BAYESIAN NETWORK META-ANALYSIS (NMA) TO ASSESS RELATIVE EFFICACY OF CANAGLIFLOZIN (CANA) VERSUS GLUCAGON-LIKE PEPTIDE-1 (GLP-1) AGONISTS IN DUAL AND TRIPLE THERAPY IN PATIENTS WITH TYPE 2 DIABETES MELLITUS (T2DM)
Author(s)
Van Sanden S1, Diels J1, Guillon P2, Nielsen AT3
1Janssen Research & Development, Beerse, Belgium, 2Janssen-Cilag France, Issy-les-Moulineaux, Cedex, France, 3Janssen-Cilag A/S, Birkerød, Denmark
OBJECTIVES: To assess the relative efficacy of CANA, an inhibitor of sodium glucose co-transporter 2, versus GLP-1 agonists, in second and third line for the treatment of T2DM patients, using NMA. METHODS: A systematic review of randomized, controlled trials and Bayesian NMA were conducted to compare HbA1c lowering of CANA versus the GLP-1 agonists liraglutide and exenatide. NMAs were conducted separately by background treatment (metformin or metformin plus sulfonylurea) and trial duration (26/52/104 weeks). Networks were based on treatment- and dose-specific nodes. Non-informative priors were used. Relative efficacy was evaluated based on absolute differences in HbA1c reductions and Bayesian probabilities. RESULTS: In dual therapy, trials were identified reporting HbA1c reduction versus baseline at Weeks 26 (n=8), 52 (n=6), and 104 (n=2). At Weeks 26 and 52, CANA 300/100mg had greater (Week 26: Δ=–0.17%/–0.28%) to similar (Week 52: Δ=–0.03%/0.09%) reductions versus exenatide 5µg, and lower/similar versus exenatide 10µg (Week 26: Δ=0.22%/0.10%). Reductions were greater for liraglutide 1.8/1.2mg, with differences versus CANA ranging between 0.20% and 0.61%. At 104 weeks, reductions were similar for CANA 100mg (Δ=0.04%/0.02%) and greater for CANA 300mg (Δ=0.13%/0.11%) compared to liraglutide 1.2/1.8mg. In triple therapy, 5 trials reporting HbA1c reduction at Week 26 were identified to compare CANA versus liraglutide 1.8mg and exenatide 10/5µg. Reductions for CANA 100mg were similar to exenatide 5µg (Δ=–0.02%) and lower versus exenatide 10µg and liraglutide 1.8mg (Δ=–0.24% and –0.33%, respectively). Reductions for CANA 300mg were greater versus exenatide 5µg, and similar versus exenatide 10µg and liraglutide 1.8mg (Δ=–0.01% and 0.08%). CONCLUSIONS: NMA results for dual therapy suggest increasing relative efficacy of CANA over time versus GLP-1 agonists, with CANA 300mg reaching similar HbA1c reductions as GLP-1 agonists at 104 weeks. NMA results for triple therapy suggest at least similar efficacy for CANA 300mg.
Conference/Value in Health Info
2015-05, ISPOR 2015, Philadelphia, PA, USA
Value in Health, Vol. 18, No. 3 (May 2015)
Code
PDB12
Topic
Clinical Outcomes
Topic Subcategory
Comparative Effectiveness or Efficacy
Disease
Diabetes/Endocrine/Metabolic Disorders