doi: 10.1161/CIRCULATIONAHA.107.695163
(Circulation. 2007;116:449-455.)
© 2007 American Heart Association, Inc.
New Drugs and Technologies |
Challenges of Establishing New Antithrombotic Therapies in Atrial Fibrillation
From the Population Health Research Institute, McMaster University, Hamilton, Ontario, Canada.
Correspondence to Dr Stuart Connolly, 237 Barton St East, Suite 504, Hamilton, Ontario, Canada L8L 2X2. E-mail connostu{at}phri.ca
 |
Introduction |
|---|
 Top Introduction Underuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Warfarin is the single most effective treatment to prevent stroke in patients with atrial fibrillation (AF) in that it reduces the risk by about two thirds compared with placebo.1 Despite its effectiveness, however, warfarin is difficult to use and its uptake into clinical practice has been constrained.2 Warfarin has a slow onset and offset of action, it exhibits considerable variability in dose response among individuals, it is subject to multiple food and drug interactions, it has a narrow therapeutic window, and considerable risk of hemorrhage exists. The slow offset of action of warfarin can be beneficial if a patient misses a dose but complicates temporary discontinuation of warfarin for surgery. Thus, warfarin requires careful laboratory monitoring and dose adjustment to maintain the international normalized ratio (INR) in the target therapeutic range. Even with careful laboratory monitoring, major bleeding occurs in 1% to 3% of AF patients on warfarin per year. Consequently, many clinicians and patients are reluctant to use warfarin. Some patients have contraindications to anticoagulation, but others choose not to use it.2 Such patients typically use aspirin.
Educational and support programs have been shown to improve knowledge of disease and treatment, INR control, and physician management of AF.3–5 Nonetheless, there have been intensive efforts to find a replacement for warfarin. Several new drugs with the potential to overcome the limitations of warfarin have been evaluated in randomized trials but appeared to be less effective or less safe than warfarin. The experience from these trials has highlighted the methodological challenges of finding a replacement for warfarin, which include a number of difficulties in design and in interpretation of results from noninferiority trials, the widespread use of warfarin in the community, availability of warfarin-naïve patients, and low event rates on warfarin. The present article will review the results of recent phase III randomized trials that evaluated new antithrombotic agents in AF, and it will focus on the methodological challenges in the evaluation of a new antithrombotic therapy.
|
Underuse of Warfarin |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
As a consequence of pharmacokinetic and pharmacodynamic limitations, warfarin requires frequent laboratory monitoring and dose adjustment. Concern about the inconvenience of the need to undergo regular laboratory monitoring and the risk of bleeding has contributed to the underuse of warfarin in patients with AF. Immediately after the benefits of warfarin for stroke prevention in AF patients were convincingly demonstrated in 1989, its use steadily increased.2 However, recent surveys6–10 suggest that the use of warfarin has plateaued in the past few years at
50% to 60% of eligible patients with AF (Table 1). New anticoagulants that do not share the pharmacological limitations of warfarin have the potential to lead to an increase in the proportion of patients with AF on effective antithrombotic therapy.
|
|
Recent Phase III Trials of Novel Antithrombotic Therapies to Prevent Stroke in AF |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Ximelagatran
The oral direct thrombin inhibitor Ximelagatran is a prodrug of melagatran, a small molecule that targets the active site of thrombin and blocks the enzyme’s catalytic activity. Ximelagatran produces a predictable anticoagulant response with no known drug or food interactions, which thereby reduces the need for laboratory monitoring. Two phase III noninferiority trials compared unmonitored Ximelagatran (36 mg twice daily) with dose-adjusted warfarin (target INR, 2.5; range, 2.0 to 3.0) in patients with nonvalvular AF and at least 1 additional risk factor for stroke. The Stroke Prevention Using the Oral Thrombin Inhibitor in Patients with Nonvalvular Atrial Fibrillation III (SPORTIF III)11 trial used an open-label design with blinded adjudication, whereas SPORTIF V12 was a double-blind trial that used a sham INR scheme to maintain blinding. The primary outcome measure in both trials was a combination of all strokes and systemic thromboembolism.
