CDEC RECORD OF ADVICE RIVAROXABAN (Xarelto Bayer Inc.) Indication: Deep Vein Thrombosis Without Symptomatic Pulmonary Embolism This document summarizes the Canadian Drug Expert Committee (CDEC) response to the Formulary Working Group Request for Advice regarding the rivaroxaban (Xarelto) Final Recommendation (August 16, 2012). Submission History: Rivaroxaban was previously reviewed by CDEC for the treatment of deep vein thrombosis (DVT) without symptomatic pulmonary embolism, and received a recommendation of list with criteria for a duration not to exceed six months (see Notice of CDEC Final Recommendation, August 16, 2012). The reasons for the recommendation were: 1. In one large randomized controlled trial of patients with acute symptomatic DVT without symptomatic pulmonary embolism (the EINSTEIN DVT trial), rivaroxaban was reported to be non-inferior to a regimen of enoxaparin plus a vitamin K antagonist (VKA) based on the incidence of recurrent DVT or pulmonary embolism. The majority of patients received treatment for six months or less; thus, there is limited comparative clinical data for treatment durations exceeding six months. 2. Based on the (CDR) re-analyses of the manufacturer s cost-utility analysis, rivaroxaban, at the submitted price, appeared to be cost saving compared with enoxaparin plus warfarin when treating patients for three months, and likely cost neutral when treating patients for six months. Request for Advice: The CDR participating drug plans submitted a Request for Advice for the following reasons: The majority of Canadian patients with DVT appear to receive treatment for more than three months, and a large percentage of patients receive treatment for more than six months. However, the majority of patients in the EINSTEIN DVT trial had a planned treatment duration of six months or less. Information regarding the comparative benefit and cost-effectiveness of rivaroxaban versus enoxaparin plus VKA based on international normalized ratio (INR) control and/or time in therapeutic range (TTR) would be of relevance. Such an assessment was not performed in the previous review, but would be of relevance to drug plans to inform coverage decisions, if relevant evidence is available. For patients requiring longer-term treatment, there is concern around the ability of drug plans to effectively switch patients to warfarin who may first have been started on rivaroxaban. CDEC Meeting March 20, 2013 Page 1 of 5
Given the above, CDR participating drug plans requested that CDEC provide additional clarity regarding: The availability and assessment of data on the comparative benefit and cost-effectiveness of rivaroxaban versus enoxaparin plus VKA for the treatment of DVT based on INR control and/or TTR. The cost-effectiveness of rivaroxaban treatment beyond three months. Summary of Committee Considerations: In response to this Request for Advice, CDEC considered the following: Materials included in the CDEC brief for the 2012 review of rivaroxaban. The original CDEC recommendation for rivaroxaban (August 16, 2012). The CDR Request for Advice Brief, including a new manufacturer-provided subgroup analysis from the EINSTEIN DVT trial related to INR control and TTR. Updated pharmacoeconomic information. The manufacturer s comments on the CDR Request for Advice Brief and the Canadian Agency for Drugs and Technologies in Health (CADTH) reviewers responses to the manufacturer s comments. Patient Input Information No patient groups responded to the CDR call for patient input for the original CDR review of rivaroxaban for the management of DVT. Efficacy In the EINSTEIN DVT trial, rivaroxaban was non-inferior to enoxaparin plus VKA for recurrent venous thromboembolism (hazard ratio [HR] 0.70; 95% confidence interval [CI], 0.44 to 1.10). The manufacturer conducted subgroup analyses to examine the relative efficacy of rivaroxaban versus enoxaparin plus VKA in relation to TTR. The TTR was determined for each patient based on the proportion of time in each INR category (< 2.0, 2.0 to 3.0, or > 3.0) divided by the total duration of the VKA period. Patients were then allocated to one of three subgroups based on TTR values: < 55.9%, 55.9% to 65.3%, and > 65.3%. Results for recurrent venous thromboembolism were similar in the three INR subgroups: < 55.9% (HR, 0.78; 95% CI, 0.38 to 1.63); 55.9% to 65.3% (HR 0.68; 95% CI, 0.29 to 1.59); > 65.3% (HR 0.69; 95% CI, 0.35 to 1.35). Harms (Safety and Tolerability) In the overall population, there was no statistically significant between-treatment difference in the incidence of clinically relevant bleeding (HR 0.97; 95% CI, 0.76 to 1.22). Clinically relevant bleeding was (confidential data regarding clinically relevant bleeding in the INR subgroups was removed at the manufacturer s request). CDEC Meeting March 20, 2013 Page 2 of 5
Cost and Cost-Effectiveness For the original CDR review of rivaroxaban, the manufacturer conducted a cost-utility analysis in adults with symptomatic DVT without pulmonary embolism, comparing rivaroxaban with enoxaparin (twice daily for eight days) plus warfarin (day nine onward) over a five-year time horizon. The Markov model was comprised of health states that describe the management and consequences of venous thromboembolism. Clinical inputs for the economic model were based on the EINSTEIN DVT trial. Treatment durations of three and six months were both considered in the manufacturer s analysis based on the clinical trial data. Long-term implications of treatment were estimated based on published literature. The manufacturer made assumptions regarding the monitoring costs associated with treatments for rivaroxaban: three family physician visits in the first three months and one every three months thereafter; for enoxaparin plus warfarin: eight physician visits in first three