Insulin Degludec Versus Insulin Glargine in Type 1 and Type 2 Diabetes Mellitus: A Meta-Analysis of Endpoints in Phase 3a Trials

Abstract

Introduction: Insulin degludec (degludec) is a basal insulin with an ultra-long, stable action profile and reduced pharmacodynamic variability. Seven phase 3a trials compared degludec with insulin glargine (glargine). Patient-level meta-analyses were performed to obtain a comprehensive overview of differences between the insulin preparations, possible because consistent outcome definitions were utilized. Methods: Three categories of trials were analyzed: basal–bolus-treated type 1 diabetes mellitus (T1DMB/B), insulin-naïve type 2 diabetes mellitus (T2DMinsulin-naïve), and basal–bolus-treated T2DM (T2DMB/B). Regression models were adjusted for baseline characteristics. Endpoints analyzed were glycosylated hemoglobin (HbA1c), fasting plasma glucose (FPG), insulin dose and hypoglycemic rates analyzed in mutually exclusive groups: non-severe nocturnal, non-severe daytime, and severe. Results: As with previous treat-to-target trials, reductions in HbA1c were similar between degludec and glargine. Reductions in FPG were significantly greater with degludec in T1DMB/B and T2DMinsulin-naïve. Total daily insulin dose was significantly lower with degludec in T1DMB/B and T2DMinsulin-naïve. Estimated hypoglycemia rate ratios for degludec/glargine were as follows for T1DMB/B, T2DMinsulin-naïve and T2DMB/B, respectively: non-severe nocturnal 0.83, 0.64, 0.75 (all P < 0.05); non-severe daytime 1.14 [not significant (ns)], 0.89 (ns), and 0.83 (P < 0.05). Rate ratios for severe events were 1.12 (ns) (T1DMB/B); 0.14 (P < 0.05) (T2DMinsulin-naïve); and not analyzed (T2DMB/B) due to too few events. Conclusions: Compared with glargine, degludec is associated with equivalent HbA1c control and significantly lower nocturnal hypoglycemia rates. In T1DMB/B and T2DMinsulin-naïve, degludec is also associated with significantly greater reductions in FPG and lower total doses of insulin versus glargine.