Author(s)


Reviewed By


Citation
Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2010;361:1139-1151

Clinicians and patients alike have been anxiously waiting for an oral alternative to warfarin. That wait appears to be finally over. On October 19, 2010 dabigatran (Pradaxa®) was approved by the FDA and became the first oral direct thrombin inhibitor available in the United States. Unlike its predecessor ximelagatran, dabigatran doesn’t seem to cause liver toxicity. And unlike warfarin, dabigatran is administered as a fixed dose and requires no ongoing laboratory monitoring. Does this mean we have found the Holy Grail for anticoagulation therapy? Probably not. Currently, dabigatran is only approved for the prevention of stroke in patients with atrial fibrillation (A-Fib). Approval for the treatment of venous thromboembolism may be on the horizon, and studies are actively under way using the drug in a variety of other conditions commonly treated with warfarin. But for now, the use of dabigatran is (should be) limited to a relatively discrete population of patients. Let’s examine the literature.

The Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY) trial was a phase 3, multicenter, international, randomized, controlled, non-inferiority trial that compared fixed dose dabigatran (at either 110 or 150 mg twice daily) with INR-adjusted dose warfarin (INR 2.0-3.0).1 Over 18,000 patients with A-Fib and at least one additional risk factor (previous stroke or transient ischemic attack, left ventricular ejection fraction < 40%, NYHA Class 2 or greater heart failure, age >=75, or age 65-74 with diabetes, hypertension, or coronary artery disease) were enrolled in the study. Overall severity of disease based on the CHADS2 scoring system was evenly represented between mild, moderate, and severe risk of thrombosis. Notable exclusion criteria included patients with severe heart-valve disorders, any condition that increased risk of hemorrhage (including history of gastrointestinal bleed in past year), blood pressure >180/100, recent or active malignancy, a specific indication for warfarin therapy other than A-Fib, and impaired renal function (CrCl < 30 mL/min). Thus the study population was not representative of the entire A-Fib population managed by many antithrombosis services. Nor were the study doses of dabigatran the same as the final FDA-approved doses. Although the 150 mg twice daily dose is recommended for those with CrCl > 30 mL/min, a dose of 75 mg twice daily for those with a CrCl of 15-30 mL/min was also approved.2 For this reason, we’ll only examine the data regarding dabigatran 150 mg twice daily vs. warfarin (INR 2.0-3.0).

Major efficacy and safety outcomes of RE-LY’s comparison of dabigatran 150 mg twice daily to warfarin (average time in INR range 64%) over a median two-year follow-up are as follows:

 

Drug favored

NNT or NNH

Event

per year

Stroke or systemic embolism (primary outcome)

D

NNT 172 for D

Hemorrhagic stroke

D

NNH 357 for W

Myocardial infarction

W

NNH 476 for D

Major bleeding (Hgb drop > 2 gm/dL; > 2 units blood needed; or critical organ bleeding site)

None

No difference

Life-threatening bleed (Hgb drop > 5 gm/dL; >4 units blood needed; intracranial; inotropes needed; or fatal)

D

NNH 286 for W

Major and minor bleeding (above plus all other bleeding)

D

NNH 58 for W

Gastrointestinal bleeding

W

NNH 204 for D

 

D = dabigatran      W = warfarin

In examining this data, dabigatran does offer an efficacy advantage over warfarin since statistical benefit was observed in the primary outcome despite the non-inferiority design of the study. Dabigatran is also less likely to cause life-threatening bleeding and hemorrhagic stroke. But of concern is the statistically higher incidence of myocardial infarction and gastrointestinal bleeding in the dabigatran group. Thus one benefit may quickly be erased by a harmful negative outcome. One must also consider if RE-LY’s outcomes were influenced by time spent in the therapeutic range (TTR) for those patients randomized to warfarin therapy. Wallentin, et al. conducted a pre-specified assessment of the RE-LY data to answer that question (TTR quartiles of <57.1%, 57.1-65.5%, 65.5-72.6%, and >72.6%)3, and the results for the lowest and highest quartiles are as follows:

 

HR (95% CI) for

p 

Event

150 mg dabigatran vs. warfarin

(interaction for all quartiles)

Stroke or systemic embolism

 

 

     TTR < 57.1%

0.57 (0.37-0.88)

0.2

     TTR > 72.6%

0.95 (0.61-1.48)

 

 

 

 

Non-hemorrhagic stroke and systemic embolism

 

 

     TTR < 57.1%

0.54 (0.34-0.88)

0.076

     TTR > 72.6%

1.21 (0.74-1.98)

 

 

 

 

Intracranial bleeding

 

 

     TTR < 57.1%

0.53 (0.25-1.15)

