A Prospective, Randomized, Double-Blind, Double-Dummy, Parallel-Group, Multicenter, Event-Driven, Non-inferiority Study Comparing the Efficacy and Safety of Once Daily Oral Rivaroxaban (BAY 59-7939) With Adjusted-Dose Oral Warfarin for the Prevention of Stroke and Non-Central Nervous System Systemic Embolism in Subjects With Non-Valvular Atrial Fibrillation (39039039AFL3001)

Rivaroxaban is a potent and highly selective oral direct factor Xa (FXa)
inhibitor. Activation of factor X to FXa plays a central role in the cascade of
blood coagulation. Therefore, its selective inhibition by rivaroxaban should
inhibit thrombin generation and may result in a better efficacy and safety
profile than available anticoagulants.

The primary objective of this study is to demonstrate that the efficacy of
rivaroxaban, a direct FXa inhibitor, is non-inferior to that of dose-adjusted
warfarin for the prevention of thromboembolic events in subjects with
non-valvular atrial fibrillation as measured by the composite of stroke and
non-central nervous system (non-CNS) systemic embolism.
Hypothesis: Rivaroxaban is non-inferior to warfarin in the prevention of the
composite endpoint of stroke and non-CNS systemic embolism in subjects with
non-valvular atrial fibrillation.
The principal safety objective of this study is to demonstrate that rivaroxaban
is superior to dose adjusted warfarin as assessed by the composite of major and
non-major clinically relevant bleeding events.
The major secondary efficacy objectives are to compare the effects of
rivaroxaban and warfarin with respect to the composite of stroke, non-CNS
systemic embolism, and vascular death and the composite of stroke, non-CNS
systemic embolism, myocardial infarction, and vascular death.
Other secondary efficacy objectives are to compare the effects of rivaroxaban
and warfarin with respect to the individual components of the composite primary
and major secondary endpoints, disabling stroke (modified Rankin Scale score of
3 to 5 inclusive), and all-cause mortality.
Other safety objectives are to compare rivaroxaban and warfarin with respect to
the individual bleeding event categories, adverse events, and clinical
laboratory evaluations (including liver function tests [LFTs]).
Exploratory objectives include pharmacokinetic (PK), pharmacodynamic (PD),
pharmacogenomic, risk marker/proteomics, treatment satisfaction and health care
resource utilization (HCRU) evaluations.

