Summary

• The ROCKET AF trial evaluated the efficacy and safety of XARELTO and warfarin for the prevention of stroke and systemic embolism (SE) in patients with nonvalvular atrial fibrillation (AF) at moderate-to-high risk for stroke.¹
  • XARELTO demonstrated noninferiority to warfarin for the time to first occurrence of stroke (any type) or non-central nervous system (CNS) SE (P<0.001), but superiority to warfarin was not demonstrated.²
  • A post hoc analysis of ROCKET AF found that XARELTO was associated with a favorable net clinical benefit (NCB) in 3 of 4 methods. Compared with warfarin, the overall NCB of XARELTO was owing to a decreased risk of ischemic events and a decreased risk of fatal or critical organ bleeding; however, XARELTO was associated with a higher risk of less clinically significant bleeding.³
  • A post hoc analysis of the ROCKET AF trial found that patients with persistent or paroxysmal AF treated with XARELTO had a similar number of stroke and SE compared with those treated with warfarin.⁴
• The J-ROCKET AF study compared the efficacy and safety of XARELTO with warfarin for the prevention of stroke in Japanese patients with nonvalvular AF.⁵
  • A subgroup analysis of J-ROCKET AF found a favorable trend of NCB of XARELTO compared with warfarin.⁶
  • Another subgroup analysis found the safety and efficacy of XARELTO and warfarin were similar, regardless of CHADS₂ (congestive heart failure, hypertension, age ≥75, diabetes, stroke [doubled], vascular disease, age 65-74, and sex category [female]) score.⁷
• Additional citations identified during a literature search have been included in the REFERENCES section for your review.^8-13

CLINICAL STUDIES

ROCKET AF

The ROCKET AF trial was a phase 3, randomized, double-blind, double-dummy, multi-center, event-driven, noninferiority study to evaluate the efficacy and safety of fixed-dose XARELTO and dose-adjusted warfarin for the prevention of stroke and SE in patients with nonvalvular AF at moderate-to-high risk for stroke.^1,14,15

Inclusion/Exclusion Criteria

Inclusion Criteria: Patients ≥18 years of age with documented nonvalvular AF at moderate-to-high risk of stroke. Elevated risk was indicated by history of prior stroke, transient ischemic attack (TIA), SE, or a CHADS₂ score >2.

The percentage of patients who did not have a previous stroke, TIA or SE and had no more than 2 risk factors was limited to 10%; the remainder of the patients were required to have a previous thromboembolism or ≥3 risk factors.

Exclusion Criteria: Patients with hemodynamically significant mitral stenosis; any valve prosthesis; planned cardioversion; atrial myxoma or left ventricular thrombus; active internal bleeding; any prior or current condition associated with an increased risk of bleeding; platelet count <90,000/μL; severe disabling stroke within 3 months or any stroke within 14 days of randomization; TIA within 3 days of randomization; anemia; creatinine clearance (CrCl) <30 mL/min; or those with significant hepatic impairment or alanine aminotransferase (ALT) >3x the upper limit of normal (ULN).

Patients that received the following treatments were excluded: aspirin (ASA) >100 mg per day; ASA in combination with thienopyridines or use of intravenous (IV) antiplatelets within 5 days of randomization; fibrinolytics within 10 days of randomization; long-term nonsteroidal anti-inflammatory drug (NSAID) use; or strong Cytochrome P450 3A4 inhibitors or inducers within 4 days of randomization or during the study.

The use of ASA (≤100 mg) monotherapy and thienopyridine monotherapy was allowed.

Study Design

The study consisted of a screening period, a double-blind treatment period, and a 30-day post-treatment observation period.¹⁴ See Figure: Study Design for ROCKET AF.

• Patients in each group were given a placebo tablet to maintain blinding. For patients receiving XARELTO, sham international normalized ratio INR results were generated to reflect the distribution of INR values from a population of patients using warfarin with characteristics similar to the study population.

