Author + information
- Jonathan P. Piccini, MD, MHS∗ ()
- Duke Center for Atrial Fibrillation, Electrophysiology Section, Cardiology Division, Duke University Medical Center, Durham, North Carolina
- ↵∗Address for correspondence:
Dr. Jonathan P. Piccini, Electrophysiology Section, Duke University Medical Center, Duke Clinical Research Institute, P.O. Box 17969, Durham, North Carolina 27710.
The treatment of atrial fibrillation (AF) has benefited significantly from clinical trials conducted over the past 5 decades. Whereas the first electrocardiogram of AF was published by Einthoven in 1906 (1), the first clinical trial in AF did not occur until the latter half of the 20th century. Among the first (if not the first) double-blind, placebo-controlled randomized clinical trial for the treatment of AF was Byrne-Quinn and Wing’s trial (2) comparing long-acting quinidine bisulfate versus placebo for the maintenance of sinus rhythm after direct current cardioversion. This landmark trial conducted nearly 50 years ago demonstrated that quinidine led to an absolute 33% lower incidence of recurrent AF. Unlike trials today, there was no pre-trial power or sample size determination, no statistical testing of the endpoints, and the study had a withdrawal rate of 18%. Certainly, clinical trials in AF have come a long way since 1970—in both quality and quantity. For example, between 2001 and 2014, the number of clinical trials in AF increased by more than 240% from 60 trials in 2001 to 200 trials in 2014 (3).
Clinical trials are a core focus for guideline recommendations in the management and treatment AF. The latest version of the American College of Cardiology/American Heart Association/Heart Rhythm Society guidelines includes 113 recommendations, yet only 9% (n = 10) of these are supported by Level of Evidence: A (supported by more than 1 randomized trial), in contrast to 51% that are supported by Level of Evidence: C (expert consensus) (3). Even among Class I recommendations, only 12% are supported by Level of Evidence: A. Given the number of unanswered questions in AF care and the limited number of recommendations supported by randomized clinical trials, there is a need to systematically evaluate our experience, progress, and future directions for AF clinical trials.
In this issue of JACC: Clinical Electrophysiology, Patel et al. (4) present the results of a comprehensive analysis of all completed phases II to IV interventional trials enrolling patients with AF or atrial flutter through October 10, 2016. Ultimately, of 67,117 studies, the Patel et al. (4) identified 348 studies (0.5%) that directly studied AF or atrial flutter. Approximately one-half of the studies were open label. Overall, 50% (n = 173) have been published and among the completed trials, 29% remain unpublished. The investigators also found that the majority of studies were funded by industry (54%) whereas a minority were funded solely by governmental agencies (9%).
Across the 348 trials, about a one-third were each focused on anticoagulation, pharmacological therapies, and device or ablation interventions. Recurrent AF was the most common endpoint (45% of trials), while only 5% of endpoints assessed quality of life, functional status, or length of stay. AF burden was an uncommon primary endpoint, utilized in only 8% of antiarrhythmic drug trials and 16% of ablation or device studies. Patel et al. (4) also assessed quality among the trials focused on medical therapies. Using the 3-item Jadad scale (0 to 5), the investigators found that the methodologic quality was high in those studies focused on anticoagulation (3.9 ± 1.4) and other pharmacologic therapies (4.0 ± 1.3). Interestingly, high-quality studies were most commonly funded by industry (77%). When restricting the focus to device or ablation studies, the investigators found that most were randomized (69%); however, no trials used a sham-comparator, and event ascertainment blinding was employed in less than one-half (n = 27 of 61, 44%).
The analysis and critical appraisal of AF clinical trials by Patel et al. (4) provides several important observations. The expansion in the number of clinical trials focused on AF is an important sign of progress in the quest for better informed care of patients with AF. However, the analysis also identifies several areas where we can do better. The fact that 3 of 10 completed AF trials remain unpublished after a median of 59 months following study completion is troubling. Although it may be more challenging to publish a study that does not meet its primary endpoint, such a “negative” trial is rarely without important insight. Given the commitment and sacrifice of the patients who participate in these trials and the enormous financial costs of these studies, all clinical trial results should be shared and disseminated throughout the scientific and clinical communities.
