Cost-Effectiveness Analysis of Exenatide versus GLP-1 Receptor Agonists in Patients with Type 2 Diabetes Mellitus

Objective The aim of this study was to assess the efficiency of exenatide 2 mg/week compared with other glucagon-like peptide-1 (GLP-1) receptor agonists (dulaglutide 1.5 mg/week, liraglutide 1.2 mg/day, liraglutide 1.8 mg/day and lixisenatide 20 mu g/day) in adult patients with type 2 diabetes mellitus (T2DM) not adequately controlled on metformin alone from the perspective of the Spanish National Health System (NHS). Methods Quality-adjusted life-years (QALYs) gained and total costs of each assessed drug combined with metformin (2 g/day) were estimated over a 40-year time horizon using the Cardiff Diabetes Model (based on UK Prospective Diabetes Study [UKPDS] 68 equations), which simulates disease progression considering the T2DM-related micro- and macrovascular complications, hypoglycaemia, nausea, body mass index (BMI) changes and treatment discontinuation due to adverse effects (AEs). Drug efficacy derived from an indirect comparison performed in a network meta-analysis. Patient characteristics were obtained from the literature. The baseline utility value (0.80) was derived from the PANORAMA study, applying utility decrements to micro- and macrovascular complications, hypoglycaemia episodes and changes in BMI. Treatment discontinuation due to AEs or poorly controlled diabetes (HbA1c > 7.5%) involved switching to second-line (basal insulin) or third-line (basal-bolus insulin) treatment. Total cost (euro, 2018) included the costs of drug acquisition, hypoglycaemia, weight gain, micro- and macrovascular complications, nausea and treatment discontinuation due to AEs. An annual discount rate of 3% was applied to costs and outcomes. Deterministic and probabilistic sensitivity analyses (SA) were performed. Results In base-case, exenatide 2 mg/week resulted in more QALYs (8.26) than dulaglutide 1.5 mg/week (8.19 QALYs), liraglutide 1.2 mg/day (8.10 QALYs), liraglutide 1.8 mg/day (8.20 QALYs) and lixisenatide 20 mu g/day (8.13 QALYs). Total cost/patient was euro20,423.27 (exenatide 2 mg/week), euro22,611.94 (dulaglutide 1.5 mg/week), euro21,065.97 (liraglutide 1.2 mg/day), euro24,865.69 (liraglutide 1.8 mg/day) and euro21,334.58 (lixisenatide 20 mu g/day). Deterministic SA confirmed the robustness of the model. In the probabilistic SA, 95-99% of the 1000 Monte Carlo iterations performed were under a hypothetical willingness-to-pay threshold of euro20,000/QALY gained. Conclusions Exenatide 2 mg/week would be a dominant strategy (more effective and less costly) versus the other GLP-1 receptor agonists assessed for the treatment of T2DM patients who are not adequately controlled on metformin alone.