Time to ENGAGE Edoxaban in Atrial Fibrillation?

Edoxaban is the latest direct oral anticoagulant (DOAC) that is about to be released in the Canadian market.   It is currently under review by Health Canada. Should we be excited about it? Or is it another “me-too” drug that is added to the list of DOAC options for stroke prophylaxis in atrial fibrillation?

According to the prescribing information based in US, edoxaban is a direct inhibitor of Factor Xa indicated for non-valvular atrial fibrillation as well as thromboembolism.   

How is edoxaban compared to warfarin for stroke prevention in non-valvular atrial fibrillation?

This was addressed in ENGAGE AF TIMI 48In this landmark trial, edoxaban was compared to warfarin to evaluate the long term efficacy and safety in patients with atrial fibrillation. It was a randomized, double-blind, double-dummy study comparing warfarin to low dose and high dose edoxaban in patients with moderate to high risk of stroke in atrial fibrillation (mean CHAD2 score 2.8± 1.0).  The median follow up was 2. 8 years (median duration of treatment of 907 days)

ENGAGE AF TIMI 48  Study Design

The primary end point of the study was the time to the first stroke or systemic embolic event. Here’s a look at the primary end point results of the three treatment groups:

  • Warfarin group = 232 patients (1.5% per year)
  • High dose edoxaban group = 182 events (1.18% per year)
  • Low dose edoxaban group = 253 events (1.61% per year)

And the key results from the modified intention-to-treat analysis are as follow:

  • Warfarin & High-dose Edoxaban: The hazard ratio is 0.79 (97.5% confidence interval , 0.63-0.99; p< 0.001 for non-inferiority)
  • Warfarin & Low-dose Edoxaban: The hazard ratio is 1.07 (97.5% confidence interval, 0.87-1.31; p=0.005 for non-inferiority)

So my interpretation is that the high dose edoxaban  is  “just as good as” warfarin, whereas it is possible that the low dose edoxaban group does not perform as well when compared to the warfarin group.

I won’t sugar coat the results. These results are not impressive, especially after the latest DOAC, apixaban has demonstrated its superiority to warfarin as well as better survival benefits in ARISTOTLE study.  But overall, ENGAGE TIMI 48 is a well designed trial with large sample size, great follow up time, low drop out rate, good patient representation from different parts of the world as well as having a decent 68% in TIT (Time in Therapeutic Range) for the warfarin group as a comparator.

Here are my thoughts from this study:

