Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) Trial: Design and Rationale Renato D

Trial Design

Apixaban for Reduction In Stroke and Other ThromboemboLic Events in Atrial Fibrillation (ARISTOTLE) trial: Design and rationale Renato D. Lopes, MD, PhD, a John H. Alexander, MD, MHS, a Sana M. Al-Khatib, MD, MHS, a Jack Ansell, MD, b Raphael Diaz, MD, c J. Donald Easton, MD, d Bernard J. Gersh, MB, ChB, DPhil, e Christopher B. Granger, MD, a Michael Hanna, MD, f John Horowitz, MD, g Elaine M. Hylek, MD, MPH, h John J. V. McMurray, MD, i Freek W. A. Verheugt, MD, PhD, j and Lars Wallentin, MD, PhD k on behalf of the ARISTOTLE Investigators Durham, NC; New York, NY; Santa Fe, Argentina; Providence, RI; Rochester, MN; Princeton, NJ; Adelaide, Australia; Boston, MA; Glasgow, United Kingdom; Nijmegen, The Netherlands; and Uppsala, Sweden

Atrial fibrillation (AF) is associated with increased risk of stroke that can be attenuated with vitamin K antagonists (VKAs). Vitamin K antagonist use is limited, in part, by the high incidence of complications when patients' international normalized ratios (INRs) deviate from the target range. The primary objective of ARISTOTLE is to determine if the factor Xa inhibitor, apixaban, is noninferior to warfarin at reducing the combined endpoint of stroke (ischemic or hemorrhagic) and systemic embolism in patients with AF and at least 1 additional risk factor for stroke. We have randomized 18,206 patients from over 1,000 centers in 40 countries. Patients were randomly assigned in a 1:1 ratio to receive apixaban or warfarin using a double-blind, double-dummy design. International normalized ratios are monitored and warfarin (or placebo) is adjusted aiming for a target INR range of 2 to 3 using a blinded, encrypted point-of-care device. Minimum treatment is 12 months, and maximum expected exposure is 4 years. Time to accrual of at least 448 primary efficacy events will determine treatment duration. The key secondary objectives are to determine if apixaban is superior to warfarin for the combined endpoint of stroke (ischemic or hemorrhagic) and systemic embolism, and for all-cause death. These will be tested after the primary objective using a closed test procedure. The noninferiority boundary is 1.38; apixaban will be declared noninferior if the 95% CI excludes the possibility that the primary outcome rate with apixaban is N1.38 times higher than with warfarin. ARISTOTLE will determine whether apixaban is noninferior or superior to warfarin in preventing stroke and systemic embolism; whether apixaban has particular benefits in the warfarin-naïve population; whether it reduces the combined rate of stroke, systemic embolism, and death; and whether it impacts bleeding. (Am Heart J 2010;159:331-9.)

Atrial fibrillation (AF) is a growing public health be accompanied by substantial morbidity and mortality.2,3 problem worldwide and is the most common arrhythmia Stroke is considered to be the most significant morbidity requiring hospitalization in the United States.1 The in patients with AF. Approximately 15% of strokes occur incidence of AF has increased over the past 2 to 3 in those with AF and the risk of stroke in untreated AF decades. The number of people with AF in the United patients averages 5% per year.3,4 The annual risk of stroke States is projected to exceed 10 million by 20502; this will in AF patients is related to age, increasing from 1.5% for patients aged 50 to 59 years to 23% for those aged 80 to 89 years.5 When compared with stroke from other

From the aDuke Clinical Research Institute, Duke University Medical Center, Durham, NC, causes, ischemic stroke secondary to AF carries twice the 5 bLenox Hill Hospital, New York, NY, cEstudios Cardiologicos Latinoamerica, Santa Fe, risk of death. Argentina, dRhode Island Hospital, Alpert Medical School, Brown University, Providence, RI, eMayo Clinic College of Medicine, Rochester, MN, fBristol-Myers Squibb, Princeton, NJ, gUniversity of Adelaide, Adelaide, Australia, hBoston University Medical Center, Boston, MA, iWestern Infirmary, Glasgow, United Kingdom, jUniversity Medical Center, Nijmegen, The Warfarin Netherlands, and kUppsala Clinical Research Center, Uppsala University, Uppsala, Sweden. Warfarin, a vitamin K antagonist (VKA), reduces the RCT reg no. NCT00412984. risk of stroke by approximately 62%6; however, VKAs Submitted February 6, 2009; accepted July 7, 2009. Reprint requests: Christopher B. Granger, MD, Duke Clinical Research Institute, Duke have major limitations and are underused in clinical 7 University Medical Center, 2400 Pratt Street, Room 0311 Terrace Level, Durham, NC 27707. practice. Patients are frequently outside the optimal E-mail: [email protected] target anticoagulation range when VKAs are prescribed, 0002-8703/$ - see front matter 8,9 © 2010, Mosby, Inc. All rights reserved. exposing them to the risk of thrombosis or bleeding. doi:10.1016/j.ahj.2009.07.035 Clinical trial data show that among patients receiving American Heart Journal Lopes et al 332 March 2010

