Thyroid and GLP-1: what is known

Why there is a warning at all

If you have read the American labelling for these medicines you will have seen a boxed warning about thyroid C-cell tumours. It is the single most frightening thing in the leaflet, it arrives with no context, and it sends a steady stream of people into our thyroid circle at two in the morning.

Here is where it comes from. In rodent toxicology studies, long-acting GLP-1 receptor agonists caused thyroid C-cell tumours — including medullary thyroid carcinoma — in rats and mice, in a dose-dependent and duration-dependent way. That is a real finding in animals and it is why the warning exists.

The reason it may not translate is also real. Rodent thyroid C-cells carry a far higher density of GLP-1 receptors than human C-cells do, and the pathway that produced tumours in rodents does not appear to be similarly active in human thyroid tissue. That is the standard explanation and it is well supported by receptor studies, but it is an argument from mechanism, not a demonstration of safety in people.

Mira’s framing: the warning is a statement about what happened in rats and about what regulators do with an unresolved question. It is not a statement about what has been shown to happen in humans, in either direction.

What has been seen in humans so far

Medullary thyroid carcinoma is a rare cancer. That rarity is the whole problem, because a trial large enough to detect a change in the rate of something that rare would need to be enormous and long, and no such trial exists.

The large cardiovascular outcome trials — LEADER for liraglutide, SUSTAIN-6 for semaglutide — followed thousands of people for years and reported thyroid adverse events in their safety tables. They did not show an alarming pattern of medullary thyroid carcinoma. But they were not designed or powered to answer this question, and the follow-up was measured in a few years rather than the decades that would matter for cancer risk.

Outside trials, pharmacoepidemiology has produced genuinely conflicting results. Some analyses of prescribing and cancer registries have reported a signal for thyroid cancer with GLP-1 use; other large cohort studies have found no association. Observational work here is difficult to do well, because people prescribed these drugs are investigated more often than the general population, and more scanning finds more thyroid nodules regardless of whether anything has changed — a bias well documented across thyroid cancer research generally.

The honest summary, and Sunil agrees with this phrasing: no clear human signal has emerged, the data cannot yet exclude a small effect, and anybody claiming certainty in either direction is going beyond the evidence.

Who is generally advised against these medicines

Independent of how the rodent question resolves, prescribing guidance in several countries excludes two groups from GLP-1 medicines on the basis of that unresolved risk.

• People with a personal or family history of medullary thyroid carcinoma.
• People with multiple endocrine neoplasia syndrome type 2, a hereditary condition that carries a high lifetime risk of that cancer.

These are specific and uncommon conditions. Medullary thyroid carcinoma is not the common kind of thyroid cancer — papillary carcinoma is far more frequent, arises from different cells, and is not what the warning is about. A great many people arriving in our thyroid circle worried about this turn out to have a family history of something entirely different, and the relief when that is unpicked is considerable.

If you are not sure which type ran in your family, that is a completely reasonable thing to find out before starting, and it is a factual question rather than a judgement call. Ask, and if nobody knows, say so.

If you already have a thyroid condition

Hypothyroidism is common, so a lot of members here take levothyroxine and want to know whether the two interact. A few things that come up repeatedly.

Having treated hypothyroidism or Hashimoto’s is not in itself a reason these medicines are withheld, and most members in this position report that their prescriber treated it as unremarkable. What does get discussed is timing and absorption. Levothyroxine is absorbed in the small intestine and its absorption is sensitive to stomach conditions and to what else is in there. These medicines slow gastric emptying substantially, particularly early on and after each increase, which is a plausible reason to keep your levothyroxine routine consistent rather than moving it about.

What that means in practice — whether your dose needs checking, when to retest, whether anything changes at all — is a conversation with whoever manages your thyroid replacement. We are not going to give you a timing rule, because the interaction has not been characterised well enough for anybody to give you one honestly.

Bea, in the thyroid circle, described the version that worked: she told her GP she was starting, asked whether her thyroid function should be rechecked at any particular point, and got a plan rather than a shrug. Asking for the recheck to be diarised is the useful move.

Symptoms worth mentioning, and calibrating the worry

Things it is reasonable to have looked at, without assuming the worst:

• A new lump in the front of the neck, or a swelling that persists.
• Persistent hoarseness that is not a passing cold.
• Difficulty swallowing that is new and does not settle.
• Symptoms of thyroid dysfunction that are new or changed — persistent cold intolerance, marked fatigue, palpitations, tremor — noting that fatigue in the early months of these medicines is extremely common and usually has nothing to do with your thyroid. See fatigue in the first months.

Routine calcitonin monitoring is not generally recommended for people on these drugs, and there is a good reason for that beyond cost. Calcitonin has a meaningful false-positive rate, and in a low-risk population a screening test with false positives generates a great deal of investigation, anxiety and occasional harm for very little detection. If you want to discuss monitoring anyway, that is legitimate — but go in knowing that the answer may be a considered no rather than a dismissal.

And keep the base rate in view. Medullary thyroid carcinoma is rare. Most neck lumps are not cancer. Most fatigue on these drugs is the drugs. Worry is not evidence, and neither is reassurance.