OBJECTIVES: A network meta-analysis (NMA) of sodium glucose coransporter-2-nhibitor (SGLT2I) monotherapy was conducted in adults with uncontrolled T2DM on diet/exercise or metformin.

METHODS: Systematic search were conducted on MEDLINE, Cochrane, and Google Scholar for randomized controlled trials of SGLT2Is in T2DM. Studies published in English language between January-2005 to November-2018, with approved dosage of SGLT2Is monotherapy and a minimum follow-up period of 24 weeks, were included. Summary measure was mean change in HbA1c (%) from baseline. Pairwise meta-analyses (PMA) were conducted to assess homogeneity amongst studies. A frequentist NMA was conducted using R package netmeta (3.5.1), based on graph-theoretical method. The netrank function of netmeta package was used to rank the treatments using P-scores, based on the point estimates and standard errors of the network estimates. Both fixed-effect and random-effects models were used for NMA, results of random-effects model are reported.

RESULTS: Of 511 retrieved studies, 12 met the inclusion criteria. Two studies were identified as a source of high statistical heterogeneity based on the results of PMAs and excluded; thus 10 studies were included. Canagliflozin 300mg showed the greatest reduction (mean difference [95% CI]) in HbA1c compared to other SGLT2Is: -0.150 (-0.157, -0.143) vs canagliflozin 100mg, -0.130 (-0.151, -0.108) vs canagliflozin 200mg, -0.521 (-0.643, -0.400) vs dapagliflozin 10mg, -0.754 (-0.908, -0.599) vs dapagliflozin 2.5mg, -0.626 (-0.755, -0.497) vs dapagliflozin 5mg, -0.400 (-0.425, -0.376) vs empagliflozin 10mg, -0.290 (-0.314, -0.266) vs empagliflozin 25mg, -0.710 (-0.946, -0.474) vs ipragliflozin 50mg, -0.411 (-0.644, -0.178) vs tofogliflozin 10mg, -0.191 (-0.272, -0.040) vs tofogliflozin 20mg, -0.338 (-0.442, -0.084) vs tofogliflozin 40mg, -1.180 (-1.201, -1.159) vs placebo.

CONCLUSIONS: Canagliflozin 300 mg monotherapy was the most effective compared to other SGLT2I monotherapies for HbA1c reduction in adults with uncontrolled T2DM on diet/exercise or metformin. These results may be validated in head-to-head clinical studies, if feasible.