Long-term use and discontinuation of GLP-1 therapy — what the trials show

What happens to body weight when GLP-1 therapy stops?

The clearest evidence comes from randomised withdrawal trials, in which participants who had already achieved significant weight loss were randomly allocated to either continue their medicine or switch to placebo. This design removes the confound of selecting for people who were already choosing to stop and allows a direct comparison of continued versus discontinued treatment.

Both STEP 4 and SURMOUNT-4 found the same directional result: stopping therapy led to substantial weight regain. In STEP 4, participants who had achieved approximately 10.6% weight loss during the open-label run-in on semaglutide 2.4 mg were then randomised to either continue semaglutide or switch to placebo for 48 weeks. Those who continued semaglutide lost a further 7.9% of their body weight. Those who switched to placebo regained 6.9% — recapturing roughly two-thirds of what they had lost.

The between-group difference in weight change at 48 weeks was approximately 14.8 percentage points — a large and statistically robust separation. This finding was accompanied by reversals in cardiometabolic improvements including waist circumference, blood pressure and glycated haemoglobin among those who discontinued.

What did STEP 4 show about discontinuation?

Published in JAMA in 2021 (Rubino et al.), STEP 4 enrolled 803 adults with a BMI of ≥30 kg/m² or ≥27 kg/m² with one or more weight-related comorbidities. After a 20-week run-in period on semaglutide 2.4 mg subcutaneous once weekly (titrated to maintenance dose), those who had lost at least some weight were randomised 2:1 to continue semaglutide (535 participants) or switch to placebo (268 participants).

The primary endpoint — change in body weight from randomisation to week 68 — showed clearly divergent trajectories. Among those continuing semaglutide, the medicine's weight-lowering effect continued to accumulate. Among those switched to placebo, weight returned at a consistent rate from the first post-randomisation visit onward. By week 68, the continuous semaglutide group had lost more weight than when they were randomised; the placebo group had almost fully returned to their original starting weight.

What does SURMOUNT-4 show about discontinuation of tirzepatide?

SURMOUNT-4 applied a similar withdrawal design to tirzepatide (Mounjaro), the dual GIP/GLP-1 receptor agonist. Participants spent 36 weeks in an open-label lead-in period on tirzepatide at the maximum tolerated dose (10 mg or 15 mg), achieving a mean weight loss of approximately 20.9% from the start of the lead-in. They were then randomised to continue tirzepatide (312 participants) or switch to placebo (311 participants) for a further 52 weeks.

Those continuing tirzepatide lost a further 5.5% from the point of randomisation. Those switching to placebo regained 14.8% — meaning they had returned close to their original weight within a year of stopping treatment. The difference between the two groups in total weight change from the start of the lead-in phase to week 88 was approximately 21 percentage points. The findings confirm that tirzepatide's weight management effect similarly depends on continued treatment.

Is GLP-1 therapy intended to be lifelong?

The STEP 4 and SURMOUNT-4 findings are consistent with the clinical framing of obesity as a chronic, relapsing condition with underlying physiological drivers. The implication — that pharmacological support may need to be ongoing for sustained benefit — mirrors the established understanding of antihypertensive or statin therapy, where stopping a medicine that is working tends to result in the return of the condition it was managing.

However, "lifelong" as a concept in prescribing is governed by evolving NICE guidance, individual NHS commissioning policies, and ongoing clinical assessment of benefit versus risk. NICE TA875 and TA1026 set out eligibility criteria and review requirements for NHS-funded prescribing. The question of how long a specific individual should remain on therapy is a clinical one, not answerable by a population-level trial.

Why does weight return after stopping?

GLP-1 receptor agonists act on hypothalamic and brainstem circuits involved in appetite regulation, promoting satiety and reducing energy intake. They also slow gastric emptying, which prolongs the feeling of fullness after eating. GIP receptor agonism (tirzepatide) additionally affects energy metabolism and adipose tissue function. When these pharmacological effects are removed, the neurobiological signals that drove the original excess weight tend to reassert — a phenomenon consistent with the broader set-point and allostatic load theories of body weight regulation.

This is not a failure of willpower or adherence; it reflects a pharmacologically mediated suppression of appetite signals that returns to baseline when the drug is cleared.

How long can you stay on Mounjaro or Wegovy?

