18 October 2011
The published data from ROCKET AF in 14,264 patients with nonvalvular atrial fibrillation (AF), who were at moderate to high risk of stroke, found that rivaroxaban daily was non-inferior to dose-adjusted warfarin for the prevention of the primary endpoint of stroke or systemic embolism. There was no difference between the drugs in the risk of major bleeding, but intracranial haemorrhage and fatal bleeding occurred less frequently with rivaroxaban.
Level of evidence:
Level 1 (good quality patient-oriented evidence) according to the SORT criteria.
Rivaroxaban▼ (Xarelto) is currently in licensing for the prevention of strokes in patients with AF. Once marketing approval is given then the number of eligible patients could be large. NICE guidance is anticipated in May 2012 and in order to prepare local decision-making bodies may need to:
- identify those patients for whom the drug might be appropriate
- plan for and assess the implications of potential changes to anticoagulation services. It should be borne in mind that the infrastructure of warfarin clinics will still be necessary for patients well established on warfarin.
This study represents the best available evidence relating to this new medicine. These data, together with their limitations, should be taken into account by local decision-makers when managing the entry of this new medicine and planning the commissioning of appropriate services. Local decision-makers may need to consider additional published data about this product as it emerges.
What is the background to this?
Patients with AF are known to be at increased risk of stroke. AF affects about 1.2% of the population with prevalence increasing with age. NICE guidance for the management of patients with AF recommends thromboprophylaxis with either warfarin or aspirin depending on their level of risk, and after consideration of individual patient factors. Dose-adjusted warfarin is the most effective and preferred option for people who are at high risk of stroke. However, it is associated with a higher bleeding risk than aspirin, and is not recommended for those at low risk of stroke. An update of this guideline is currently in the process of being scheduled into the NICE work programme. Details of any update will be found here.
Rivaroxaban▼ (Xarelto) is an oral direct factor Xa inhibitor administered as a fixed-dose that does not require laboratory monitoring and, unlike warfarin, has no known food and few drug interactions. It is currently licensed for short-term use (maximum 35 days) for the prevention of venous thromboembolism (VTE) in adult patients undergoing elective hip or knee replacement surgery. The European Medicines Agency (EMA) has recently granted a positive opinion for two new strengths (15 mg and 20 mg tablets) for the prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation with one or more risk factors. NICE guidance for this additional indication is anticipated in May 2012.
What does this study claim?
The per-protocol, as treated primary analysis was designed to determine whether rivaroxaban was non-inferior to dose-adjusted warfarin (target INR of 2.0 to 3.0) in preventing stroke or systemic embolism. The primary efficacy endpoint was a composite of stroke (ischaemic or haemorrhagic) and systemic embolism. Among patients in the warfarin group INR values were within the therapeutic range for a mean of 55% of the time. Over a median 590 days of treatment exposure the event rates for the primary endpoint were 1.7% per year in the rivaroxaban daily group and 2.2% per year in the warfarin group (hazard ratio [HR] 0.79, 95% confidence interval [CI] 0.66 to 0.96, p < 0.001 for non-inferiority).
The trial protocol specified that if non-inferiority was achieved in the primary analysis a test for superiority should be carried out in the safety population during treatment i.e. those patients who received at least one dose of study drug and were followed up while they received the drug or within 2 days after discontinuation. In this population rivaroxaban was shown to be statistically superior to warfarin with an event rate for stroke and systemic embolism of 1.7% per year compared with 2.2% per year for warfarin (HR 0.79, 95% CI 0.65 to 0.95, p < 0.02 for superiority).
Testing for non-inferiority and superiority was also performed in the intention to treat (ITT) population as part of a sensitivity analysis. The event rate for stroke and systemic embolism was 2.1% per year for rivaroxaban and 2.4% per year for warfarin (HR 0.88, 95% CI 0.75 to 1.03, p < 0.001 for non-inferiority, p = 0.12 for superiority).
