OBJECTIVES To evaluate the cost of using the first-in-class sodium-glucose co-transporter-2 (SGLT-2) inhibitor dapagliflozin compared to the glucagon-like peptide-1 (GLP-1) analogue lixisenatide, when added to insulin for the treatment of patients with type 2 diabetes mellitus (T2DM) who are inadequately controlled on insulin alone. METHODS Bucher indirect comparisons and a network meta-analysis were performed in the add-on to insulin population with regards to the key T2DM outcomes of HbA1c, weight and hypoglycaemia. Dapagliflozin had similar treatment effects to lixisenatide, with no statistically significant differences between dapagliflozin and lixisenatide in these outcomes. Therefore, a cost-minimisation analysis over a 1-year time horizon was developed from a UK healthcare perspective. Drug and administration costs, including a nurse consultation, were considered in the model. Drug costs were those of the British National Formulary (September 2013), device costs were taken from the Information Services Division Drug Tariff 2013 and the cost of a nurse consultation was estimated using the Unit Costs of Health and Social Care report compiled by the Personal Social Services Research Unit. The application of an annual discount rate of 3.5% and use of a longer time horizon (up to 5 years) were explored in scenario analyses. RESULTS Dapagliflozin was associated with a yearly cost of £476.96 per patient. The cost per patient for lixisenatide was £765.76 in the first year and £752.07 in subsequent years. Therefore, dapagliflozin was cost-saving compared to lixisenatide, driven by the difference in drug and administration costs. Applying the annual discount rate and using a longer time horizon, dapagliflozin was associated with savings of £1,255.30 over 5 years compared to lixisenatide. CONCLUSIONS Dapagliflozin in combination with insulin was shown to be a cost-saving treatment option compared to lixisenatide from a UK healthcare perspective for patients with T2DM who are inadequately controlled on insulin alone.