warfarin or other VKAs, even in the highly structured greater convenience of use, and/or greater safety. When setting of a trial, the international normalized ratio (INR) compared with warfarin in the SPORTIF III and V trials, is outside the therapeutic range (2.0-3.0) about one third the oral direct thrombin inhibitor ximelagatran was of the time.10-13 Moreover, a meta-analysis shows that the found to be effective with an acceptable bleeding proportion of time outside the therapeutic INR range in risk,10,12 but its development was stopped due to liver community practice is 43%.11 Thus, a large number of toxicity. A number of new anticoagulants, including patients with AF who should be on warfarin are either not factor Xa inhibitors and direct thrombin inhibitors, are deriving full benefit from warfarin or are not receiving it now being developed and evaluated as alternatives to at all. Not only do VKAs have a narrow therapeutic warfarin for stroke prevention in AF. The ideal agent window but they also exhibit a highly variable dose should be oral and free from variation in absorption response that is attributable to genetic, disease-related, related to food intake and should have predictable and environmental factors. Their dosing can be particu- pharmacokinetics, few drug interactions, and minimal larly challenging among elderly individuals due to toxicity. It would not require therapeutic monitoring changes in the pharmacokinetics and pharmacodynamics and its efficacy and safety would be comparable with or that occur with age. Other factors influencing drug effect superior to warfarin at preventing thromboembolism in include prescription and nonprescription drugs, dietary AF subjects. vitamin K, and botanical products.14,15 The need for regular and lifelong therapeutic monitoring is an incon- venience for many patients. These limitations of VKAs Apixaban illustrate the need for new oral anticoagulants for the Factor Xa occupies a pivotal role in the clotting prevention of thromboembolism in patients with AF. cascade because it promotes conversion of prothrom- Current guidelines, including those from the American bin to thrombin. Developed as an anticoagulant agent, College of Cardiology/American Heart Association/Euro- apixaban is an orally active selective inhibitor of the pean Society of Cardiology, recommend a risk-based coagulation factor Xa. The direct mechanism of this approach for antithrombotic therapy for stroke preven- drug does not require the presence of antithrombin. It tion.16 Aspirin or VKAs are recommended for patients has a predominantly nonrenal (75%) clearance and a with AF who have one moderate risk factor for stroke half-life around 12 hours in healthy volunteers. The (CHADS2 risk score of 1), and VKAs are recommended for effect is independent of vitamin K intake, and with the patients with more than one moderate risk factor exception of strong CYP3A4 inhibitors, there is minimal 16 (CHADS2 risk score ≥2) for stroke. Vitamin K potential for drug-to-drug interaction. This compound antagonists are efficacious at preventing stroke even in has shown efficacy in preclinical animal models for the lower risk AF population17; however, their routine venous and arterial thrombosis and has no organ- use is not currently recommended because the overall net specific toxicity in animal models of up to 12 months benefit in this low-risk group is uncertain. Because the of exposure.20,21 oral factor Xa inhibitors avoid many of the limitations of In a phase II dose-ranging study of deep vein VKAs, it is important to evaluate these new agents across thrombosis (DVT) prevention in patients undergoing the spectrum of risk including patients with a low to knee replacement surgery, pooling of all doses of intermediate risk of stroke. apixaban showed a 21% reduction in venous thrombo- Another important issue is that in patients starting embolism (VTE) or all-cause death when compared with warfarin for the first time, the risk of thromboembolism enoxaparin and a 53% reduction in VTE or all-cause death and bleeding is high for the first year due in part to the when compared with warfarin.22 Importantly, the time taken to establish an adequate and stable INR. In frequency of major bleeding events was low (0%-3.3%) one observational study, patients starting warfarin had a and comparable among all apixaban arms. A dose-ranging 3-fold increased risk of bleeding in the first 90 days of trial in DVT treatment including 520 patients with treatment7 compared with patients already on warfarin.7 proximal DVT also showed favorable efficacy and safety, The ACTIVE-W study showed that, in patients with AF at including the 5 mg twice a day apixaban dose arm.23 high risk of stroke, warfarin was superior to clopidogrel The ADVANCE-1 study, a phase III VTE prevention trial, plus aspirin for prevention of vascular events.18 compared apixaban, 2.5 mg, twice daily with the Food Additional analyses also showed that the rates of and Drug Administration-approved dose of enoxaparin discontinuation of warfarin therapy during the first (30 mg twice daily).24 The primary efficacy outcome was year of the trial were higher (15%) in the warfarin-naïve a composite of symptomatic or asymptomatic DVT, patients when compared with the warfarin-experienced pulmonary embolism, and death from any cause. In a patients (8%).19 preliminary analysis, the rate of the primary efficacy In antithrombotic therapy for the prevention of endpoint in the apixaban arm was similar to that thromboembolism in patients with AF, there is a need observed with enoxaparin (9.0% vs 8.9%, P = .64). The for an alternative to warfarin that has greater efficacy, major bleeding rate for apixaban, however, tended to be American Heart Journal Lopes et al Volume 159, Number 3 333