SPORTIF III (open-label n=3407) showed a trend to fewer strokes in favor of Ximelagatran. The annual rates of stroke and systemic embolism were 1.6% on Ximelagatran and 2.3% on warfarin. SPORTIF V (double-blind n=3922), on the other hand, showed results that tended to favor warfarin. The annual rates of stroke and systemic embolism were 1.6% on Ximelagatran and 1.2% on warfarin. In both trials, rates of major bleeding were somewhat lower on Ximelagatran compared with warfarin. Although some heterogeneity between the results of the 2 SPORTIF trials was observed for the impact on stroke, a prespecified analysis of the data pooled from SPORTIF III and SPORTIF V was reported. The rates of stroke and systemic embolic events were almost identical with Ximelagatran and warfarin (1.6% per year and 1.6% per ear; risk ratio 0.98; 95% confidence interval [CI], 0.73 to 1.30; P=0.98). The rate of major bleeding (on treatment) in the pooled analysis tended to be lower with Ximelagatran (1.9% per year versus 2.5% per year; risk ratio 0.76; 95% CI, 0.56 to 1.03; P=0.07).13
Ximelagatran failed to receive approval from the US Food and Drug Administration. The main reason was concern about safety; Ximelagatran was associated with raised transaminase levels in 4% to 6% of patients and with a few cases of fulminant hepatic failure.14 However, regulatory authorities and other commentators14 also concluded that noninferiority had not been established, even though both trials satisfied their prespecified noninferiority hypotheses. This conclusion was based on re-analysis of the noninferiority margin with more conservative estimates of efficacy of warfarin based on a somewhat different meta-analysis of previous warfarin trials, which applied a more conservative rule for determination of noninferiority margin and focused on the double-blind SPORTIF 5 results in preference to the open-label SPORTIF 3 results because of the heterogeneity between SPORTIF III and V results.
Idraparinux
Idraparinux is a synthetic analog of the active pentasaccharide sequence contained in unfractionated heparin and low molecular weight heparin that selectively inhibits activated coagulation factor X. It has a plasma half-life of 80 hours, which allows it to be administered subcutaneously once weekly.15 The Atrial Fibrillation Trial of Monitored, Adjusted Dose Vitamin K Antagonist, Comparing Efficacy and Safety With Unadjusted
SanOrg 34006/Idraparinux (AMADEUS) Study was an open-label phase III trial that aimed to recruit >7000 patients to compare subcutaneous idraparinux (given subcutaneously once weekly without coagulation monitoring) with dose-adjusted warfarin (target INR, 2.5; range, 2.0 to 3.0) in patients with AF and at least 1 additional risk factor for stroke. Designed as a noninferiority trial, the primary efficacy outcome measure was a composite of all strokes (ischemic and hemorrhagic) and systemic embolic events. The trial was terminated early because of safety concerns with regard to bleeding.
Aspirin and Clopidogrel
The Atrial Fibrillation Clopidogrel Trial With Irbesartan for Prevention of Vascular Events (ACTIVE)-W trial, an open-label randomized controlled trial (n=6706 patients), compared combination aspirin (75 to 100 mg recommended) and clopidogrel (75 mg) therapy to oral anticoagulation (INR 2 to 3) in patients with AF and at least 1 risk factor for stroke.16 Designed as a noninferiority trial, ACTIVE-W was terminated early when oral anticoagulation was found to be superior to combination antiplatelet therapy for prevention of stroke. After a median follow-up of 1.28 years, the annual rate of the primary outcome (a composite of stroke, systemic thromboembolism, myocardial infarction, and vascular death) was 3.9% in the oral anticoagulation group and 5.6% in the combination antiplatelet therapy group (P=0.0003).
|
New Agents in Phase III Trials |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Dabigatran
Dabigatran is a novel oral direct thrombin inhibitor and a prodrug of dabigatran etexilate. Like Ximelagatran, dabigatran produces a predictable anticoagulant response with no known drug or food interactions, which thereby reduces the need for coagulation monitoring. The half-life of dabigatran is
8 hours after single-dose administration and up to 14 to 17 hours after multiple doses. It is usually given twice daily.17 Dabigatran is compared with warfarin (INR 2 to 3) in a very large, ongoing, phase III, noninferiority, randomized controlled trial, Randomized Evaluation of Long-Term Anticoagulant Therapy (RELY). This trial will enroll 15 000 patients randomized to warfarin, or to 1 of 2 doses of dabigatran (110 mg or 150 mg, twice daily). The trial will balance enrollment between warfarin-experienced patients and those who have received it for <2 months. The warfarin to dabigatran comparison is unblinded and the 2 doses of dabigatran are blinded.