NICE TA875 (semaglutide 2.4 mg for obesity) and TA1026 (tirzepatide for obesity) both specify treatment continuation criteria. Both appraisals expect prescribers to assess ongoing eligibility and response at defined intervals. The NICE criteria require discontinuation if meaningful clinical response (typically at least 5% weight loss) has not been achieved and maintained. The appraisals do not impose a maximum treatment duration for those who continue to benefit and remain eligible.

In practice, NHS access via Tier 3 weight management services and specialist pathways involves periodic clinical review. For a detailed breakdown of NHS access and NICE criteria, see Mounjaro vs Wegovy — NHS access and NICE eligibility.

Can I taper off rather than stop suddenly?

Neither STEP 4 nor SURMOUNT-4 investigated gradual dose tapering as a discontinuation strategy; both used abrupt withdrawal from full maintenance dose to placebo. There is therefore no direct trial evidence to determine whether tapering reduces the rate or extent of weight regain. Some clinical guidance suggests tapering may be preferable for tolerability reasons (avoiding rebound GI symptoms) but this is based on pharmacological reasoning rather than randomised evidence on weight outcomes. Any tapering approach should be discussed with the prescribing clinician.

Frequently asked questions

What happens to body weight when GLP-1 therapy is stopped?

Both STEP 4 and SURMOUNT-4 demonstrated significant weight regain after randomised withdrawal of semaglutide and tirzepatide respectively. In STEP 4, participants who switched to placebo regained approximately two-thirds of their lost weight within 52 weeks. SURMOUNT-4 showed similar directional findings for tirzepatide.

How long can you stay on Mounjaro or Wegovy?

Current NICE technology appraisals (TA875 for semaglutide, TA1026 for tirzepatide) specify time-limited treatment periods linked to continuing clinical benefit. The prescribing indication does not impose a pharmacological upper time limit, but NHS and clinical governance frameworks require periodic review of continued eligibility. Discuss duration with your prescribing clinician.

Is GLP-1 therapy intended to be lifelong?

The clinical evidence — including STEP 4 and SURMOUNT-4 — suggests these medicines work while they are being taken and that discontinuation is associated with weight regain. Whether this implies lifelong use for a specific individual is a clinical judgement that depends on therapeutic goals, eligibility, ongoing benefit and individual circumstances.

Why does weight return after stopping GLP-1 therapy?

The proposed mechanism relates to the physiological role of GLP-1 and GIP signalling in appetite regulation and energy homeostasis. When pharmacological support for these pathways is withdrawn, hunger signalling tends to return towards pre-treatment levels. This is consistent with the understanding that obesity is a chronic condition with underlying physiological drivers.

What does STEP 4 specifically show about discontinuation?

STEP 4 randomised participants who had already lost weight on semaglutide 2.4 mg to either continue semaglutide or switch to placebo for 48 weeks. Those who continued semaglutide lost an additional 7.9% of body weight; those switched to placebo regained 6.9%. The between-group difference in weight change was approximately 14.8 percentage points.

What does SURMOUNT-4 show about discontinuation of tirzepatide?

SURMOUNT-4 employed a similar randomised withdrawal design in participants who had achieved significant weight loss on tirzepatide. Those continuing tirzepatide maintained and extended their weight loss; those switched to placebo experienced substantial weight regain. The trial reinforced the finding from STEP 4 that the weight-management benefit of incretin therapy requires continued treatment.

Can I taper off GLP-1 therapy rather than stopping suddenly?

There is no trial evidence establishing that gradual dose tapering reduces the degree or speed of weight regain compared with abrupt discontinuation. This is a clinical question for your prescribing clinician, who can advise based on your specific circumstances and the evidence available at the time of your review.

Does stopping GLP-1 therapy affect cardiovascular risk?

There is no randomised trial evidence specifically addressing the cardiovascular consequences of GLP-1 discontinuation. The SELECT trial showed cardiovascular benefit from ongoing semaglutide use in a high-risk population; whether and how quickly those benefits attenuate after stopping is not established in the current evidence base.

Are lifestyle changes enough to maintain weight loss after stopping?

STEP 4 participants who switched to placebo also received lifestyle counselling; this did not prevent the majority of weight regain. This finding is consistent with the broader evidence that intensive lifestyle intervention alone is insufficient to sustain pharmacologically-achieved weight loss for most individuals.

What does 'weight regain' actually mean in these trials?

In trial terms, weight regain is the percentage of body weight lost during the run-in phase that is subsequently recovered after randomisation to placebo or after treatment ends. In STEP 4, roughly two-thirds of the weight lost was regained within a year. Absolute figures vary by individual baseline characteristics.