GLP-1 agonists in adolescent obesity: NMA

A network meta-analysis (NMA) published on 4 July 2026 in Pharmacological Research compared six GLP-1 receptor agonists head-to-head in children and adolescents with obesity, pooling data from 17 randomised controlled trials and 1,230 participants^[1]. The headline finding: semaglutide 2.4 mg produced the largest body-weight reduction in the network, at a mean of 18.00 kg (95% credible interval: 24.34 to 11.69 kg), alongside a BMI reduction of 5.9 kg/m² and a waist circumference reduction of 12.2 cm. The authors caution that most between-drug comparisons are indirect and the findings should be treated as hypothesis-generating.

This article explains what a network meta-analysis is and is not, what the paper found for each agent studied, and what remains unresolved in the paediatric obesity pharmacotherapy literature. This is educational reporting, not clinical guidance. If you are considering pharmacological treatment for a child or adolescent with obesity, the appropriate starting point is a paediatric endocrinologist or obesity-medicine specialist.

What a network meta-analysis does, and where it falls short

A standard meta-analysis pools results from trials that compare the same two treatments. A network meta-analysis goes further: it uses a statistical framework to make indirect comparisons between treatments that have never been compared head-to-head, by routing the comparison through a common comparator (usually placebo). This makes NMAs useful for generating ranked efficacy estimates across a whole drug class, but the indirect comparisons carry more uncertainty than direct trial evidence. Li et al. note explicitly that "most comparisons were indirect" and describe their findings as hypothesis-generating rather than definitive.

The 17 trials pooled were heterogeneous in design, follow-up duration, age range, and background lifestyle support, all factors that affect observed weight loss. The 1,230 total participants is a small number by adult-obesity-trial standards, and the credible intervals for several comparisons were wide. Readers should not interpret the ranked efficacy table as a clinical hierarchy.

Efficacy findings by outcome domain

Weight and BMI outcomes were led by semaglutide 2.4 mg^[1]. The paper reports a mean body-weight reduction of 18.00 kg (95% CrI: 24.34 to 11.69), a BMI reduction of 5.9 kg/m², and a waist-circumference reduction of 12.2 cm. These numbers are consistent with, and contextualised by, the STEP TEENS randomised trial (Weghuber et al., NEJM 2022), which enrolled 201 adolescents aged 12 to 17 and reported a mean BMI change of 16.1% with semaglutide 2.4 mg versus +0.6% with placebo over 68 weeks^[3].

Glycaemic control told a different story. Dulaglutide 1.5 mg showed the largest HbA1c reduction in the network (-1.50%), while lixisenatide 20 mcg produced the largest fasting glucose reduction (-3.98 mmol/L)^[1]. Neither of these agents led on weight outcomes. The divergence between weight-ranked and glycaemia-ranked agents is a key practical point: an agent that is optimal for one outcome may not be optimal for another, and treatment choice in clinical practice depends on which outcome matters most for a given patient.

For blood pressure, exenatide 20 mcg was associated with the largest systolic blood pressure reduction in the network (-6.64 mmHg). On lipid profiles, the paper reports that no agent produced statistically significant improvements. This is notable given that atherogenic dyslipidaemia is common in adolescents with obesity and represents an independent cardiovascular risk factor.

Safety signals in the paediatric population

The paper reports that semaglutide was generally well tolerated in the NMA safety analysis, a finding that aligns with the STEP TEENS direct trial data, where gastrointestinal adverse events occurred in 62% of the semaglutide group versus 42% in the placebo group, and cholelithiasis developed in 4% of semaglutide participants versus none in the placebo arm^[3]. These rates are consistent with the adult STEP programme adverse-event profile.

A second NMA published in JAMA Pediatrics in June 2026, covering 42 trials and 3,835 participants, found that pharmacotherapy combined with lifestyle intervention was more effective than either alone, and that semaglutide plus structured counselling was associated with a BMI reduction of 8.31 kg/m² and a BMI z-score reduction of 1.80^[2].

What this is not

This NMA does not establish that semaglutide is the right treatment for every adolescent with obesity. It does not adjudicate safety over long treatment horizons; the included trials range widely in follow-up length. It does not address the durability question: the STEP-4 discontinuation data in adults showed substantial weight regain when semaglutide was stopped, and equivalent discontinuation trial data in adolescents is limited. And it does not resolve the cost, access, or regulatory picture, which varies considerably by country.

Paediatric obesity pharmacotherapy is an evolving clinical area. The EMA extended the Mounjaro (tirzepatide) label to adolescents and children aged 10 and over with type-2 diabetes in late 2025, adding a second GLP-1 class agent to the European paediatric landscape. Country-by-country regulatory status for semaglutide is documented on the semaglutide regulation pages.

What we do not yet know

Several questions remain open. Long-term efficacy and safety data beyond the trial durations covered in this NMA are absent. The optimal age of initiation and the appropriate duration of treatment in adolescents are unresolved. The effect of pharmacotherapy on growth, bone development, and pubertal progression over multi-year periods has not been systematically characterised. And the interaction between pharmacotherapy and the structured lifestyle support that nearly all included trials provided makes it difficult to isolate the drug effect in real-world settings where that support is not always available.

This uncertainty is the reason the editorial team frames this as a research summary rather than a clinical recommendation. For the underlying adult weight-loss data and the wider regulatory landscape, see the semaglutide peptide page.