The principal safety endpoint was a composite of major and nonmajor clinically relevant bleeding events. This occurred in 1475 patients in the rivaroxaban group and in 1449 patients in the warfarin group; an event rate of 14.9% and 14.5% per year, respectively (HR for rivaroxaban 1.03, 95% CI 0.96 to 1.11, p = 0.44). Intracranial haemorrhage occurred less frequently with rivaroxaban (0.5% vs. 0.7% per year, p = 0.02), as did fatal bleeding (0.2% vs. 0.5% per year, p = 0.003).
How does this relate to other studies?
Dabigatran etexilate▼(Pradaxa) is a direct thrombin inhibitor that has been approved for oral use to prevent stroke and systemic embolism in adult patients with nonvalvular AF and one or more risk factors. This licence approval was based on data from the RE-LY study that the NPC discussed here. The NPC have also covered an editorial in the British Medical Journal which discussed the role of dabigatran etexilate in patients with AF and concluded that whilst the drug has some benefits, evidence on long-term efficacy and safety is required. A NICE appraisal of dabigatran etexilate for the management of patients with AF is underway and due to be published in December 2011.
Rivaroxaban▼ was shown to be non-inferior to warfarin in preventing strokes or systemic embolism in people with atrial fibrillation who are at moderate to high risk for a stroke, while demonstrating a comparable risk of major and nonmajor clinically significant bleeding. Intracranial haemorrhage occurred less frequently than with warfarin, but the incidence of gastrointestinal bleeding increased.
The trial methodology increases the complexity in interpreting the efficacy data. In non-inferiority trials an ITT analysis should be conducted also. Such an analysis avoids the bias that occurs with per-protocol on-treatment analyses when patients discontinue their randomised treatment for reasons related to the treatment and therefore the patients who do so have a different risk profile from those who do not.
In ROCKET AF the on-treatment analysis was based on observations that were truncated at 2 days after discontinuation of randomised treatment; a timeframe likely to miss events related to inadequate coagulation during the transition to alternative treatment. There may be a greater risk of such events with rivaroxaban as it has a short half-life of about 5-9 hours compared to that of warfarin of 40 hours. The ITT population included all patients who underwent randomisation and were followed for events during treatment or after premature discontinuation.
The test for superiority in the “as-treated safety population” demonstrated that the p value was significant for rivaroxaban; however the ITT analysis did not show superiority for rivaroxaban over warfarin.
The new oral anticoagulants do not require regular anticoagulant monitoring therefore there might be the opportunity to reduce some costs associated with anticoagulation services and reduce the impact of monitoring on patients’ lives. However, many of the costs associated with the current infrastructure for warfarin will be fixed due to the need to maintain the existing service for patients well established on warfarin. The price of rivaroxaban 20mg daily is not yet known but the annual cost of this treatment for the prevention of stroke is likely to be less than a £1000; still considerably more expensive than warfarin.
Another source of concern is that to date clinical trials have not addressed the absence of antidotes to rapidly reverse the anticoagulant effect of rivaroxaban or dabigatran in case of life-threatening haemorrhage e.g. in surgery.
A phase III, multicentre, randomised, double-blind, double-dummy, event-driven study.
The study involved 14,264 patients with nonvalvular AF who were at moderate to high risk of stroke. Elevated risk was indicated by a history of stroke, transient ischaemic attack (TIA) or systemic embolism, or at least two of the following risk factors: heart failure or a left ventricular ejection fraction ≤ 35%, hypertension, an age of 75 years, or more or the presence of diabetes mellitus. Due to violations in Good Clinical Practice guidelines 93 patients were excluded from all efficacy analyses.
Exclusion criteria included heart valve disorders, a severe stroke within 3 months or any stroke within 14 days, creatinine clearance of less than 30 ml/min or significant liver disease.
