Home / Therapies / GLP-1 Agonist Therapy Center / A New Medication Class for Children with Diabetes

A New Medication Class for Children with Diabetes

Nov 16, 2019
Editor: Steve Freed, R.PH., CDE

Author: Adam Chalela, B.S. Doctor of Pharmacy Candidate USF College of Pharmacy

The results of a recent phase III study on the use of liraglutide for children with type 2 diabetes may offer a new route for better glycemic control. 

The incidence of childhood obesity is increasing and exceptionally prevalent in disadvantaged minority populations. Metformin is the current first-line treatment option of choice for patients with newly diagnosed diabetes or mild forms of the disease and has been shown in studies to be a safe option for children and adolescents. Aggressive forms of the disease are due primarily to inadequate blood sugar and HbA1C control with metformin monotherapy in these children and adolescent populations. Currently, the only other drug class approved for children with diabetes who have failed metformin monotherapy is injectable insulin, a frightening option for small children.


Liraglutide, sold under the brand name Victoza in the United States, is a glucagon-like peptide-1 receptor agonist that has recently shown adequate efficacy and safety profiles in adolescents with type 2 diabetes at standard, FDA-approved, adult doses, prompting the swift transition to phase 3 study.

The phase 3 study, named Evaluation of Liraglutide in Pediatrics with Diabetes (Ellipse), was a multicentered, randomized, placebo-controlled, double-blinded trial with an open-label extension period during the second half of the study period that aimed to determine the degree of superiority, if any, that Liraglutide demonstrates in the control of adolescent diabetes as compared to placebo. Patients with type 2 diabetes who were between ages 10 and 17 with an HbA1C of between 6.5% and 11% (patients treated with metformin with or without insulin had stricter HbA1C goals) and a body mass index in the 85th percentile or higher were enrolled into this study; patients with type 1 diabetes were excluded as they are usually insulin-dependent and have a much more complicated disease state to control. A total of 134 patients met this criterion and entered in the Ellipse study between November of 2012 and May of 2018.

Determination of efficacy through patient HbA1C at the end of the 26-week double-blinded phase was the primary endpoint of the Ellipse study. Exploratory secondary endpoints of change in fasting blood glucose, body weight, fasting lipid values, and blood pressure, as well as the total incidence of any A1C values that improved to a value of 7% or lower, were also analyzed. Patients who received at least one dose of either study drug (liraglutide or placebo) were part of the primary safety and efficacy analysis. Out of the 118 (87.4%) patients who completed the entire treatment duration of 26 weeks, 17 (28.8%) patients in the liraglutide group and 29 (49.3%) patients in the placebo group required rescue medications for hypo- or hyperglycemic events at least once.

Patients who underwent treatment with either liraglutide or placebo had their respective doses escalated weekly throughout the first three weeks of randomization until they were at either the maximum dose of study drug or their fasting blood glucose values were stable (110 mg/dL or lower) at the dose they were already on. Unwanted side-effects resulted in only six patients from the liraglutide group and four from the placebo group being unable to increase the dose of study drug past the 0.6mg daily baseline.

Mean HbA1c values decreased by 0.64% within the liraglutide group and increased by 0.42% within the placebo group at week 26 for an ultimate difference of 1.06%. The total difference increased between the two groups at week 52 to 1.3%. Patients randomized to liraglutide experienced more adverse events per patient-year than those randomized to placebo; those in the liraglutide group had a 118% higher relative risk of developing nausea and a 192% higher relative risk of developing emesis during treatment, as compared to placebo. Patients in the liraglutide group were also at an 82% increased risk of any ADA-classified hypoglycemic event occurring during treatment, as expected with any glucose-lowering drug when compared to placebo.

Practice Pearls:

  • Overall, liraglutide demonstrated superior HbA1C lowering capabilities over placebo in children with type 2 diabetes who had inadequate glycemic control with metformin.
  • Gastrointestinal and hypoglycemic events were more prevalent in children treated with liraglutide than those treated with a placebo.
  • One hundred thirty-four children with diabetes were part of the primary analysis and study results may not be generalizable to the majority population.

 Tamborlane WV, Barrientos-Perez M, Fainberg U, et al. Liraglutide in children and adolescents with type 2 diabetes. N Engl J Med 2019;[EPub Ahead of Print].DOI: 10.1056/NEJMoa1903822

Adam Chalela B.S., PharmD Candidate, USF College of Pharmacy Class of 2020


For more information about diabetes in children, visit our Pediatrics center.