SELECT Trial Analysis Finds Baseline Characteristics Cannot Predict How Much Weight a Person Will Lose on Semaglutide

Diabetes Care, a journal of the American Diabetes Association, published a prespecified secondary analysis of the SELECT cardiovascular outcomes trial on September 24, 2026, addressing a question that matters more to individual patients than to the trial's headline result: who loses the most weight on semaglutide, and can that be predicted in advance. SELECT enrolled 17,604 adults with overweight or obesity and established atherosclerotic cardiovascular disease but not diabetes, with a mean age of 62 and 72% men. The analysis was led by Ildiko Lingvay of UT Southwestern Medical Center, with co-authors from Northwestern University, the Cleveland Clinic, Harvard Medical School, University College London and other academic centers, alongside several Novo Nordisk employees; Novo Nordisk manufactures semaglutide and funded the trial.

The analysis covered 7,594 SELECT participants who were randomized to semaglutide and continued treatment for at least one year, representing 86.3% of everyone assigned to the drug. Researchers tested a wide range of baseline candidate predictors, including demographics, medical history, laboratory variables, and clinical measurements, against each participant's peak percentage body weight loss (%BWL) using a general linear model.

Median peak weight loss across the group was 12.5% of body weight (interquartile range 7.9% to 17.9%), reached after a median of 19 months of treatment (IQR 10 to 30 months). But the full predictive model, built from all the baseline characteristics tested, explained less than 10% of the variability in how much weight any individual participant ultimately lost. Female sex was the single largest contributor to that limited explanatory power, accounting for 85.3% of the model's predictive value, which itself explained only about 8% of the overall variability.

Two factors were associated with clearly larger differences in outcome. Dose mattered: participants who reached the maximum semaglutide dose of 2.4 mg had a median peak weight loss of 13.3% (IQR 9.0% to 18.4%), compared with 6.0% (IQR 1.9% to 10.9%) for those capped at the lowest dose of 0.24 mg. Sex mattered too: women reached a higher median peak weight loss, 16.6% (IQR 10.9% to 23.4%), than men, at 12.4% (IQR 8.4% to 16.9%).

The authors concluded that the amount of weight loss an individual can expect from semaglutide cannot be predicted from the clinically available baseline characteristics they examined, and wrote that unmeasured biological, behavioral, or genetic differences are more likely responsible for the wide variation between patients. The finding does not change SELECT's previously reported cardiovascular results, but it underscores that, outside of dose and sex, doctors currently have little clinical information to forecast how a given patient will respond before starting treatment.

Semaglutide is already approved by the FDA and the European Medicines Agency and registered with Australia's Therapeutic Goods Administration, marketed as Wegovy for chronic weight management and cardiovascular risk reduction and as Ozempic and Rybelsus for type 2 diabetes. This analysis does not change that regulatory status in any market, does not establish a new mechanism of action, and is not treatment or dosing guidance; decisions about starting, adjusting, or stopping any GLP-1 therapy should be made with a qualified clinician.