Pradaxa, Boehringer Ingelheim’s (“BI”) competition to warfarin (brand name: Coumadin) in the anti-coagulation market, was approved for use to reduce the risk of stroke in those patient who have non-valvular atrial fibrillation, a heart rhythm disorder. The FDA approved Pradaxa (dabigatran) based on BI’s RE-LY study, an open-label clinical trial that studied more than 18,000 patients. According to BI, the benefits of Pradaxa are a 35% reduced risk of stroke, as compared to warfarin, and greater convenience, in that Pradaxa does not require the blood tests or dietary restrictions that accompany warfarin therapy, all in the context of safety comparable to warfarin. Further study of the RE-LY data, however, calls these claimed benefits into question. Here’s the rest of the story.
To begin, BI’s claimed context of safety appears to be inaccurate. RE-LY data shows that patients on Pradaxa are more likely to suffer severe bleeds; often, these have occurred as GI bleeds. Further, unlike warfarin, Pradaxa has no reversal agent, meaning that doctors cannot administer a drug to reverse Pradaxa’s effects. In the event a Pradaxa patient experiences a bleed, a traumatic injury, or the need for emergency surgery, doctors must often delay treatment (especially surgery) until Pradaxa has been eliminated from the patient’s body. Compromised kidney function increases the risk further. Pradaxa, unlike warfarin, is eliminated through the kidneys. Compromised kidney function slows the body’s availability to eliminate Pradaxa, meaning that needed treatment may be delayed even further. In some cases, delayed treatment has led to death. In other cases, treatment could not be delayed and patients bled to death because doctors could not reverse Pradaxa’s effects. Given the increased risk of bleeds and the increased difficulty in reversing Pradaxa’s effects, one might question whether Pradaxa really is as safe as warfarin. Even if Pradaxa is not as safe, a patient might still chose to use it if Pradaxa provides a benefit that outweighs the increased risk.
BI touts Pradaxa as being 35% more effective than warfarin. A full reading of the RE-LY data reveals that the 35% reduction in stroke risk applies only to patients who were poorly controlled. If a patient is poorly controlled on anti-coagulation therapy, the blood may be too thin, leading to an increased risk of severe bleeding, or too thick, leading to an increased risk of stroke. Among RE-LY patients who were well-controlled on warfarin, Pradaxa was no more effective at preventing stroke. If Pradaxa is no more effective at preventing stroke in a well-controlled warfarin patient, why should the patient switch to Pradaxa?
Pradaxa is more convenient than warfarin, according to BI’s promotional material. This is, BI says, because patients on Pradaxa need not undergo regular blood tests to determine whether they are well–controlled, while warfarin patients must be regularly tested so doses can be adjusted to keep the blood properly thinned. Warfarin also requires certain dietary restrictions. Certain foods, including green vegetables, contain vitamin K, which counteracts the effects of warfarin. Pradaxa carries no dietary restriction. Are these tests and dietary restrictions such an inconvenience that patients should switch to Pradaxa? Not according to the RE-LY data. According to a recent evaluation of RE-LY data ("Assessing the Impact of Dabigatran and Warfarin on Health-Related Quality of Life: Results from an RE-LY Sub-study" published in the International Journal of Cardiology on May 10, 2013):
"anticoagulated patients without outcome events (e.g. strokes or major bleedings) had stable [health related quality of life]. Scores between dabigatran and warfarin were comparable, which was unexpected given the known complexities of warfarin treatment.”
According to BI’s own data, excluding those patients who are successfully treated (meaning the treatment didn’t fail, resulting in stroke, or cause a bleed) with these two anti-coagulants, there is no difference in quality of life as related to health and medical treatment. That is to say, according to the patients successfully treated, using Pradaxa is no more inconvenient than using warfarin. It should also be noted that this data was gathered at a time when BI did not recommend any testing for patients on Pradaxa. BI now recommends that Pradaxa patients undergo kidney function testing.
We’ve asked before, and we ask again: why should a patient who is well-controlled on warfarin switch to Pradaxa? In such a patient, according to the available data, Pradaxa is no more effective, no more convenient, more likely to cause a bleed, and harder to reverse in an emergency. While Pradaxa may be appropriate for patients poorly-controlled with warfarin, the data seems to suggest that, among those well-controlled with warfarin, Pradaxa provides no clinical benefit.
Here’s another question: why did BI not release all of this information when it released Pradaxa in 2010? RE-LY, the source of all of this data, was completed long before Pradaxa hit the market. BI had the data, and certainly could have – almost certainly should have – publicized this data. What could BI have to lose by keeping this data to themselves for years?