Summary

• Premature discontinuation of any oral anticoagulant, including XARELTO, in the absence of adequate alternative anticoagulation increases the risk of thrombotic events. An increased rate of stroke was observed during the transition from XARELTOto warfarin in clinical trials in atrial fibrillation (AF) patients. If XARELTOis discontinued for a reason other than pathological bleeding or completion of a course of therapy, consider coverage with another anticoagulant.¹
• The protocol for ROCKET AF did not stipulate anticoagulation after study drug discontinuation, but warfarin patients who completed the study were generally maintained on warfarin. XARELTOpatients were generally switched to warfarin without a period of coadministration of warfarin and XARELTO, so that they were not adequately anticoagulated after stopping XARELTOuntil attaining a therapeutic international normalized ratio (INR).¹
• After the end of the study when patients discontinued blinded study drug and were transitioned to open-label vitamin K antagonist (VKA) therapy, there were significantly more strokes/non-central nervous system (CNS) embolism events in the XARELTO group compared to the warfarin group (22 vs 6; P=0.0044).² Patients initially in the XARELTO group took longer to reach therapeutic INR than those in the warfarin group (median time to reach therapeutic INR: XARELTO 13 days vs warfarin 3 days).^3,4
• An analysis of the ROCKET AF study by Mahaffey et al (2013) supports the hypothesis that the excess number of events was likely attributable to a period of inadequate anticoagulation in XARELTO patients transitioned to open-label VKA therapy at the end of the study. Although no evidence of rebound phenomenon was observed, it cannot be entirely excluded.²
• Additional studies related to real-world evidence are summarized below.^5-8
• Additional citations identified during a literature search are included in the REFERENCES section for your review.^9-13

CLINICAL STUDIES

Stroke Prevention in Nonvalvular Atrial Fibrillation (NVAF)

The ROCKET AF (XARELTO, Once-daily, oral, direct Factor Xa inhibition Compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation) trial¹⁴ was a phase 3, randomized, double-blind, double-dummy, activecontrolled, parallel-group, multicenter, event-driven, noninferiority study to evaluate the efficacy and safety of oral fixed-dose XARELTO 20 mg once daily (15 mg for patients with creatinine clearance [CrCl] 30-49 mL/min) and dose-adjusted warfarin (target INR: 2.0-3.0) for the prevention of stroke and systemic embolism in patients with NVAF at moderate-to-high risk for stroke.

• The primary efficacy endpoint was the composite of stroke and non-CNS systemic embolism. The primary safety endpoint was the composite of major and nonmajor clinically relevant bleeding.¹⁴
• The study consisted of a screening period, a double-blind treatment period, and a 30day posttreatment observation period (See Figure: Study Design of ROCKET AF).¹⁵
• Patient visits were scheduled for weeks 1, 2, and 4 after randomization and for every month thereafter. A standardized questionnaire and examination were used to screen for stroke symptoms and potential clinical events during follow-up.¹⁵
• When the requisite number of events was accrued, site investigators were notified to perform an end of study (EOS) visit, during which information was collected and participants were transitioned to open-label VKA.^2,15
• An excess of thromboembolic events occurred during the 30 days after the EOS visit when patients were transitioned from blinded therapy to open-label VKA.¹⁴ Patients originally in the XARELTO group took longer to reach therapeutic INR than those in the warfarin group (median time to reach therapeutic INR: XARELTO 13 days vs warfarin 3 days).³
• Based on this information, the prescribing information for XARELTO includes a black box warning cautioning healthcare providers that discontinuing XARELTO prematurely could increase the risk of thrombotic events.^1,2

Events Occurring After Treatment Discontinuation

Mahaffey et al (2013)² evaluated the excess of thromboembolic events during the 30 days after the EOS when patients were transitioned from blinded study therapy to openlabel VKA therapy.