Factor 10a Inhibitors
Rivaroxaban is a member of a new class of orally available, small molecule, active-site–directed factor Xa inhibitor. It has high bioavailability (60% to 80%) and rapid onset of action, and it is excreted by renal (two thirds) and fecal (one third) routes. Rivaroxaban has a half-life of 6 to 9 hours. Apixaban is another oral factor Xa inhibitor. Rivaroxaban and apixaban are both compared with warfarin in large phase III trials for the prevention of stroke or systemic embolism in patients with AF.18,19 Both trials use a single dose of investigational product and are double blinded. Patient enrollment in each trial is about 14 000 to 15 000 patients.
|
Challenges in Evaluation of New Antithrombotic Drugs to Replace Warfarin in AF |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Several challenges exist in establishing a new antithrombotic therapy in AF. These are reviewed in Table 2 and will be discussed in detail below.
|
Trial Design
Three options exist in the design of trials for the evaluation of a new antithrombotic agent in AF. Placebo-controlled trials are desirable and ethical in low-risk patients, but low event rates and the high probability that bleeding risk will outweigh benefit make them unattractive. Alternatively, it is possible to evaluate whether a new agent is superior to aspirin because many high-risk AF patients currently receive aspirin. The final alternative is to compare a new drug to warfarin. Because it is unlikely that any new treatment will be more effective than warfarin for stroke prevention, such trials use a noninferiority design. This has been the most commonly used approach, so we will discuss it first.
Noninferiority Trials Against Warfarin
Ideally, one would like to be able to ensure that the new agent is totally comparable to the established agent. From the perspective of statistics, a new agent cannot be shown to be identical to standard therapy because all estimates of treatment effects have some uncertainty (CIs) around them. Therefore, the concept of noninferiority is logical and useful in trials that seek to establish a new agent by means of comparison to existing therapy. In a noninferiority trial, the goal is to demonstrate that a new agent is both superior to placebo and preserves a prespecified fraction of the benefit of standard therapy over control or placebo.18 Inherent in a noninferiority design is the potential for a loss of benefit. Reduced benefit of a new drug compared with the standard treatment can be justified if the loss of benefit falls within prespecified limits and if the new antithrombotic drug offers ancillary benefits such as greater ease of use, reduced need for laboratory monitoring, or possibly reduced risk of bleeding.
A key challenge in the design of any noninferiority trial is the selection of the noninferiority margin, which quantifies the minimally acceptable potential loss in efficacy of the new agent compared with standard therapy. Ideally, clinical judgment should play a key role in the decision of how much loss of efficacy is acceptable. However, given how difficult it is to formalize such judgements, a preference exists for methods that derive the margin from statistical reasoning.13 In AF, this approach relies on meta-analysis of prior randomized controlled AF trials to provide an estimate (and range of plausible effects through the 95% CI) of the efficacy of warfarin. Several randomized trials exist of warfarin against placebo or control in AF. Patients varied in baseline risk of stroke, and many patients were at low risk. Comparison of the new antithrombotic drug against warfarin makes it possible to estimate what the treatment effect of the new antithrombotic drug would have been if it had been tested directly against placebo. However, to show that a new agent is better than placebo is only the first step. The physician is also interested in whether the new agent is "as good as" the existing treatment.20 To do this, a trial sets a stricter noninferiority margin that specifies that a part of the known benefit of warfarin over placebo is to be preserved by the new therapy.
Although the statistical approach provides a measure of objectivity in determination of noninferiority margins, important uncertainties still exist. There are different ways to estimate the least likely treatment effect of warfarin, although the upper boundary of the 95% CI is usually chosen.21,22 It is still a matter of judgment to decide what fraction of the historically determined benefit of warfarin the agent should preserve. A conservative approach, which chooses a low estimate of the benefit standard therapy and also chooses to preserve a high fraction (
50%) of the margin of benefit of standard therapy, leads to a very narrow noninferiority margin. This in turn increases the likelihood of false-negative results, which means that effective therapies with potential benefits will not be adopted. More liberal noninferiority margins increase the likelihood that weakly effective or even ineffective therapies will be adopted into clinical practice. An overly liberal approach is unacceptable when outcomes such as stroke and death are prevented. A conservative approach to setting the noninferiority margin is currently favored by regulatory agencies. This results in the need for very large trials in AF (typically 10 000 to 15 000 patients or more). Although a Bayesian approach might be used to reduce the size of study needed, such methods have not been accepted by authorities. Use of an outcome event composite will increase event rates and can reduce sample size in superiority trials. However, in noninferiority trials in AF, little such benefit is derived from the use of a cluster of total vascular events as primary outcome because the benefit of increased events is offset by a more conservative noninferiority margin, as a result of a wider CI for total vascular events (than for stroke) in placebo-controlled trials.