lower than for enoxaparin (0.7% vs 1.4%, P = .053), and Table I. Inclusion and exclusion criteria the composite rate of clinically relevant nonmajor Inclusion criteria bleeding and major bleeding was significantly lower in Age ≥18 y patients assigned to apixaban than those assigned to Permanent or persistent AF or atrial flutter on ECG at enrollment; or AF enoxaparin (2.9% vs 4.3%, P = .034). or atrial flutter documented by ECG or as an episode ≥1 min on APPRAISE was a phase II, randomized, double-blind, rhythm strip, Holter monitor, or intracardiac recording on 2 separate occasions at least 2 wk apart in 12 mo before enrollment placebo-controlled, parallel arm study that enrolled 1,715 One or more of the following risk factors for stroke patients with acute coronary syndrome (ACS) from Age ≥75 y 25 Europe and North America. This study demonstrated Prior stroke, TIA, or systemic embolus that the addition of apixaban to current antiplatelet Symptomatic CHF within 3 mo or LV dysfunction with LVEF ≤40% by therapy for 6 months after ST elevation or non-ST echocardiography, radionuclide study, or contrast angiography Diabetes mellitus elevation ACS results in a dose-dependent increase in Hypertension requiring pharmacologic treatment bleeding and a trend toward a reduction in ischemic Women of childbearing potential must use contraception to avoid events. Based on these data, apixaban at a total daily dose pregnancy during treatment period or for 2 wk after last dose of of between 5 and 10 mg is being tested in patients with study medication, whichever is longer ACS receiving aspirin or dual antiplatelet therapy in the All subjects must provide signed written informed consent phase III trial APPRAISE 2. Exclusion criteria These findings indicate that oral factor Xa inhibition at AF or atrial flutter due to reversible causes (eg, thyrotoxicosis, pericarditis) a dose of 5 to 10 mg daily may be an effective and safe Clinically significant (moderate or severe) mitral stenosis approach toward anticoagulation. Based on the experi- Increased bleeding risk believed to be a contraindication to oral ence in DVT prevention and treatment trials, a dose of 5 anticoagulation (eg, previous intracranial hemorrhage) Conditions other than AF that require chronic anticoagulation mg twice a day of apixaban was chosen for comparison (eg, prosthetic mechanical heart valve) with warfarin for prevention of stroke and systemic Persistent uncontrolled hypertension (SBP N180 mm Hg or DBP N100 mm Hg) embolism in patients with AF in the ARISTOTLE trial. Active infective endocarditis Planned major surgery Planned AF or atrial flutter ablation procedure Study objectives Use of unapproved investigational drug or device in past 30 d Required aspirin N165 mg/d Primary objective Simultaneous treatment with both aspirin and a thienopyridine The primary objective of this study is to determine (eg, clopidogrel, ticlopidine) whether apixaban is noninferior to warfarin (INR target Severe comorbid condition with life expectancy ≤1y range 2.0-3.0) at reducing the combined outcome of Active alcohol or drug abuse or psychosocial reasons that make study participation impractical stroke (ischemic or hemorrhagic) and systemic embo- Recent stroke (within 7 d) lism in subjects with AF and at least 1 additional risk Severe renal insufficiency (serum creatinine level N2.5 mg/dL or calculated factor for stroke. creatinine clearance b25 mL/min) ALT or AST N2 × ULN or a total bilirubin ≥1.5 × ULN (unless an alternative causative factor [eg, Gilbert's syndrome] is identified) Secondary objectives 3 Platelet count ≤100,000/mm The key secondary objectives are to determine if Hemoglobin level b9 g/dL apixaban is superior to warfarin for the combined endpoint Inability to comply with INR monitoring of stroke (ischemic or hemorrhagic) and systemic embo- ECG, Electrocardiogram; CHF, congestive heart failure; LV, left ventricular; LVEF, left lism, and for all-cause death. These will be tested after the ventricular ejection fraction; SBP, systolic blood pressure; DBP, diastolic blood pressure; primary objective using a closed test procedure. ALT, alanine transaminase; AST, aspartate transaminase; ULN, upper limit of normal; wk, weeks; mo, months. Other secondary objectives are to compare apixaban and warfarin with respect to: • The composite endpoint of ischemic stroke, hem- • The composite endpoint of ischemic stroke, orrhagic stroke, systemic embolism, myocardial hemorrhagic stroke, systemic embolism, and all- infarction, and all-cause death, cause death, • Major bleeding • The composite endpoint (in warfarin-naïve patients) of ischemic stroke, hemorrhagic stroke, systemic embolism and major bleeding, Study population • The composite endpoint of ischemic stroke, In this double-blind study, we have randomized 18,206 hemorrhagic stroke, systemic embolism, and patients with AF from over 1,000 centers in about 40 major bleeding, countries. Eligible subjects were randomly assigned in a • The composite endpoint of ischemic stroke, 1:1 ratio to receive either apixaban or warfarin titrated to a hemorrhagic stroke, systemic embolism, major target INR range of 2.0 to 3.0. Both warfarin-naïve and bleeding, and all-cause death, warfarin-experienced patients are being recruited, with a American Heart Journal Lopes et al 334 March 2010

