Weight regain after stopping a GLP-1: the evidence

The GLP-1 drugs reduce appetite while you take them. They do not reset the body's defence of its weight. When they stop, appetite returns over the weeks it takes the drug to clear, and weight follows. This is now one of the best-measured findings in obesity medicine, because the manufacturers ran trials that stopped the drug deliberately. The practical side of stopping is on the stopping a GLP-1 page. This page is the evidence.

The studies in one table

STEP 1 extension: one year off semaglutide

At the end of STEP 1's 68 weeks, both the drug and the lifestyle programme stopped. A representative subset of 327 participants, from sites in Canada, Germany, the UK, the US and Japan, was followed for another year. They had lost a mean of 17.3% on semaglutide by week 68. By week 120 they had regained 11.6 percentage points, leaving a net loss of 5.6%. That is about two thirds of the loss back within a year. Blood pressure, cholesterol, HbA1c and other cardiometabolic measures that had improved on the drug moved back towards baseline for most variables. These analyses were exploratory, and the extension had no randomised comparison of its own, but the direction is unambiguous.

STEP 4: randomised withdrawal of semaglutide

STEP 4 is the cleaner test. 902 people took semaglutide for 20 weeks and lost 10.6% on average. The 803 who reached the 2.4 mg dose were randomised, two to one, to keep taking it or switch to a placebo injection, without knowing which. From week 20 to week 68, those who continued lost a further 7.9%. Those switched to placebo regained 6.9%. The gap between them, 14.8 percentage points, is the effect of continuing the drug. Waist circumference, blood pressure and physical functioning also favoured continuing.

SURMOUNT-4: randomised withdrawal of tirzepatide

SURMOUNT-4 did the same for tirzepatide, with a longer lead-in. 783 adults took tirzepatide at the highest tolerated dose, 10 or 15 mg, for 36 weeks. The 670 randomised had lost a mean of 20.9%. Over the next 52 weeks, those who continued lost a further 5.5%. Those switched to placebo regained 14.0%. At week 88, 89.5% of the continuing group had kept at least 80% of their lead-in loss, against 16.6% of the placebo group. Measured from the very start, the totals were -25.3% for continuing and -9.9% for stopping. People who stopped kept some benefit at 88 weeks, but most of it was gone and the curve was still rising.

The 2026 BMJ meta-analysis

In January 2026 the BMJ published a systematic review and meta-analysis of every study it could find that followed people after stopping a weight-loss medicine: 37 studies, 63 treatment arms and 9,341 participants. Average treatment lasted 39 weeks and average follow-up after stopping 32 weeks. Across all medicines, weight came back at an average of 0.4 kg a month, faster than after stopping a behavioural weight-loss programme by 0.3 kg a month, independent of how much had been lost. Cardiometabolic markers were projected to return to baseline within 1.4 years.

For semaglutide and tirzepatide specifically, regain was faster, 0.8 kg a month, with return to baseline weight projected at about 1.5 years. The authors are clear that data for these newer drugs ran to at most 12 months after stopping, so the 1.5-year figure is a projection rather than an observation. The review included randomised trials, non-randomised trials and observational studies; the observational data show association, not cause.

Why it happens

Weight loss of any kind triggers adaptations that push weight back up: hunger rises, fullness signals fall, and energy expenditure drops by more than the lost mass explains. On the drug, the GLP-1 signal overrides much of the appetite side. Off it, those adaptations act unopposed. The faster regain after GLP-1s than after diet programmes in the BMJ analysis fits this: the drug held back a stronger counter-pressure. Lean mass lost during treatment may also matter, which is one reason the muscle loss page stresses protein and resistance training.

What the studies do not say

• That everyone regains. These are averages. In the STEP 1 extension some participants kept most of their loss. The trials did not set out to identify who.
• What happens with a planned taper. STEP 4 and SURMOUNT-4 stopped abruptly onto placebo. Whether tapering to a lower dose, or extending the interval, keeps more weight off is being studied, and the maintenance dose page covers what is known.
• Whether a structured exercise and diet programme after stopping changes the curve. The STEP 1 extension stopped the lifestyle programme too.
• What happens in years three and four off the drug. Follow-up is short.

What it means in practice

The labels describe these drugs as long-term treatment, and the withdrawal trials are the reason. That does not mean stopping is wrong: cost, side effects, pregnancy and preference are all good reasons. It means stopping is a decision to plan, with the prescriber, rather than an event. The stopping page covers how people approach it, and the master table puts these trials beside the rest.

If you do stop, the most useful thing is a record of where you started: your weight, waist, and if possible your labs on the day of the last dose. Regain is easier to respond to early, and a log makes it visible in weeks rather than months. Peptide AI keeps dose, weight and labs on one timeline, with the dose log free. See the free GLP-1 tracker or download the app.