Uncertainties With Noninferiority Designs
Two factors related to the design of noninferiority trials justify a conservative approach to setting the noninferiority margin: first, the potential that there will be gradual slippage of standards of approvability of new agents as a result of a series of noninferiority trials; and second, the need to assume constancy of clinical treatment over time. Noninferiority trial designs explicitly accept a potential loss of efficacy of the new agent compared with standard therapy. If the new agent becomes the new standard, this opens the possibility of further loss of efficacy when the next new agent is tested against the new standard with a noninferiority design. Eventually, some new agent could achieve noninferiority to the previous standard (which in turn was noninferior to a former standard), when in fact the new agent is no better than placebo. The solution to this concern, which is to have very conservative noninferiority margins, may be so onerous an obstacle to evaluation of the first new agent in AF that the sorts of trials needed may simply be impractical.
A more fundamental concern is the assumption of constancy of clinical practice. The critical margin of noninferiority in trials performed today is set on the basis of results of trials done in the past. One therefore implicitly assumes that the efficacy of standard therapy has remained generally constant into the current era. This constancy assumption cannot be proven because a placebo arm does not exist in the noninferiority trial. The constancy assumption may not hold because clinical practice in fact does evolve over time. A conservative approach to the setting of noninferiority margins has been recommended as a solution23,24 and is especially justifiable when concern exists that the standard treatment is less effective today than in the past. Under such circumstances, a new drug might be able to demonstrate noninferiority to a now less effective standard therapy even though the new drug is not better than placebo.
The management of patients with AF has changed since randomized controlled trials of warfarin were first performed >15 years ago and now include better control of blood pressure and increased use of drugs such as statins, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers.11,12,16 Evidence also exists, based on lower stroke rates in recent trials,12,16 that warfarin may be used more effectively today than in the previous generation of trials (see first paragraph of Improved INR Control and the Challenge for Modern Trials and Table 3). Such a change might justify a more liberal noninferiority margin, as the standard therapy may have become harder to match. Most researchers and regulators feel uncomfortable with this approach, but the margins chosen for noninferiority trials should be based on all available evidence. In summary, important regulatory concerns about noninferiority trial design and its long-term implications for the drug approval process have led to a conservative approach to setting the noninferiority margin. This sets a high standard for new agents to achieve, compared with the original standards where trials have evaluated the superiority of agents versus placebo therapy. Table 2 summarizes key features of noninferiority trial design.
|
|
Superiority Trials Against Aspirin |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
A superiority trial of a new agent against aspirin is ethical and feasible in high-risk patients with AF. As shown in Table 1, many patients exist who cannot or will not take warfarin. Some of these have increased risk of bleeding and are not suitable for a trial of antithrombotic therapy. However, others have compliance issues specifically related to vitamin K antagonists, and these patients may be able to safely use an antithrombotic agent that does not require monitoring. Recent guideline changes25 indicate that AF patients with a single risk factor may receive either warfarin or aspirin. These patients are at lower risk of events, and this would mean that, if they were the only patients enrolled, the study size could be very large. This is offset in part by the fact that superiority trials designed to detect fairly large treatment effects (35% to 45% relative risk reduction) require far fewer patients and events than noninferiority trials in AF.
An important synergy exists in the performance of a superiority trial against aspirin in support of a noninferiority trial against warfarin, with the same therapeutic agent, in the 2 populations of AF patients at high risk for stroke. Trials that demonstrate superiority of a new agent over an existing agent or placebo provide the most direct way to establish the efficacy of the new agent. Unlike noninferiority trials, superiority trials do not require interpretation of a meta-analysis of previous trials, do not require agreement on a noninferiority margin, and do not make assumptions about historical constancy of treatment effects. Also unlike noninferiority trials, superiority trials do not raise concerns about the balance of the potential loss of efficacy and also face less serious problems when noncompliance with the study drug is prominent. Although most investigators would be hesitant to accept a superiority trail against aspirin as pivotal for development of a new antithrombotic agent, few would disagree that clear evidence of substantial efficacy of a new agent against aspirin would provide support to an application for approval based on a noninferiority trial against warfarin.
