Pierre Ambrosi1,2, Aurélie Daumas3, Patrick Villani3, Roch Giorgi4. 1. Department of Therapeutics, Marseille, France, pierre.ambrosi@ap-hm.fr. 2. Department of Cardiology, Hôpital de la Timone, Marseille, France, pierre.ambrosi@ap-hm.fr. 3. Department of Therapeutics, Marseille, France. 4. Aix Marseille Univ, APHM, INSERM, IRD, SESSTIM, Sciences Economiques et Sociales de la Santé et Traitement de l'Information Médicale, Hop Timone, BioSTIC, Biostatistique et Technologies de l'Information et de la Communication, Marseille, France.
Abstract
BACKGROUND AND OBJECTIVES: The value of glycosylated hemoglobin (HbA1c) as a surrogate marker for the prevention of cardiovascular outcomes on antidiabetic drugs is debated. The 2008 FDA guidance led to multiple large clinical trials to evaluate the effect of new antidiabetic drugs versus placebo on major adverse cardiac events (MACE). The aim of this study was to evaluate the relation between MACE and HbA1c decrease between antidiabetic drug and placebo across the spectrum of cardiovascular outcome trials (CVOT). METHODS: In this systematic review, we included randomized controlled trials that compared an antidiabetic drug to placebo in addition to current standard of care with the primary intention of demonstrating cardiovascular safety. We investigated the relationship between MACE decrease on antidiabetic drug and HbA1c reduction on antidiabetic drug using the coefficient correlation. We also studied the effects of potential confounders on MACE decrease. RESULTS: Fourteen eligible trials including 128,149 patients were included, 12,114 of whom experienced MACE. Mean achieved HbA1c absolute reductions on antidiabetic treatment versus placebo varied from 0.29 to 1%. The decrease of MACE on antidiabetic drug was significantly correlated with mean HbA1c reduction (r = 0.88, 95% CI: 0.67-0.96, p < 0.001) and weight loss (r = 0.81, 95% CI: 0.46-0.94, p < 0.001). In a bivariate model including weight loss, only HbA1c reduction remained significantly correlated with the decrease of MACE on antidiabetic drug (p = 0.019). CONCLUSION: Across CVOT, the decrease in MACE incidence on various antidiabetic drugs is significantly correlated with HbA1c reduction. This meta-analysis supports HbA1c as an appropriate surrogate endpoint for cardiovascular events. Our analysis supports that changes in HbA1c should be taken into account while interpreting effects of new antidiabetic drugs on cardiovascular outcomes.
BACKGROUND AND OBJECTIVES: The value of glycosylated hemoglobin (HbA1c) as a surrogate marker for the prevention of cardiovascular outcomes on antidiabetic drugs is debated. The 2008 FDA guidance led to multiple large clinical trials to evaluate the effect of new antidiabetic drugs versus placebo on major adverse cardiac events (MACE). The aim of this study was to evaluate the relation between MACE and HbA1c decrease between antidiabetic drug and placebo across the spectrum of cardiovascular outcome trials (CVOT). METHODS: In this systematic review, we included randomized controlled trials that compared an antidiabetic drug to placebo in addition to current standard of care with the primary intention of demonstrating cardiovascular safety. We investigated the relationship between MACE decrease on antidiabetic drug and HbA1c reduction on antidiabetic drug using the coefficient correlation. We also studied the effects of potential confounders on MACE decrease. RESULTS: Fourteen eligible trials including 128,149 patients were included, 12,114 of whom experienced MACE. Mean achieved HbA1c absolute reductions on antidiabetic treatment versus placebo varied from 0.29 to 1%. The decrease of MACE on antidiabetic drug was significantly correlated with mean HbA1c reduction (r = 0.88, 95% CI: 0.67-0.96, p < 0.001) and weight loss (r = 0.81, 95% CI: 0.46-0.94, p < 0.001). In a bivariate model including weight loss, only HbA1c reduction remained significantly correlated with the decrease of MACE on antidiabetic drug (p = 0.019). CONCLUSION: Across CVOT, the decrease in MACE incidence on various antidiabetic drugs is significantly correlated with HbA1c reduction. This meta-analysis supports HbA1c as an appropriate surrogate endpoint for cardiovascular events. Our analysis supports that changes in HbA1c should be taken into account while interpreting effects of new antidiabetic drugs on cardiovascular outcomes.