BackgroundIncidence of type 2 diabetes (T2D) in children and adolescents is increasing, but treatment options are limited.MethodsThis was a 26-week, phase 3 trial with a 26-week extension among patients (10 to 17 years of age) with uncontrolled T2D (A1C 6.5 to 10.5%) receiving metformin, insulin, or both. Participants were randomly assigned 1:1:1 to 5 mg of dapagliflozin (N=81), 2.5 mg of saxagliptin (N=88), or placebo (N=76). Patients in active treatment groups with A1C ≥7% at week 12 were further randomly assigned 1:1 at week 14 to continue the dose or up-titrate to a higher dose (10 mg of dapagliflozin or 5 mg of saxagliptin). The primary end point was change in A1C at week 26. Safety was assessed over 52 weeks.ResultsAt week 26, the difference versus placebo in adjusted mean change in A1C was −1.03 percentage points (95% confidence interval [CI], −1.57 to −0.49; P<0.001) for dapagliflozin and −0.44 percentage points (95% CI, −0.93 to 0.05; P=0.078) for saxagliptin. Adverse events (AEs) and serious AEs occurred in 72.8% and 8.6% of patients receiving dapagliflozin, 69.3% and 8.0% of patients receiving saxagliptin, and 71.1% and 6.6% of patients receiving placebo. Severe hypoglycemia occurred in 4.9%, 4.5%, and 7.9% of patients in each group, respectively. Over 52 weeks, the most common AE was headache (dapagliflozin 14.8%; placebo 5.3%). Most events were mild and none was considered serious or resulted in discontinuation.ConclusionsDapagliflozin, but not saxagliptin, showed significant improvement in A1C compared with placebo. Nonserious headaches were more common in participants treated with dapagliflozin than in those receiving placebo. (Funded by AstraZeneca; ClinicalTrials.gov number, NCT03199053.)