This randomized, placebo-controlled, double-blind, parallel study assessed the safety, tolerability, pharmacokinetics, and pharmacodynamics of exenatide once weekly (QW) in 30 Japanese patients with type 2 diabetes (T2D) suboptimally controlled by diet and exercise alone or combined with biguanide, sulfonylurea, thiazolidinedione, or combinations of these agents (58.6% male; 58±9 years; body mass index 26.3±2.9 kg/m²; hemoglobin A_1c [HbA_1c] 7.4±0.8%; fasting plasma glucose [FPG] 156.1±29.1 mg/dL; duration of T2D 6±5 years; means ± SD). Patients were randomized in a 1:1:1 ratio to subcutaneous placebo QW, exenatide QW 0.8 mg, or exenatide QW 2.0 mg for 10 weeks. All evaluable patients were analyzed (placebo QW, n=10; exenatide QW 0.8 mg, n=10; exenatide QW 2.0 mg, n=9), unless otherwise stated. Steady-state plasma exenatide concentrations were observed by Week 8 of the study. For the evaluable pharmacokinetic population, geometric mean (90% confidence interval) steady-state plasma concentrations (pg/mL) were 81.2 (68.3-96.4) and 344.5 (256.5-462.7) with exenatide QW 0.8 mg (n=8) and exenatide QW 2.0 mg (n=5), respectively. Baseline-to-Week 10 glycemic improvements with placebo QW, exenatide QW 0.8 mg, and exenatide QW 2.0 mg, respectively, were: HbA_1c (%): -0.4±0.3, -1.0±0.7, and -1.5±0.7; FPG (mg/dL): -20.5±20.4, -25.2±10.9, and -50.8±27.8; and 2-hour postprandial plasma glucose excursions (mg/dL): -8.8±26.9, -50.0±41.1, and -59.7±26.8 (means ± SD). No serious adverse events (AEs) were reported and no AEs led to study discontinuation in any group. The most frequent AE observed was mild-to-moderate injection site induration. No serious hypoglycemia was reported. Exenatide QW for 10 weeks was well tolerated and improved short-term glycemic control in Japanese patients with suboptimally controlled T2D.