Are GLP-1s Safe Long-Term?

GLP-1 receptor agonists have been prescribed for type 2 diabetes since the mid-2000s, so regulators and clinicians have a long real-world track record with the drug class in general. But the high-dose, weight-loss-focused use of semaglutide and tirzepatide is much newer, and the honest answer to “are they safe long-term” is: safer than critics feared, less proven than a decade of marketing might suggest.

Plums, fruits, food — illustrating Are GLP-1s Safe Long-Term?

What “long-term” actually means here

Exenatide and liraglutide have been used in diabetes care for close to twenty years, and that history is reassuring for the drug class’s general safety profile. But the obesity-dose versions — Wegovy (semaglutide) and Zepbound (tirzepatide) — only reached the market in 2021 and 2023. The most relevant long-term safety data specific to weight management comes from the SELECT trial, which followed roughly 17,600 adults with overweight or obesity and established cardiovascular disease for an average of about three to four years. It found that semaglutide reduced major adverse cardiovascular events compared with placebo, which is reassuring for heart risk over that window. It is not the same as multi-decade safety data, which simply doesn’t exist yet for this dosing and population.

The risks that are already well documented

Several safety signals are well characterized, even if not fully resolved:

Gastrointestinal effects — nausea, vomiting, diarrhea, and constipation are the most common side effects, occur in a substantial minority of users, and are usually worst during dose escalation. They’re the leading reason people discontinue treatment.
Pancreatitis — a rare signal has been observed across the drug class since early diabetes trials. Absolute risk appears low, and a clear causal link hasn’t been firmly established, but it’s monitored closely and is a reason to stop the drug if symptoms appear.
Gallbladder disease — cholelithiasis and cholecystitis occur more often in GLP-1 users, most likely as a downstream effect of rapid weight loss itself rather than a direct drug toxicity.
Thyroid C-cell tumors — this is the boxed warning that gets the most attention, and it’s worth being precise about its basis. Rodent studies found dose-dependent increases in thyroid C-cell tumors, including medullary thyroid carcinoma, in rats and mice given GLP-1 agonists. Human thyroid C-cells appear to express far fewer GLP-1 receptors than rodent cells do, and no clear signal of increased medullary thyroid cancer has emerged in the diabetes population’s much longer usage history. The warning remains on the label out of caution, and the drugs are contraindicated in anyone with a personal or family history of medullary thyroid carcinoma or MEN2 syndrome.
Lean mass loss — body composition sub-studies suggest a meaningful share of weight lost on these drugs — commonly cited estimates run up to roughly a quarter to a third — is lean mass rather than fat, which matters more for older adults and anyone at risk of sarcopenia. This is an active area of research, not a settled number.

None of the known risks are dealbreakers for most people, but “no major red flags in three to four years of trial data” is a different statement than “safe for decades,” and it’s worth not confusing the two.

Sandwich, food, flying food — illustrating Are GLP-1s Safe Long-Term?

The chronic-use problem

Perhaps the biggest long-term question isn’t a side effect at all — it’s that these drugs appear to require ongoing use to keep their benefits. Extension data from semaglutide trials shows that people who stop the drug tend to regain a substantial portion of the weight they lost within about a year, along with some reversal of metabolic improvements. That reframes the “long-term safety” question: it’s not just about tolerating the drug for a few years, but about the implications of tolerating it indefinitely, since stopping largely undoes the benefit.

Lemon, vitamins, meal — illustrating Are GLP-1s Safe Long-Term?

What we don’t know yet

Nobody has 20- or 30-year randomized data on high-dose semaglutide or tirzepatide in people without diabetes, because the drugs haven’t been used that way for that long. Questions that remain open include the cumulative effect of years of appetite suppression and altered eating patterns on nutritional status, the long-term trajectory of muscle mass in people who stay on these drugs for a decade or more, and whether cardiovascular benefits seen in high-risk populations extend to the healthier, younger people increasingly using these drugs for weight loss alone.

What’s known	Evidence status
Diabetes-dose safety over 10-20 years	Substantial real-world and trial data
Obesity-dose cardiovascular safety (SELECT)	~3-4 years, one major RCT
GI side effects	Well characterized, common, usually manageable
Thyroid C-cell tumor risk in humans	Rodent-based warning; no confirmed human signal after years of use
Muscle/lean mass loss over a decade of use	Preliminary, still being studied
Effects of stopping and restarting repeatedly	Not well studied

The takeaway

GLP-1 drugs have a reassuring safety record for the years they’ve been studied at obesity doses, and a genuinely long track record at diabetes doses. But “long-term” studies in medicine usually mean five, ten, or twenty years, and the newer high-dose obesity indications simply haven’t been around that long. That’s not a reason for alarm, but it is a reason for humility about what’s actually known. Anyone considering years of GLP-1 use — especially alongside other medications, or with a personal or family history of thyroid or pancreatic disease — should work through that decision with a clinician who can weigh their individual risk profile against the current evidence, not just the marketing.