Consistent Benefit of Apixaban, Even in Patients at Highest Risk of Bleeding: ARISTOTLE

October 04, 2012

October 3, 2012 (Durham, North Carolina) — An analysis of Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) has demonstrated consistent stroke and major bleeding reductions with apixaban (Eliquis, Bristol-Myers Squibb/Pfizer) compared with warfarin in patients with differing levels of baseline risk [1]. In an analysis stratified by baseline risks of stroke and bleeding with CHADS2, CHA2DS2-VASc, and HAS-BLED scores, researchers found that apixaban consistently outperformed warfarin in terms of reducing the risk of clinical events, including stroke, bleeding, and mortality.

Apixaban also appeared to be even more effective in reducing the risk of intracranial bleeding in patients with higher HAS-BLED scores. In individuals with HAS-BLED scores of >3, apixaban reduced the risk of intracranial bleeding 78% compared with warfarin, whereas the risk of intracranial bleeding was reduced 34% compared with warfarin in patients with HAS-BLED scores of 0 to 1.

"This is very important to highlight," lead investigator Dr Renato Lopes (Duke Clinical Research Institute, Durham, NC) told heartwire , "because every time we decide to give an anticoagulant to patients with atrial fibrillation, the risk of intracranial bleeding is always a major concern. It is a devastating complication. Here, we showed that in patients with a HAS-BLED score of 3 or greater--in other words, the patients at a high risk for bleeding in clinical practice--the magnitude of the reduction in bleeding risk was even greater than in patients at lower risk for bleeding. Apixaban resulted in less intracranial bleeding than warfarin regardless of the patients' risk for bleeding, but it caused even less bleeding than warfarin in patients with higher HAS-BLED scores."

This, say the researchers, is one of the major findings of the study, one that should help physicians in their daily decision-making when treating AF patients.

In an editorial accompanying the study [2], Drs Vassilis Vassiliou and Paul Flynn (Cambridge University National Health Service Trust, UK) note that the decision to give an anticoagulant to a patient with atrial fibrillation is a difficult one, with thromboembolism being pitted against hemorrhage, and both risks imperfectly predicted by even the best risk-stratification scores. Given the reduction in hemorrhage among apixaban-treated patients with a HAS-BLED score of >3, the latest analysis suggests that "after many years of searching, physicians might now face a less agonizing choice on behalf of high-risk patients who have not been managed well in the past."

The new analysis and editorial are published online October 2, 2012 in the Lancet.

Secondary Analysis of the Successful ARISTOTLE Study

In ARISTOTLE, first presented at the European Society of Cardiology 2011 Congress in Paris, France, and reported by heartwire at that time, the investigators randomized 18 201 AF patients to apixaban (5 mg orally twice daily) or warfarin (target INR of 2.0 to 3.0). After a median follow-up of 1.8 years, results showed that apixaban was associated with a 21% reduction in the risk of stroke or systemic embolism, a 31% reduction in bleeding, and an 11% reduction in all-cause mortality. The new analysis attempted to determine how the results differed according to the patients' baseline risk of ischemic events and bleeding using the risk scoring systems.

"We know that the scores have basically two objectives in clinical practice," said Lopes. "One is to predict risk. We know that the scores used in atrial fibrillation are good for that, with CHADS2 and CHA2DS2-VASc good for predicting stroke risk and HAS-BLED and other bleeding scores good for predicting the risk of bleeding. However, for the second objective, which is using the scores to guide therapy, the scores don't help us that much."

Given the overlap between variables used to calculate the risk scores, it can be difficult to pinpoint the exact trade-off between risk and benefit that would allow physicians to prescribe anticoagulation or hold off treatment, said Lopes. The alternative to this, he added, would be a drug that worked across the board.

With this in mind, the researchers wanted to assess the safety and efficacy of apixaban across the different stroke and bleeding risk categories. Overall, apixaban significantly reduced the risk of stroke, systemic embolism, mortality, International Society of Thrombosis and Hemostasis (ISTH) major bleeding, intracranial bleeding, and any bleeding compared with warfarin irrespective of CHADS2 and CHA2DS2-VASc score. Similarly, irrespective of HAS-BLED score, patients treated with apixaban had lower rates of stroke or systemic embolism, mortality, ISTH major bleeding, TIMI major or minor bleeding, GUSTO severe or moderate bleeding, and any bleeding.

As noted, the reduction in intracranial bleeding with apixaban compared with warfarin was larger in patients with a HAS-BLED score >3 (hazard ratio 0.22; 95% CI 0.10–0.48) than those with a HAS-BLED score 0 to 1 (HR 0.66; 95% CI 0.39–1.12), a differential reduction between the two HAS-BLED groups that bordered on statistical significance (p=0.0604 for interaction).

"It is important to use these scores in clinical practice to quantify the risk for events that patients have and to identify patients who would benefit from treatment," Lopes told heartwire . "Remember, all these recent large AF trials excluded patients with a CHADS2 score of zero. When using these risk scores to guide treatment decisions--in other words, to guide whether or not to give the drug to a patient--it's nice to have a new drug like apixaban that has consistent results showing that it decreases stroke, causes less bleeding, and causes less mortality, regardless of the bleeding and stroke risk-score categories. This is the first and an important step toward making physicians' lives easier in the day-by-day management of stroke prevention in patients with AF."

The editorialists echo this statement but add that now is probably not the time to replace warfarin with apixaban as the drug of choice in patients with atrial fibrillation. As noted, participants in ARISTOTLE had at least one risk factor for thromboembolism in addition to atrial fibrillation, so another trial is needed to identify the role of apixaban when no additional risk is present. In addition, cost-effectiveness analyses will need to be performed, although Vassiliou and Flynn emphasize that not only was apixaban associated with a clinical advantage over warfarin, it significantly reduced all-cause mortality.