• The ROCKET AF protocol did not specify a prescribed algorithm for EOS transition to open-label VKA therapy. This allowed physicians to manage patients according to local standards. However, site investigators were advised about the appropriate transition to open-label VKA via webinars, frequently asked question documents, and a clinical helpline. If open-label VKA was indicated, investigators were instructed to start therapy on the day of the EOS visit and to discontinue study medication. To maintain the study blind, an unblinded INR was collected on the third day after the EOS visit.
• At EOS, 92.2% of patients still taking study drug were transitioned to open-label VKA therapy within 30 days (XARELTO: 4,232/4,591 [92.2%]; warfarin: 4,292/4,657 [92.2%]). A total of 92% of patients were transitioned to VKA therapy on the same day as the EOS visit or the day after.
• Bridging therapy with low molecular weight or unfractionated heparin was not used frequently (XARELTO: 83/4,591 [1.8%]; warfarin: 56/4,657 [1.2%]).
• For days 3-30, there were more strokes in the XARELTO group compared to the warfarin group (XARELTO 22 events, event rate 6.42; warfarin 6 events, event rate 1.73; hazard ratio [HR], 3.72; confidence interval [CI], 1.51-9.16; P=0.0044). See Figure: First Primary Event During Transition to Open-Label Therapy for Patients Completing the Study.
• Efficacy and safety results are presented in Table: Efficacy and Safety Endpoints from Day 3 to 30 After Last Dose of Study Medication.
• In patients with documented INR values, 83% and 52% of the warfarin and XARELTO group, respectively, had 1 or more INR values ≥2.0 during the first 30 days after the last dose of study drug (P<0.0001). Sixty-three percent (1,748/2,775) of warfarin patients and 33.0% (928/2,814) of XARELTO patients had 2 or more INR value ≥2 (P<0.0001) measurements within 30 days of study drug discontinuation. Regions that reported more INR assessments had numerically fewer stroke or systemic embolic events.
• These analyses support the hypothesis that the excess number of events (XARELTO 22 vs warfarin 6) may be attributable to a period of inadequate anticoagulation of XARELTO patients transitioned to VKA therapy. The reason for the period of anticoagulation is not clear. The protocol did not stipulate a prescribed algorithm for transition, but allowed physicians to manage patients using local standards. However, investigators were advised through use of study memos, flyers, and Webcasts about appropriate transition to open-label VKA.
• Although no evidence of rebound phenomenon was observed, it cannot be entirely excluded.

Treatment of Deep Vein Thrombosis (DVT), Pulmonary Embolism (PE), and Reduction in the Risk of Recurrence of DVT and of PE

Prins et al (2013)¹⁶ compared the incidence of recurrent venous thromboembolism (VTE) in patients who continued anticoagulant treatment compared with those who stopped over the course of the 1-month poststudy medication observational period. Patients were included if they entered the 1-month poststudy medication observational period in the EINSTEIN-DVT and EINSTEIN-PE studies. After discontinuing study medication, continuation of anticoagulant treatment was at the discretion of the treating physician. Anticoagulant therapy given during the 1-month poststudy observational period were validated as full anticoagulant treatment or no anticoagulant treatment. Patients randomized to XARELTO therapy were divided into those who received bridging low molecular weight heparin therapy and those who did not.

Full anticoagulant treatment was received by 24.2% (861/3,565) and 40.2% (1,422/3,536) patients in the XARELTO and VKA therapy groups, respectively (P>0.0001). During the 1-month poststudy medication observational period, 0.9% (32/3,587) and 0.6% (22/3,582) patients developed a recurrent VTE in the XARELTO and enoxaparin/VKA groups, respectively. The incidence of recurrent VTE was 0.4% (9/2,283) and 0.9% (45/4,818) among patients who received any full anticoagulant treatment and patients who did not, respectively (risk difference -0.5%, 95% CI, 0.8% to 0.1%). The incidence of recurrent VTE in patients who had received XARELTO and continued with VKA treatment was similar among those who did not receive low molecular weight heparin bridging (0.4%; 2/498) and those who did receive low molecular weight heparin bridging (0.3%; 1/363) (risk difference 0.1%, 95% CI, -0.6% to 0.6%).

REAL-WORLD EVIDENCE

Álvaro et al (2024)⁵ conducted a nested case-control analysis of data from the Spanish database, BIFAP, to estimate the association between discontinuation of direct oral anticoagulants (DOACs; including XARELTO) and risk of ischemic stroke (IS) among patients with NVAF.

• Patients were followed from the date of the first DOAC prescription up to IS (index date).
  • IS was defined as having an entry code of IS or unspecified stroke in the absence of any code suggesting hemorrhagic stroke within ±90 days.
  • Exposure to DOACs was determined from the date of the first DOAC prescription or start of the follow-up to the index date. According to the most recent episode of DOAC use before the index date, exposure of DOAC was classified as:
    • Current use: When the prescription supply lasted until/over the index date or ended within the past 7 days.
    • Past use: When the prescription supply ended 31-180 days before the index date.
• A total of 59,020 patients were included in the cohort of new DOAC users with NVAF. XARELTO was the most frequently prescribed DOAC, with 32.1% current users in the control group.
• The risk of IS associated with past users compared with current users for XARELTO was 2.13 (95% CI, 1.34-3.39).