The primary objective of the study was to demonstrate noninferiority of XARELTO to that of dose-adjusted warfarin in the per-protocol (PP) population during treatment (PP, as-treated population).

• The PP population included all patients who received >1 dose of study drug, did not have a major protocol violation, and were followed for events while receiving study drug or within 2 days after discontinuation.
• If non-inferiority was achieved, a superiority analysis was conducted in the safety population during treatment (safety, as-treated population), which included patients who received >1 dose of a study drug and were followed for events while they were receiving study drug or within 2 days after discontinuation, regardless of adherence to study protocol.
• An analysis for non-inferiority and a sensitivity test for superiority was also performed in the intentto-treat (ITT) population, which included all patients who underwent randomization and were followed for events during treatment or after premature discontinuation. Additionally, a post hoc analysis was performed in the ITT population, and events occurring during the end-of-study transition to open-label treatment with conventional anticoagulants.

Endpoints

Primary efficacy endpoint: composite of stroke (ischemic or hemorrhagic) and SE.

• Stroke: a sudden focal neurologic deficit with a cerebrovascular cause that lasted for ≥24 hours
• SE: an abrupt vascular insufficiency associated with clinical or radiological evidence of arterial occlusion in the absence of another likely mechanism (e.g., atherosclerosis, instrumentation, or trauma).

Secondary endpoints: composite of stroke, SE, vascular death, and myocardial infraction (MI); the composite of stroke, SE, and vascular death; individual components of the composite end points were also included.

Primary safety endpoint: composite of both major and nonmajor clinically relevant (NMCR) bleeding

• Major bleeding (MB): clinically overt bleeding that was associated with: fatal outcome, occurred at a critical site (intracranial, intraspinal, intraocular, pericardial, intra-articular, intramuscular with compartment syndrome, or retroperitoneal site); decrease of ≥2 g/dL in hemoglobin (Hb); required transfusion of ≥2 units of whole blood or packed red blood cells; or permanent disability.
• NMCR bleeding: overt bleeding not meeting the criteria for MB but requiring medical intervention, unscheduled contact with a physician, temporary interruption of study drug, pain, or impairment of daily living.

Sample Characteristics and Study Duration

Baseline characteristics were similar between treatment groups:

• The median age was 73 years, 60.3% of study participants were male, the mean CHADS₂ score was 3.5, and 62.4% of the patients had previously been treated with a vitamin K antagonist (VKA).
• At baseline, 54.8% of patients had a previous stroke, SE, or TIA; 62.5% had heart failure; 90.5% had hypertension; and 40% had diabetes.
• 87% of patients in each treatment group had a CHADS₂ score of ≥3.
• The mean baseline CrCl was 72.75 mL/min.¹⁶

The median duration of treatment exposure was 590 days; median follow-up period was 707 days. 32 patients were lost to follow-up. In the patients who were receiving warfarin, the mean time in therapeutic range (TTR) for INR values was 55%.

Efficacy Results

See Tables: Primary Efficacy Outcome: Composite of Stroke (Ischemic or Hemorrhagic) and SE and Secondary Efficacy Outcomes (Safety, As-treated Population).

• XARELTO was non-inferior to warfarin for prevention of stroke or SE in the PP as-treated population (P<0.001).
• In the safety as-treated population and during treatment in the ITT population, patients in the XARELTO group had lower event rates than those in the warfarin group (both with P=0.02 for superiority).
• In patients that ended treatment before the end of the prescribed period, 51.2% of XARELTO patients and 48.8% of warfarin patients were transitioned to open-label warfarin.

After sites were notified to end study treatment, 92% of patients in both groups still on the assigned study drug were transitioned to VKA.^1,17

• The median time to reach therapeutic INR was 13 days for those in the XARELTO group compared to 3 days for those in the warfarin group.
• Significantly more patients that transitioned from XARELTO than from warfarin developed primary events during the first month after discontinuing randomized treatment (22 vs. 6; P=0.008).

