OBJECTIVES: The aim of this analysis was to estimate the cost-effectiveness of exenatide once weekly (EQW) versus exenatide BID therapy, two formulations of the same glucagon-like peptide-1 receptor agonist molecule, in type 2 diabetes patients in the UK. Pooled data from DURATION-1 and DURATION-5, phase 3, randomised, open label clinical trials in 295 and 252 patients respectively, were used.  EQW was associated with greater LS mean HbA1c reduction (-1.7% versus -1.2%, respectively, p<0.001) and weight reduction (-2.9 kg versus -2.4 kg, respectively, p=0.126). METHODS: A previously published and validated diabetes model (IMS CORE Diabetes Model) was used to make 50 year projections of clinical and cost outcomes based on pooled DURATION-1 and 5 baseline patient characteristics (age 55.3 years, duration of diabetes 7 years, HbA1c 8.36%) and study results. Costs were derived from published sources and expressed in 2010 UK Pounds. A discount rate of 3.5 % was applied to both costs and outcomes. Various sensitivity analyses were performed. RESULTS: EQW treatment was projected to improve quality-adjusted life expectancy by 0.18 quality-adjusted life years (QALYs) (95% confidence interval 0.10 to 0.25) and life expectancy by 0.16 years (95% confidence interval 0.07 to 0.26) versus exenatide BID. EQW was associated with delayed onset of any diabetes-related complication versus exenatide BID by almost 6 months on average. Due to the lower incidence of diabetes-related complications during treatment with EQW, and hence reduction in their treatment costs, EQW was associated with direct medical cost savings (difference of -£305, 95% confidence interval -£715 to £35). EQW was therefore projected to be dominant versus exenatide. This result was robust to all sensitivity analysis. CONCLUSIONS: Based on DURATION-1 and 5, EQW was projected to be less costly and more effective than exenatide BID over a patients’ lifetime in the UK setting.