In the big withdrawal trials, people who stopped semaglutide or tirzepatide regained a large share of the weight they had lost, and the gains in blood pressure and blood sugar faded with it. One year after stopping semaglutide in the STEP 1 extension, people had regained about two-thirds of their lost weight. In SURMOUNT-4, people switched from tirzepatide to placebo gained back 14.0% of their body weight over a year. This guide walks through the three trials that tested stopping, with the exact numbers, and explains what they can and cannot tell you.
Every number here comes from the published trials or the FDA labels for Wegovy and Zepbound. Where we do our own arithmetic, we say so.
A withdrawal trial asks what happens when you take the drug away. There are two common designs.
The second design is the stronger test, because both groups start from the same point and only one thing changes. In both, the switch to placebo was abrupt, with no taper.
STEP 1 randomized 1,961 adults with obesity or overweight and without diabetes. They took semaglutide 2.4 mg (the Wegovy dose) or placebo once a week for 68 weeks, along with a lifestyle program. At week 68, the drug and the lifestyle program both stopped. A subset of 327 people was then followed for another year off treatment (Wilding et al., 2022).
Not everyone ended in the same place. At week 120, 48.2% of the former semaglutide group (95 of 197) were still at least 5% below their starting weight, compared with 22.6% of the former placebo group. People who had lost the most also tended to regain the most, but they kept a larger net loss. Those who had lost 20% or more by week 68 were still 12.1% below baseline at week 120. Those who had lost less than 5% ended 4.2% above where they began.
The authors also tracked health markers. During treatment, blood pressure, blood sugar, blood fats and inflammation all improved. After stopping, they reported that these improvements "reverted towards baseline" for most measures. The table shows the semaglutide group's observed averages from the paper.
| Measure (STEP 1 extension, semaglutide arm) | Start (week 0) | End of treatment (week 68) | One year off (week 120) |
|---|---|---|---|
| Body weight, kg | 105.6 | 87.5 | 99.0 |
| Systolic blood pressure, mmHg | 129 | 121 | 131 |
| HbA1c, % | 5.7 | 5.2 | 5.6 |
| Triglycerides, mg/dL | 131.1 | 95.7 | 122.4 |
| C-reactive protein, mg/L | 2.95 | 1.28 | 1.83 |
Systolic blood pressure (the top number) went back to its starting level or slightly above. HbA1c, a three-month average of blood sugar, nearly did too. Triglycerides and C-reactive protein, a marker of inflammation, rose but stayed below their starting values.
Two limits matter. All analyses in the extension were labeled exploratory, meaning they were not the trial's main tested question. And because the lifestyle program stopped at the same time as the drug, the extension cannot separate the effect of stopping semaglutide from the effect of stopping the program.
STEP 4 enrolled 902 adults without diabetes. Everyone took semaglutide during a 20-week run-in: 16 weeks of dose increases, then 4 weeks at 2.4 mg. The 803 people who reached the full dose had lost an average of 10.6% of their weight. They were then randomized 2 to 1 to keep taking semaglutide (535 people) or switch to placebo (268 people) for 48 more weeks. Both groups kept the lifestyle program (Rubino et al., 2021).
The Wegovy label describes STEP 4 as "Study 5" and gives the raw changes. From randomization to week 68, waist grew by 3.3 cm on placebo and shrank by 6.4 cm on semaglutide. Systolic blood pressure rose 4.4 mmHg on placebo and 0.5 mmHg on semaglutide (Wegovy label, 14.2).
The label adds a fair caution. Only people who tolerated the drug and reached the full dose were randomized, so the results "may not reflect the experience of patients in the general population who are first starting" Wegovy.
Stomach and gut side effects were reported in 49.1% of people who kept taking semaglutide versus 26.1% on placebo.
SURMOUNT-4 tested the same question with tirzepatide, the drug in Zepbound and Mounjaro. It enrolled 783 adults without diabetes. Everyone took the highest dose they could tolerate, 10 mg or 15 mg a week, for a 36-week open-label lead-in. The 670 people who finished the lead-in had lost an average of 20.9% of their weight. They were then randomized 1 to 1 to keep tirzepatide or switch to placebo for 52 weeks, with diet and activity counseling throughout (Aronne et al., 2024).
The Zepbound label calls this "Study 4" and lists the health markers. Weight, waist, blood pressure and HbA1c had all improved in the lead-in, and all moved back the other way after the switch to placebo (Zepbound label, 14.1, Table 7):
| Change from week 36 to week 88 (SURMOUNT-4) | Switched to placebo | Stayed on tirzepatide |
|---|---|---|
| Body weight | +14.0% | -5.5% |
| Waist, cm | +7.8 | -4.3 |
| Systolic blood pressure, mmHg | +8.2 | +2.0 |
| Diastolic blood pressure, mmHg | +3.2 | -0.7 |
| HbA1c, % | +0.3 | -0.0 |
A year after stopping, the placebo group was still below its starting weight on average, but blood pressure and blood sugar were rising with the weight.
| Trial | Time on drug before stopping | Weight lost on drug | What happened after |
|---|---|---|---|
| STEP 1 extension (semaglutide 2.4 mg) | 68 weeks | 17.3% | Regained 11.6 points in 1 year off; net loss 5.6% |
| STEP 4 (semaglutide 2.4 mg) | 20 weeks | 10.6% | +6.9% over 48 weeks on placebo vs -7.9% on drug |
| SURMOUNT-4 (tirzepatide 10 or 15 mg) | 36 weeks | 20.9% | +14.0% over 52 weeks on placebo vs -5.5% on drug |
The pattern is the same in all three: within a year, much of the lost weight came back.
The STEP 1 authors say their findings "confirm the chronicity of obesity," meaning it behaves like a long-term condition, and "suggest ongoing treatment is required" to keep the benefits.
Stopping is not like flipping a switch, because these drugs clear slowly.
We cover the numbers in depth in how long Ozempic stays in your system and how long tirzepatide stays in your system. You can also try the semaglutide or tirzepatide half-life calculator.
We did not find a taper schedule in the Wegovy or Zepbound labels, and none of the three trials used one. So there is no label-backed way to step down. If you want to stop or lower your dose, that is a plan to make with your prescriber.
Restarting is covered, at least for Wegovy. The label says that if 2 or more doses in a row are missed, you should restart the dose increases at a lower dose to reduce the risk of stomach side effects (Wegovy label, 2.4). The Zepbound label says to skip a missed dose if more than 4 days have passed (Zepbound label, 2.3). For longer breaks on either drug, ask your prescriber how to restart rather than jumping back to your old dose. The missed-dose pages for Wegovy, Ozempic, Zepbound and Mounjaro lay out each label's rule.
People stop for cost, supply, side effects, pregnancy plans or surgery. The trials do not say you can never stop. They show what happened on average when people did.
Dose Track can help with the record keeping. It logs every dose across your compounds, tracks your weight and side effects, and keeps vial and supply counts if you use compounded vials. If you are hitting a stall rather than thinking about stopping, read our guide to the GLP-1 weight-loss plateau. If side effects are the reason, see GLP-1 nausea: what the labels say.
This article is general information, not medical advice. Talk to your prescriber before you start, change or stop any medicine.