Safety Results

• Bleeding results are presented in the Table: Event Rate of Bleeding Outcomes.
• The event rates of the composite of major and NMCR bleeding events and of MB rates were comparable between XARELTO and warfarin.
• Fatal bleeding and bleeding involving a critical site were more common in warfarin-treated patients, while rates of bleeding associated with a decrease of ≥2 g/dL in Hb or bleeding that resulted in a transfusion of ≥2 units of blood were higher in XARELTO-treated patients.
• Intracranial hemorrhage (ICH) rates were significantly lower with XARELTO than with warfarin (P=0.02).
• MB from a gastrointestinal site was significantly greater with XARELTO compared to warfarin (P<0.001).

• Other safety outcomes were reported more frequently with XARELTO than with warfarin (epistaxis: 10.1% vs. 8.6%, respectively; hematuria: 4.2% vs. 3.4%, respectively; P<0.05)
• The rates of other treatment-emergent adverse events (AEs) were similar between the two groups.
• Increases in ALT levels >3 times the ULN and total bilirubin levels >2 times the ULN occurred in 0.47% and 0.50% of patients receiving XARELTO and warfarin, respectively.

TTR

• TTR is presented in Table: Treatment Effects by Quartiles of Center TTR (Safety Population).
• The mean TTR for warfarin was 55%.
• An analysis of quartiles for TTR found the event rates in the XARELTO arm was similar between study sites with good control of INR and those with poor control of INR. Regional differences and the high-risk patient population played a role in the TTR result.¹

Barnett et al (2018)³ conducted a post hoc analysis of the ROCKET AF trial to evaluate the the NCB of XARELTO compared with warfarin using different methods in the on-treatment population (N=14,236).

Results

• A total of 14,236 patients received at least 1 dose of the study drug (on-treatment population), of whom 7111 and 7125 received XARELTO and warfarin, respectively.
• In both the XARELTO and warfarin groups, the mean CHADS₂ score was 3.5 (standard deviation [SD], 0.9) and the median age was 73 years.

Efficacy Results

• The composite of all-cause death, stroke (ischemic or hemorrhagic), MI, or SE occurred in 471 (4.25 per 100 PY) and 575 (5.12 per 100 PY) patients who received XARELTO and warfarin (rate difference, -86.8 per 10,000 PY; 95% CI, -143.6 to -30.0), respectively.

Safety Results

• The composite of major bleeding occurred in 395 (3.60 per 100 PY) and 386 (3.45 per 100 PY) patients who received XARELTO and warfarin, respectively. Rates of major bleeding were not statistically significant (rate difference, 14.2 per 10,000 PY; 95% CI, -35.2 to 63.7).

NCB

• Method 1: NCB was defined as an unweighted composite of all-cause death, stroke, MI, SE, or MB.
  • XARELTO had an NCB of -35.5 per 10,000 PY (95% confidence interval [CI],
    -108.4 to 37.3).
• Method 2: NCB was defined as an unweighted composite of all-cause death, stroke, MI, SE, fatal bleeding, or critical organ bleeding.
  • Compared with warfarin, XARELTO had an NCB of -96.8 per 10,000 PY (95% CI, 157.0 to -36.8); therefore, approximately 97 AEs per 10,000 PY were expected to be prevented.
• Method 3: NCB was calculated using the following formula: (ischemic stroke/systemic embolism [SSE]_XARELTO - SSE_warfarin) + 1.5 × (ICH_XARELTO - ICH_warfarin).
  • Compared with warfarin, XARELTO had fewer stroke equivalents per 10,000 PY, with an NCB of -65.2 per 10,000 PY (95% CI, -112.3 to -17.8), see Table: NCB of XARELTO Based on Different Methods (1-3) (Ontreatment Population).
• Method 4: NCB was calculated using the following formula: NCB = Death × (death_XARELTO - death_warfarin) + SSE × (SSE_XARELTO - SSE_warfarin) + ICH × (ICH_XARELTO - ICH_warfarin) + MB × (MB_XARELTO - MB_warfarin). Each rate difference was multiplied by a weighting factor which reflects the relative impact of that event type in terms of disability and death.
  • Compared with warfarin, XARELTO had fewer death equivalents per 10,000 PY, with an NCB of -54.8 per 10,000 PY (95% CI, -96.0 to -10.2), see Table: NCB of XARELTO Based on Method 4 (On-treatment Population, N=14,236).
  • The use of smaller weights for event types other than death reduced the NCB to -44.5 per 10,000 PY, but this was still significant (95% CI, -82.0 to -4.8), whereas higher weights increased the NCB to -65.1 per 10,000 PY (95% CI, -112.2 to -15.8).
• The overall NCB of XARELTO was owing to a decreased risk of ischemic events and a decreased risk of fatal or critical organ bleeding; however, XARELTO was associated with a higher risk of less clinically significant bleeding.