  1. Only the high dose edoxaban group is as good as warfarin.  This is clear from the main results of the analysis for non-inferiority. In the intention to treat analysis for superiority, there is a trend favouring high-dose edoxaban group versus the warfarin group but because the result does not achieve statistical significance, one cannot make any conclusion that it is “better” than warfarin. It is interesting that the low-dose edoxaban is “worse” than the warfarin group. I wonder if a similar subgroup analysis is conducted for rivaroxaban and apixaban, will we find similar results?  That is, is the low dose rivaroxaban or apixaban as good as warfarin?  In practice, many patients are on the low dose DOAC due to reduced renal function. I think it’s an important question, I am not sure we will see a study conducted to find such answer.
  • There is a difference in its efficacy in the risk reduction in hemorrhagic stroke versus ischemic stroke. If you focus on the outcomes related to the different types of stroke, it may be impressive to see that both the low-dose and high-dose edoxaban groups out-performed the warfarin group in terms of the risk reduction in hemorrhagic stroke – a trend that we consistently see in clinical trial results of other DOACs.  However, the results aren’t so impressive with the ischemic stroke which is the most common type of stroke we see in practice. Here’s a quick look at the rates of ischemic stroke:
    • Warfarin group – 1.25%
    • High dose Edoxaban group – 1.25%
    • Low dose Edoxaban group – 1.77%
  • Overall, the bleeding risks are lower with both the low and high-dose edoxaban groups except for GI bleed as compared to the warfarin group.  There is also a dose-related trend with the bleeding risk; that is, the high-dose edoxaban group had higher bleeding risk compared to the low-dose edoxaban group in all types of bleeds.  The question is why GI bleed is higher for edoxaban. We also see similar trend that Gl bleeds were higher with dabigatran and rivaroxaban but not with apixaban in their landmark trials compared to warfarin.
  • Impact of Renal Function on Outcomes. One key discussion topic that has come out of edoxaban is that renal function can impact on outcomes. This makes theroetically sense, given 50% of edoxaban is excreted renally. So impaired renal function may increase the drug exposure in the body, possibly improving outcome with more anticoagulation. The same can be said in people with very efficient renal function, that drug exposure may be reduced due to the quick renal elimination, rendering possible inadequate anti-coagulation and increasing risk of stroke and thromboembolism.
    • These outcomes have been evaluated in their subgroup analysis here.   In this analysis, the relative risk of stroke / systemic embolism with the high-dose edoxaban group is similar between patients with CrCL > 50mL/min to those with CrCL ≤ 50mL/min. (HR, 0.87; 95% CI, 0.65–1.18; P for interaction=0.94).
    • However, there is a trend towards lower efficacy with higher levels of CrCL:
      • CrCL ≤ 50mL/min: HR =0.87; 95% CI, 0.65-1.18
      • CrCL > 50-95mL/min: HR =0.78; 95%, 0.64-0.96
      • CrCL > 95mL/min: HR= 1.36; 95%, 0.88-2.10 
    • The authors suggest that given the bleeding risks are similar between groups with different levels of CrCL, the net clinical outcomes are more favourable for High-Dose Edoxaban Group across the range of CrCL.
    • However, the US product monograph clearly specifies that edoxaban is not to be used in individuals where CrCL is above 95mL/min as a result of this subgroup analysis.

Edoxaban is the latest direct inhibitor of factor Xa that will be available soon in the Canadian market. Although the results are not very impressive from its landmark trial of ENGAGE AF-TIMI 48, the study is well designed. We will likely continue to see different subgroup analyses to emerge to identify its best place of therapy for atrial fibrillation.  I think these results don’t necessarily mean edoxaban is not as good as rivaroxaban and apixaban but it implies  there are some unanswered questions, or that we may need a head-to-head trial comparing all three direct inhibitors of factor Xa, perhaps in the near future?

Edoxaban PK

Dosing Considerations & Recommendation (as per Lexicomp Online for US only):

  • Prior to initiation of edoxaban, assess creatinine clearance (CrCl) using the Cockcroft-Gault equation.
  • For patients with nonvalvular atrial fibrillation, do not use edoxaban if CrCl is >95 mL/minute.
  • Deep vein thrombosis and pulmonary embolism: Oral: 60 mg once daily after 5 to 10 days of initial therapy with a parenteral anticoagulant.
    • Patient weight ≤60 kg: 30 mg once daily
    • Concomitant therapy with specific P-gp inhibitors (ie, verapamil, quinidine; the short-term use of azithromycin, clarithromycin, erythromycin, oral itraconazole, oral ketoconazole): 30 mg once daily
  • Nonvalvular atrial fibrillation (NVAF) (to prevent stroke and systemic embolism): Oral: 60 mg once daily
    • Conversion from warfarin or other vitamin K antagonists: Discontinue warfarin and initiate edoxaban as soon as INR falls to ≤2.5.
    • Dosage reduction necessary in all patients with CrCl 15 to 50 mL/minute to 30mg po daily



Published by


My name is Cynthia Leung and I am a practicing pharmacist in Kingston Ontario, Canada. This blog is for me to share my ideas, opinions and perspectives on how medications are used in our health care system. Note that these posts are my own opinions and do not represent the opinions of my current or former employers and / or organizations that I may belong to. Any possible case scenarios described in my posts would be modified to maintain patient confidentiality. This blog is not a platform for professional advise for patients or health care providers and the content is not meant to support any clinical decisions or replace professional opinions. Also the images are either taken or created by the author, or adapted with permission. I hope you will enjoy reading my posts!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s