 |
 |
14th February 2011
Two phase III EINSTEIN trials of rivaroxaban▼ (Xarelto®) have recently been published as a single paper. The EINSTEIN–DVT (Acute) Study (n=3,449) found that rivaroxaban was non-inferior to standard therapy (enoxaparin followed by Vitamin K antagonist) at preventing symptomatic, recurrent venous thromboembolism (VTE) when anticoagulation therapy was administered for up to 12 months for the acute event. In the EINSTEIN-Extension (Continued Treatment) Study (n=1,196) rivaroxaban was more effective than placebo when patients had continued anticoagulation after the acute treatment. The risk of major bleeding in both trials was not raised by rivaroxaban, but clinically relevant non-major bleeding was seen more frequently than placebo in the Extension Study.
Level of evidence:
Level 1 (good quality patient-oriented evidence) according to the SORT criteria.
Action
We have discussed previously that dabigatran etexilate▼ and rivaroxaban▼, unlike warfarin, do not require regular anticoagulant monitoring. Although the costs of dabigatran etexilate and rivaroxaban for the management of acute VTE are not yet known, they are likely to be higher than that for warfarin. Reduced monitoring is attractive for patients and the NHS. However, commissioners may wish to include in their planning that many of the costs associated with current warfarin services will be fixed, due to the need to maintain the existing infrastructure for long-standing patients using warfarin, whilst the place in therapy of newer antithrombotics is established.
What is the background to this?
Standard treatment of acute VTE (deep vein thrombosis [DVT] or pulmonary embolism [PE]) is with parenteral heparin with overlapping administration of a vitamin K antagonist (VKA). In the UK, warfarin is the most commonly used VKA. Duration of treatment will vary depending upon the individual patient, risk factors and the underlying cause of the VTE. Further information on the management of VTE can be found in the NPC eLearning materials on VTE.
Rivaroxaban, an oral direct factor Xa inhibitor administered as a fixed-dose, does not require laboratory monitoring and, unlike warfarin, has no known food and few drug interactions.
Rivaroxaban is currently licensed for short-term use for the prevention of VTE in adult patients undergoing major hip surgery (treatment duration of 5 weeks is recommended) and for major knee surgery (treatment duration of 2 weeks is recommended). It is recommended by NICE as one of the options to consider for this indication.
NICE guidance on the use of dabigatran etexilate for the treatment of acute VTE is planned. NICE is also to appraise rivaroxaban for the treatment and secondary prevention of VTE. No timelines are available yet for either appraisal.
What do these studies claim?
In the acute trial, the primary efficacy outcome of symptomatic, recurrent VTE was seen in 2.1% of rivaroxaban patients and 3.0% of patients on standard therapy (subcutaneous enoxaparin followed by a VKA), hazard ratio (HR) 0.68 (95% confidence interval [CI] 0.44 to 1.04), P<0.001 for non-inferiority.
Clinically-relevant bleeding (a composite of major or clinically-relevant non-major bleeding) was the principal safety outcome and was experienced by 8.1% of patients in each group; HR 0.97 (95%CI 0.76 to 1.22), P=0.77.
A pre-defined secondary outcome included net clinical benefit (a composite of the primary efficacy outcome or major bleeding) and was seen in 2.9% of rivaroxaban and 4.2% of standard treatment patients: HR 0.67 (95%CI 0.47 to 0.95), P=0.03.
In the Continued Treatment Study patients who had completed 6 to 12 months of treatment for VTE, and were deemed to require further anticoagulation, received either rivaroxaban or placebo for another 6 or 12 months. The same primary outcome as for the Acute Study was used and occurred in 1.3% of rivaroxaban and 7.1% of placebo patients (HR 0.18, 95%CI 0.09 to 0.39, P<0.001).The principal safety outcome, major bleeding, was seen in 0.7% of rivaroxaban but in none of the placebo patients (P=0.11). Hence 34 recurrent venous thromboembolic events were prevented at the cost of 4 major bleeding events.
Net clinical benefit (a composite of recurrent VTE or major bleeding) was seen in 2.0% of rivaroxaban and 7.1% of placebo patients (HR 0.28, 95%CI 0.15 to 0.53, P<0.001).
How does this relate to other studies?
Dabigatran etexilate, a further oral anticoagulant, was studied in the RE-COVER trial and found to be as effective as dose-adjusted warfarin in preventing recurrent VTE and related deaths in patients with verified, symptomatic, acute VTE initially treated with parenteral anticoagulants. The risk of major bleeding and other adverse effects were similar in the two groups. Further details can be found in our rapid review.
The Acute PE Study (EINSTEIN-PE) of rivaroxaban is ongoing and expected to complete in April 2011.
So what?
Oral rivaroxaban, as a single treatment, was found to be non-inferior to enoxaparin plus oral VKA in the Acute DVT Trial and better than placebo in preventing recurrent VTE in the Continued Treatment Study, with an acceptable risk of bleeding. The latter study explored the benefit-to-risk ratio when treatment with rivaroxaban was administered for an additional 6 to 12 months in patients who had completed 6 to 12 months of anticoagulation. Clinically the long-term risk of recurrent VTE after anticoagulation is stopped must be balanced against the risks of ongoing therapy. In the Continued Treatment Study rivaroxaban prevented 34 VTE recurrent events at a cost of 4 major bleeds.
One concern is that Factor Xa inhibitors have shorter half-lives than warfarin, which may result in less protection if doses are missed. Generally, trials of these agents have not included older patients, patients at high risk of bleeding, those with complex medical illnesses or on concomitant antiplatelet treatment. Experience using these newer agents in clinical practice may be different from the trial setting.
