Weight Regain After Stopping GLP-1 Drugs

When patients stop taking semaglutide or tirzepatide, weight does not stay put. A reconstructed aggregate-data Bayesian longitudinal meta-analysis published in September 2026 in Endocrinology, Diabetes and Metabolism provides the most granular picture yet of how weight trajectories unfold in the months and years after GLP-1 class drugs are stopped^[1]. The findings add precision to what smaller trials had already suggested: discontinuation triggers a consistent pattern of weight return that differs in pace and magnitude between the two leading agents.

The meta-analysis drew on reconstructed individual-level data across multiple randomised controlled trials and used Bayesian modelling to estimate weight trajectories at the aggregate level over time. The approach lets researchers describe not just how much weight returns on average, but the probabilistic shape of the regain curve^[1]. That granularity matters clinically: a gradual slope and a rapid one carry different implications for how soon after stopping a patient needs a follow-up intervention.

What the foundational trial data show

The meta-analysis builds on a foundation of withdrawal data from key registration trials. The European Medicines Agency's assessment of semaglutide (Wegovy) describes one of the most direct discontinuation experiments in the clinical programme. In a study involving 902 adults, participants who had completed a 20-week semaglutide run-in period and then continued the drug lost a further 8% of their body weight over the remaining observation period. Those switched to placebo instead regained 7% of their weight^[2]. The divergence between continued treatment and discontinuation was visible early and widened as the follow-up extended.

For tirzepatide, the EMA's assessment of Mounjaro documents that in the pivotal 72-week weight management study, participants lost at least 15% of their body weight on average compared with 3% on placebo, with over 85% of tirzepatide-treated participants achieving at least a 5% reduction^[3]. That starting point matters for interpreting regain: the higher the initial loss, the larger the absolute weight that can be recovered even if the percentage recovered is similar.

The STEP 4 trial, one of the landmark semaglutide continuation studies, found that participants who had lost weight during a 20-week run-in and were then switched to placebo regained more than half of their initial weight loss over the subsequent follow-up period^[5]. Continued semaglutide, by contrast, produced additional weight loss. The pattern underlines a consistent theme across the drug class: the weight-lowering effect depends on maintained treatment.

Why weight returns after stopping

GLP-1 receptor agonists act on central appetite pathways and slow gastric emptying, reducing energy intake during treatment. When the drug is cleared, those pharmacological effects dissipate and the biological drivers of weight gain that existed before treatment resume. The Bayesian meta-analysis^[1] is consistent with this mechanism: weight return is not a failure of willpower or behaviour but a predictable physiological response to stopping a drug that was actively modifying appetite and energy regulation.

A 2026 clinical review in the Journal of Clinical Endocrinology and Metabolism found that up to half of patients discontinue obesity medication within one to two years of starting, for reasons including side effects, cost, insurance coverage gaps, and inadequate perceived response^[4]. The same paper described cases in which patients who restarted treatment after a gap showed a blunted response, gaining less with re-introduction than they had lost initially. This observation, while based on limited case data, raises questions about whether early discontinuation carries costs beyond the immediate weight regain.

Comparing the two agents after stopping

One of the contributions of the 2026 Bayesian meta-analysis is placing semaglutide and tirzepatide side by side within the same modelling framework^[1]. Because tirzepatide targets both GIP and GLP-1 receptors while semaglutide targets GLP-1 alone, their on-treatment efficacy differs, and there was a reasonable hypothesis that their post-discontinuation trajectories might also differ. The meta-analysis's aggregate modelling approach lets researchers estimate whether the rate or depth of weight return varies between agents even when head-to-head withdrawal trial data are not available.

In practice, the distinction between a 15% initial loss followed by partial regain and a 20% initial loss followed by partial regain may matter more in absolute terms than any difference in the percentage of lost weight that returns. A patient who lost 20 kg and regains half is in a different clinical position from one who lost 12 kg and regains half, even if the relative regain rate is identical. The Bayesian framework is well suited to capturing and communicating this kind of uncertainty around individual trajectories^[1].

What the data mean for treatment decisions

The accumulating evidence on weight regain after stopping GLP-1 class drugs has shifted how obesity specialists frame the duration question. A 2022 perspective in the Journal of Clinical Endocrinology and Metabolism noted that the beneficial effects of semaglutide on body weight appear to require maintenance of treatment^[5], and that some degree of weight return is observed across GLP-1 receptor agonists when treatment extends beyond one year without continuation. This is not unique to GLP-1 drugs: similar dynamics occur after bariatric surgery and with other weight-management interventions, though the timescale and magnitude differ.

For patients and clinicians, the practical upshot is that stopping semaglutide or tirzepatide for reasons of cost, supply, or side effects is likely to result in meaningful weight return, and the speed of that return may be faster than the pace of the original loss. The 2026 meta-analysis provides the trajectory data that can inform shared decision-making: not just whether weight returns, but at what rate and to what probable endpoint over a defined time window^[1]. That information belongs in the conversation between a patient considering stopping and the clinician managing their care.

This article is for informational purposes only and does not constitute medical advice. Consult a healthcare provider before making any treatment decisions.