STEP 5 answers a simple question about STEP 1: does the weight come back while you are still taking the drug? Over 104 weeks, on average, it did not.
What the trial asked
Whether semaglutide 2.4 mg once weekly, with behavioural intervention, produces and sustains weight loss over two years compared with placebo.
Design
Double-blind, randomised 1:1 to semaglutide 2.4 mg or placebo for 104 weeks, both with behavioural intervention. Efficacy was assessed in all randomised participants regardless of treatment discontinuation or rescue intervention (treatment-policy estimand; the full text also reports a trial-product estimand in its supplement).
Who was enrolled
304 participants randomised between October 2018 and February 2019: 152 to semaglutide, 152 to placebo. Adults with obesity, or overweight with at least one weight-related comorbidity, without diabetes. 236 (77.6 percent) female, 283 (93.1 percent) white, mean age 47.3 (SD 11.0), mean BMI 38.5 (SD 6.9), mean weight 106.0 kg (SD 22.0). 92.8 percent attended the end-of-trial safety visit.
Primary endpoint and result
Coprimary endpoints at week 104: percent change in body weight, and weight loss of 5 percent or more.
| Arm | Mean weight change at week 104 |
|---|---|
| Semaglutide 2.4 mg (n=152) | -15.2% |
| Placebo (n=152) | -2.6% |
Estimated treatment difference -12.6 percentage points (95% CI -15.3 to -9.8; P less than 0.0001). Weight loss of 5 percent or more: 77.1 versus 34.4 percent (P less than 0.0001).
Set that beside STEP 1 at week 68: -14.9 versus -2.4 percent. The two-year number is essentially the same as the 68-week number. The confidence interval is wider because the trial is a sixth of the size.
Key secondary results
The abstract does not list secondary endpoints beyond the responder rate.
Adverse events and discontinuation
Gastrointestinal adverse events, mostly mild to moderate: 82.2 percent with semaglutide versus 53.9 percent with placebo. From Table 3 of the PMC full text: adverse events led to trial-product discontinuation in 9 participants (5.9 percent) on semaglutide and 7 (4.6 percent) on placebo; gastrointestinal events led to permanent discontinuation in 6 (3.9 percent) versus 1 (0.7 percent). Serious adverse events: 12 (7.9 percent) versus 18 (11.8 percent).
Funding and registration
The abstract does not name the funder; the full text and the trial registration identify Novo Nordisk as sponsor. ClinicalTrials.gov NCT03693430.
What this trial does not show
- 304 people, 93 percent white, 78 percent female. It is a small, homogeneous sample and its confidence interval reflects that.
- It shows the average did not drift up during continued treatment. It does not show what happens after stopping; that is STEP 4.
- Two years is still not long-term for a chronic condition. Longer exposure data come from the cardiovascular trial SELECT, which measured events rather than weight as its primary endpoint.
Where the numbers come from
PubMed abstract (PubMed 36216945) and the open-access PMC full text (PMC9556320), fetched 4 September 2026.
Sources
- Two-year effects of semaglutide in adults with overweight or obesity: the STEP 5 trial. Nat Med 2022;28(10):2083-2091. PubMed 36216945 Accessed September 4, 2026.
- PMC full text, PMC9556320 (Table 3, adverse events) Accessed September 4, 2026.
- ClinicalTrials.gov NCT03693430 Accessed September 4, 2026.
Canonical URL: https://formblendsresearch.com/trials/step-5. Written by the FormBlends editorial team. This page is educational and is not medical advice; see the medical disclaimer.