special emphasis on the former. Our aim is for 40% of INR monitoring patients to be warfarin naïve. The inclusion and exclusion The INR monitoring for the warfarin-naïve patients criteria are summarized in Table I. begins on the fourth day after initiation of study drug and is performed twice a week for 2 weeks, once a week for 2 weeks, and monthly thereafter once a stable INR is Randomization and study drug attained. For the warfarin-experienced patients who Subjects who were on warfarin before randomization have been on stable dosing of VKA for at least 3 months, discontinued the drug 72 hours before randomization and INR monitoring visits are required on day 1, week 1, were not dosed with study drug until the INR was b2.0. week 2, and then monthly. An investigator may increase Randomization is stratified by investigative site and prior the frequency of INR monitoring if it is considered warfarin use status (experienced or naïve). Patients are clinically indicated. classified as warfarin naïve if they have never used Titration of the study drug is based on central warfarin (or other VKAs) or if they have used it for ≤30 monitoring of INR measurements using encrypted point- consecutive days. Otherwise, patients are considered of-care (POC) devices, centralized dosing recommenda- warfarin experienced. tions, and sham apixaban titration. The POC device To maintain blinding, study medications are packaged delivers an encrypted result to the investigator who using a double-dummy design. The 2 sets of tablets each telephones or electronically transmits the result along subject receives are distinguishable by color and size, with the subject's identification number, date, and time to but active apixaban tablets match placebo apixaban a central response system. This system processes the tablets and active warfarin tablets match placebo information in a blinded manner and returns either a true warfarin tablets to ensure blinding of the patient and INR value (if a subject is receiving warfarin) or a sham INR investigator. After randomization, patients receive either value (if a subject is receiving apixaban). Although apixaban and warfarin placebo or apixaban placebo and investigators will need to obtain open label INR values warfarin. During the titration phase, we recommend the when clinically indicated, efforts will be made to minimize use of a dosing algorithm with initial daily dose of up to nonstudy INR assessments. The final dosing decision rests 6 mg of warfarin (or warfarin placebo) (unless already with the investigator. Assessments of outcomes and study on a stable dose of warfarin, in which case that may be medication compliance are performed at each INR visit. continued) and dose of apixaban (or apixaban placebo) of 5 mg twice a day. For patients who are estimated to Follow-up have higher apixaban drug exposure, we will use a The follow-up period will last until the attainment of at lower dose of 2.5 mg twice a day of apixaban. Subjects least 448 primary study events. Follow-up of subjects who fulfill any 2 of the following criteria at baseline will who discontinue study drug will occur quarterly until the receive the lower apixaban dose of 2.5 mg twice a day: end of the study. age ≥80 years, body weight ≤60 kg, and serum creatinine level ≥1.5 mg/dL (133 μmol/L). Subsequent warfarin doses are recommended based upon an Outcome definitions algorithm provided to the investigators; however, the Efficacy outcomes final dose decision is left to the discretion of the The primary efficacy outcome is the time to first investigator. Subjects, investigators, members of the occurrence of stroke (ischemic or hemorrhagic) or steering and adjudication committees, and the sponsor's systemic embolism. staff conducting the study do not have access to In this study, stroke is defined as a nontraumatic abrupt individual subject treatment assignments. onset of a focal neurologic deficit lasting at least 24 hours. A retinal ischemic event (embolism or thrombosis) will be considered a stroke. A cerebral imaging study Study design and duration (computed tomographic scan or magnetic resonance The trial is event driven, thus, the number of subjects imaging) is recommended for all suspected strokes. required and length of treatment were estimated based Strokes will be classified as ischemic, ischemic with on event rates in similar trials. The final duration of the hemorrhagic transformation, hemorrhagic, or of uncer- trial will be determined by the time required to accrue at tain type. Hemorrhagic strokes will be subclassified as least 448 primary efficacy events (see statistical methods). subdural, subarachnoid, or intraparenchymal. All subjects will be followed from randomization until the A transient ischemic attack (TIA) is defined as a study end date. nontraumatic abrupt onset of a focal neurologic deficit The study includes a screening period of up to 14 days. lasting b24 hours. Stroke and TIA will be further Subjects with AF and at least ≥1 risk factors for stroke will subclassified based on whether there is imaging evidence be evaluated for study eligibility. The design of the study of a new cerebral infarction that correlates with the is shown in Figure 1. clinical presentation of the subject. American Heart Journal Lopes et al Volume 159, Number 3 335

Figure 1

Study design.

