Metabolic & Cardiometabolic
stopping GLP-1 receptor agonists causes substantial weight regain
In plain terms: If I stop a GLP-1 drug, will I gain the weight back?
Strong support Metabolic & Cardiometabolic
Part of: 💊 GLP-1 Drugs
RefutedContestedStrong support
consensus score 0.91
Yes — trials and pooled data consistently show people regain about two-thirds of lost weight within ~1 year of stopping, so these drugs work like chronic therapy rather than a cure.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
7 support 0 contradict 0 tested null 2 mixed · 9 sources, 7 independent groups · 3 superseded — pooled inside a review, counted once
The evidence (20)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Kolli RT et al 2025 · study_type: meta-analysis | meta-analysis | supports | moderate | Meta-analysis of 36 studies (PubMed/Cochrane/Scopus/ScienceDirect/Web of Science, Jan 2010-Oct 2024) quantifying weight regain after peak weight loss: semaglutide MD=-5.15kg, exenatide -3.06kg, liraglutide -1.50kg, orlistat -1.66kg after di |
| Wilding 2022 · Diabetes Obes Metab | RCT | supports | moderate | STEP-1 extension: after withdrawal, semaglutide group regained 11.6 of 17.3 percentage-pts lost by wk120 (~two-thirds); BP/lipids reverted too. ↩ SUPERSEDED — pooled in the review above, counted once |
| Wilding JPH, Batterham RL, Calanna S, Davies M, Kushner RF 2021 · N Engl J Med | RCT | supports | high | OFF-SCOPE for the regain claim. STEP 1 is a 68-week ON-TREATMENT trial: 'The coprimary end points were the percentage change in body weight and weight reduction of at least 5%', and the result is 'The mean change in body weight from baseline to week 68 was -14.9% in the semaglutide group as compared with -2.4% with placebo'. No post-discontinuation weight measurement appears anywhere in the available text; the abstract's only mention of stopping is estimand language - 'The primary estimand ... assessed effects regardless of treatment discontinuation or rescue interventions' - which is an analysis convention, not a withdrawal follow-up. The prior extract conceded this itself by describing the paper as a 'baseline for extension regain'. The regain evidence belongs to the STEP 1 extension (Wilding 2022), which must share group_id step1 with this note. Read abstract-only 2026-08-19; paper is paywalled (no PMCID, isOpenAccess N). ↩ SUPERSEDED — pooled in the review above, counted once |
| Lau D et al] # corrected 2026-08-17 from PubMed; was [Rubino 2021 · Ann Intern Med | RCT | supports | moderate | Synopsis of STEP-4: continued semaglutide maintained loss while withdrawal produced regain; corroborates primary. |
| Khanna A et al 2026 · study_type: observational | observational | supports | low | Review of STEP 1-5 trials: STEP 4 highlighted notable weight regain following treatment withdrawal ('reflecting the relapsing nature of obesity'); STEP 5 confirmed maintenance of ~15% loss with continued treatment over 2 years. No specific |
| Yates N et al 2025 · study_type: observational | observational | supports | low | Narrative review/prescriber guide states regaining weight is common when GLP-1/GIP drugs are stopped, hence usually need long-term continuation. No quantitative regain data given. |
| Amorim 2026 · Chronic Dis Transl Med | observational | mixed | low | Review flags durability/duration as key gap; benefits wane without continued GLP-1RA exposure. |
| Damen 2026 · Ann Intern Med | meta-analysis | supports | low | The prior extract ('weight benefits of obesity pharmacotherapy are not durable after discontinuation') has NO basis in this paper. The review's own outcome set is 'mortality, weight loss, and quality of life' plus MACE and serious adverse events; the single occurrence of the word in the entire record is a harms outcome during treatment - 'In meta-analyses, nearly all studied interventions were more effective than placebo and/or LI in reducing weight, but more discontinuations due to adverse events were observed.' The words regain, cessation and withdrawal do not appear. The prior extract appears to have been generated from the string 'discontinuations'. OFF-SCOPE: wrong outcome. Quality dropped to low as this source carries no information on the claim at all. Read from abstract only (2026-08-19). |
