In STEP 5, the two-year trial of semaglutide 2.4 mg, the authors wrote that reductions in weight, waist circumference, blood pressure and HbA1c appeared to plateau around week 60 and were then maintained for the rest of the study (PMID 36216945). Week 60 is about 14 months in.
So a plateau somewhere past the first year is the documented shape of the curve rather than a sign that something has gone wrong. What follows is when each trial flattened, the two things that make it flatten, and the one trial that shows what holding a plateau actually takes.
- STEP 5 reported the plateau explicitly: around week 60, with the mean at -15.2% holding through week 104 in 304 adults on semaglutide 2.4 mg without diabetes.
- The flattening has two causes arriving together: the dose stops escalating, and weight loss itself drives appetite up by roughly 100 kcal a day per kilogram lost.
- A plateau is not the drug failing. In STEP 4, people who switched to placebo at week 20 regained 6.9% while those who continued lost a further 7.9%.
- Three months in is not a plateau. The first 16 to 20 weeks of either label are escalation steps, and a real-world cohort was still at 5.9% at three months against 10.9% at six.
When each curve actually flattened
Every row names its drug, dose, population and duration, because a plateau at 68 weeks and a plateau at 20 weeks are different claims.
| Trial | Drug and dose | Population | Duration | Result and curve shape |
|---|---|---|---|---|
| STEP 1 (PMID 33567185) | Semaglutide 2.4 mg | Overweight or obesity, no diabetes | 68 weeks | -14.9%; fast through the escalation, slower to the endpoint |
| STEP 4 (PMID 33755728) | Semaglutide 2.4 mg | Overweight or obesity, no diabetes | 20-week run-in, then 48 weeks | -10.6% by week 20, then -7.9% more on continued treatment |
| STEP 5 (PMID 36216945) | Semaglutide 2.4 mg | Overweight or obesity, no diabetes | 104 weeks | -15.2%; plateau reported around week 60, then maintained |
| SURMOUNT-1 (PMID 35658024) | Tirzepatide 5 / 10 / 15 mg | Obesity, diabetes excluded | 72 weeks | -15.0% / -19.5% / -20.9% at the endpoint |
| SURMOUNT-1 at 3 years (PMID 39536238) | Tirzepatide 5 / 10 / 15 mg | Obesity with prediabetes | 176 weeks | -12.3% / -18.7% / -19.7%; held, slightly below the 72-week figures |
The last two rows are worth reading carefully. They are the same trial, but the 176-week analysis covers only the 1032 participants who had prediabetes at randomisation, while the 72-week figures are the full 2539. That makes the small difference between them a population difference as much as a time one. What it does establish is that three years of treatment held close to 20% at the top dose rather than drifting back up.
Three months in is not a plateau
Most searches about a plateau come earlier than any of these trials flattened. Both labels build the dose in four-week steps: semaglutide passes through 0.25, 0.5, 1 and 1.7 mg before reaching 2.4 mg at week 17, and tirzepatide climbs in 2.5 mg increments. SURMOUNT-1 allowed 20 weeks for escalation alone.
During that window the dose is still rising, so the appetite effect is still rising with it. A real-world cohort of 175 patients on semaglutide averaged 5.9% at three months and 10.9% at six, which is a curve still falling steeply. A flat fortnight inside that window is noise, and what the first weeks actually look like is covered in your first month on a GLP-1.
What makes the curve flatten
The more interesting mechanism is not a drop in metabolism, which is the usual explanation offered. It is appetite.
A 2016 analysis used a validated method to back-calculate energy intake from the weight curves of 153 patients in a 52-week placebo-controlled trial of canagliflozin, a drug that causes weight loss by excreting glucose in urine, so participants were not aware of the energy deficit being created. Weight loss drove a proportional increase in appetite worth about 100 kcal a day of extra eating per kilogram of weight lost, which the authors noted was more than threefold larger than the corresponding adaptation in energy expenditure (PMID 27804272).
Applied to a 15 kg loss, that is a pull of roughly 1500 kcal a day acting against the deficit. A GLP-1 blunts that pull rather than removing it, so the curve flattens at the point where the drug’s appetite suppression and the body’s counter-pressure balance out. Reduced energy expenditure contributes, but on these numbers it is the smaller of the two forces.
The dose ceiling arrives at the same time
The second cause is simpler. Once the maintenance dose is reached, the dose stops going up, so the drug effect stops growing while the counter-pressure keeps growing with every kilogram lost. That is why the plateau tends to land some months after the final escalation step rather than immediately at it.
See where your own curve should be
Put a starting weight and a timeline against the STEP and SURMOUNT trial figures.
A plateau is not the drug stopping working
STEP 4 is the trial that separates those two ideas. All 902 participants took semaglutide through a 20-week run-in, and the 803 who reached the 2.4 mg maintenance dose, having lost a mean of 10.6%, were randomised to continue it or switch to placebo for 48 weeks. Both groups kept the same lifestyle support.
Mean weight change from week 20 to week 68 was -7.9% on continued semaglutide and +6.9% on placebo, a 14.8 percentage point gap. The people who looked like they had stopped responding were, in fact, holding a position that required the drug to hold.
So a flat line on the scale at month 14 and a flat line because the treatment has failed look identical week to week, and the trial evidence says the first is far more likely. STEP 5’s thresholds at 104 weeks make the same point from the other direction: 77.1% of the semaglutide group were still at least 5% below baseline, 52.1% at least 15% below, and 36.1% at least 20% below.
