OBJECTIVES: To compare the cost-effectiveness of canagliflozin (CANA) and dapagliflozin (DAPA), two compounds with sodium glucose co-transporter 2 (SGLT2) activity, in dual therapy as add-on to metformin from the Spanish National Health System perspective.  Network meta-analyses (NMAs) have found that CANA 300 mg lowers HbA1c more than DAPA 10 mg in dual and triple therapy.  Pharmacokinetic and pharmacodynamic differences support these results.  Specifically, CANA 300 mg has been shown to reduce the renal threshold for glucose excretion more than DAPA 10 mg, resulting in ~25% greater 24-hour urinary glucose excretion.  In addition, CANA 300 mg may transiently block intestinal SGLT1, delaying glucose absorption and reducing postprandial glucose. METHODS: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of CANA 100 and 300 mg versus DAPA 10 mg using Spanish-specific utilities and cost data.  Direct costs were reported in euros and an annual discount rate of 3% was applied to costs and effects.  The time horizon used for the economic evaluation was 40 years to reflect the chronic nature of the disease.  A randomised, controlled trial of CANA in dual therapy and an NMA were sourced for initial patient characteristics and treatment effects.  Results were compared with the willingness-to-pay (WTP) threshold reported for Spain (€30,000/QALY). RESULTS: CANA 100 mg dominated DAPA in dual therapy, with 0.061 quality-adjusted life years (QALYs) gained.  CANA 300 mg was cost-effective compared to DAPA 10 mg in dual therapy with a cost-effectiveness ratio below the WTP threshold in Spain; QALYs gained were 0.084. CONCLUSIONS: These results suggest that adding CANA 100 or 300 mg instead of DAPA 10 mg in patients inadequately controlled on metformin would result in more efficient use of healthcare resources in the Spanish setting.