This is a prospective, randomized, double-blind, double-dummy, parallel-group,
active-controlled, multicenter, event-driven study comparing the efficacy and
safety of rivaroxaban with warfarin for the prevention of stroke and non-CNS
systemic embolism in subjects with non-valvular atrial fibrillation. An
independent blinded Clinical Enpoint Committee (CEC) will apply the protocol
definitions and adjudicate and classify the following endpoints: stroke,
non-CNS systemic embolism, death, myocardial infarction, transient ischemic
attack (TIA), major bleeding event and non major clinically relevant bleeding
event.
The study will be divided into a screening period, a double-blind treatment
period closing with an end*of study visit, and a posttreatment observation
period. At the end-of-study visit or at an early study medication
discontinuation visit for premature discontinuation of study therapy, subjects
will be transitioned from study drug to an open-label vitamin K antagonist
(VKA) or other appropriate therapy. At the end of the posttreatment observation
period, a follow-up visit will occur. The duration of the treatment period for
a given subject will depend on the time required to accrue 405 adjudicated
primary efficacy endpoint events, i.e., stroke, non-CNS systemic embolism, in
the per protocol population. As a result, the time on study drug will vary from
subject to subject depending upon the time of the subject*s enrollment. The
expected maximum duration of the study is 32 months, but may extend to a
maximum of 4 years depending upon the rate of subject recruitment and endpoint
event rates. Approximately 14,000 subjects are expected to enroll in this
study. However, the sample size may be increased to a maximum of 16,000
subjects if needed.
Subjects with non valvular atrial fibrillation, who are at risk for stroke and
non CNS systemic embolism will be randomized into the study using an
Interactive Voice Response System (IVRS). Eligible subjects are those with a
prior stroke, TIA or non CNS systemic embolism or who have 2 or more of the
following risk factors: age >_75 years, hypertension, heart failure and/or left
ventricular ejection fraction <_35% or diabetes mellitus. Once all of the
inclusion criteria and none of the exclusion criteria have been met, subjects
will be randomly assigned to 1 of 2 treatment groups: rivaroxaban or warfarin.
Those subjects assigned to rivaroxaban will receive rivaroxaban 20 mg p.o. once
daily plus warfarin placebo p.o. once daily titrated to a target sham
International Normalized Ratio (INR) of 2.5 (range 2.0 to 3.0, inclusive).
Subjects with moderate renal impairment at screening (defined as calculated
creatinine clearance [CLCR] between 30 and 49 mL/min, inclusive) will have a
dose adaptation to rivaroxaban 15 mg p.o. once daily. Those subjects assigned
to warfarin will receive warfarin p.o. once daily titrated to a target INR of
2.5 (range 2.0 to 3.0, inclusive) plus rivaroxaban placebo p.o. once daily.
Sham INRs will be generated according to an algorithm reflecting the
distribution of INR values from a population of warfarin-treated subjects
similar to the study population, and periodically updated in the IVRS based on
the distribution of INR values from study subjects assigned to the warfarin
treatment. The sham INRs will be based on real subject data that take previous
warfarin doses, age, and sex into account.
To maintain the blind, warfarin and its matching placebo will be dose adjusted
based on either real or sham INR results, respectively. To accomplish this,
study sites will be provided with a specially designed point-of-care INR device
that displays a code number instead of the actual INR value. This code number
is entered into the IVRS along with the subject*s study identification number.
The IVRS will decode the INR code number and then issue a standardized report
containing the actual INR value if the subject is assigned to receive warfarin
or a sham value if the subject is assigned to receive rivaroxaban. The INR
provided by the IVRS must be used by the investigator to adjust the
warfarin/warfarin placebo dose in all subjects from the time of randomization
until transition from study drug to an open-label VKA or other appropriate
therapy. While on study drug, unblinded INR measurements must not be performed
except in case of a medical emergency. It is especially important to limit site
personnel knowledge of any unblinded INR values to a minimum. Furthermore, it
is important to always use the point-of-care device provided to measure the INR
to ensure consistency of warfarin dosing. In addition, during the study, it is
recommended that INR monitoring (using the point-of-care device provided) occur
as clinically indicated but at least every 4 weeks.
The screening period will begin up to 30 days before randomization of the
subject into the study. At the screening visit, study personnel will explain
the nature of the study and obtain informed consent from the subject before the
initiation of any study-related procedures. Any adverse events that occur after
the signing of the informed consent will be recorded. Once a subject is
determined to be eligible for the study, the subject will be instructed to
discontinue their VKA (if applicable); in this case, daily unblinded INRs
should be performed at the treating institution (i.e., not using the
point-of-care device) and randomization of the subject should occur within 36
hours as the INR falls below 3.0. Once the subject*s eligibility for the study
has been reconfirmed, the subject will be randomized (Day 1) and study drug
will be dispensed. The first dose of study drug will be taken that evening with
food. A blood sample will be collected on the day of randomization (Day 1 is
preferred, any time after informed consent is obtained is acceptable) from
subjects consenting to the optional pharmacogenomic component of the study.
Subjects will return for visits at Week 1, 2, 4, and then every 4 weeks
thereafter for the duration of the double-blind treatment period. After Week 1,
all visits during the first year will be Full Visits. Double blind treatment
visits occurring after 1 year will take place every 4 weeks and either a Brief
Visit or a Full Visit will be performed according to the Visit Schedule
provided in the Time and Events Schedule. A 12-lead electrocardiogram (ECG) and
clinical laboratory tests will be performed annually and at the screening visit.
The frequency and timing of PK, PD, efficacy, safety, or other measurements are
provided in the Time and Events Schedules. All randomized subjects will be
followed until the study ends (405 adjudicated endpoint events reached followed
by study closure activities) even if they did not take study drug or
prematurely discontinued study drug. Every effort will be made to contact any
subjects lost to follow up and collect information on the occurrence of
efficacy endpoint events and the reason for discontinuation. When the
pre-specified number of adjudicated primary efficacy endpoint events has been
reached in the per protocol population, the sites will be notified by the
sponsor to schedule each subject still receiving blinded study medication for
an end-of study visit. This visit should be completed as soon as possible, but
within 30 days after site notification. At the end-of-study visit, subjects
will be transitioned from blinded study drug to an open label VKA or other
appropriate therapy and followed in the posttreatment observation period. The
posttreatment observation period ends with a follow-up visit, which will be
performed approximately 30 days (+/-5 days) after the end of study visit.
Subjects who have previously prematurely discontinued study drug will be
contacted for a final assessment of efficacy endpoint events within 30 days of
site notification.
An Executive Committee (EC) will be formed that has overall responsibility for
the conduct and reporting of the study. An independent Data Monitoring
Committee (DMC) will be commissioned for this study. This committee will
monitor the progress of the study and ensure that the safety of subjects is not
compromised. Any recommendations from the DMC will be made to the EC. An
independent blinded Clinical Endpoint Committee (CEC) will apply the protocol
definitions and adjudicate and classify the following endpoints: stroke, non
CNS systemic embolism, death, myocardial infarction, TIA, major bleeding event,
and non major clinically relevant bleeding event. Adjudicated results will be
used for final analyses.
An interim review of efficacy and safety data will be performed when 50% of the
primary efficacy events as reported by the investigators have occurred to
assess the option of stopping early for futility. All available data will be
used for the interim analysis.

