancer Society, MTC accounts for less than 5% of all thyroid cancers. Papillary thyroid cancer — the most frequently diagnosed form — accounts for approximately 8 in 10 thyroid cancer cases, and arises from an entirely different cell type.
This distinction matters for interpreting the Boxed Warning. When the label discusses thyroid C-cell tumor risk, it is describing a specific and uncommon cancer subtype — not thyroid cancer as a broad category. A reader who processes "GLP-1 and thyroid cancer" as a statement about common thyroid malignancies is working from a different frame than the one the warning actually describes.
Because MTC originates in C cells, calcitonin — the hormone those cells produce — serves as a clinically relevant biomarker for MTC. Elevated calcitonin may indicate abnormal C-cell activity. This is biologically separate from thyroid-stimulating hormone (TSH) and the standard thyroid function tests most patients are familiar with.
What the Rodent Studies Found — and What They Did Not
GLP-1 receptor agonists bind to the GLP-1 receptor, which is expressed on various cells throughout the body. Thyroid C cells are among them. In rodent studies — primarily conducted in rats and mice over extended durations at high doses — researchers observed thyroid C-cell hyperplasia (abnormal cell proliferation) and tumor formation, including carcinomas. The mechanism appears to be GLP-1 receptor-mediated: prolonged stimulation of the GLP-1 receptor on C cells, at doses substantially exceeding standard clinical levels, was linked to these cellular changes.
A 2024 systematic literature review by Feier C, Vonica R, Faur A, Streinu D, and Muntean C, published in the International Journal of Molecular Sciences, reported that long-term exposure to GLP-1 receptor agonists in rodents was linked to C-cell hyperplasia and tumor formation through a receptor-mediated mechanism, while similar findings were not observed in primates.
Two features of the rodent data limit direct translation to human relevance.
First, the doses used in rodent toxicology studies were substantially higher than the levels used in clinical treatment. Preclinical safety studies are typically designed to test at dose extremes in order to identify hazard potential — this is standard in drug development and does not necessarily predict what happens at therapeutic exposures.
Second, and more mechanistically important: the density of GLP-1 receptors expressed on thyroid C cells differs between rodents and primates. Rodent thyroid C cells appear to carry GLP-1 receptors at notably higher density than human thyroid C cells. This biological difference may explain why the C-cell tumor signal appeared in rodents but not in primate studies — a finding noted in both the scientific literature and reflected in the FDA's own acknowledgment that the relevance of rodent findings to humans has not been determined.
What the rodent studies did not do is establish that GLP-1 receptor agonists produce thyroid cancer in people. The signal from rodents is the reason the Boxed Warning exists; the absence of the signal in primates and in human data so far is the reason the warning says "not determined" rather than "confirmed risk."
What Human Data Has and Has Not Established
Human clinical data on GLP-1 receptor agonists and thyroid cancer have been generated through randomized controlled trials and real-world epidemiological studies, and the picture they present differs from the rodent findings.
The 2024 systematic review by Feier et al., analyzing 10 randomized controlled trials with 14,550 total participants — including 7,830 who received semaglutide — found that the rate of thyroid cancer among participants in the semaglutide-treated groups was below 1%, with no pattern suggesting a significant elevation in risk.
Large real-world database studies have similarly not shown a statistically significant increase in thyroid cancer risk among people taking GLP-1 receptor agonists compared to those on other medications used for similar indications. The European Medicines Agency, following a review of available evidence, concluded that there was no established causal relationship between GLP-1 receptor agonists and thyroid cancer.
This is not the same as a definitive conclusion that there is no risk. Several important caveats apply.
Most cardiovascular outcome trials and safety studies were not statistically powered to detect rare cancer outcomes. MTC is uncommon enough — less than 5% of thyroid cancers — that even a moderately elevated relative risk might not be detectable in a trial of tens of thousands of patients with years of follow-up. The follow-up periods in most trials also may not capture a slow-developing malignancy signal.
