STEP 12 reports greater 44-week weight loss with semaglutide in mainland China and Taiwan

What happened

The STEP 12 trial found greater average weight loss with once-weekly semaglutide 2.4 mg than with placebo among adults enrolled in mainland China and Taiwan. Both groups also received a lifestyle intervention during the 44-week treatment period.

The randomized, double-blind phase 3b trial enrolled 242 adults at 19 sites. Eligibility began at a BMI of 24 for people who also had a weight-related condition. Adults in the 28 to under 30 BMI range could enroll whether or not they had type 2 diabetes. The study assigned 161 people to semaglutide and 81 to placebo. Forty-seven participants, or 19.4%, had type 2 diabetes.

At week 44, average body weight had fallen 12.1% in the semaglutide group and 2.2% in the placebo group. The estimated difference was 9.9 percentage points. At least 5% weight loss was recorded for 80.5% of participants assigned semaglutide and 24.4% assigned placebo. Both comparisons met the trial's statistical threshold.

Adverse events were reported by 87.6% of semaglutide participants and 75.3% of placebo participants. Problems involving the digestive system appeared most often. Safety results were analyzed descriptively, which means the trial did not test whether the difference in overall adverse-event rates was statistically significant.

What it means

This study supplies randomized evidence from adults enrolled under BMI thresholds used in mainland China and Taiwan. Under the study conditions, adding semaglutide to the lifestyle program produced greater average weight loss than adding placebo.

The study also shows why regional trial evidence matters. Entry thresholds, body composition, health profiles, clinical practice, and treatment access are not identical across populations. A trial designed for the population it enrolls can answer questions that a broad global result may not resolve.

What it does not mean

No active obesity medicine served as the comparator, so the results cannot rank semaglutide against another treatment. The study also does not establish what happens after several years of therapy or after treatment is stopped. Follow-up lasted 44 weeks, and the 242-person sample was too small to characterize uncommon adverse effects.

The 12.1% result is a group average, not a promised individual outcome. Missing week-44 values were handled with washout multiple imputation, so the estimate depends in part on assumptions about those missing outcomes. The PubMed abstract does not report separate weight or safety results for the subgroup with type 2 diabetes.

Novo Nordisk funded the trial. Several authors reported employment, stock ownership, consulting work, or other industry relationships. Those disclosures do not erase the randomized result, but they are part of the evidence record.