The diagnosis of systemic embolism requires a clinical Bleeding events are also classified by the TIMI26 and history consistent with an acute loss of blood flow to a GUSTO criteria.27 peripheral artery (or arteries) supported by evidence of The secondary safety outcome for this trial is a composite embolism from surgical specimens, autopsy, angiogra- of major bleeding and clinically relevant nonmajor phy, vascular imaging, or other objective testing. bleeding. Other safety outcome measures include minor bleeding, fractures, and other adverse events. Safety outcomes The primary safety endpoint is time to first occurrence Clinical Events Committee adjudication of confirmed major bleeding. An independent, blinded, clinical events committee The definition of major bleeding described below is (CEC) adjudicates all suspected hemorrhagic and non- adapted from the protocol and the International Society hemorrhagic strokes, TIAs, systemic emboli, major and on Thrombosis and Hemostasis definition. The baseline clinically relevant non-major bleeding, myocardial infarc- hemoglobin level is defined as the closest hemoglobin tion, and cause of death. level value before the bleeding event. Using prespecified event definitions and agreed upon Major bleeding is defined as acute or subacute clinically event adjudication criteria, the CEC adjudicates suspected overt bleeding accompanied by ≥1 of the following: (1) a events based on the preponderance of the evidence and decrease in hemoglobin level of ≥2 g/dL over a 24-hour the clinical knowledge and experience of the physician period; (2) a transfusion of ≥2 U of packed red blood reviewers. Event adjudication in ARISTOTLE occurs in 2 cells; and/or (3) bleeding that is fatal or occurs in at least 1 phases. All suspected events are adjudicated in phase I. of the following critical sites: intracranial, intraspinal, Each stroke, systemic embolism, and bleeding event are intraocular, pericardial, intraarticular, intramuscular with reviewed by 2 independent physician reviewers. Each compartment syndrome, retroperitoneal. myocardial infarction and death event are reviewed by 1 Clinically relevant nonmajor bleeding is defined as reviewer. Significant disagreements between phase I acute or subacute clinically overt bleeding that does not reviewers for stroke, systemic embolism, and bleeding satisfy the criteria for major bleeding and leads to hospital cases are identified as needing phase II review. In admission for bleeding, physician-guided medical or addition, a quality control sample undergoes phase II surgical treatment for bleeding, or a change in antith- review. All phase II reviews are conducted by committee rombotic therapy (including study drug) for bleeding. with the final adjudicated result being a consensus of the All acute clinically overt bleeding events not meeting committee members present. In both phase I and phase II the criteria for either major bleeding or clinically relevant reviews, all stroke events are evaluated by at least nonmajor bleeding are classified as minor bleeding. 1 neurologist. American Heart Journal Lopes et al 336 March 2010

Statistical analysis and sample Figure 2 size calculation A meta-analysis of 6 AF studies estimated a relative reduction in the risk of stroke or systemic thromboembolism of 62% for warfarin versus placebo.6 Based on historical trials, the primary noninferiority hypothesis of ARISTOTLE is that apixaban will preserve at least 50% of the benefit of warfarin in preventing stroke and systemic embolism. This gives an upper CI of 1.88 of the apixaban versus warfarin relative risk (Figure 2A). To be more certain that apixaban is noninferior to warfarin based on this single clinical trial, more stringent boundaries are defined. In response to different international regulatory bodies, 2 noninferiority tests will be applied. First, the 95% CI should not include ≥1.38 to declare noninferiority (Figure 2B). In addition, the 99% CI should not include ≥1.44 to declare noninferiority. Because the event rate is lower than initially expected, it is estimated that approximately 18,000 randomized patients with sufficient risk of stroke and sufficient treatment duration will result in 448 patients with primary outcome events needed for 90% power to meet the primary objective of the study. We originally estimated that the follow-up would be an average of 1.8 years, assuming a stroke or systemic embolism rate of 1.67 per 100 subject-years. An independent Data Monitoring Committee is charged with monitoring the accumulating trial data. A formal interim analysis will be performed once 50% of the primary efficacy endpoint events have been confirmed by the CEC. The objective of this interim analysis is to determine whether apixaban is superior to warfarin for the primary efficacy endpoint. No interim testing for noninferiority will be performed.

Pharmacokinetic biomarkers The main objectives of the biomarker and genetic Illustration and interpretation of noninferiority boundaries for substudy program are to correlate genetic polymorph- warfarin comparator AF trials. A, Noninferiority boundaries for isms and levels of biomarkers with clinical outcomes, to warfarin comparator AF trials. B, Apixaban preserves some benefit improve risk stratification for stroke among patients (A), apixaban preserves N50% of warfarin benefit (B), apixaban with AF, and to relate the effects of apixaban and preserves more than about 40% of warfarin benefit (C), and warfarin to these disease biomarkers. Several biomarkers apixaban is superior to warfarin (D). will be analyzed at baseline for as many of the 18,206 patients as possible (Table II), and a second blood sample will be collected for approximately 5,000 scientific direction and input, addresses policy issues patients at 2 months. This will allow the analysis of regarding the protocol, and meets periodically to assess changes in biomarker levels. the trial progress. The ARISTOTLE executive committee, composed of a subset of senior leaders from the steering committee, is Organizational structure responsible for evaluating the progress and safety of the The ARISTOTLE trial is led by an academic steering trial and making decisions regarding early termination or committee composed of 2 cochairs, national coordina- continuation of the trial. tors from each participating country, and a representa- A subset of the steering committee will form the tive from the trial sponsor. This committee provides publications committee that will oversee the publication American Heart Journal Lopes et al Volume 159, Number 3 337

