Semaglutide weight loss cannot be predicted from baseline features

Design

Prespecified secondary analysis of a randomised trial (SELECT)

Population

7,594 adults with obesity and atherosclerotic disease, no diabetes, on semaglutide ≥1 year

Primary outcome

Peak percentage body weight loss and its baseline predictors

Effect

Median peak loss 12.5% at 19 months; baseline model explained <10% of variation

This prespecified SELECT analysis followed 7,594 adults without diabetes but with established atherosclerotic disease who stayed on semaglutide for at least a year — 86.3% of those randomised to it.

Median peak weight loss was 12.5% (IQR 7.9–17.9), reached after a median of 19 months. A model built from demographics, history, laboratory values and measurements explained under 10% of the variation, and female sex accounted for most of that. Women reached a median of 16.6% against 12.4% in men. Those whose maximum dose was 2.4 mg lost a median 13.3%, against 6.0% at 0.24 mg.

In practice, a patient's age, BMI or HbA1c will not tell you how well they will respond. Response has to be judged on treatment, and the plateau comes later than many expect — a review at three or six months is too early to call the final result.

• Do not promise a patient a particular weight loss from their baseline profile; response varied widely and unpredictably.
• Expect the peak around a year and a half in; judge the response over that horizon, not at the first review.
• Women in SELECT lost more weight than men on average.
• Patients who stayed on low doses lost much less — dose tolerance is worth working on before concluding a poor response.

Why it matters

Clinicians and payers want to select likely responders in advance, and this says the usual clinical variables cannot do it.

Don't overread it

The dose comparison is observational within the semaglutide arm — people who tolerate full dose may differ in other ways.

The statistics, in plain English

Explaining under 10% of the variability means that if you lined patients up by predicted weight loss, the actual results would be scattered almost as widely as if you had not predicted at all. The interquartile range of 7.9–17.9% means the middle half of patients fell somewhere in that ten-point span.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.