Abstract 17392: Gastrointestinal Bleeding With Edoxaban versus Warfarin: Results From the ENGAGE AF-TIMI 48 Trial

Be careful out there.
http://circ.ahajournals.org/content/132/Suppl_3/A17392.short

1. James Aisenberg1;
2. Kathryn Friedman1;
3. Jay Desai1;
4. Jeffrey I Weitz2;
5. Robert Giugliano3;
6. Christian T Ruff3;
7. Francesco Nordio3;
8. Michele Mercuri4;
9. Youngsook Choi4;
10. Laura Bower5;
11. Elliot Antman3;
12. Eugene Braunwald3

+ Author Affiliations

1. ¹Internal Medicine, Icahn Sch of Medicine, Mount Sinai Med Cntr, New York, NY
2. ²Internal Medicine, McMaster Univ and Thrombosis and Atherosclerosis Rsch Institute, Hamilton, Canada
3. ³TIMI Study Group, Div of Cardiovascular Medicine, Brigham and Women’s Hosp and Harvard Med Sch, Boston, MA
4. ⁴Pharma Development, Daiichi Sankyo, Edison, NJ
5. ⁵Pharma Development, Daiichi Sankyo, New York, NY

Abstract

Background: There is more major gastrointestinal bleeding (M-GIB) with most NOACs than with warfarin. The clinical impact of this finding is poorly understood.

Methods: The ENGAGE AF-TIMI 48 trial compared the efficacy and safety of higher-dose (HD-E: 60 mg/ 30mg) or lower-dose (LD-E: 30 mg/15 mg) edoxaban regimens with dose-adjusted warfarin for prevention of stroke and systemic embolism in non-valvular atrial fibrillation. ISTH definitions for major and life-threatening (LT) bleeding events were utilized. In this pre-specified analysis, we investigated the adjudicated M-GIB events utilizing pre-defined severity and outcome endpoints.

Results: Although the risk of M-GIB was higher with HD-E than with warfarin, the risk of LT or fatal GIB was similar (figure), and surgery for GIB was required less frequently with HD-E than warfarin (HR 0.37; 95%CI, 0.16-0.88; p=0.03). The risk for M-GIB-related permanent drug discontinuation was similar with HD-E and warfarin (HR 0.96; 95% CI, 0.67-1.37, p=0.8), as were the risks of hospitalization (HR 1.14; 95%CI, 0.92-1.40, p=0.2) and Hgb decrease > 5 g/dL (HR 1.01; 95%CI, 0.74-1.38; p=0.9). The risks of M-GIB and of LT or fatal GIB were lower with LD-E than with warfarin (Figure), as were the risks of M-GIB-related permanent drug discontinuation (HR 0.51; 95% CI, 0.33-0.78, p=0.002), hospitalization (HR 0.67; 95%CI, 0.53-0.85; p=0.001), Hgb decrease > 5 g/dL (HR 0.58; 95%CI, 0.40-0.84; p=0.003), and M-GIB requiring transfusion of > 2 units of red cells (HR 0.66; 95%CI, 0.50-0.89; p=0.006). Fatal GIB occurred in 2, 3 and 7 patients treated with HD-E, LD-E and warfarin, respectively.

Conclusions: M-GIB and LT-GIB with edoxaban are dose-dependent. Although the risk of M-GIB is higher with HD-E than warfarin, the risk of LT-GIB or fatal GIB is similar and the severity and outcomes are no worse than with warfarin. The risks of M-GIB, and of LT-GIB or fatal GIB, are lower with LD-E than warfarin.