Evaluation of the feasibility of a superiority trial against aspirin in AF patients at moderate to high risk is a goal of the ongoing Atrial Fibrillation Clopidogrel Trial With Irbesartan for Prevention of Vascular Events (ACTIVE A) trial, which compares clopidogrel plus aspirin to aspirin therapy alone. It has completed enrollment of 7645 high-risk AF patients in >30 countries and 500 centers worldwide and will report results in 2008.26
|
Declining Event Rates |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Rates of strokes in AF trials appear to be declining over time (Table 3). The primary prevention randomized controlled trials before 1995 did not select high-risk AF patients.27 In the pooled analysis of these trials, 35% of patients were <65 years of age and had no risk factors for stroke. With the publication in 1996 of Stroke Prevention in Atrial Fibrillation (SPAF) 3, AF trials have only included patients with risk factors for stroke.28 SPAF 3 reported an ischemic stroke risk on warfarin of 1.9% per year. Trials published recently generally report much lower rates of ischemic stroke on warfarin or vitamin K antagonists: 1.9% per year in SPORTIF 3 (2003), 1.1% per year in SPORTIF 5 (2005), and 1.0% per year in ACTIVE W (2006) (Table 3). The lower risk of ischemic stroke on warfarin, despite selection of high-risk patients in recent trials, may be related to more effective concomitant therapies, more aggressive control of systemic hypertension, and improved control of oral anticoagulant therapy. The last decade has seen an increased use of statin therapy and more aggressive treatment of blood pressure with increased use of angiotensin-converting enzyme inhibitors and angiotensin receptor blocking drugs. The mean systolic blood pressure at baseline in the pooled analysis of the controlled trials of warfarin in AF27 was 141 mm Hg and in SPAF 3 was 140 mm Hg.28 The corresponding baseline mean systolic blood pressures in SPORTIF 3, SPORTIF 5, and ACTIVE W were 139 mm Hg, 132 mm Hg, and 133 mm Hg, respectively. Very low event rates on warfarin create a challenge for study design. If the total rate of ischemic hemorrhagic stroke and systemic embolism is only 1.5% per year, 30 000 patient years of follow-up are needed to have the 450 events required in a typical noninferiority trial in AF. By contrast, a total of only 108 ischemic strokes (from 3706 patients) are reported in the pooled data from the first 5 trials of warfarin against placebo or control in AF.1
|
Improved INR Control and the Challenge for Modern Trials |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
INR control of warfarin therapy has improved since the first trials with warfarin were performed. At the time of the first 5 placebo-controlled trials, use of the INR to control warfarin was just being introduced clinically in North America, and 3 of the 5 studies controlled warfarin by means of the prothrombin time ratio.29–31 This method is inferior to the use of the INR because it does not correct for the variable characteristics of different thromboplastin reagents and is likely to lead to less optimal INR control. INR was used to monitor anticoagulation in 2 of the original placebo-controlled trials (Copenhagen Atrial Fibrillation, Aspirin, and Anticoagulation [AFASAK] study32 and Canadian Atrial Fibrillation Anticoagulation [CAFA] study33), in which the percentages of INR values in the target range were 42% and 44%, respectively. In the recent SPORTIF 3, SPORTIF 5, and ACTIVE W trials, the time in therapeutic range of the INR was 66%, 68%, and 64%, respectively.
The relationship between INR control and both embolic and hemorrhagic outcomes is well established, and improved INR control will result in lower event rates on warfarin.34,35 This makes modern warfarin therapy a formidable comparator therapy and creates a dilemma for noninferiority trial design and conduct. In a noninferiority trial, should one target the rates of INR control observed in the original placebo-controlled trials, or a level typically achieved in clinical practice, or the best possible level of INR control? Insistence on the latter may demand a level of rigor that is artificial and inconsistent with the constancy principle of noninferiority trials. If one uses the former approach, the investigators may run the risk of not using best possible therapies.
