Skip to main content
glpverdict
Editorial medical reviewFDA & peer-reviewed sourcesUpdated weeklyMedically reviewed content

Weight management

GLP-1 rebound weight gain: what STEP-4 and SURMOUNT-4 actually show — and how to prevent it

Two randomized trials answer the question directly: what happens when you stop semaglutide or tirzepatide? Roughly two-thirds of the weight lost comes back within a year. This guide translates STEP-4 (JAMA 2021), SURMOUNT-4 (NEJM 2024), and the AACE 2024 obesity guidelines into an evidence-based prevention playbook.

By Marisa Chen, RDRegistered dietitian · 12 years GLP-1 clinic work10 min read

Medically reviewed by Jane Novak, MD, MPH, Endocrinology · Internal medicineUpdated Jul 17, 2026

GLP-1 rebound weight gain: what STEP-4 and SURMOUNT-4 actually show — and how to prevent it — guide hero illustration

Why regain happens — mechanism, not willpower

GLP-1 receptor agonists work while circulating. Semaglutide half-life ≈ 1 week; tirzepatide half-life ≈ 5 days. Discontinuation drops systemic exposure to sub-therapeutic within 4-6 weeks. Appetite regulation, gastric emptying, and mesolimbic reward-signal dampening all return to baseline.

Regain is biology, not personal failure. Human physiology defends the highest previously-established body-weight setpoint via elevated ghrelin, suppressed leptin, and increased food-noise. This is a survival mechanism, not a willpower deficit. That is the point the AACE 2024 guidelines make when they treat obesity as chronic disease.

STEP-4: what the numbers actually said

STEP-4 randomized 803 adults who had completed 20 weeks of semaglutide 2.4 mg (reaching ≈10.6% mean weight loss) to either continue semaglutide or switch to placebo for the following 48 weeks. Everyone stayed on the lifestyle-intervention arm.

By week 68, the continued-semaglutide group had lost an additional 7.9% (total ~17.7%). The switched-to-placebo group regained 6.9% — net loss of only ~5.0% from baseline. About 66% of the loss achieved during the semaglutide-run-in phase came back within 48 weeks of stopping.

Regain was measurable within 12 weeks of discontinuation and continued at a roughly linear rate through week 48. The rate did not level off during the trial window.

SURMOUNT-4: tirzepatide regain is larger

SURMOUNT-4 randomized 670 adults who had completed 36 weeks of tirzepatide (reaching ≈20.9% mean weight loss) to either continue tirzepatide or switch to placebo for the following 52 weeks.

The continued-tirzepatide group achieved an additional 5.5% loss (total ~25.3%). The switched-to-placebo group regained 14.0% — net loss of only ~9.9% from baseline. About 67% of the loss achieved during the run-in phase came back within 52 weeks.

The absolute regain is larger for tirzepatide (14.0%) than for semaglutide (6.9%) because the pre-discontinuation loss was larger (20.9% vs 10.6%) — the proportion regained is nearly identical.

The prevention playbook — what actually reduces regain

Nothing fully prevents regain after a full stop — the trials are unambiguous. But two levers reduce it materially, and a third avoids the full-stop scenario.

1. Protein target 0.7-1.0 g per lb of goal body weight, sustained through the taper and post-discontinuation window. Lean-mass preservation directly reduces resting metabolic rate loss, which directly reduces regain. Sub-0.7 g/lb accelerates regain.

2. Two resistance-training sessions per week, minimum. Resistance training preserves the fat-free mass that leptin-signaling depends on. Aerobic-only training does not have the same effect on regain in the trial extension data.

3. Do not stop — taper instead. AACE 2024 recommends dose-reduction over 12-24 weeks rather than a hard stop, when clinical circumstances allow. Individual-clinician evidence supports this but no head-to-head RCT has published yet. If cost forces you off, see the taper/bridge options below.

If cost forces stopping — real bridge options

The regain data assumes complete discontinuation. If cost pressure is what would force stopping, you have three realistic bridge options before hitting placebo.

Manufacturer savings cards: Wegovy at $499/mo via NovoCare (cash) is available without insurance in the US. Zepbound at $349-499/mo via LillyDirect (cash). Both cut the effective cost roughly in half vs list price.

Compounded 503A telehealth: patient-specific semaglutide or tirzepatide via a licensed 503A pharmacy runs $99-249/mo. Post-shortage the pathway is still legal (individualized formulation, per-patient clinical need) but is not identical to brand-name product. Our /best-glp1-providers directory lists currently-compliant options.

Naltrexone-bupropion (Contrave) as a bridge: a lower-efficacy oral obesity medication ($50-99/mo generic), roughly 5-6% mean loss vs 15% for semaglutide, but non-injectable and cash-affordable. May slow regain compared to a full-placebo drop, per limited observational data.

What regain actually looks like week-by-week

Regain is not instant — the trials showed it accumulates.

Weeks 0-4 after stopping: minimal scale change. Appetite noticeably returns; food-noise resumes within 10-14 days.

Weeks 4-12: 2-4% of body weight regained on average. Some patients describe rapid appetite override — hunger becomes urgent rather than steady.

Weeks 12-24: 4-7% regained. Portion sizes return to pre-treatment norms. Clothing fit changes noticeable.

Weeks 24-52: total regain converges to the STEP-4 / SURMOUNT-4 published numbers (7-14% depending on drug). Rate levels off around month 12 as body weight approaches a new setpoint.

This trajectory is not inevitable if you sustain protein + resistance training. Both trials studied protocol-standardized lifestyle intervention that was likely below what a motivated patient can sustain — real-world outcomes with aggressive lifestyle can be somewhat better than trial mean.

Frequently asked questions

Will I gain back everything I lost if I stop Wegovy?
Not everything, but roughly two-thirds within a year, per STEP-4. If you lost 30 lbs on Wegovy, expect ≈20 lbs back over the 12 months after discontinuation. Regain is measurable within 8-12 weeks of stopping.
Is Zepbound regain worse than Wegovy regain?
The absolute number of pounds is larger for Zepbound (~14% of body weight vs ~7% for semaglutide), but the proportion regained (~67% of prior loss) is nearly identical between the two drugs. Zepbound just enables larger loss to begin with, so there is more to lose back.
Can I keep the weight off with just diet and exercise after stopping?
Some patients do — but they are a minority. The STEP-4 and SURMOUNT-4 trials both included lifestyle-intervention arms during the placebo phase. Even with structured lifestyle intervention, mean regain was 6.9% and 14.0% respectively. Individual outliers exist; the average patient does not maintain full loss without ongoing pharmacotherapy.
What is the best time to try tapering off?
AACE 2024 does not endorse a routine taper period — the current guideline is chronic therapy for chronic disease. When taper is chosen, common protocols step from 2.4 mg to 1.7 mg then 1.0 mg then 0.5 mg over 12-24 weeks, monitoring for weight-regain acceleration. If regain exceeds 1% per month on a lower dose, most clinicians recommend re-escalating rather than continuing the taper.
Does resistance training really help prevent regain?
Yes — the mechanism is fat-free-mass preservation, which sustains resting metabolic rate. Direct RCT evidence in the GLP-1 discontinuation context is limited, but the broader weight-maintenance literature (National Weight Control Registry, ADA-endorsed reviews) is unambiguous that resistance training is the single most protective activity against long-term regain.

Disclosure: Glpverdict is affiliate-funded but editorially independent. Some links above are sponsored — we earn commission on signups with a partner provider. This does not affect our editorial rankings or medical review.

Read this next

GLP-1 rebound weight gain: what STEP-4 and SURMOUNT-4 actually show — and how to prevent it | Glpverdict