Table II. Biomarkers

Biomarkers Area Characteristic hs-Troponin I Myocardial necrosis, Strong predictor for raised mortality in ACS and in healthy elderly myocardial function NT-proBNP Myocardial necrosis, ↑ in patients with AF; high level indicates increased risk for thromboembolism myocardial function ADMA Endothelial function ↑ in patients with vascular disease vWF Endothelial function ↑ levels in AF patients; predictor for vascular events hs-CRP, IL-6 Inflammation Associated with AF and risk of future CV events in healthy individuals Soluble CD40 ligand Platelet activity ↑ levels related to inflammatory activity, coagulation, and platelet activation Fragment 1+2 Coagulation ↑ in patients with AF and related to risk factors for stroke; reduced with warfarin treatment D-dimer Coagulation ↑ in patients with AF and associated with new thrombotic events; reduced with warfarin treatment Cystatine C Renal function ↑ in patients with reduced renal function; poorer prognosis in patients with CV disease; better marker for endogenous GFR than creatinine clearance HbA1C, Apo A, Apo B Metabolism and lipoproteins Lipoproteins and diabetes mellitus risk factors for CV disease; diabetes mellitus predictor of an increased risk for complications

BNP, Brain natriuretic peptide; CV, cardiovascular; GFR, glomerular filtration rate. process for the main manuscript and all secondary and the drug was withdrawn from all markets in 2006. In presentations and manuscripts resulting from the trial. addition, there was an unexplained higher rate of major The ARISTOTLE Trial is sponsored and funded by bleeding in SPORTIF V when compared with SPORTIF III. Bristol-Myers Squibb (Princeton, NJ, USA) and Pfizer Recently the RE-LY open label trial reported the results (New York, NY, USA). comparing 110 mg twice daily and 150 mg twice daily of the oral thrombin inhibitor dabigatran versus warfarin in 18,113 AF paitents with at least one risk factor for stroke. Discussion In this study dabigatran 150 mg twice daily was superior Atrial fibrillation is a very common arrhythmia and is and 110 mg twice daily noninferior concerning stroke associated with an increased risk of mortality and while simultaneously both doses reduced intracranial and morbidity, particularly stroke, which is the third most life-threatening hemorrhage.32 Therefore, there is now a common cause of death in developed countries and the larger focus on testing for potential superiority of leading cause of serious long-term disability worldwide.28 apixaban over warfarin in the current trial. The ACTIVE-W study included 6,706 patients with AF Other phase III trials are testing inhibitors of factor Xa and at least ≥1 risk factors for stroke.18 ACTIVE-W was for stroke prevention in patients with AF. ROCKET-AF is designed to compare oral anticoagulation therapy with comparing warfarin with rivaroxaban in a double-blind aspirin plus clopidogrel for prevention of vascular events trial of 14,000 patients with AF at high risk for stroke.33 in patients with AF at high risk for stroke. This study was AVERROES is another double-blind trial of apixaban in not blinded. Moreover, there was no requirement for a comparison with aspirin in 5,600 patients with AF and certain proportion of warfarin-naïve patients. At study moderate risk of stroke or intolerance to warfarin.34 entry, approximately 76% of the patients were on BOREALIS-AF is testing subcutaneous weekly injections warfarin. The study was stopped early because of the of biotinylated idraparinux, a subcutaneous indirect superiority of warfarin over aspirin plus clopidogrel for factor Xa inhibitor, in patients with AF and high risk of prevention of vascular events in patients with AF. These stroke.35 This trial is designed as a double-blind compar- results were driven by the higher rates of stroke on ison of idraparinux with warfarin in 9,600 patients. clopidogrel plus aspirin. ARISTOTLE will define whether apixaban is noninfer- Several oral direct thrombin inhibitors are being tested ior or/and superior to warfarin in the overall population for stroke prevention in patients with AF such as and whether it may be superior solely in the warfarin- ximelagatran and dabigatran.29 This class of drug has a naïve population, which is more susceptible to compli- wider therapeutic range than warfarin, low potential for cations until a stable warfarin dose has been established food and drug interactions, and no requirements for dose and more prone to warfarin discontinuation. adjustments or regular monitoring. The SPORTIF III and V Important features of the ARISTOTLE trial include the trials involving 3,407 and 3,922 patients, respectively, investigation of a drug with a relatively long half-life and showed no difference in stroke prevention between robust phase II to III data on safety and efficacy of the ximelagatran and warfarin.10,12 However, there was a tested dose for DVT prophylaxis, the double-blind design significant increase in the level of liver enzymes in with use of encrypted POC INR monitoring devices, good patients on ximelagatran. Moreover, major adverse representation of warfarin-naïve patients, inclusion of the 30,31 cardiovascular events were observed in other studies, full risk-spectrum of patients (CHADS2 score ≥1), American Heart Journal Lopes et al 338 March 2010

