S-LiTE reports the only controlled test of exercise around GLP-1 cessation

A one-year follow-up of a Danish randomized trial, published in eClinicalMedicine, offers the closest thing researchers have to controlled evidence on what happens after people stop a GLP-1 medication — and whether exercise during treatment changes the outcome ^[1][2].

The trial, run in Copenhagen, enrolled adults with obesity aged 18–65 with a starting body mass index of 32–43 kg/m². Participants first lost an average of 13.1 kg (about 29 pounds) during an eight-week low-calorie diet. They were then randomly assigned, in equal numbers, to one of four weight-maintenance groups for 52 weeks: a supervised exercise program plus placebo; the GLP-1 receptor agonist liraglutide 3.0 mg injected once daily plus usual physical activity; exercise plus liraglutide; or placebo plus usual physical activity ^[1].

What happened after treatment stopped

166 participants finished the 52-week maintenance phase. Everyone who had been randomized was invited back for assessments at week 104 — one year after all treatment ended. During that post-treatment year there was no contact between participants and study staff. 109 people attended the follow-up visits, which took place between Dec 17, 2018 and Dec 17, 2020 ^[1].

Measured from randomization (week 0, after the diet-induced loss) to week 104, the group that had received combined exercise and liraglutide weighed 5.1 kg less than the group that had received liraglutide alone (95% CI −10.0 to −0.2; P = 0.040) and had a body-fat percentage 2.3 points lower (−4.3 to −0.3; P = 0.026) ^[1].

The study also looked at how many people held onto a meaningful loss. Counting from the very start of the study (week −8) to week 104, participants from the combination group were more likely to still be at least 10% below their initial body weight than those from the placebo group (odds ratio 7.2, 95% CI 2.4 to 21.3) or the liraglutide-alone group (OR 4.2, 1.6 to 10.8). Those from the exercise group were also more likely than placebo participants to keep a 10% loss (OR 3.7, 1.2 to 11.1) ^[1].

Regain during the drug-free year told a similar story. Between week 52 and week 104, weight regain was 6.0 kg greater after stopping liraglutide than after stopping supervised exercise (95% CI 2.1 to 10.0). Compared with the combination group, the liraglutide-alone group regained 2.5 kg more, but that estimate ranged from −1.5 to 6.5 kg — a confidence interval that crosses zero, meaning the difference was not statistically clear ^[1].

The authors concluded that adding supervised exercise to obesity drug treatment "seems to improve healthy weight maintenance after treatment termination," and that body weight and body composition were maintained a year after exercise stopped, in contrast to regain after stopping medication alone ^[1].

Why it matters for patients

Weight regain after stopping a GLP-1 drug is well documented, and this trial was built specifically to test whether anything done during treatment changes that trajectory. The answer here points toward exercise, but several limits matter.

First, the drug studied was liraglutide 3.0 mg, an older once-daily GLP-1 medication, not semaglutide (Ozempic, Wegovy, Rybelsus) or tirzepatide (Mounjaro, Zepbound). Whether the same pattern holds with the newer, more potent drugs is not addressed by this study.

Second, the exercise arm was a supervised program delivered inside a research trial — structured contact that most people paying for their own care do not have. The trial materials provided here do not describe the program's exact content, frequency or intensity.

Third, the numbers are small and the follow-up was incomplete. Of the participants originally randomized, 109 returned at week 104 ^[1]. Small samples produce wide confidence intervals, which is visible in the 5.1 kg difference whose range stretched from 10.0 kg down to just 0.2 kg ^[1].

The trial was funded by Helsefonden and the Novo Nordisk Foundation, and several authors reported financial ties to drug makers including Novo Nordisk, which markets liraglutide and semaglutide ^[1].

What happens next

The study is registered under EudraCT 2015-005585-32 and ClinicalTrials.gov NCT04122716 ^[1]. No further follow-up beyond week 104 is described in the published report, and no comparable controlled test of exercise around cessation of semaglutide or tirzepatide appears in these sources. Until that exists, this remains the main randomized evidence on the question — informative, but drawn from one drug, one city and about a hundred people who came back.