STEP 4: continuing versus stopping semaglutide after 20 weeks

STEP 4 is the semaglutide withdrawal trial. It is shorter on the front end than SURMOUNT-4 (20 weeks of drug before randomisation instead of 36) and the regain it measured is correspondingly smaller in absolute terms, but the shape of the result is the same.

What the trial asked

Whether continuing semaglutide 2.4 mg, compared with switching to placebo, maintains weight loss after a 20-week run-in on the drug.

Design

Randomised, double-blind, 68-week phase 3a withdrawal study at 73 sites in 10 countries, June 2018 to March 2020. All participants took semaglutide during a 20-week run-in (16 weeks of escalation, 4 weeks at 2.4 mg). Those who reached the 2.4 mg maintenance dose were randomised 2:1 to continue semaglutide (n=535) or switch to placebo (n=268) for 48 weeks, with lifestyle intervention in both groups.

Who was enrolled

902 adults with a BMI of at least 30, or 27 or more with at least one weight-related comorbidity, without diabetes, entered the run-in. 803 (89.0 percent) reached 2.4 mg and were randomised. Mean age 46 (SD 12); 634 (79 percent) women. The abstract gives a mean body weight of 107.2 kg (SD 22.7) for the randomised group without saying whether it is the week 0 or week 20 figure, so it is reported as written. Mean weight loss during the run-in was 10.6 percent. 787 (98.0 percent) completed the trial and 741 (92.3 percent) completed treatment.

Primary endpoint and result

Primary endpoint: percent change in body weight from week 20 to week 68.

Difference -14.8 percentage points (95% CI -16.0 to -13.5; P less than 0.001).

Key secondary results

Confirmatory secondary endpoints, continued semaglutide versus placebo, all P less than 0.001:

• Waist circumference: -9.7 cm (95% CI -10.9 to -8.5).
• Systolic blood pressure: -3.9 mm Hg (95% CI -5.8 to -2.0).
• SF-36 physical functioning score: +2.5 (95% CI 1.6 to 3.3).

Adverse events and discontinuation

Gastrointestinal events: 49.1 percent with continued semaglutide versus 26.1 percent with placebo. Discontinuation because of adverse events was similar: 2.4 percent versus 2.2 percent.

Funding and registration

The abstract does not name the funder. ClinicalTrials.gov NCT03548987.

What this trial does not show

• Only participants who tolerated and reached 2.4 mg were randomised. The 11 percent who did not are not in the primary result, and the low adverse-event discontinuation rate after randomisation partly reflects that selection.
• The placebo arm regained 6.9 percent over 48 weeks from a 10.6 percent run-in loss. The trial does not follow them long enough to say whether the regain would have continued to baseline.
• No taper, no reduced-dose arm, no switch to a different drug.
• It cannot separate the effect of stopping the drug from the effect of knowing one might have stopped; blinding limits but does not remove that.

Where the numbers come from

PubMed abstract of the primary paper (PubMed 33755728), fetched 4 September 2026. The tirzepatide equivalent is SURMOUNT-4; two-year continuous treatment is STEP 5.