It should be noted that the dose of enoxaparin (1mg/kg twice daily) used in the acute study reflects the US dose, which differs from that used in the UK. Whether there are any significant implications of this are uncertain.
Study details – EINSTEIN-DVT (Acute) Study
The EINSTEIN Investigators. Oral Rivaroxaban for Symptomatic Venous Thromboembolism. N Engl J Med 2010; 363:2499-510
Design: Randomised, event-driven, non-inferiority, open-label study with concealed allocation. Suspected outcome events were assessed by a blinded adjudication committee.
Patients: Patients had acute, symptomatic, objectively confirmed proximal DVT, but without symptomatic PE. Patients at high risk of bleeding and those with clinically significant liver disease were amongst those excluded. Patients had a mean age of 56.1 years.
Intervention and comparison: 1,731 patients were assigned to receive rivaroxaban, 15mg twice daily for the first three weeks and then 20mg daily for a total of three, six or twelve months (duration determined by the treating physician prior to randomisation). 1,718 patients were assigned to receive standard therapy: at least five days of subcutaneous enoxaparin 1mg/kg twice daily and a VKA. Enoxaparin was stopped when the international normalized ratio (INR) was at least 2.0 for two consecutive days. The median duration of treatment with enoxaparin was eight days (interquartile range six to eleven). The VKA dose was adjusted to maintain an INR of 2.0 to 3.0. Overall the INR was in this therapeutic range for 57.7% of the time.
Outcomes and results: It had been pre-defined that enrolment would stop once it had been estimated that enough events had occurred. Therefore, duration of treatment was shorter than anticipated for 5.9% of rivaroxaban and 5.5% of standard therapy patients. 15 rivaroxaban patients (0.9%) and 18 standard therapy patients (1.0%) were lost to follow-up.
The primary efficacy outcome of objectively documented symptomatic, recurrent VTE (a composite of DVT or non-fatal or fatal PE) assessed using ITT was seen in 2.1% of rivaroxaban patients and 3.0% of patients on standard therapy: HR 0.68 (95% CI 95% 0.44 to 1.04), P<0.001 for non-inferiority.
A secondary outcome, net clinical benefit (a composite of the primary efficacy outcome or major bleeding) was seen in 2.9% of rivaroxaban and 4.2% of standard treatment patients: HR 0.67 (95%CI 0.47 to 0.95), P=0.03.
4.3% of the rivaroxaban patients and 3.9% of the standard treatment group withdrew due to adverse events. The principal safety outcome was clinically-relevant bleeding, defined as a composite of major or clinically-relevant non-major bleeding. This was experienced by 8.1% of patients in each group; HR 0.97 (95%CI 0.76 to 1.22), P=0.77.
The combination of an alanine aminotransferase level exceeding 3x upper limit of normal (ULN) and a bilirubin level exceeding twice the ULN was seen in 0.1% of rivaroxaban patients and in 0.2% on standard therapy.
Study details –EINSTEIN-Extension (Continued Treatment) Study
The EINSTEIN Investigators. Oral Rivaroxaban for Symptomatic Venous Thromboembolism. N Engl J Med 2010; 363:2499-510
Design: Double-blind, event-driven, placebo-controlled randomised trial with concealed allocation. Suspected outcome events were assessed by a blinded adjudication committee.
Patients: Patients (mean age 58.3 years) had objectively confirmed symptomatic DVT or PE treated for 6 to 12 months with a VKA or rivaroxaban and were deemed to require continued anticoagulation. Patients at high risk of bleeding and those with clinically significant liver disease were amongst those excluded.
Intervention and comparison: Patients were randomised to rivaroxaban 20mg daily (n=602) or matching placebo (n=595) for six or 12 months, determined by the treating physician prior to randomisation. Of the rivaroxaban group, 429 (71%) had previously received a VKA and the rest had received rivaroxaban. One patient from the placebo group was excluded from the results due to invalid consent.
Outcomes and results: As in the previous study, duration of treatment was shorter than intended due to event-driven termination: in 25.9% of rivaroxaban and 24.9% of placebo patients. 0.2% of patients in each group were lost to follow-up.
The primary efficacy end point was objectively documented symptomatic, recurrent VTE (the composite of DVT or non-fatal or fatal PE) using the ITT population. This was confirmed in 1.3% of rivaroxaban and 7.1% of placebo patients (HR 0.18, 95%CI 0.09 to 0.39, P<0.001).
A secondary outcome, net clinical benefit (a composite of the primary efficacy outcome or major bleeding) was seen in 2.0% of rivaroxaban and 7.1% of placebo patients (HR 0.28, 95%CI 0.15 to 0.53, P<0.001).
The principal safety outcome, major bleeding, was seen in 0.7% of rivaroxaban patients and in none of the placebo patients (P=0.11). No fatal haemorrhages occurred. Clinically relevant non-major bleeding, predominantly mucosal, was more frequent in rivaroxaban patients: 5.4% versus 1.2% (P value not quoted). All four rivaroxaban patients with major bleeding and 6 of the 32 with clinically relevant non-major bleeding withdrew from treatment.
No patient in either group had the combination of an alanine aminotransferase level exceeding 3x upper limit of normal (ULN) and a bilirubin level exceeding twice the ULN.
Sponsorship for both studies
Bayer Schering Pharma and Ortho-McNeil
Please comment on this blog in the NPC discussion rooms, or using our feedback form.
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