Steinberg et al (2015)⁴ conducted a post hoc analysis of the ROCKET AF trial to compare outcomes between patients with persistent (n=11,548) and paroxysmal (n=2514) AF receiving oral anticoagulation. Patients with new-onset AF were excluded from the analysis.

The primary efficacy endpoint was stroke (ischemic or hemorrhagic) or SE in the ITT population. Half of patients with paroxysmal and persistent AF were randomized to XARELTO and half were randomized to warfarin.

Compared with warfarin, the rate of stroke or SE in patients treated with XARELTO was consistent among patients with paroxysmal AF (1.73% XARELTO vs 1.74% warfarin) and persistent AF (2.03 vs 2.32%; P_interaction = 0.60). The rate of MB events was consistent in patients treated with XARELTO compared to warfarin with paroxysmal AF (3.43% XARELTO vs. 3.19% warfarin) and persistent AF (3.61 vs. 3.49%; P_interaction = 0.94).

J-ROCKET AF

Hori et al (2012)⁵ conducted the J-ROCKET AF study to confirm non-inferiority of XARELTO versus warfarin with regard to the principal safety outcome (major or NMCR bleeding) for the prevention of stroke in patients in Japan with nonvalvular AF.

• Doses of XARELTO were optimized for the pharmacokinetic and pharmacodynamic profile of Japanese patients.
• Patients were randomized to receive 15 mg XARELTO once daily (or 10 mg in patients if CrCl 30-49 mL/min at study entry) or warfarin (target INR 2.0-3.0 for patients aged <70; 1.6-2.6 for those aged ≥70).

With the exception of age (patients >20 years of age were eligible), J-ROCKET had the same inclusion criteria as ROCKET AF. The exclusion criteria of J-ROCKET AF was the same as for ROCKET AF. Antiplatelet therapy was not allowed within 5 days of randomization or during the study. However, for patients who received a coronary stent, thienopyridines or cilostazol were allowed.

The primary efficacy endpoint was the composite of stroke and non-CNS SE. Secondary efficacy endpoints were the same as defined in ROCKET AF.

Efficacy Results:

• There were 1,274 patients in the PP population. Baseline demographics were similar among both treatment groups.
• Moderate renal impairment was present in 22.1% of patients randomized to receive XARELTO 10 mg once daily.
• The event rate of stroke and non-CNS SE in the PP population was significantly lower for XARELTO compared with warfarin (1.26% vs. 2.61%, respectively; hazard ratio [HR] 0.49, 95% CI 0.24-1.00; P=0.05).
• The event rate of stroke and non-CNS SE in the ITT population including 30-day follow-up was 2.38 per year for XARELTO compared with 2.91 per year for warfarin (HR=0.82; 95% CI 0.46, 1.45).
• Compared with warfarin, there were fewer ischemic stroke events in patients treated with XARELTO (warfarin, n=17; XARELTO, n=7; HR 0.40, 95% CI 0.17-0.96) and a lower rate of all-cause stroke (warfarin, n=21; XARELTO, n=10; HR 0.46; 95% CI 0.22-0.98).
• The occurrence of primary hemorrhagic stroke was similar between groups.

