ABSTRACT: Abstract Background: In people with overweight or obesity, long-term maintenance of weight loss is challenging. Subcutaneous (s.c.) semaglutide, a glucagon-like peptide-1 analogue, has shown clinically-relevant weight loss in a phase 2 trial in people with obesity. STEP 4 investigated the impact of continued semaglutide 2.4 mg treatment, vs switching to placebo, on maintenance of weight loss in participants who reached 2.4 mg of semaglutide during a run-in period. Methods: This was a 68-week withdrawal trial (NCT03548987) in 902 subjects aged ≥18 years with body mass index (BMI) ≥30 kg/m2 (or BMI ≥27 kg/m2 with ≥1 weight-related comorbidity), without diabetes. Following a 20 week run-in period, 803 subjects who reached the maintenance dose of once-weekly (OW) s.c. semaglutide 2.4 mg were randomized 2:1 to continue treatment with semaglutide 2.4 mg or switch to placebo for 48 weeks, both as adjunct to lifestyle intervention. The primary endpoint was percentage change in body weight between randomization (week 20) and week 68. Confirmatory secondary endpoints included change in waist circumference and systolic blood pressure. Two estimands were defined: treatment policy and trial product; results are presented for the treatment policy estimand, unless stated otherwise. Results: Mean body weight (±SD) was 107.2 ±22.7 kg at week 0 and 96.1 ±22.6 kg at randomization (week 20; mean change -10.6%). Randomized participants were mostly female (79%) and white (84%); mean age was 46 years and mean BMI was 34.4 kg/m2. Between weeks 20–68, estimated mean body weight change was −7.9% vs +6.9% for semaglutide 2.4 mg vs placebo (estimated treatment difference [ETD]: −14.8%; 95% confidence interval [CI]: −16.0, −13.5; p<0.0001), and -8.8% vs 6.5%, respectively, for the trial product estimand (ETD: -15.3%; 95% CI: -16.5, -14.1; p<0.0001). For participants randomized to continue semaglutide, the estimated change in body weight from week 0–68 was -17.4% (-18.2% for trial product estimand). Continued semaglutide treatment (weeks 20–68) led to clinically-relevant improvements in waist circumference, systolic blood pressure, BMI, HbA1c, FPG, and lipids (total cholesterol, LDL, VLDL, and triglycerides) vs switching to placebo (p<0.0001 for all). During the run-in period, 5.3% of participants discontinued treatment due to adverse events; during the randomized period, 2.4% (semaglutide) and 2.2% (placebo) discontinued. Nausea, diarrhea and constipation (mostly transient and mild-to-moderate) were the most frequent adverse events with semaglutide. Conclusion: In adults with overweight or obesity, continued treatment after dose escalation with OW s.c. semaglutide 2.4 mg until week 68 led to clinically-relevant weight loss, while switching to placebo led to significant weight regain; these data underscore the chronicity and relapsing nature of obesity, and the need for continued treatment.