Efficacy and Safety of Dapagliflozin in Patients With Inadequately Controlled Type 1 Diabetes (the DEPICT-2 Study): 24-Week Results From a Randomized Controlled Trial

OBJECTIVE: This 24-week, double-blinded, phase 3 clinical trial (DEPICT-2; ClinicalTrials.gov, NCT02460978) evaluated efficacy and safety of dapagliflozin as adjunct therapy to adjustable insulin in patients with inadequately controlled type 1 diabetes (HbA1c 7.5-10.5%). RESEARCH DESIGN AND METHODS: Patients were randomized 1:1:1 to dapagliflozin 5 mg (n = 271), dapagliflozin 10 mg (n = 270), or placebo (n = 272) plus insulin. Insulin dose was adjusted by investigators according to self-monitored glucose readings, local guidance, and individual circumstances. RESULTS: Baseline characteristics were balanced between treatment groups. At week 24, dapagliflozin significantly decreased HbA1c (primary outcome; difference vs. placebo: dapagliflozin 5 mg -0.37% [95% CI -0.49, -0.26], dapagliflozin 10 mg -0.42% [-0.53, -0.30]), total daily insulin dose (-10.78% [-13.73, -7.72] and -11.08% [-14.04, -8.02], respectively), and body weight (-3.21% [-3.96, -2.45] and -3.74% [-4.49, -2.99], respectively) (P < 0.0001 for all). Mean interstitial glucose, amplitude of glucose excursion, and percent of readings within target glycemic range (>70 to ≤180 mg/dL) versus placebo were significantly improved. More patients receiving dapagliflozin achieved a reduction in HbA1c ≥0.5% without severe hypoglycemia compared with placebo. Adverse events were reported for 72.7%, 67.0%, and 63.2% of patients receiving dapagliflozin 5 mg, dapagliflozin 10 mg, and placebo, respectively. Hypoglycemia, including severe hypoglycemia, was balanced between groups. There were more adjudicated definite diabetic ketoacidosis (DKA) events with dapagliflozin: 2.6%, 2.2%, and 0% for dapagliflozin 5 mg, dapagliflozin 10 mg, and placebo, respectively. CONCLUSIONS: Dapagliflozin as adjunct therapy to adjustable insulin in patients with type 1 diabetes was well tolerated and improved glycemic control with no increase in hypoglycemia versus placebo but with more DKA events.