The median age was 73 years and 60.3% were men. The mean CHADS2 score at baseline was 3.5; 54.8% of patients had had a previous stroke, systemic embolism or TIA; 90.5% had hypertension and 62.5% had heart failure. In the rivaroxaban group 16.6% had had a previous myocardial infarction compared with 18.0% in the warfarin group (p < 0.05). Previous use of vitamin K antagonists was reported by 62.4% of patients.
Intervention and comparison
Patients were randomised to receive either rivaroxaban (20mg daily or 15mg daily in patients with a creatinine clearance of 30 to 49 ml per minute) or dose-adjusted warfarin (target INR of 2.0 to 3.0). INR values were within the therapeutic range in the warfarin group for a mean of 55% of the time. At some time during the study 34.9% of the rivaroxaban group and 36.2% of the warfarin group took aspirin concurrently.
Outcomes and results
The primary efficacy endpoint was the composite of stroke (ischaemic and haemorrhagic) and systemic embolism. The primary analysis was pre-specified to be performed in the per-protocol population. If non-inferiority was achieved in the primary analysis then a prespecified test for superiority was carried out in the safety population during treatment i.e. those patients who received at least one dose of study drug and were followed up while they received the drug or within 2 days after discontinuation. The median follow-up period for both these analyses was 590 days. A sensitivity analysis for non-inferiority and superiority in the ITT population was also performed with a median follow-up period of 707 days.
In the per-protocol population 188 primary endpoint events occurred in the rivaroxaban group (n = 6958) compared with 241 events in patients taking warfarin (n = 7004) giving an event rate of 1.7% and 2.2% per year, respectively. The HR for rivaroxaban was 0.79, 95% CI 0.66 to 0.96, p < 0.001 for non-inferiority.
In the as-treated safety population 189 primary endpoint events occurred in patients in the rivaroxaban group (n = 7061) and 243 primary endpoint events in the warfarin group (n = 7082). The event rate was 1.7% and 2.2% per year, respectively (HR for rivaroxaban 0.79, 95% CI 0.65 to 0.95, p < 0.02 for superiority). Details of secondary efficacy outcomes are available for this group. There were 101 myocardial infarctions in patients taking rivaroxaban and 126 in the warfarin group (an event rate of 0.9% and 1.1% per year, respectively; HR 0.81, 95%CI 0.63 to 1.06, p =0.12). Death from all-causes occurred in 208 patients in the rivaroxaban group and 250 in the warfarin group. (1.9% and 2.2% per year, respectively; HR 0.85, 95%CI 0.70 to 1.02, p = 0.07).
For the ITT population 269 primary endpoint events occurred in the rivaroxaban group (n = 7081) and in 306 in those patients taking warfarin (n = 7090). An event rate of 2.1% and 2.4% per year, respectively (HR 0.88, 95% CI 0.75 to 1.03, p < 0.001 for non-inferiority, p = 0.12 for superiority).
The principal safety endpoint was a composite of major and nonmajor clinically relevant bleeding events. There were 1475 events in the rivaroxaban group and 1449 events in the warfarin group (an event rate of 14.9% and 14.5% per year, respectively, HR for rivaroxaban 1.03, 95% CI 0.96 to 1.11, p = 0.44). Intracranial haemorrhage occurred less frequently with rivaroxaban (0.5% vs. 0.7% per year, p =0.02), as did fatal bleeding (0.2% vs. 0.5% per year, p =0.003).
Major bleeding from a gastrointestinal site was more common in the rivaroxaban group (3.2%) compared with 2.2% for warfarin, p ≤ 0.001.
The rates of adverse events was similar in both groups, except for epistaxis which occurred more frequently in patients treated with rivaroxaban (10.14%) than with warfarin (8.55%), p < 0.05. Elevated serum levels of hepatic enzymes occurred with equal frequency in the two groups, 0.47 % for rivaroxaban and 0.50% for warfarin.
This study was sponsored by Johnson & Johnson, and Bayer Healthcare.
Please comment on this rapid review using our feedback form.
Useful links to other organisations that provide information on managing the entry of new medicines can be found here
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you