Share this article on:

FDA Approval Sought for Rivaroxaban to Prevent Stroke in Atrial Fibrillation

Hurley, Dan

doi: 10.1097/01.NT.0000394980.53862.0c
Back to Top | Article Outline




Investigators reported that a new oral agent, rivaroxaban, requires none of the frequent testing for anticoagulation associated with warfarin, and also appears to lower the risk of intracranial bleeding events.

Neurologists seeking to prevent stroke in patients with atrial fibrillation may soon have a second new alternative to warfarin to work with, in addition to the recently approved drug dabigatran, if the FDA likes what it sees in the results of a large international trial unveiled at the November meeting of the American Heart Association (AHA).

Like dabigatran, the new oral agent, rivaroxaban (Xarelto, Bayer/Johnson & Johnson), requires none of the frequent testing for anticoagulation associated with warfarin, and also appears to lower the risk of intracranial bleeding events. Rivaroxaban, an oxazolidinone derivative that inhibits Factor Xa, interrupts the intrinsic and extrinsic pathway of the blood coagulation cascade, inhibiting both thrombin formation and development of thrombi.

On Jan. 5, a new drug application for rivaroxaban was submitted to the FDA for the prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation (AF).

As described at the AHA meeting, rivaroxaban met the pre-specified criteria for non-inferiority to warfarin in the ROCKET AF trial. Whether the drug is in fact superior to warfarin in preventing both hemorrhagic and ischemic strokes was less clear on the basis of the presentation, but its greater ease of use drew enthusiastic views from neurologists who specialize in stroke.

“The fact that it may be equivalent to warfarin is very promising to hear,” said Sean I. Savitz, MD, associate professor of neurology and co-director of the stroke program at the University of Texas Medical School at Houston. “A lot of us have been waiting for these new anticoagulant agents with great anticipation. Warfarin is generic and the cost to buy the drug is low, but patients don't want to have to undergo so many blood tests to check their INR [international normalized ratio] level. Also, on warfarin, patients are often under or over the desired therapeutic INR level. It would be great to have drugs that have more stable dosing and that do not require blood test monitoring.”

While awaiting the kind of details only publication in a peer-reviewed journal will bring forth, he said that based on the slides presented at the AHA meeting, ROCKET AF appeared to have “all the makings of a good study.”

Back to Top | Article Outline


The clinical trial randomized 14,264 patients in 45 countries to either 20 mg daily of rivaroxaban, or warfarin with an INR target of 2.5. To be eligible, patients had to have AF and either a prior stroke, transient ischemic attack or systemic embolus, or at least three high-risk factors for stroke: congestive heart failure, hypertension, age of at least 75 years, or diabetes. The primary outcome was stroke or systemic embolism during a mean follow-up of 1.9 years.

No matter how the outcome data was statistically sliced, it reached a statistically significant p value for non-inferiority of rivaroxaban compared to warfarin, but not always for superiority.

On the primary efficacy endpoint — the “per protocol” analysis, which includes all randomized patients except those with pre-defined major protocol deviations (such as study drug compliance of less than 60 percent ) — the study found a rate of stroke or non-CNS embolism of 1.71 per 100 patient years on rivaroxaban, compared to 2.16 on warfarin, for a non-inferiority p value of less than 0.001, and a superiority p value of 0.018.

But on the most conservative measure, the intention-to-treat analysis (which includes all randomized patients followed until the end of the study), the event rate per 100 patient years was 2.12 on rivaroxaban, compared to 2.42 on warfarin, for a non-inferiority p value of less than 0.001, but a non-significant superiority p value of 0.117.

A member of the ROCKET-AF executive study committee noted that 55 percent of the patients in the trial had a prior stroke or TIA or systemic embolism. In general, the outcomes in these patients appear similar to the overall trial findings but more analyses of this important population are ongoing.

“These analyses are important for neurologists who will be making decisions about anticoagulation for their patients who had a prior stroke and have atrial fibrillation,” said Kenneth W. Mahaffey, associate professor of medicine in the division of cardiology at Duke University School of Medicine and co-director of cardiovascular research at the Duke Clinical Research Institute.

The time-in-therapeutic range for warfarin-assigned patients was just 57.8 percent, an unusually low value for a clinical trial and one that led discussants at the AHA presentation to question whether the warfarin arm was a fair comparator. But the figure did not provoke great concern from Mitchell S.V. Elkind, MD, associate professor of neurology and associate chairman of neurology for clinical research and training at Columbia University.

“Warfarin is often very difficult to manage in clinical practice as well as in clinical trials,” Dr. Elkind said. “That's one of its disadvantages.”

One of the more heartening findings of the study was the low number of intracranial hemorrhages seen in the rivaroxaban arm, a total of just 55, for a rate of 0.5 percent per year, compared to 84 in the warfarin arm, or 0.7 percent per year, for a relative risk reduction of 33 percent (95% CI, 6% to 53%, p=0.2).



“Intracranial hemorrhage is the most feared and devastating complication of anticoagulation of old people,” said Robert G. Hart, MD, professor of neurology at the University of Texas Health Science Center in San Antonio. “It is the deal-breaker in the benefit/risk equation. I think that the message for neurologists is that the era of novel oral anticoagulants has arrived.”

Dr. Hart pointed to last year's approval of dabigatran, and the results of not only rivaroxaban but also apixaban (presented at the European Society of Cardiology meeting in August) as evidence that 2010 will go down in history as marking “the beginning of the end for our old feisty friend warfarin.”

Apixaban, under development by Bristol-Myers Squibb and Pfizer, was compared to aspirin in a phase 3 trial involving over 5,500 atrial fibrillation patients at risk for stroke but unsuitable for treatment with warfarin. The trial was terminated early when the drug's efficacy over aspirin was established after a mean follow-up of 1.1 years, with first strokes reduced 54 percent over aspirin (p<0.0001). Among the 765 patients who had a prior stroke or TIA, the rate of stroke during the study was 8.0 percent per year on aspirin, compared to 2.7 percent per year on apixaban (66 percent reduction, p=0.003).

But with only dabigatran yet approved as an alternative to warfarin, Dr. Savitz said that in his clinical practice, he began using it soon after its approval by the FDA in October.

“A lot of issues have come up,” he told Neurology Today. “If somebody has an intracerebral hemorrhage, since we know the INR with warfarin, we can reverse it by giving blood products. But in the situation of dabigatran, we're not aware if anything will work to reverse its effects. It will take time to accumulate experience with the use of the drug.”

Dr. Elkind said that he, too, has begun using dabigatran, but said he was less concerned about the rare risk of bleeding compared to its effectiveness in preventing clots. “We'll probably make decisions about treatment based on the vast majority of patients who will be helped by it, rather than by the few who will have a bleeding event,” he said.



On balance, Dr. Elkind said: “What the studies are showing us is these new agents are probably about as good as warfarin. They may be marginally better in some of the analyses, but nothing is a slam dunk against Coumadin. We're just going to have a lot more options.”

He added, however, that not all his patients have been eager to switch. “We hear a lot about the inconvenience of checking the INR, but I've had some patients tell me they like to get their INR checked,” Dr. Elkind said. “It gives them the sense that they're in the right range and they're being watched over.”

Back to Top | Article Outline


Ezekowitz MD, Wallentin L, Connolly SJ, et al. Dabigatran and warfarin in vitamin K antagonist-naive and experienced cohorts with atrial fibrillation. Circulation 2010; 122 (22): 2246–53.
©2011 American Academy of Neurology