|
Potential for Different Treatment Effects in Warfarin-Naïve and Warfarin-Experienced Patients |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Another challenge for the design of antithrombotic trials in AF is the widespread use of warfarin therapy in specialists’ practices in Western countries. Because patients are most easily recruited into clinical trials from such practices (especially from anticoagulant clinics), a tendency exists to enroll patients already on successful warfarin treatment for years. Such patients are thus likely to be self-selected "responders" to warfarin because patients with a tendency to hemorrhage on warfarin, or in whom INR control is difficult, will self-identify during prestudy exposure to warfarin. In recent studies such as SPORTIF 3, SPORTIF 5, and ACTIVE W,
80% of patients were warfarin experienced (Table 3). In ACTIVE W, considerable evidence was produced that previous warfarin experience predicted a better response to warfarin with lower risk of hemorrhage.16 The upper 2 panels of the Figure show the cumulative risk of a primary event for patients on and not on oral anticoagulant therapy at entry to ACTIVE W, and the lower 2 panels show the same data from the outcome of major bleeding. For the primary outcome, a trend exists toward a greater benefit with oral anticoagulant therapy (compared with clopidogrel plus aspirin) in patients on oral anticoagulant at entry. The interaction is more impressive for major bleeding where patients with prior use of oral anticoagulant therapy had relatively less bleeding on oral anticoagulants, and patients not on oral anticoagulants at entry had relatively more bleeding on oral anticoagulants (Pinteraction=0.03). Unpublished data from the SPORTIF trials also support the idea that evaluation of a new agent against warfarin is more challenging in warfarin-experienced patients (personal communication, J. Halperin, MD, 2006). Various surveys2,6,7–10 have documented that only about half of AF patients receive warfarin; therefore, it is appropriate to enroll balanced numbers of anticoagulation-experienced and anticoagulation-naïve patients into future trials. This opens the possibility to evaluate a new therapy as an alternative to ongoing warfarin and as initial therapy in AF.
|
|
Conclusion |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
Evaluation of new antithrombotic agents for AF is challenging. We have discussed the commonly used trial design of noninferiority and alternatives to this approach. We have discussed the assumptions inherent to noninferiority trial design and some of the complexities of trial conduct, both of which complicate interpretation of results. Noninferiority trial design is still evolving, and regulatory agencies mandate conservative noninferiority margins. This, together with declining stroke rates in AF patients on warfarin, leads to a requirement for very large patient enrollments (
15 000) and a real risk of false-negative results. The enrollment challenge is compounded because the easiest approaches (eg, anticoagulation clinics) to identify high-risk patients for these trials leads to the enrollment of warfarin-experienced individuals. These patients have had a chance to discover whether or not they respond well to warfarin but not to the new agent. The efficacy of warfarin may have improved over time with improved INR control and advances in concomitant therapy for hypertension. A superiority trial design, which compares a new agent against aspirin, could be performed as a complement to a noninferiority trial. Three new antithrombotic strategies that use noninferiority designs have been recently evaluated in AF without success. The experience from these trials has highlighted some of the challenges and should enable the design of better programs for the evaluation of new antithrombotic drugs in AF. Promising new agents will be tested in the next few years, and hopefully some of these will be demonstrated to be effective, safe, and easier to use than warfarin. |
Acknowledgments |
|---|
Disclosures
Drs Connolly and Yusuf have received research grant support from Sanofi Aventis, Bristol-Myers Squibb, and Boehringer Ingelheim for management of the ACTIVE and RELY trials, and both have given lectures and performed other consulting tasks for these companies. Dr Yusuf has served as a consultant/advisory board member for GlaxoSmithKline. The other authors report no conflicts.
|
References |
|---|
|
TopIntroductionUnderuse of WarfarinRecent Phase III Trials...New Agents in Phase...Challenges in Evaluation of...Superiority Trials Against...Declining Event RatesImproved INR Control and...Potential for Different...ConclusionReferences |
|---|
- Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: A meta-analysis. Ann Intern Med. 1999; 131: 492–501.
[Abstract/Free Full Text] - Bungard TJ, Ghali WA, Teo KK, McAlister FA, Tsuyuki RT. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med. 2000; 160: 41–46.
[Abstract/Free Full Text] - Lane DA, Ponsford J, Shelley A, Sirpal A, Lip GYH. Patient knowledge and perceptions of atrial fibrillation and anticoagulant therapy: effects of an educational intervention programme: the West Birmingham Atrial Fibrillation Project. Intern J of Cardiol. 2006; 110: 354–358.[CrossRef]
- Claes N, Buntinx F, Vijgen J, Arnout J, Vermylen J, Fieuws S, Van Loon H. The Belgian Improvement Study on Oral Anticoagulation Therapy: a randomized clinical trial. Eur Heart J. 2005; 26: 2159–2165.