multinational representation of various health care 9. Wan Y, Heneghan C, Perera R, et al. Anticoagulation control and systems and ethnic groups, and modification of dose of prediction of adverse events in patients with atrial fibrillation: a the study drug for patients with the highest drug systematic review. Circulation Cardiovascular Quality and Outcomes exposure (2 criteria of older age, renal insufficiency, or 2008;1:84-91. low body weight). In ARISTOTLE, the protocol provides 10. Albers GW, Diener HC, Frison L, et al. Ximelagatran vs warfarin for stroke prevention in patients with nonvalvular atrial fibrillation: a guidance on management of bleeding. Most bleeding can randomized trial. JAMA 2005;293:690-8. be managed by discontinuation of antithrombotic thera- 11. van Walraven C, Jennings A, Oake N, et al. Effect of study setting on py (including study medication), local hemostatic mea- anticoagulation control: a systematic review and metaregression. sures, and fresh frozen plasma as needed. Patients with Chest 2006;129:1155-66. ongoing serious or life-threatening bleeding may require 12. Olsson SB. Stroke prevention with the oral direct thrombin inhibitor unblinding to guide reversal with vitamin K, and ximelagatran compared with warfarin in patients with non-valvular consideration of use of prothrombin complex concen- atrial fibrillation (SPORTIF III): randomised controlled trial. Lancet trate, if on warfarin. For patients on apixaban, reversal of 2003;362:1691-8. the anticoagulant effects will occur relatively rapidly over 13. Singer DE, Albers GW, Dalen JE, et al. Antithrombotic therapy in atrial fibrillation: American College of Chest Physicians Evidence- time given its half-life of around 12 hours. Apixaban has Based Clinical Practice Guidelines (8th Edition). Chest 2008;133: no specific antidote or reversal agent. Other agents, such 546S-92S. as recombinant activated factor VII (NovoSeven, Novo 14. Holbrook AM, Pereira JA, Labiris R, et al. Systematic overview of Nordisk A/S, Bagsvaerd, Denmark), have not been studied warfarin and its drug and food interactions. Arch Intern Med 2005; in this setting and are not recommended. 165:1095-106. 15. Cheng TO. Warfarin interaction with herbal drugs and food. Int J Cardiol 2007;119:107-8. Conclusion 16. Fuster V, Rydén LE, Cannom DS, et al. ACC/AHA/ESC 2006 In conclusion, the ARISTOTLE trial will answer many Guidelines for the Management of Patients with Atrial Fibrillation: a important questions related to stroke prevention in AF. report of the American College of Cardiology/American Heart Most importantly, it will compare apixaban with Association Task Force on Practice Guidelines and the European warfarin for stroke prevention in a wide range of Society of Cardiology Committee for Practice Guidelines (Writing patients with AF. It will provide information about the Committee to Revise the 2001 Guidelines for the Management of efficacy of apixaban both in warfarin-experienced and Patients with Atrial Fibrillation): developed in collaboration with the warfarin-naïve patients. Finally, it will generate a better European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006;114:e257-e354. understanding of this common disease and the risks of 17. Healey JS, Hart RG, Pogue J, et al. Risks and benefits of oral stroke and bleeding based on large scale substudies on anticoagulation compared with clopidogrel plus aspirin in patients genetic and protein biomarkers. with atrial fibrillation according to stroke risk: the atrial fibrillation clopidogrel trial with irbesartan for prevention of vascular events (ACTIVE-W). Stroke 2008;39:1482-6. References 18. The ACTIVE Writing Group on behalf of the ACTIVE Investigators. 1. Wattigney WA, Mensah GA, Croft JB. Increasing trends in Clopidogrel plus aspirin versus oral anticoagulation for atrial hospitalization for atrial fibrillation in the United States, 1985 fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for through 1999: implications for primary prevention. Circulation 2003; prevention of Vascular Events (ACTIVE W): a randomised controlled 108:711-6. trial. Lancet 2006;367:1903-12. 2. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of 19. Connolly SJ, Pogue J, Eikelboom J, et al. Benefit of oral anticoagulant atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and over antiplatelet therapy in atrial fibrillation depends on the quality of implications on the projections for future prevalence. Circulation international normalized ratio control achieved by centers and 2006;114:119-25. countries as measured by time in therapeutic range. Circulation 3. Benjamin EJ, Wolf PA, D'Agostino RB, et al. Impact of atrial 2008;118:2029-37. fibrillation on the risk of death: the Framingham Heart Study. 20. Wong PC, Crain EJ, Xin B, et al. Apixaban, an oral, direct and highly Circulation 1998;98:946-52. selective factor Xa inhibitor: in vitro, antithrombotic and antihemo- 4. Wolf PA, D'Agostino RB, Belanger AJ, et al. Probability of stroke: a static studies. J Thromb Haemost 2008;6:820-9. risk profile from the Framingham Study. Stroke 1991;22:312-8. 21. Pinto DJ, Orwat MJ, Quan ML, et al. 1-[3-Aminobenzisoxazol-5′-yl]- 5. Lairikyengbam SK, Anderson MH, Davies AG. Present treatment 3-trifluoromethyl-6-[2′-(3-(R)-hydroxy-N-pyrrolidinyl)methyl-[1, 1′]- options for atrial fibrillation. Postgrad Med J 2003;79:67-73. biphen-4-yl]-1,4,5,6-tetrahydropyrazolo-[3,4-c]-pyridin-7-one 6. Hart RG, Benavente O, McBride R, et al. Antithrombotic therapy to (BMS-740808) a highly potent, selective, efficacious, and orally prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann bioavailable inhibitor of blood coagulation factor Xa. Bioorg Med Intern Med 1999;131:492-501. Chem Lett 2006;16:4141-7. 7. Hylek EM, Evans-Molina C, Shea C, et al. Major hemorrhage and 22. Lassen MR, Davidson BL, Gallus A, et al. The efficacy and safety of tolerability of warfarin in the first year of therapy among elderly apixaban, an oral, direct factor Xa inhibitor, as thromboprophylaxis patients with atrial fibrillation. Circulation 2007;115:2689-96. in patients following total knee replacement. J Thromb Haemost 8. Goto S, Bhatt DL, Röther J, et al. Prevalence, clinical profile, and 2007;5:2368-75. cardiovascular outcomes of atrial fibrillation patients with athero- 23. Buller HR. A dose-finding study of the oral direct factor Xa inhibitor thrombosis. Am Heart J 2008;156:855-63. apixaban in the treatment of patients with acute symptomatic deep American Heart Journal Lopes et al Volume 159, Number 3 339