Even though studies focused on anticoagulation and pharmacologic therapies were generally of high quality, as with most endeavors in clinical investigation, there remains room for improvement. Whereas effective blinding in device and ablation trials is challenging, the fact that only 44% of device or ablation studies featured blinding in event ascertainment or adjudication is something we should seek to improve. The recently completed CABANA (Catheter Ablation vs. Anti-Arrhythmic Drug Therapy for Atrial Fibrillation Trial) trial (NCT00911508) did not show definitive superiority of catheter ablation over medical therapy for the composite endpoint of all-cause mortality, disabling stroke, severe bleeding, or cardiac arrest by intention-to-treat analysis. Following the study’s presentation, many have advocated for a comparison of catheter ablation to a sham control. Such a trial would help avert some of the challenges of an open-label trial. In theory, a sham-controlled trial would limit the magnitude of crossover, limit bias in outcomes ascertainment, and would help determine the magnitude of the placebo effect. However, sham-controlled trials are not without significant difficulties, including more difficult enrollment, the potential for biased enrollment (e.g., sicker patients may be reluctant to be randomized to a sham procedure), and concerns over harm during sham procedures, particularly if left atrial instrumentation is performed (5). Ultimately, sham-controlled trials for AF may become a reality despite their difficulties and ethical challenges to help demonstrate or provide a more refined estimate of the magnitude of benefit of catheter ablation for AF.
Another notable finding in the survey by Patel et al. (4) was the limited number of trials that utilized AF burden as a primary endpoint. Only 8% of antiarrhythmic drug studies and only 16% of ablation/device studies included AF burden as the primary endpoint. Consistent with current regulatory recommendations (6), the standard measure of recurrent AF is most often a time-to-event analysis. Time to recurrence is not a useful metric in clinical care, and a binary or time to recurrence approach ignores the complexity of AF. Moreover, it is important to note that rhythm control interventions can lead to improved outcomes, even when there is some degree of residual AF, as was seen in the CASTLE-AF (Catheter Ablation vs. Standard Conventional Treatment in Patients With LV Dysfunction and AF) trial (7). There is a growing body of evidence demonstrating that AF burden is independently associated with many outcomes in AF, including incident heart failure, thromboembolic events, all-cause mortality, and perhaps cognitive impairment (8). A recent scientific statement from the American Heart Association highlights the need for future studies focused on the link between AF burden and outcomes (8). At a minimum, ongoing and future clinical trials should collect and report data on AF burden whenever possible. Moreover, given the existing data on the relationship between AF burden and outcomes, inclusion of AF burden as a secondary endpoint should be strongly considered—if not routine—in AF clinical trials.
Finally, Patel et al. (4) found that only 10% of the AF pharmacotherapy trials studied rate-controlling strategies. This is noteworthy, especially when there are no guideline recommendations for rate control currently supported by Level of Evidence: A (9). Said another way, there is no firm evidence that defines best practice for rate control of AF. Even though rate control is not an area with a high degree of innovation in the current landscape, it should be an area of urgency for the cardiovascular community. Rate control is an important priority for all patients with AF, especially those with long-standing persistent AF. Recent evidence has further highlighted the challenges and limitations of many rate-control therapies, especially in patients with heart failure (10). Moving forward, clinical trials should help determine both optimal methods of rate control and optimal targets for rate control based on large pragmatic outcomes trials.
The journey toward better evidence to guide AF management will continue. Over the past 10 years, there has been an extraordinary amount of progress. After several decades, the AF clinical trial enterprise has performed well and continues to move in the right direction. However, like an athlete interested in improving performance, we need to carefully evaluate the box score. Innovation and improvements in trial design can help achieve higher quality. Future trials should incorporate more clinically relevant endpoints and target more pragmatic questions that address the difficulties encountered in the day-to-day management of AF.
↵∗ Editorials published in JACC: Clinical Electrophysiology reflect the views of the authors and do not necessarily represent the views of JACC: Clinical Electrophysiology or the American College of Cardiology.
Dr. Piccini receives funding for clinical research from Abbott Medical, ARCA Biopharma, Boston Scientific, Gilead, and Janssen Pharmaceuticals; and serves as a consultant to Allergan, Bayer, Biotronik, Johnson and Johnson, Medtronic, Sanofi, and Phillips.
The author attests he is in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Byrne-Quinn E.,
- Wing A.J.
- Barnett A.S.,
- Lewis W.R.,
- Field M.E.,
- et al.
- Patel R.B.,
- Venkateswaran R.V.,
- Singh A.,
- et al.
- Center for Devices and Radiological Health
- Marrouche N.F.,
- Brachmann J.,
- Andresen D.,
- et al.
- Chen L.Y.,
- Chung M.K.,
- Allen L.A.,
- et al.
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.