Characteristics and outcomes of patients with type 2 diabetes mellitus treated with canagliflozin: a real-world analysis

BACKGROUND Canagliflozin, an oral agent that inhibits sodium glucose co-transporter 2, improves glycemic control, body weight, and blood pressure and is generally well tolerated in patients with type 2 diabetes mellitus (T2DM). This study extends the scope of previous analyses by evaluating outcomes associated with the use of canagliflozin over a 6-month period in a real-world setting. METHODS This retrospective cohort study used data obtained from a large health plan database for patients (≥18 years) with a diagnosis of T2DM who filled at least one canagliflozin prescription between April 1, 2013 and October 30, 2013 (first 7 months canagliflozin was commercially available in the USA) and were continuously enrolled in the health plan for 6 months prior to (baseline) and 6 months following the first canagliflozin prescription claim (follow-up). Changes in glycemic control were evaluated, along with characteristics of enrolled patients and changes in treatment patterns. RESULTS 4017 patients (mean age 56 years, 43 % female) met the study inclusion criteria. Of these, at the time of first canagliflozin claim, 21 % used canagliflozin concomitantly with three or more other antihyperglycemic agents (AHAs), 29 % with two other AHAs, 30 % with one other AHA, and 20 % without other AHAs. During follow-up, patients received 3.4 (average) canagliflozin prescription fills and a mean of 148 total days of supply; median adherence (interquartile range [IQR]) was 86 % (66-98 %) for patients with ≥2 fills. Among patients with available glycated hemoglobin (A1C) measurements at baseline and follow-up (n = 826, baseline A1C 8.59 %), mean A1C reduction was 0.81 % (P < 0.001). Mean A1C reduction during the follow-up period was greatest in patients with the highest baseline A1C levels. Of the patients who used canagliflozin concomitantly with other AHAs, 20 % were observed to discontinue one or more other AHAs during follow-up. The most commonly discontinued baseline AHAs were: glucagon-like peptide-1 receptor agonists (16 %), dipeptidyl peptidase-4 inhibitors (15 %), insulin (13 %), sulfonylureas (13 %), and metformin (11 %). CONCLUSIONS This real-world study on canagliflozin use in a range of patients with T2DM demonstrated significant improvements in mean A1C from baseline following the first canagliflozin prescription. In patients concomitantly using one or more additional AHAs at baseline, there appears to be a trend toward lower other AHA use after canagliflozin initiation.