vein thrombosis. Presented as a late-breaking clinical trial abstract 30. Hermans C, Claeys D. Review of the rebound phenomenon in new at: 21st Congress of the International Society on Thrombosis and anticoagulant treatments. Curr Med Res Opin 2006;22:471-81. Haemostasis; July 9, 2007; Geneva, Switzerland; 2007. 31. Fiessinger JN, Huisman MV, Davidson BL, et al. Ximelagatran vs low- 24. Bristol-Myers Squibb and Pfizer provide update on apixaban clinical molecular-weight heparin and warfarin for the treatment of deep vein development program [press release]. Princeton, NJ & New York, thrombosis: a randomized trial. JAMA 2005;293:681-9. NY: Business Wire; August 26, 2008. 32. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin 25. Alexander JH, Becker RC, Bhatt DL, et al. Apixaban, an oral, direct, in patients with atrial fibrillation. N Engl J Med 2009;361:1139-51. selective factor Xa inhibitor, in combination with antiplatelet therapy 33. Randomized, Double-Blind Study Comparing Once Daily Oral after acute coronary syndrome: results of the Apixaban for Prevention Rivaroxaban With Adjusted-Dose Oral Warfarin for the Prevention of of Acute Ischemic and Safety Events (APPRAISE) trial. Circulation Stroke in Subjects with Non-Valvular Atrial Fibrillation. ClinicalTrials. 2009;119:2877-85. gov Identifier: NCT00403767. ClinicalTrials.gov Web site. http:// 26. Bovill EG, Terrin ML, Stump DC, et al. Hemorrhagic events during clinicaltrials.gov/ct2/show/NCT00403767?term = therapy with recombinant tissue-type plasminogen activator, heparin, nct00403767&rank=1. Last accessed February 4, 2009. and aspirin for acute myocardial infarction. Results of the Thrombo- 34. A Phase III Study of Apixaban in Patients With Atrial lysis in Myocardial Infarction (TIMI), Phase II Trial. Ann Intern Med Fibrillation (AVERROES). ClinicalTrials.gov Identifier: 1991;115:256-65. NCT00496769. Clinical Trials.gov Web site. http://clinicaltrials. 27. An international randomized trial comparing four thrombolytic gov/ct2/show/NCT00496769?term=nct00496769&rank=1. Last strategies for acute myocardial infarction. The GUSTO Investigators. accessed February 4, 2009. N Engl J Med 1993;329:673-82. 35. Evaluation of Weekly Subcutaneous Biotinylated Idraparinux Versus 28. Lloyd-Jones D, Adams RJ, Brown TM, et al. Heart Disease and Stroke Oral Adjusted-Dose Warfarin to Prevent Stroke and Systemic Statistics—2010 Update. A Report From the American Heart Thromboembolic Events in Patients with Atrial Fibrillation (BOREALIS- Association. Circulation 2009. AF). ClinicalTrials.gov Identifier: NCT00580216. ClinicalTrials.gov 29. DiNisio M, Middeldorp S, Büller HR. Direct thrombin inhibitors. N Web site. http://clinicaltrials.gov/ct2/show/NCT00580216? Engl J Med 2005;353:1028-40. term=nct00580216&rank=1. Last accessed February 4, 2009.