Effect of liraglutide treatment on body mass index and weight parameters in children and adolescents with type 2 diabetes: Post hoc analysis of the ellipse trial

Abstract

Background: Weight loss in children and adolescents with type 2 diabetes (T2D) is associated with improved glycaemic control.

Objectives: To assess the effects of liraglutide vs placebo on body mass index (BMI) and weight parameters in children and adolescents with T2D using data from the ellipse trial (NCT01541215).

Methods: The ellipse trial randomized participants (10-<17 years old, BMI >85th percentile, T2D, glycated haemoglobin [HbA_1c ] 7.0%-11.0% [if diet- and exercise-treated] or 6.5% to 11.0% [if treated with metformin, basal insulin or both]) to liraglutide or placebo. This post-hoc analysis evaluated changes from baseline to weeks 26 and 52 in absolute BMI, percent change in BMI and other weight-related parameters. Changes were assessed by liraglutide overall (all doses) and liraglutide by dose (0.6, 1.2 and 1.8 mg/day) vs placebo using a pattern mixture model of observed data, with missing observations imputed from each treatment group.

Results: In total, 134 participants were included. There were statistically significant differences between groups in certain parameters, including absolute BMI (estimated treatment difference [ETD] -0.89 kg/m² ; 95% confidence interval [CI] -1.71,-0.06) and percent change in BMI (ETD -2.73%; 95% CI -5.15,-0.30) at week 52, but none at week 26. Dose-dependent effects were not observed for liraglutide vs placebo for all BMI/weight parameters.

Conclusions: Compared with placebo, liraglutide was associated with statistically significant reductions in BMI/weight parameters at week 52, but not week 26, in children and adolescents with T2D.

Keywords: anti-obesity agents; body mass index; liraglutide; paediatric obesity; weight.

FIGURE 1

Differences in mean change from baseline in BMI and weight parameters between placebo and liraglutide overall and by liraglutide dose, at week 52 of the ellipse trial. Data analysed using a PMM of observed data with missing observations imputed within each randomized treatment group based on multiple (×10,000) imputations. Data were then analysed with ANCOVA model containing treatment (consisted of four groups: placebo and the three doses of liraglutide), sex and age group as fixed effects and baseline value as covariate. ETDs and CIs were combined using Rubin's formula. Participants categorized by dose taken for the longest time during the maintenance period, which consisted of the double‐blind and open‐label parts of the trial. %BMIp95, percentage of the 95th percentile for BMI; ΔBMIp95, difference in BMI from 95th percentile BMI; BMI, body mass index; CI, confidence interval; ETD, estimated treatment difference from baseline; PMM, pattern mixture model; TMI, tri‐ponderal mass index; WC, waist circumference