Systematic Review of the Cost Effectiveness of Dapagliflozin Used to Treat Patients with Type 2 Diabetes Mellitus

Abstract

S75the United States (US). MethodS: The United Kingdom Prospective Diabetes Study (UKPDS) Outcomes Model was used to assess the costs (2017 US dollars) and qualityadjusted life-years (QALYs) associated with SGLT2i or GLP-1RA add-on therapy in T2D after metformin failure. The model was run over 3-and 5-year time horizons from a US payer perspective. The UKPDS Outcomes Model predicts diabetes-related outcomes, including: ischemic heart disease, chronic heart failure, amputation, blindness, renal failure, stroke, myocardial infarction, and death. Baseline characteristics were derived from published literature. Simulated hemoglobin A1c (A1C) reductions for the first two years of the model were obtained from long-term outcomes studies and were extrapolated using the model's default algorithm. Costs were obtained from RedBook for drugs and US national sources and published literature for others. Utilities were obtained from published literature. Costs and QALYs were discounted 3% annually. ReSultS: At 5 years, SGLT2i dominated GLP-1RA; SGLT2i cost $66,000 and yielded 3.54 QALYs, whereas GLP-1RA cost $91,000 and yielded 3.54 QALYs. Despite a greater initial reduction in A1C, GLP-1RA had a greater increase in A1C at two years in published clinical trials, which resulted in similar modeled mean A1C values at five years (SGLT2i 7.75% vs. GLP-1RA 7.64%) but higher drug costs. A 3-year time horizon yielded similar results, but the absolute difference in total costs was smaller (SGLT2i $41,000 vs. GLP-1RA $57,000). ConCluSionS: In the short-term, SGLT2i are cost-effective compared to GLP-1RA. Treatment with these agents for a longer time horizon may yield differences in diabetes-related outcomes that may affect their cost-effectiveness, but these differences may not have been captured from a short US payer perspective.