0.89

     TTR > 72.6%

0.39 (0.18-0.34)

 

 

 

 

Major bleeding

 

 

     TTR < 57.1%

0.71 (0.52-0.96)

0.03

     TTR > 72.6%

1.16 (0.88-1.54)

 

 

 

 

Major gastrointestinal bleeding

 

 

     TTR < 57.1%

1.08 (0.70-1.66)

0.019

     TTR > 72.6%

2.00 (1.25-3.21)

 

 

 

 

Total bleeding

 

 

     TTR < 57.1%

0.89 (0.78-1.01)

0.15

     TTR > 72.6%

1.00 (0.89-1.12)

 

 

 

 

Stroke, systemic embolism, PE, MI, death, and major bleeding

 

 

     TTR < 57.1%

0.67 (0.56-0.80)

0.0006

     TTR > 72.6%

1.11 (0.91-1.35)

 

 

 

 

 

This data clearly suggests that as warfarin management improves, the advantage moves towards warfarin – particularly with regard to non-hemorrhagic stroke, systemic embolism, and gastrointestinal bleeding.

There may be other disadvantages to dabigatran. First, in the RE-LY trial, dabigatran was more likely than warfarin to cause dyspepsia (11.3% vs. 5.8%, p<0.001), and discontinuation rates were higher in the dabigatran group at both one and two years primarily due to serious adverse effects and gastrointestinal symptoms (15.5% vs. 10.2% at one year; 21.2% vs. 16.6% at two years).1 Second, although drug interactions are less likely with dabigatran as compared to warfarin, P-glycoprotein inhibitors and inducers can alter dabigatran levels.4 The clinical significance of these interactions is unknown as the RE-LY study did not include information about concurrent use of many of these drugs. Moreover, the population of patients taking a proton pump inhibitor in the RE-LY study did not appear to gain additional benefit from dabigatran as did other subgroups (which could simply be due to the small sample size, but further evaluation is warranted). Third, there is no reversal agent for dabigatran. Fourth, long-term effects of the drug are still unknown. And finally, there is the question of cost. A recent cost analysis published in the Annals of Internal Medicine suggested that dabigatran may be cost-effective when compared to warfarin if the 150 mg dose is priced at less than $13.70 per day5 (of note, recent news articles report the average wholesale price of both strengths of dabigatran to be $6.75 per day6). “Cost effectiveness” in this study was reported as $45,372 per quality adjusted life year gained.  However, information was not given regarding the estimated drug cost of warfarin – leading one to wonder if the cost of brand name Coumadin or generic warfarin (available for $4 per month a many community pharmacies) was used in the analysis. Further, the manufacturer of dabigatran has not released information about whether the drug will be available through a patient assistance program. In my opinion, that clearly leaves warfarin as the preferred drug for uninsured patients.  In our current healthcare system, it is often easier to help these patients access anticoagulation management programs than expensive medications.

So is dabigatran the Holy Grail we’ve been waiting for? For me, it’s not even close. The first consideration is affordability, since in my opinion inability to access the drug eliminates any need to even have a debate about dabigatran vs. warfarin. And even in patients who can afford the drug’s cost, I would only recommend it for those patients who closely meet RE-LY’s inclusion and exclusion criteria AND meet all of the following characteristics:

  1. Normal renal function (since the FDA-approved 75 mg dose was not studied in RE-LY and no outcome data is available in other published clinical trials)
  2. Low gastrointestinal bleed risk (due to the higher incidence in the dabigatran group and the lower efficacy trend observed in patients on proton pump inhibitors)
  3. Inability to gain good INR control on warfarin and/or lack of access to routine INR testing.

Otherwise, I feel there are simply too many unknowns about this new therapy compared to our standard of care. Warfarin remains the best bet for most patients! What do you think?

[collapsed]

1. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2010;361:1139-1151.

2. Pradaxa [package insert]. Ridgefield, CT: Boehringer Ingelheim Pharmaceuticals, Inc; 2010.

3. Wallentin L, Yusuf S, Ezekowitz MD, et al. Efficacy and safety of dabigatran compared with warfarin at different levels of international normalised ratio control for stroke prevention in atrial fibrillation: an analysis of the RE-LY trial. Lancet. 2010;376:975-83.

4. New drug: Pradaxa (dabigatran). Pharmacist’s Letter/Prescriber’s Letter 2010;26(11):261101.

5. Freeman JV, Zhu RP, Owens DK, et al. Cost-effectiveness of dabigatran compared with warfarin for stroke prevention in atrial fibrillation. Ann Int Med. 2010, November 2.

6. Earthtimes. Boehringer Ingelheim’s Pradaxa Available in U.S. Pharmacies Starting Wednesday, November 3. Accessed November 2, 2010.