Does GLP-1 weight loss last? What two- and three-year trials found

Why longer trials matter

The trials most people have heard of, STEP 1 and SURMOUNT-1, measured their main results at 68 and 72 weeks. That leaves an obvious question. Does the weight stay off if you keep taking the medicine, or does it creep back after the first year? STEP 5 and the three-year SURMOUNT-1 analysis were designed to answer it.

Both trials kept people on their assigned treatment, medicine or placebo, the whole time. Both also gave everyone lifestyle support. They tell you about continued use. What happens after stopping is a different question, covered in what happens when you stop a GLP-1.

STEP 5: two years of semaglutide

STEP 5 enrolled 304 adults without diabetes who had obesity, or overweight plus at least one weight-related health problem. On average they were 47 years old, weighed 106.0 kg and had a BMI of 38.5. Most were women (77.6%) and most were white (93.1%). Half took semaglutide 2.4 mg once a week and half took placebo, for 104 weeks.¹

Weight loss on semaglutide levelled off at about week 60 and was kept for the rest of the trial. The small difference between year one and year two led the authors to say there was, on average, minimal weight regain while treatment continued.¹

How many people reached each milestone at two years

Blood sugar also moved in a good direction. Among people who started the trial with prediabetes, 79.7% on semaglutide had normal blood sugar at week 104, vs 37.0% on placebo. This was an exploratory result, and the authors say it needs confirming.¹

Most people stayed the course: 92.8% completed the trial. Stomach and gut side effects were reported by 82.2% on semaglutide and 53.9% on placebo, mostly mild to moderate. Nausea was the most common, in 53.3% vs 21.7%. Side effects led 5.9% of people on semaglutide to stop treatment, vs 4.6% on placebo.¹

SURMOUNT-1 at three years: tirzepatide and prediabetes

SURMOUNT-1 enrolled 2,539 adults with obesity. Its main results, at 72 weeks, showed average weight loss of 15.0% to 20.9% depending on dose, vs 3.1% on placebo.³ Of all the participants, 1,032 also had prediabetes. That group stayed on their assigned dose of tirzepatide (5, 10 or 15 mg a week) or placebo for 176 weeks, a little over three years, followed by 17 weeks off treatment.²

The second question was whether tirzepatide slowed the move from prediabetes to type 2 diabetes. It did, by a wide margin.

In statistical terms, the hazard ratio for developing diabetes during treatment was 0.07, which means a much lower risk with tirzepatide than with placebo. After 17 weeks off treatment, the gap had narrowed a little, but the risk was still clearly lower (hazard ratio 0.12).²

Apart from COVID-19, the most common side effects were in the stomach and gut. Most were mild to moderate and happened mainly in the first 20 weeks, while the dose was being raised. The authors reported no new safety concerns over the longer follow-up.²

Reading these results

• Who was studied. STEP 5 enrolled adults without diabetes. The three-year SURMOUNT-1 results apply to adults with obesity and prediabetes. Results may differ for people with type 2 diabetes; see GLP-1 weight loss in type 2 diabetes.
• A plateau is normal. In STEP 5, weight came down until about week 60 and then held steady. SURMOUNT-1 also reported weight loss that was sustained over three years. A plateau after the first year matches what the trials saw.
• Continued treatment. These results come from people who kept taking the medicine. In SURMOUNT-1, diabetes diagnoses rose a little in the 17 weeks after stopping.
• Averages hide a wide range. In STEP 5, about a third of people on semaglutide lost 20% or more at two years, while nearly a quarter lost less than 5%.

How long to stay on treatment is a decision to make with your prescriber, based on your own results and health.