Health & Wellness 4 min read

Weekly Semaglutide Cut BMI in Children Aged 6 to 11, but Part of the Drop Returned While Still on Treatment

A Novo Nordisk-funded phase 3 trial of 167 young patients found a 16.2-point gap over placebo at week 68. By week 104, before any washout, 4.1 points of the fall had come back.

Priya Nair
Data, AI Governance & Policy Analyst
Published 17 Nov 2026, 10:39 PM (SGT)
Share:
Two teddy bears with a stethoscope, a common symbol of children's medicine Two teddy bears with a stethoscope, a common symbol of children's medicine Photo by Myriams-Fotos on Pixabay
Advertisement

17 NOV 2026 — A phase 3 trial published online in the New England Journal of Medicine on 16 November tested weekly semaglutide in children aged 6 to 11 with obesity. At 68 weeks, the semaglutide group had lowered body-mass index (BMI) by 16.2 percentage points more than a placebo, the authors report. The abstract's week 104 numbers show part of that reduction coming back while the children were still on treatment.

The trial was funded by Novo Nordisk, which makes the drug, and four of the 11 named authors give the company as their affiliation.

Who took part, and on what?

The double-blind trial, according to its study abstract, split 167 children two to one by chance: 112 got semaglutide and 55 a placebo. Both groups had lifestyle counselling. Injections went under the skin once a week for 104 weeks, followed by a seven-week washout with no treatment. The authors note that the drug is already approved for obesity in adolescents and adults, while data for young children "are lacking".

The trial's registry record lists the entry criteria. A child needed a BMI at or above the 95th percentile for age and sex, a body weight above 45 kg, and at least one unsuccessful attempt to lose weight after a structured diet-and-exercise programme lasting three months or more. A hormonal, hypothalamic or syndromic cause of obesity ruled a child out. The dose rose in steps over 16 weeks, from 0.25 mg to 1.7 mg, then stayed at 2.4 mg for the remaining 88 weeks.

How far did BMI fall by week 68?

At week 68, the main end point, mean BMI had fallen 13.5% in the semaglutide group and risen 2.7% in the placebo group. The estimated difference of 16.2 percentage points had a 95% confidence interval of 12.9 to 19.4 points.

Of the children on semaglutide, 79% had dropped at least one BMI category by then, against 23% on placebo. Some 38% of the semaglutide group had moved below the obesity line the trial used, the 95th percentile, compared with 2% of the placebo group.

Side effects were common in both arms. Adverse events were recorded in 96% of children on semaglutide and 93% on placebo. Gastrointestinal problems were the most frequent and more common with the drug, and they led one child on semaglutide to stop treatment.

Did the effect hold to week 104?

Partly. On semaglutide, mean BMI was 9.4% below baseline at week 104, against 13.5% at week 68. Over those 36 weeks, the group gave back 4.1 percentage points, or about 30% of its week-68 reduction (4.1 divided by 13.5). The washout only began after week 104, so this happened while the children were still receiving weekly injections.

The placebo group went on rising, from 2.7% to 8.4%, a further 5.7 points. Because placebo climbed faster than semaglutide drifted back, the gap between the arms widened a little, from 16.2 to 17.8 points.

Advertisement
Mean BMI change from baselineWeek 68Week 104Week 68 to 104 (our arithmetic)
Semaglutide (112 children)-13.5%-9.4%4.1 points back towards baseline
Placebo (55 children)+2.7%+8.4%5.7 points further up
Estimated difference-16.2 points-17.8 pointsGap 1.6 points wider

The abstract does not explain the drift. A child's obesity threshold here is a percentile for age and sex, not a fixed number, so a change in a growing child's BMI is harder to read than an adult's.

Where does the evidence stop?

The full article was not open to us, so its tables, the washout results and any subgroup figures are not reported here.

The children in this paper make up one half of a larger study. The registry also enrols a teenage group aged 12 to under 18, records 212 participants enrolled in all, and names Novo Nordisk A/S as lead sponsor. It shows the study as active but no longer recruiting, with completion estimated for 7 December 2026 and no results posted.

The location list names 56 sites in 10 countries, 24 of them in the United States and the rest in Europe, Israel (three) and Mexico (one). None is in South-east, East or South Asia. The list covers both age groups and does not say which children were seen where, so the trial says nothing direct about how the drug performs in children from this region.

Advertisement
Priya Nair
Data, AI Governance & Policy Analyst

Priya Nair covers AI governance, data protection, privacy, and digital trust topics for RECATOOLS.

View author profile → · Editorial policy

About this byline Priya Nair is a RECATOOLS editorial persona for AI governance, privacy, and digital trust coverage. Articles are produced and reviewed under RECATOOLS editorial supervision.

Corrections policy

Advertisement