Long-term safety: what the trials followed

The longest trials, side by side

The main weight-reduction trials behind Wegovy®, Zepbound® and Foundayo™ lasted 68 to 72 weeks. The longest exposure comes from outcome trials, built to count heart or kidney events and run until enough had occurred, and from a few weight trials extended to two or three years. Each one collected safety data in its own way, so the table describes what each reported rather than ranking them.

Findings that only appeared with time and numbers

• Fractures: in SELECT, more hip and pelvic fractures were reported on Wegovy® than on placebo in women (1% against 0.2%) and in people aged 75 and over (2.4% against 0.6%). In a trial in adults with MASH, a fatty liver disease, fractures were reported in 4.4% against 3.3% ^[1].
• Kidney stones: urolithiasis in 1.2% on Wegovy® against 0.8% in SELECT, with serious cases in 0.6% against 0.4% ^[1]. They have their own article.
• Low blood sugar in people without diabetes: rare overall, with 3 serious episodes on Wegovy® against 1 on placebo in SELECT, and more among people who had previously had bariatric surgery ^[1].
• Age: in SELECT, people aged 75 and over reported more serious adverse reactions than younger participants, on both semaglutide and placebo ^[1].

Warnings that apply however long you take it

Some label warnings are not tied to the first weeks. Every GLP-1 label covered here carries a boxed warning about thyroid C-cell tumours, based on rodent studies; the labels state that it is unknown whether the medicines cause these tumours in people, because the relevance of the rodent findings has not been determined ^[1][5]. The thyroid warning article explains the evidence, and the cancer questions article covers other cancers; neither topic is repeated here.

Gallbladder disease: on Wegovy®, gallstones were reported in 1.6% against 0.7% on placebo in adult weight trials. The label notes that fast or large weight loss can raise this risk, but that the excess remained after allowing for the amount of weight lost ^[1]. The Zepbound® label links acute gallbladder events with weight reduction, and in the 5.4-year dulaglutide trial serious gallbladder inflammation occurred in 0.5% against 0.3% ^[5][6].

Pancreatitis: in the Wegovy® weight trials, adjudicated acute pancreatitis occurred at 0.2 cases per 100 patient-years against under 0.1 on placebo, and reports after approval include necrotizing pancreatitis, sometimes fatal ^[1]. The pancreatitis and gallbladder articles go through the signs in detail.

The eyes: findings that needed years

In SUSTAIN 6, a 2-year trial of semaglutide in people with type 2 diabetes and high heart risk, complications of diabetic eye disease occurred in 3.0% on Ozempic® against 1.8% on placebo; the absolute increase was larger in people who already had retinopathy, 8.2% against 5.2% ^[2]. The dulaglutide trial showed a similar pattern over 5.4 years: 8.5% against 6.2% in people with a history of retinopathy, and 1% in both groups without it ^[6].

The labels explain that a rapid improvement in blood sugar has been linked with a temporary worsening of retinopathy, and the Ozempic® label adds that the effect of long-term control with semaglutide on these complications has not been studied ^[2][5].

Separately, the European information for Ozempic® describes NAION, a rare optic nerve condition that causes sudden vision loss. Several large epidemiological studies suggest about a two-fold higher relative risk with semaglutide in type 2 diabetes, roughly one extra case per 10,000 person-years, with no set time after starting when it may appear ^[7]. These are observational studies, which cannot prove cause. The eye changes article covers both topics.

What has been reported since approval

Section 6.2 of each label lists events reported after the medicines reached the market. For semaglutide and tirzepatide these include pancreatitis, sometimes severe or fatal, a blocked or paralysed bowel and severe constipation, serious allergic reactions, kidney injury and inhaling stomach contents during anaesthesia or deep sedation ^[1][3].

These reports are voluntary and come from an unknown number of users, so the labels say their frequency cannot be reliably estimated and a causal link cannot always be established ^[1][3]. They are signals that shaped the warnings, not rates.

Why some risks surface later

• Supported by the labels: gallbladder events are tied to weight reduction, and retinopathy worsening to a rapid fall in blood sugar. Both depend on changes that build up over months ^[1][2][5].
• Supported by the trials: some differences are small in absolute terms, such as four more kidney stone reports per thousand people over about three years in SELECT, and only become visible in very large or long trials ^[1].
• Not explained by the sources: the Wegovy® label reports the fracture imbalance but gives no cause, and the SELECT abstract does not discuss it. Weight loss, falls and age are possible factors, but none was tested as the explanation in the documents we used ^[1][8].
• General background: older adults reported more serious adverse reactions on both treatments in SELECT, which shows that age itself raises the background rate ^[1].

When long-term treatment stops

Stopping has its own consequences for weight and health measures. A year after STEP 1 ended, participants had regained about two-thirds of the weight they had lost, and cardiometabolic improvements had moved back towards baseline ^[12]. In SURMOUNT-4, people switched from tirzepatide to placebo after 36 weeks saw their weight rise by 14.0% over the next year, while those who continued lost a further 5.5% ^[13]. The stopping article and the article on how long a GLP-1 can be taken cover this in full.

What is still unknown, and what may help

The longest controlled data reach about five years for dulaglutide and three to four years for semaglutide and tirzepatide ^{[3][6][8][9][11]}. The large outcome trials collected only part of the safety picture, and their entry criteria excluded some groups, so they say less about people who were not enrolled ^[1][2]. Safety beyond these periods rests on continued reporting and on new studies as they are published.

What the labels ask for over time is regular review rather than a single check: monitoring for eye changes in people with a history of diabetic retinopathy, attention to heart rate with Wegovy® and Saxenda®, and kidney checks when stomach side effects cause fluid loss ^[1][2][4]. A personal record of doses, side effects, weight and any falls or eye tests makes those reviews easier.

Wegovy®, Ozempic® and Saxenda® are registered trademarks of Novo Nordisk A/S. Mounjaro®, Zepbound® and Trulicity® are registered trademarks and Foundayo™ is a trademark of Eli Lilly and Company. PepLabo is not affiliated with or endorsed by Novo Nordisk or Eli Lilly and Company.

When to get help