Mehmood et al (2023)⁶ conducted a retrospective cohort study to evaluate the safety and efficacy of XARELTO in pediatric patients with VTE in Pakistan.

• A total of 45 patients with a median age of 16 (3-18) years were included.
• XARELTO was used as the primary anticoagulant in 14 patients, and the median duration of XARELTO therapy was 8 (7-201) days.
• After discontinuation of XARELTO, recurrence of VTE occurred in 4.4% (2/45) of patients.

Hu et al (2022)⁷ conducted a network meta-analysis of 10 observational cohort studies to compare the safety and efficacy of resuming DOACs (dabigatran, XARELTO, apixaban, or edoxaban) or warfarin in patients with AF and prior gastrointestinal bleeding (GIB).

• A total of 59,244 patients with AF and prior GIB were included; among these patients, 27,793 resumed DOACs, 24,635 resumed warfarin, and 6,816 did not resume anticoagulation.
• Compared with patients who did not resume anticoagulation, in patients who resumed DOACs, only XARELTO was associated with an increased risk of recurrent GIB (pooled HR, 1.67; 95% credible interval, 1.16-2.65).
• Compared with patients who resumed dabigatran, those who resumed XARELTO were associated with an increased risk of recurrent GIB (pooled HR, 1.88; 95% credible interval, 1.06-3.75).

Khorana et al (2017)⁸ conducted a retrospective study of claims data in the Truven Health Analytics MarketScan Databases dated February 2011 to April 2015 to assess the risk for recurrent VTE and major bleeding associated with extended XARELTO treatment in a clinical practice setting in patients with VTE who continued and discontinued treatment. All patients were aged > 18 years and started XARELTO treatment within 7 days after a first VTE (index VTE). Available data was required from at least 12 months before the index date (the end of the initial treatment period, 3 or 6 months) and patients had to be anticoagulant naïve prior to the index VTE. A diagnosis of AF, a major bleeding event during the 12-month baseline period, or a recurrent VTE between the index VTE and the index date was cause for exclusion. The study population was categorized into discontinued cohort (treatment ended within the 30-day window after the index date) and continued cohort (continued treatment ended after the 30 day window).

• To assess the relationship between the cohorts, inverse probability of treatment weight (IPTW) was used based on a propensity score method and for first recurrent VTE and time to first major bleeding IPTW-adjusted Kaplan-Meier rates were utilized. Patients in the 3-month treatment group had significantly lower rates of recurrent VTEs in the continued cohort (n=5,933) vs the discontinued cohort (n=1,536) at 3 (0.70% vs 1.70% P=0.003), 6 (1.41% vs 2.34% P=0.016), 9 (1.82% vs 3.01% P=0.012), and 12 (1.97% vs 3.01% P=0.017) months of follow-up. The 6-month treatment group saw similar results in the continued cohort (n=2,676) vs the discontinued cohort (n=1,127) at 3 (0.82% vs 1.41% P=0.133), 6 (1.22% vs 2.96% P=0.023), 9 (1.35% vs 3.02% P=0.023), and 12 (1.72% vs 3.70% P=0.024) months of follow-up. The difference in the cumulative event rates for major bleeding events in the 3-month treatment group was not statistically significant. The continued and the discontinued cohorts at 3, 6, 9, and 12 months, respectively, were 0.58% and 0.82% (P=0.367), 0.91% and 0.88% (P=0.754), 1.33% and 1.18% (P=0.870), and 1.44% and 1.44% (P=0.813). The results of the 6-month treatment group, the continued and the discontinued, at 3, 6, 9, and 12 months, respectively, were 0.38% and 0.37% (P=0.903), 0.72% and 0.80% (P=0.854), 0.82% and 1.32% (P=0.544), and 1.53% and 1.32% (P=0.794). These results were consistent to those in the 3 month treatment group.

LITERATURE SEARCH

A literature search of MEDLINE^®, EMBASE^®, BIOSIS Previews^®, DERWENT^® (and/or other resources, including internal/external databases) was conducted on 04 March 2024.