[Abstract/Free Full Text] - Khan TI, Kamali F, Kesteven P, Avery P, Wynne H. The value of education and self-monitoring in the management of warfarin therapy in older patients with unstable control of anticoagulation. Br J Haematol. 2004; 126: 557–564.[CrossRef][Medline]
[Order article via Infotrieve]
- Nieuwlatt R, Capucci A, Lip GYH, Olsson SB, Prins MH, Nieman FH, Lopez-Sendon J, Vardas PE, Aliot E, Santini M, Crijns H; on behalf of the Euro Heart Survey Investigators. Antithrombotic treatment in real-life atrial fibrillation patients: a report from the Euro Heart Survey on Atrial Fibrillation. Eur Heart J. 2006; 27: 3018–3026.
[Abstract/Free Full Text] - Hylek EM, D’Antonio J, Evans-Molina C, Shea C, Henault LE, Regan S. Translating the results of randomized trials into clinical practice: the challenge of warfarin candidacy among hospitalized elderly patients with atrial fibrillation. Stroke. 2006; 37: 1075–1080.
[Abstract/Free Full Text] - Birman-Deych E, Radford MJ, Nilasena DS, Gage BF. Use and effectiveness of warfarin in Medicare beneficiaries with atrial fibrillation. Stroke. 2006; 37: 1070–1074.
[Abstract/Free Full Text] - Waldo AL, Becker RC, Tapson VF, Colgan KJ; for the NABOR Steering Committee. Hospitalized patients with atrial fibrillation and a high risk of stroke are not being provided with adequate anticoagulation. J Am Coll Cardiol. 2005; 46: 1729–1736.
[Abstract/Free Full Text] - Go AS, Hylek EM, Borowsky LH, Phillips KA, Selby JV, Singer DE. Warfarin use among ambulatory patients with non-valvular atrial fibrillation: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. Ann Intern Med. 1999; 131: 927–934.
[Abstract/Free Full Text] - Executive Steering Committee on behalf of the SPORTIF III Investigators. Stroke prevention with the oral direct thrombin inhibitor Ximelagatran compared with warfarin in patients with non-valvular atrial fibrillation (SPORTIF III): randomised controlled trial. Lancet. 2003; 362: 1691–1698.[CrossRef][Medline]
[Order article via Infotrieve]
- SPORTIF Executive Steering Committee for the SPORTIF V Investigators. Ximelagatran vs warfarin for stroke prevention in patients with non-valvular atrial fibrillation. JAMA. 2005; 293: 690–698.
[Abstract/Free Full Text] - Kaul S, Diamond GA, Weintraub WS. Trials and tribulations of non-inferiority: the Ximelagatran experience. J Am Coll Cardiol. 2005; 46: 1986–1995.
- Lawrence J, Hung J, Mahjoob K. Statistical review and evaluation, clinical studies, NDA 21-686 (2004). FDA website. Available at: http://www.fda.gov/ohrms/dockets/ac/04/briefing/2004-4069B1_07_FDABackgrounder-C-R-stat%20Review.pdf.
- Koopman MMW, Büller HR(Academic Medical Centre, Amsterdam, The Netherlands). Short- and long-acting synthetic pentasaccharides (minisymposium). J Intern Med. 2003; 254: 335–342.[CrossRef][Medline]
[Order article via Infotrieve]
- The ACTIVE Writing Group on behalf of the ACTIVE Investigators. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the atrial fibrillation clopidogrel trial with irbesartan for prevention of vascular events (ACTIVE W): a randomised controlled trial. Lancet. 2006; 367: 1903–1912.[CrossRef][Medline]
[Order article via Infotrieve]
- Gustafsson D. Oral direct thrombin inhibitors in clinical development. J Int Med. 2003; 254: 322–334.[CrossRef][Medline]
[Order article via Infotrieve]
- Eriksson BI, Borris LC, Dahl OE, Haas S, Huisman MV, Kakkar AK, Muehlhofer E, Dierig C, Misselwitz F, Kalebo P; for the ODIXa-HIP Study Investigators. A once-daily, oral, direct factor Xa inhibitor, Rivaroxaban (BAY 59-7939), for thromboprophylaxis after total hip replacement. Circulation. 2006; 114: 2374–2381.
[Abstract/Free Full Text] - Lassen MR, Davidson BL, Gallus A, Pineo G, Ansell J, Deitchman D, Apixaban investigators. A phase II randomized double-blind eight-arm parallel-group dose-response study of Apixaban, a new oral factor Xa inhibitor in knee replacement surgery. Blood. 2006; 108: 574. Abstract.