| Berg 2025 · Obes Rev | meta-analysis | supports | high | SR/MA (n=2372): after stopping, regain proportional to loss — liraglutide +2.20kg, semaglutide/tirzepatide +9.69kg; regardless of lifestyle. |
| Aronne 2024 · JAMA | RCT | supports | high | SURMOUNT-4: stopping tirzepatide gave +14.0% vs -5.5% continued (difference 19.4 pts); only 16.6% of placebo kept >=80% loss. |
| Tzang CC et al 2025 · study_type: meta-analysis | meta-analysis | supports | high | Systematic review and meta-analysis of 18 RCTs (3771 participants, PROSPERO-registered): in obesity, GLP-1RA discontinuation caused body weight gain of 5.63kg (95% CI 3.52-7.73) and HbA1c rise of 0.25%; in T2D, weight gain 2.03kg. Greater r |
| Budini B et al 2026 · study_type: meta-analysis | meta-analysis | supports | high | Systematic review + nonlinear meta-regression of 48 studies (6 RCTs, n=3236 in the pooled model): at 1 year post-cessation 60% of lost weight was regained, plateauing at an estimated 75.3% (95% CI 68.9-81.6) of weight lost. |
| Ritsinger 2026 · Diabetes Obes Metab | observational | mixed | moderate | Real-world context: sustained risk reduction presupposes sustained treatment/weight control, undermined by discontinuation. |
| Lasik L et al 2026 · study_type: observational | observational | supports | low | Narrative review: clinical trial data indicate GLP-1/GIP cessation is associated with biologically driven weight regain and partial loss of metabolic benefits, attributed to adaptive thermogenesis and hunger-satiety axis changes. No new qua |
| McGowan 2025 · Nat Med | meta-analysis | supports | high | SR/MA of obesity pharmacotherapy: maximal weight loss is treatment-duration dependent; cessation reverses gains. |
| Kramer 2024 · J Clin Endocrinol Metab | observational | mixed | moderate | GLP-1RA after bariatric surgery: weight trajectory depends on continued treatment; supports chronic-therapy framing. |
| Rubino 2021 · JAMA | RCT | supports | high | STEP-4 withdrawal: switching to placebo at wk20 led to +6.9% regain vs -7.9% with continued semaglutide (difference ~14.8 pts). ↩ SUPERSEDED — pooled in the review above, counted once |
| Quarenghi M et al 2025 · study_type: observational | observational | supports | moderate | Narrative review of 13 RCTs (search of PubMed/Cochrane/Google Scholar) found rapid regain of weight after cessation of liraglutide, semaglutide, or tirzepatide regardless of treatment duration, substantially mitigating metabolic benefits. |
| Codella R et al 2025 · study_type: observational | observational | supports | low | Narrative review states stopping GLP-1 therapy alone often leads to weight regain, contrasted with exercise-supported maintenance. No specific n or % regain reported. |
| Hamarsheh S, Jaber AR, Abu-Khazneh O, Andonie CR, Majadleh S, Zahran A, Ayesh H 2026 · Endocrinol Diabetes Metab | meta-analysis | tested-null | low | OFF-SCOPE for weight regain. This network meta-analysis measures on-treatment outcomes only: 'The primary outcome was the percentage change in body weight from baseline', with secondary outcomes absolute weight, waist circumference, BMI, >=5/10/15/20% responder proportions, HDL-C, and safety - 'Safety outcomes comprised the incidence of any adverse event, serious adverse events, adverse events resulting in treatment discontinuation and gastrointestinal adverse events'. 'Treatment discontinuation' here means stopping because of an adverse event (e.g. semaglutide 7.2 mg RR = 3.09, 95% CI 1.45-6.59), NOT weight trajectory after stopping. No included trial is a randomised-withdrawal design and none contributes off-treatment follow-up; eligibility asked only for 'a minimum treatment duration of 12 weeks'. The words regain, rebound, washout, off-treatment and withdrawal do not occur anywhere in the full text or in the complete Supporting Information. The paper's only statement about durability is a call for future research - 'include longer follow-up to clarify the durability of weight loss and its translation into sustained cardiometabolic benefit' - which is a declaration that the question is open, not a finding on either side of it. The previous extract's 'durability requires ongoing therapy, consistent with regain on stopping' has no anchor in this paper. Full text + full Supporting Information read 2026-08-19. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.