Longer-term observational data are still accumulating. The FDA Boxed Warning remains in place because the uncertainty has not been resolved, not because a risk has been confirmed.
"Available human data have not confirmed the rodent signal" is a precise statement. Claiming that GLP-1 medications carry no thyroid risk at all is a different statement — and drawing that conclusion from the current evidence would outrun what the evidence actually supports.
Calcitonin: What It Is and Why It Comes Up
Calcitonin is a hormone produced by thyroid C cells. In routine medical care, most people are familiar with thyroid-stimulating hormone (TSH) and T4 measurements as part of standard thyroid function testing. Calcitonin is different: it is not part of a standard thyroid panel for most people, but it functions as a specific biomarker for C-cell activity and MTC.
Because MTC arises from calcitonin-producing cells, elevated calcitonin levels can be an indicator of abnormal C-cell activity. A provider evaluating a patient for MTC, or monitoring someone with known C-cell concerns, would look at calcitonin as a meaningful signal in that clinical context.
Some clinical discussions around initiating GLP-1 receptor agonist therapy have included reference to baseline calcitonin measurement — primarily in patients where C-cell risk is clinically relevant. Whether routine calcitonin monitoring adds value for the general population of people starting GLP-1 receptor agonists is a question that remains under review in clinical guidelines and has not been universally adopted as a standard protocol. The decision to check calcitonin, and what to do with the result, belongs in a conversation with a prescribing provider who can evaluate the full clinical picture.
A calcitonin result does not independently diagnose MTC. There are benign causes of mild calcitonin elevation, and a result outside the reference range requires clinical interpretation — not a self-driven conclusion.
Who the Label Says Should Not Take These Medications
The FDA label identifies two specific categories of patients for whom GLP-1 receptor agonists are contraindicated, specifically because of the thyroid C-cell tumor concern from rodent data:
People with a personal history of medullary thyroid carcinoma (MTC), and people with a personal or family history of Multiple Endocrine Neoplasia type 2 (MEN2) syndrome.
MEN2 is a rare hereditary condition that markedly elevates the lifetime risk of MTC. It is caused by mutations in the RET proto-oncogene and can be identified through genetic testing. Because MEN2 markedly raises the baseline probability of C-cell malignancy, adding a drug with an uncertain C-cell signal is not considered acceptable under current FDA labeling for this population.
A personal history of MTC — meaning the patient has already been diagnosed with this cancer type — similarly places the patient outside the indicated use, given the rodent signal and the direct relevance of C-cell biology to that history.
These contraindications are evaluated by a prescribing provider before initiating therapy, as part of standard medical history-taking. This is not an assessment patients make for themselves. A prescriber reviewing whether a GLP-1 receptor agonist is appropriate will screen for the histories that fall under these categories.
For people without these specific histories, the label's contraindication does not apply — though the Boxed Warning disclosure remains part of the prescribing conversation for all patients.
MTC is rare, which is why most people taking GLP-1 medications will not develop it. But rare is not the same as impossible, and that is precisely why the FDA maintains the Boxed Warning and why providers screen for relevant history before prescribing. A warning that covered only common outcomes would not need to be a Boxed Warning.
For guidance on discussing any side effect concerns with your provider — including when to reach out and what to report — our guide on GLP-1 side effects and when to contact a provider covers that territory. If you are preparing for a first conversation with a prescriber about starting a GLP-1 medication, questions to bring to that appointment can help you prepare.
Using our GLP-1 side-effect and progress tracker gives you a way to keep a structured record of anything you notice — so your provider has the context they need at your next appointment. Start your log here.
This article is for informational and educational purposes only and is not intended as medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider before making decisions about your health, medications, or any concerns related to thyroid health. The GLP-1 Journal does not provide medical guidance and cannot advise you on your individual health situation.
Written by Joon, a health writer managing type 2 diabetes and hypertension, for The GLP-1 Journal.