- Brown D, Volkers P, Day S. An introductory note to CHMP guidelines: choice of the non-inferiority margin and data monitoring committees. Stat Med. 2006; 25: 1623–1627.[CrossRef][Medline]
[Order article via Infotrieve]
- Siegel JP. Equivalence and non-inferiority trials. Am Heart J. 2000; 139: S166–S170.[CrossRef][Medline]
[Order article via Infotrieve]
- Lewis JA. The European regulatory experience. Stat Med. 2002; 21: 2931–2938.[CrossRef][Medline]
[Order article via Infotrieve]
- Hung HMJ, Wang SJ, O’Neill R. A regulatory perspective on choice of margin and statistical inference issue in non-inferiority trials. Biom J. 2005; 47: 28–36.[CrossRef][Medline]
[Order article via Infotrieve]
- Hung HMJ, Wang SJ, Tsong Y, Lawrence J, O’Neill R. Some fundamental issues with non-inferiority testing in active controlled trials. Stat Med. 2003; 22: 213–225.[CrossRef][Medline]
[Order article via Infotrieve]
- Fuster V, Ryden LE, Asinger RW, Cannom DS, Crijns HJ, Frye RL, Halperin JL, Kay GN, Klein WW, Levy S, McNamara RL, Prystowsky EN, Wann LS, Wyse DG, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Gregoratos G, Hiratzka LF, Jacobs AK, Russell RO, Smith Jr SC, Alonso-Garcia A, Blomstrom-Lundqvist C, de Backer G, Flather M, Hradec J, Oto A, Parkhomenko A, Silber S, Torbicki A. ACC/AHA/ESC Guidelines for the Management of Patients With Atrial Fibrillation: Executive Summary A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation) Developed in Collaboration With the North American Society of Pacing and Electrophysiology. Circulation. 2001; 104: 2118–2150.
[Free Full Text] - The Active Steering Committee; ACTIVE Investigators. Rationale and design of ACTIVE: the atrial fibrillation clopidogrel trial with irbesartan for prevention of vascular events. Am Heart J. 2006; 151: 1187–1193.[CrossRef][Medline]
[Order article via Infotrieve]
- Atrial Fibrillation Investigators. The efficacy of aspirin in patients in atrial fibrillation: analysis of pooled data from the three randomized trials. Arch Int Med. 1997; 157: 1237–1240.[Abstract]
- Stroke Prevention in Atrial Fibrillation Investigators. Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: stroke prevention in atrial fibrillation III randomised clinical trial. Lancet. 1996; 348: 633–638.[CrossRef][Medline]
[Order article via Infotrieve]
- The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with non-rheumatic atrial fibrillation. New Engl J Med. 1990; 323: 1505–1511.[Abstract]
- Stroke Prevention in Atrial Fibrillation Investigators. Stroke prevention in atrial fibrillation study: final results. Circulation. 1991; 84: 527–539.
[Abstract/Free Full Text] - Ezekowitz MD, Bridgers SL, James KE, Carliner NH, Colling CL, Gornick CC, Karuse-Steinrauf H, Kurtzke JF, Nazarian SM, Radford MJ, Rickles FR, Shabetai R, Deykin D; for the Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators. Warfarin in the prevention of stroke associated with non-rheumatic atrial fibrillation. NEJM. 1992; 327: 1406–1412.[Abstract]
- Petersen P, Boysen G, Godtfredsen J, Andersen ED, Andersen B. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation. The Copenhagen AFASAD Study. Lancet. 1989; 333: 175–179.[CrossRef]
- Connolly SJ, Laupacis A, Gent M, Roberts RS, Cairns JA, Joyner C; for the CAFA Study Co-investigators. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol. 1991; 18: 349–355.[Abstract]
- Hylek EM, Skates SJ, Sheehan MA, Singer DE. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients with non-rheumatic atrial fibrillation. N Engl J Med. 1996; 335: 540–546.
[Abstract/Free Full Text] - Reynolds MW, Fahrbach K, Hauch O, Wygant G, Estok R, Cella C, Nalysnyk L. Warfarin anticoagulation and outcomes in patients with atrial fibrillation: a systematic review and meta-analysis. Chest. 2004; 126: 1938–1945.
[Abstract/Free Full Text]
This article has been cited by other articles:
 |
|
 |
|
  M. D. Ezekowitz and R. Nagarakanti Letter Regarding Article by Connolly et al, "Challenges of Establishing New Antithrombotic Therapies in Atrial Fibrillation" Circulation, February 5, 2008; 117(5): e149 - e149. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||