months (95% with the family physician and 5% at anticoagulation clinic) with 19% of patients requiring a home care nurse for the administration of enoxaparin for the eight days of treatment and three physician visits every three months thereafter. The manufacturer reported that rivaroxaban is dominant when compared with enoxaparin plus warfarin when patients are treated for three or six months; however, the gains in quality-adjusted life-years (QALYs) were small (0.001 to 0.003). CDR conducted a number of sensitivity analyses to account for assumptions used by the manufacturer in their economic model, including the removal of the assumed cost savings arising from early discharge for rivaroxaban patients, varying the number of days of enoxaparin used from 5 to 10, varying the dosing strategy of enoxaparin from 1.0 mg/kg twice daily to 1.5 mg/kg once daily, and varying the assumed cost of INR monitoring and family physician visits. Given the similar clinical effects and benefits in quality of life between rivaroxaban and enoxaparin plus warfarin, all re-analyses focused on an evaluation of costs alone (an assumption of similar clinical effects and harms cost minimization analyses). The CDR re-analyses of the manufacturer s cost-utility analysis suggested that rivaroxaban, at the submitted price, appeared to be cost saving compared with enoxaparin plus warfarin when treating patients for three months, likely cost neutral when treating patients for six months, and more costly when used beyond six months. The daily cost of rivaroxaban is $5.68 for the first 21 days (15 mg twice daily) and $2.84 thereafter (20 mg daily). The alternative treatment is a low molecular weight heparin (enoxaparin) plus warfarin: the daily cost of enoxaparin ranges from $20.50 (1.5 mg/kg daily) to $30.75 (1 mg/kg twice daily) administered for the first 5 to 10 days (assuming a 70 kg patient), followed by treatment with warfarin ($0.07 daily, not including monitoring costs). CDEC Meeting March 20, 2013 Page 3 of 5
Response to the Request for Advice: Based on the above considerations, CDEC s response to the Request for Advice is as follows: In response to the CDR participating drug plans request that CDEC provide additional clarity regarding the availability and assessment of data on the comparative benefit and costeffectiveness of rivaroxaban versus enoxaparin plus VKA for the treatment of DVT based on INR control and/or TTR, CDEC concluded that there is insufficient evidence to assess the relative efficacy and safety of rivaroxaban versus enoxaparin plus VKA for the treatment of DVT based on INR control and/or TTR. CDEC members noted their disappointment that they were unable to address the question due to insufficient evidence. The committee noted that this is an important question and would like the opportunity to reconsider it in the future, should sufficient evidence become available. In response to the CDR participating drug plans request that CDEC provide additional clarity regarding the cost-effectiveness of rivaroxaban treatment beyond three months, the committee concluded that the use of rivaroxaban versus enoxaparin plus VKA for the treatment of DVT is likely cost saving at three months, likely cost neutral when treating patients for six months, and more costly when used beyond six months. CDEC noted that the cost-effectiveness of rivaroxaban coverage is highly sensitive to the duration of anticoagulation and to the proportion of patients who transition to warfarin after rivaroxaban therapy. Discussion Points: CDEC discussed that patients who require treatment for DVT for a period exceeding six months may be switched to a VKA, employing a transition period as described in the product monograph. The committee noted that the safety and efficacy of transitioning patients from rivaroxaban to warfarin are inadequately studied. The committee noted that there are significant limitations with the subgroup analyses based on TTR, including the loss of randomization and a reduction in statistical power. The committee noted that the cost neutrality associated with six months of rivaroxaban treatment is dependent on the dose of enoxaparin and the monitoring regimen for warfarin. CDEC Members: Dr. Robert Peterson (Chair), Dr. Lindsay Nicolle (Vice-Chair), Dr. Ahmed Bayoumi, Dr. Bruce Carleton, Ms. Cate Dobhran, Mr. Frank Gavin, Dr. John Hawboldt, Dr. Peter Jamieson, Dr. Julia Lowe, Dr. Kerry Mansell, Dr. Irvin Mayers, Dr. Yvonne Shevchuk, Dr. James Silvius, and Dr. Adil Virani. CDEC Meeting March 20, 2013 Page 4 of 5
March 20, 2013 Meeting Regrets: One CDEC member could not attend the meeting. Conflicts of Interest: None About This Document: CDEC provides formulary listing recommendations or advice to CDR participating drug plans. CDR clinical and pharmacoeconomic reviews are based on published and unpublished information available up to the time that CDEC deliberated on a review and made a recommendation or issued a record of advice. Patient information submitted by Canadian patient groups is included in the CDR reviews and used in the CDEC deliberations. The manufacturer has reviewed this document and has requested the removal of confidential information in conformity with the CDR Confidentiality Guidelines. The CDEC recommendation or record of advice neither takes the place of a medical professional providing care to a particular patient nor is it intended to replace professional advice. CADTH is not legally responsible for any damages arising from the use or misuse of any information contained in or implied by the contents of this document. The statements, conclusions, and views expressed herein do not necessarily represent the view of Health Canada or any provincial, territorial, or federal government or the manufacturer. CDEC Meeting March 20, 2013 Page 5 of 5