Safety Results:

• There were 1,278 patients included in the safety population (XARELTO = 639, warfarin = 639).
• The composite of major or NMCR bleeding was similar between XARELTO and warfarin groups (18.04% vs. 16.42%, respectively; HR 1.11; 95% CI 0.87-1.42).
• ICH occurred in 5 and 10 patients treated with XARELTO and warfarin, respectively.
• Treatment-emergent fatal bleeding occurred in 1 XARELTO-treated patient and 3 patients treated with warfarin.
• Principal safety outcome rates between groups were not significantly different in patients with moderate renal impairment and patients with mild or no renal impairment and baseline CrCl ≥ 50 mL/min (HR 1.07; 95% CI 0.80, 1.43; interaction P-value = 0.628).
• Treatment-emergent AEs, including drug-related AEs, occurred at a similar incidence between treatment groups.
  • Treatment-emergent serious AEs occurred in 23.6% of XARELTO patients and 24.3% of warfarin patients.
  • Epistaxis (XARELTO 16.3%; warfarin 9.4%) and subcutaneous hemorrhage (XARELTO 10.5%; warfarin 12.5%) were the most frequently reported treatment-emergent bleeding events.

Non-inferiority of XARELTO on the principal safety outcome versus warfarin was confirmed, and data was consistent with the global ROCKET AF trial results.

Uchiyama et al (2013)⁶ performed a subgroup analysis of J-ROCKET AF to evaluate two methods to quantify NCB of XARELTO versus warfarin in the safety population.

• Method 1: NCB was calculated as the difference between ischemic strokes avoided and the excess ICH cases, with a weight of 1.5 to account for the increased morbidity associated with ICH.
  • XARELTO was nominally significantly favorable over warfarin (difference in incidence rate, -2.13; 95% CI, -0.259 to -3.99).
  • Similar results were obtained without weighing and with a 2.0 weight in a sensitivity analysis.
• Method 2: NCB was calculated as the composite end point of MB events and secondary efficacy endpoints.
  • The event rate of the composite end point tended to be lower in patients treated with XARELTO compared with warfarin (difference in incidence rate, -1.14; 95% CI, -3.40 to 1.12).
  • Similar results were found in an analysis of cardiovascular death compared to a composite endpoint for all-cause death.
• An analysis by CHADS2 score or CrCl was also conducted, and demonstrated that XARELTO had a positive NCB trend in patients with a CHADS₂ score of > 2 and in patients with moderate renal insufficiency, respectively. There was no difference between XARELTO and warfarin.

Hori et al (2014)⁷ evaluated the safety and efficacy of XARELTO and warfarin in relation to CHADS₂ scores in a subanalysis of the J-ROCKET AF trial.

• The mean CHADS2 score was 3.25. See Table: NMCR Bleeding and MB Outcomes Stratified by CHADS₂ Scores.

• The P-value for interaction for the rate of major and NMCR bleeding between the XARELTO and warfarin groups was 0.700.
• Regardless of risk factor stratification, there were no further differences between XARELTO and warfarin groups (moderate-risk group HR, 1.06, 95% CI, 0.58-1.95; high-risk group HR, 1.11, 95% CI, 0.86-1.45, P=0.88).
• Event rates for stroke or non-CNS SE (primary efficacy endpoint) were lower in patients treated with XARELTO than warfarin (moderate-risk group; HR, 0.46, 95% CI, 0.09-2.37; high-risk group; HR, 0.49, 95% CI, 0.22-1.11; P=0.935).

LITERATURE SEARCH

A literature search of MEDLINE^® EMBASE^®, BIOSIS Previews^®, and DERWENT^® (and/or other resources, including internal/external databases) was conducted on 16 December 2023.