Subjects will be randomly assigned to 1 of 2 treatment groups: rivaroxaban or
warfarin. Subjects assigned to rivaroxaban will receive rivaroxaban 20 mg p.o.
once daily plus warfarin placebo p.o. once daily titrated to a target sham INR
of 2.5 (range 2.0 to 3.0, inclusive). Subjects with moderate renal impairment
at screening (defined as calculated CLCR between 30 and 49 mL/min, inclusive)
will have a dose adaptation to rivaroxaban 15 mg p.o. once daily. Those
subjects assigned to warfarin will receive warfarin p.o. once daily titrated to
a target INR of 2.5 (range 2.0 to 3.0, inclusive) plus rivaroxaban placebo p.o.
once daily.

PHARMACOKINETIC/PHARMACODYNAMIC EVALUATIONS:
Sparse samples for PD evaluations will be collected in all subjects. The PD
measurements will include prothrombin time (PT), FXa activity, and
prothrombinase induced clotting time (PiCT). Matched PK and PD samples will be
collected at yet to be identified, selected sites that can provide the
logistical requirements necessary for rich sampling in a limited number of
subjects. Together this data will be used to describe drug exposure and
exposure/response relationships.
EFFICACY EVALUATIONS/CRITERIA:
The primary efficacy outcome is the composite of stroke and non CNS systemic
embolism. Major secondary efficacy endpoints include the composite of stroke,
non CNS systemic embolism, and vascular death; and the composite of stroke, non
CNS systemic embolism, myocardial infarction, and vascular death. Other
secondary efficacy endpoints include: individual components of the composite
primary and major secondary efficacy endpoints; disabling stroke; and all-cause
mortality.
SAFETY EVALUATIONS:
The principal safety endpoint is the composite of major and non-major
clinically relevant bleeding events. Safety will also be assessed by evaluation
of adverse events, including bleeding events, clinical laboratory tests
(including LFTs), ECGs, vital signs, and physical examinations.
HEALTH ECONOMIC EVALUATIONS:
The following HCRUs will be assessed: hospitalizations (total days length of
stay, ICU/CCU days), emergency room visits, unscheduled outpatient physician
consultations or visits related to bleeding, surgeries, other selected
procedures (inpatient and outpatient) and post stroke care status.
Anti-Clot Treatment Scale (ACTS): Subject satisfaction with treatment will be
measured with the ACTS in a subset of the population from the United States
(n=500), Germany (n=300) and Netherlands (n=300). The ACTS questionnaire will
be cross-validated against a general treatment satisfaction questionnaire
(Treatment Satisfaction Questionnaire for Medication [TSQM] version II) within
this subset.
RISK MARKERS/PROTEOMICS:
In a subset of 5,000 subjects at (yet to be identified) select sites, 2 blood
samples will be collected for assessment of risk markers (e.g., D dimer) and
proteomics.