GLP-1 Receptor Agonists and Thyroid Medullary Cancer

Rodent C-cell tumor data

Pre-clinical studies of GLP-1 receptor agonists, including liraglutide, semaglutide, and tirzepatide, showed dose-related thyroid C-cell tumors in rodents. The findings were consistent across molecules in the class and across rodent species (rats and mice)¹. C-cells (parafollicular cells) are calcitonin-producing cells in the thyroid that express the GLP-1 receptor at higher density in rodents than in humans.

The proposed mechanism is direct GLP-1 receptor activation of rodent C-cells, leading to chronic calcitonin elevation, C-cell hyperplasia, and eventual neoplastic transformation. This mechanism appears to be species-specific: humans have substantially fewer GLP-1 receptors on C-cells, and human C-cell calcitonin response to GLP-1 agonists has not been demonstrated.

Human evidence

Human evidence for MTC risk has been examined in several ways. The LEADER cardiovascular outcomes trial (liraglutide vs placebo, 9,340 participants, median 3.8 years) showed no significant difference in MTC or other thyroid cancer incidence². Pooled analyses of GLP-1 trials including SUSTAIN, REWIND, and others have similarly not detected an MTC signal³.

Observational studies have produced mixed results. A 2022 French case-control study using national health data reported a relative risk of 1.78 for thyroid cancer (any type) in GLP-1 users⁴. However, the study was limited by short follow-up, possible detection bias (GLP-1 users have more contact with the healthcare system and may receive more imaging), and lack of differentiation between cancer subtypes. A 2023 U.S. cohort study using Department of Veterans Affairs data did not find elevated thyroid cancer risk⁵.

The FDA reviewed the post-marketing data on multiple occasions and has not modified the boxed warning beyond its original 2010 issuance for liraglutide. The European Medicines Agency reached similar conclusions.

Multiple endocrine neoplasia type 2 (MEN-2)

MEN-2 is an autosomal dominant inherited syndrome caused by RET proto-oncogene mutations. Patients with MEN-2 have a near-100% lifetime risk of medullary thyroid cancer, and many also develop pheochromocytoma and hyperparathyroidism. Prevalence is approximately 1 in 30,000 in the general population.

The GLP-1 boxed warning specifically lists MEN-2 as a contraindication because patients with this syndrome already have elevated MTC risk independent of any GLP-1 exposure. Whether GLP-1 agonists increase MTC risk in MEN-2 patients specifically is not known and would be difficult to study given the rarity of the condition; the contraindication is precautionary.

Calcitonin screening

Routine calcitonin screening before or during GLP-1 therapy is not recommended in patients without a family history of MTC or MEN-2. Baseline elevated calcitonin levels are nonspecific and frequently lead to additional workup that has not been shown to improve outcomes. The number-needed-to-screen to detect a single MTC case in low-risk patients is impractically high.

Patients with a family history of MTC or MEN-2, or with personal symptoms suggestive of MTC (palpable thyroid nodule, neck mass, hoarseness, dysphagia, calcitonin-mediated diarrhea, or facial flushing), should undergo evaluation with a thyroid ultrasound and calcitonin level before considering GLP-1 therapy.

Other thyroid neoplasms

Papillary thyroid carcinoma (the most common thyroid cancer) and follicular thyroid carcinoma have a different cell of origin (follicular cells, not C-cells) and a different mechanism. The boxed warning does not extend to these subtypes, and large datasets have not identified an elevated risk.

Thyroid nodules are common in the general population (palpable in approximately 4 to 7% of adults, sonographically visible in up to 50%). The presence of nodules at baseline is not a contraindication to GLP-1 therapy in patients without MTC family history or concerning ultrasound features. Patients with newly identified nodules during GLP-1 therapy should be evaluated according to standard thyroid nodule guidelines, not differently because of GLP-1 use.

Clinical decision-making

For most patients, the GLP-1 boxed warning should be discussed but does not preclude treatment. Inform patients that the warning is based on rodent data with uncertain human translation; that human studies have not demonstrated elevated MTC risk; and that the absolute risk, even if it exists, is small given the rarity of MTC.

Patients with the following warrant additional consideration before initiating: personal or family history of MTC, known MEN-2 (absolute contraindication per labeling), known RET mutation, palpable thyroid nodule with sonographic suspicious features, or unexplained calcitonin elevation. In these cases, alternative weight-management therapies should be considered first.

Routine ongoing calcitonin or ultrasound monitoring during GLP-1 therapy is not recommended in standard-risk patients. Symptomatic concerns (new neck mass, persistent hoarseness, dysphagia) warrant evaluation at any point.

Related editorial coverage

• Semaglutide peptide profile
• Tirzepatide peptide profile
• GLP-1 contraindications and precautions
• GLP-1 side effects detailed reference

Frequently asked questions

Does semaglutide cause thyroid cancer?

Rodent studies showed C-cell tumors with semaglutide and other GLP-1 agonists, but human studies have not demonstrated elevated medullary thyroid cancer (MTC) risk. The FDA boxed warning is precautionary based on the rodent data and is not based on human clinical evidence²⁵.

Can I take semaglutide if I have thyroid nodules?

Yes, generally. Thyroid nodules without features of medullary thyroid cancer (palpable hard mass, suspicious ultrasound findings, family history of MTC) are not a contraindication to GLP-1 therapy. Standard thyroid nodule evaluation should be completed regardless of GLP-1 status.

Should I get my calcitonin level checked?

Routine calcitonin screening before or during GLP-1 therapy is not recommended for low-risk patients. Patients with a family history of MTC or MEN-2, or with thyroid symptoms, warrant evaluation. The predictive value of routine screening in average-risk patients is low.

Is the boxed warning class-wide or specific to one molecule?

The boxed warning applies to multiple GLP-1 receptor agonists including semaglutide, tirzepatide, liraglutide, and dulaglutide. The mechanism is hypothesized to be class-level (GLP-1 receptor activation on C-cells in rodents). Short-acting agents like exenatide do not carry the same boxed warning because their pre-clinical C-cell tumor signal was less consistent.