BACKGROUND: Semaglutide 2·4 mg once weekly has been investigated for weight management in global populations. Differences exist between Asian and non-Asian populations in terms of body composition and definitions of obesity. In the Semaglutide Treatment Effect in People with obesity (STEP) 6 trial, we assessed the effect of semaglutide versus placebo for weight management in adults from east Asia with obesity, with or without type 2 diabetes. METHODS: This randomised, double-blind, double-dummy, placebo-controlled, phase 3a superiority trial was done at 28 outpatient clinics in Japan and South Korea. Eligible participants were adults (aged ≥18 years in South Korea; ≥20 years in Japan) with a BMI of at least 27·0 kg/m with two or more weight-related comorbidities or a BMI of 35·0 kg/m or more with one or more weight-related comorbidity (one comorbidity had to be either hypertension, dyslipidaemia, or, in Japan only, type 2 diabetes) who had at least one self-reported unsuccessful dietary attempt to lose bodyweight. Participants were randomly assigned (4:1:2:1) to once-weekly subcutaneous semaglutide 2·4 mg or matching placebo, or semaglutide 1·7 mg or matching placebo, plus lifestyle recommendations for 68 weeks. Data for the placebo groups were pooled in statistical analyses. The primary endpoints were percentage change in bodyweight from baseline at week 68 and the proportion of participants who had achieved a reduction of at least 5% of baseline bodyweight at week 68. Change in abdominal visceral fat area was assessed as a supportive secondary endpoint using computed tomography scanning in a subset of participants. Efficacy outcomes were assessed in the full analysis set, which included all randomly assigned participants according to the intention-to-treat principle. Safety was assessed in all participants who received at least one dose of the study drug. This trial was registered with ClinicalTrials.gov, NCT03811574. FINDINGS: Between Jan 21, 2019 and June 4, 2019, 437 participants were screened, of whom 401 were randomly assigned to semaglutide 2·4 mg (n=199), semaglutide 1·7 mg (n=101), or placebo (n=101) and included in the intention-to-treat analysis. Estimated mean change in bodyweight from baseline to week 68 was -13·2% (SEM 0·5) in the semaglutide 2·4 mg group and -9·6% (0·8) in the semaglutide 1·7 mg group versus -2·1% (0·8) in the placebo group (estimated treatment difference [ETD] -11·1 percentage points [95% CI -12·9 to -9·2] for semaglutide 2·4 mg vs placebo; -7·5 percentage points [95% CI -9·6 to -5·4] for semaglutide 1·7 mg vs placebo; both p<0·0001). At week 68, a larger proportion of participants had achieved a 5% or higher reduction in baseline bodyweight in the semaglutide 2·4 mg group (160 [83%] of 193 participants) and semaglutide 1·7 mg group (71 [72%] of 98 participants) than in the placebo group (21 [21%] of 100 participants); odds ratio [OR] 21·7 [95% CI 11·3 to 41·9] for semaglutide 2·4 mg vs placebo; OR 11·1 [95% CI 5·5 to 22·2] for semaglutide 1·7 mg vs placebo; both p<0·0001). Abdominal visceral fat area was reduced by 40·0% (SEM 2·6) among participants in the semaglutide 2·4 mg group and 22·2% (3·7) among participants in the semaglutide 1·7 mg group versus 6·9% (3·8) in the placebo group (ETD -33·2% [95% CI -42·1 to -24·2] for semaglutide 2·4 mg vs placebo; -15·3% [95% CI -25·6 to -4·9] for semaglutide 1·7 mg vs placebo). 171 (86%) of 199 participants in the semaglutide 2·4 mg group, 82 (82%) of 100 participants in the semaglutide 1·7 mg group, and 80 (79%) of 101 participants in the placebo group reported adverse events. Gastrointestinal disorders, which were mostly mild to moderate, were reported in 118 (59%) of 199 participants in the semaglutide 2·4 mg group, 64 (64%) of 100 participants in the semaglutide 1·7 mg group, and 30 (30%) of 101 participants in the placebo group. Adverse events leading to trial product discontinuation occurred in five (3%) of 199 participants in the semaglutide 2·4 mg group, three (3%) of 100 participants in the semaglutide 1·7 mg group, and one (1%) of 101 participants in the placebo group. INTERPRETATION: Adults from east Asia with obesity, with or without type 2 diabetes, given semaglutide 2·4 mg once a week had superior and clinically meaningful reductions in bodyweight, and greater reductions in abdominal visceral fat area compared with placebo, representing a promising treatment option for weight management in this population. FUNDING: Novo Nordisk. TRANSLATIONS: For the Korean and Japanese translations of the abstract see Supplementary Materials section.