Where the ceiling itself moves
If the plateau is set partly by the dose ceiling, raising the ceiling should push the plateau further down, and that is what the 7.2 mg trial found. STEP UP randomised 1407 adults with obesity and no diabetes to semaglutide 7.2 mg, 2.4 mg or placebo for 72 weeks and recorded -18.7%, -15.6% and -3.9%. The trade-off showed up in the safety data: altered skin sensation was reported by 22% of the 7.2 mg arm against 6% on 2.4 mg.
This is a description of what the trials and labels contain, not a suggestion about anyone’s own dose. Which dose a person is on, and whether it changes, is a decision for their prescriber.
Plateau and regain are different questions
A plateau is what the curve does while treatment continues. What happens after it stops is a separate body of evidence: in the STEP 1 extension, participants regained 11.6 percentage points in the year after treatment ended, and SURMOUNT-4 measured a larger regain. That evidence is set out in what happens when you stop a GLP-1.
One thing worth watching through the flat stretch is body composition rather than the scale number, since weight that is not moving can still be changing in make-up. The trial measurements on that are in muscle loss on a GLP-1, and the fuller set of semaglutide percentages is in how much weight you can lose on Wegovy.
FAQ
When does GLP-1 weight loss plateau?
In STEP 5, the two-year semaglutide 2.4 mg trial, the authors reported that reductions in weight appeared to plateau around week 60, which is about 14 months, and were then maintained through week 104. STEP 1 and STEP 4 show the same shape over 68 weeks.
Why does weight loss stop on a GLP-1?
Two things arrive at once. The dose stops climbing once the maintenance dose is reached, so the appetite effect stops increasing. And weight loss itself raises appetite: a 52-week analysis quantified the feedback at about 100 kcal a day of extra eating per kilogram of weight lost, more than three times the size of the drop in energy expenditure over the same period.
Is a plateau a sign the drug has stopped working?
Not in the trial data. STEP 4 took 803 adults who had already lost a mean of 10.6% during a 20-week run-in and randomised them to continue semaglutide 2.4 mg or switch to placebo. Over the next 48 weeks the continued group lost a further 7.9% while the placebo group regained 6.9%. Holding a plateau is itself an active effect.
Can you lose more weight after a plateau?
The trial curves flatten rather than resume falling. In STEP 5 the mean was -15.2% at 104 weeks against roughly the same figure at week 60, and in the three-year SURMOUNT-1 analysis the 176-week figures were slightly smaller than the 72-week ones. Anything that would change that, including a dose change, is a decision for the prescriber.
Is a plateau at three months normal?
Three months is usually too early to be a plateau at all. On both labels the first 16 to 20 weeks are four-week escalation steps, so the maintenance dose has not been reached. A real-world semaglutide cohort averaged 5.9% at three months and 10.9% at six months, so the curve was still falling steeply through that window.
Does a higher dose move the plateau?
In the trials, a higher ceiling produced a lower plateau. STEP UP randomised 1407 adults with obesity and no diabetes to semaglutide 7.2 mg, 2.4 mg or placebo for 72 weeks and recorded -18.7%, -15.6% and -3.9%. Altered skin sensation was reported by 22% of the 7.2 mg group against 6% on 2.4 mg.
Sources
- Garvey WT, Batterham RL, Bhatta M, et al. Two-year effects of semaglutide in adults with overweight or obesity: the STEP 5 trial. Nat Med. 2022;28(10):2083-2091. PMID: 36216945
- Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1). N Engl J Med. 2021;384(11):989-1002. PMID: 33567185
- Rubino D, Abrahamsson N, Davies M, et al. Effect of Continued Weekly Subcutaneous Semaglutide vs Placebo on Weight Loss Maintenance in Adults With Overweight or Obesity: The STEP 4 Randomized Clinical Trial. JAMA. 2021;325(14):1414-1425. PMID: 33755728
- Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1). N Engl J Med. 2022;387(3):205-216. PMID: 35658024
- Jastreboff AM, le Roux CW, Stefanski A, et al. Tirzepatide for Obesity Treatment and Diabetes Prevention. N Engl J Med. 2025;392(10):958-971. PMID: 39536238
- Polidori D, Sanghvi A, Seeley RJ, Hall KD. How Strongly Does Appetite Counter Weight Loss? Quantification of the Feedback Control of Human Energy Intake. Obesity (Silver Spring). 2016;24(11):2289-2295. PMID: 27804272
- Wharton S, Davies M, Lingvay I, et al. Once-weekly semaglutide 7.2 mg in adults with obesity (STEP UP): a randomised, controlled, phase 3b trial. Lancet Diabetes Endocrinol. 2025. PMID: 40961952
- Ghusn W, De la Rosa A, Sacoto D, et al. Weight Loss Outcomes Associated With Semaglutide Treatment for Patients With Overweight or Obesity. JAMA Netw Open. 2022;5(9):e2231982. PMID: 36121652
- Wilding JPH, Batterham RL, Davies M, et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: The STEP 1 trial extension. Diabetes Obes Metab. 2022;24(8):1553-1564. PMID: 35441470
This article is for information only and is not medical advice. Dosing, suitability and any change to treatment are decisions for your prescriber.