Treatment Patterns and Persistence With GLP-1 RA Treatments Among Patients With Type 2 Diabetes in France: A Retrospective Cohort Analysis.

INTRODUCTION: In type 2 diabetes (T2D), persistence with injectable glucose-lowering therapy is associated with better outcomes. This study used real-world pharmacy data to report on persistence with glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in patients with T2D in France. METHODS: This retrospective cohort analysis presents longitudinal data from approximately 7500 French retail pharmacies that filled GLP-1-RA prescriptions for GLP-1 RA-naïve patients with T2D ('index therapy': dulaglutide; once-weekly exenatide [exenatide QW]; twice-daily exenatide [exenatide BID]; liraglutide) between January 2015 and December 2016 (follow-up ≥ 12 months). The main outcome was treatment persistence (absence of discontinuation [gap following index therapy prescription ≥ 2-fold the expected duration of that prescription] or switch [new non-index glucose-lowering prescription issued ≤ 30 days before/after index therapy discontinuation]). Persistence was calculated as the median duration through Kaplan-Meier survival analysis over the variable follow-up period and as the proportion of patients persistent at 12 months. In addition to persistence outcomes (discontinuation/switch), three other treatment modifications were assessed: augmentation/intensification with a new non-index glucose-lowering therapy; off-label dose increase (daily dose > 20 μg for exenatide BID; two consecutive prescriptions with daily dose > 1.8 mg for liraglutide); and off-label dose decrease (two consecutive prescriptions with average daily dose lower than the index dose). Off-label dose changes were not assessed for dulaglutide or exenatide QW (as single-dose, prefilled pens). RESULTS: Median persistence was longest for dulaglutide (373 days) versus liraglutide (205 days), exenatide QW (184 days) and exenatide BID (93 days). Twelve months after treatment initiation, the percentage of persistent patients ranged from 51% (dulaglutide) to 21% (exenatide BID). Overall, treatment modification occurred less commonly for dulaglutide than for the other index GLP-1 RAs. CONCLUSION: This analysis revealed marked differences in persistence among GLP-1 RAs, which was highest for dulaglutide and lowest for exenatide BID. The prospective TROPHIES study will provide additional information about persistence with dulaglutide and liraglutide, including reasons for treatment modifications.