OBJECTIVE—This study evaluates the ability of the incretin mimetic exenatide (exendin-4) to improve glycemic control in patients with type 2 diabetes failing to achieve glycemic control with maximally effective metformin doses.

RESEARCH DESIGN AND METHODS—A triple-blind, placebo-controlled, 30-week study at 82 U.S. sites was performed with 336 randomized patients. In all, 272 patients completed the study. The intent-to-treat population baseline was 53 ± 10 years with BMI of 34.2 ± 5.9 kg/m² and HbA_1c of 8.2 ± 1.1%. After 4 weeks of placebo, subjects self-administered 5 μg exenatide or placebo subcutaneously twice daily for 4 weeks followed by 5 or 10 μg exenatide, or placebo subcutaneously twice daily for 26 weeks. All subjects continued metformin therapy.

RESULTS—At week 30, HbA_1c changes from baseline ± SE for each group were −0.78 ± 0.10% (10 μg), −0.40 ± 0.11% (5 μg), and +0.08 ± 0.10% (placebo; intent to treat; adjusted P < 0.002). Of evaluable subjects, 46% (10 μg), 32% (5 μg), and 13% (placebo) achieved HbA_1c ≤7% (P < 0.01 vs. placebo). Exenatide-treated subjects displayed progressive dose-dependent weight loss (−2.8 ± 0.5 kg [10 μg], −1.6 ± 0.4 kg [5 μg]; P < 0.001 vs. placebo). The most frequent adverse events were gastrointestinal in nature and generally mild to moderate. Incidence of mild to moderate hypoglycemia was low and similar across treatment arms, with no severe hypoglycemia.

CONCLUSIONS—Exenatide was generally well tolerated and reduced HbA_1c with no weight gain and no increased incidence of hypoglycemia in patients with type 2 diabetes failing to achieve glycemic control with metformin.