Author + information
- Received January 1, 2018
- Revision received April 17, 2018
- Accepted April 19, 2018
- Published online July 16, 2018.
- Ravi B. Patel, MDa,
- Ramkumar V. Venkateswaran, MDb,
- Abhayjit Singh, BAb,
- Deepak L. Bhatt, MD, MPHb,
- Gregg C. Fonarow, MDc,
- Rod Passman, MD, MSCEa,
- Javed Butler, MD, MPH, MBAd,
- Clyde W. Yancy, MD, MSca and
- Muthiah Vaduganathan, MD, MPHb,∗ ()
- aDivision of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois
- bHeart and Vascular Center, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts
- cDivision of Cardiology, University of California Los Angeles, Los Angeles, California
- dDepartment of Medicine, University of Mississippi, Jackson, Mississippi
- ↵∗Address for correspondence:
Dr. Muthiah Vaduganathan, Brigham and Women’s Hospital Heart and Vascular Center and Harvard Medical School, 75 Francis Street, Boston Massachusetts 02115.
Objectives This analysis sought to systematically characterize trial-level patterns in atrial fibrillation/atrial flutter (AF/AFL) by using the ClinicalTrials.gov database.
Background Despite an abundance of clinical trials in this field, there is a lack of high-level evidence guiding management of AF/AFL.
Methods We queried all closed, phase II to IV interventional trials registered in the ClinicalTrials.gov database through October 2016 that enrolled patients known to have AF/AFL. Published trials were evaluated for methodological quality, using the 3-item Jadad scale (range: 0 to 5, where 5 = highest quality).
Results The initial search yielded 465 uniquely registered studies, of which 348 directly studied AF/AFL. Of those studies, 173 (50%) were published, enrolling a median of 190 patients from a median of 15 sites. The volume of published trials increased over time (7% prior to 2008 vs. 41% from 2014 to 2016; p < 0.001 for trend). Of the completed trials, 29% remain unpublished. Industry sources accounted for most funding (54%). Recurrence of AF/AFL was the most common endpoint (45%), whereas rates of primary clinical endpoints were low (13%). The mean Jadad score of published trials of pharmacological approaches (n = 112) was 4.0 ± 1.4. Of the 61 AF/AFL trials involving ablation or device therapies, 69% were randomized, 28% were single-arm studies, and patient, proceduralist, and event-ascertainment blinding was used in 16%, 4%, and 44%, respectively.
Conclusions Contemporary trials of AF/AFL are often multicenter and modest in size. The primary study endpoint is commonly recurrence of arrhythmia, even in high-quality and late-phase trials. Although methodological quality is high in trials of pharmacologic approaches, trials of AF/AFL ablation and device therapies variably employ randomization and blinding.
Atrial fibrillation/atrial flutter (AF/AFL) has evolved into a global epidemic, affecting more than 30 million individuals worldwide (1). Although the overall prevalence of AF/AFL is estimated to be 1%, its burden in older patients ranges from 5% to 9%, and its incidence is expected to double by 2030 (1–3). Furthermore, despite effective preventive therapies against stroke and systemic embolism, nearly half of all deaths in AF/AFL remain cardiovascular in nature (4–6). Although a number of AF/AFL devices and ablation are used in clinical practice, the quality of evidence supporting their use is uncertain. Recognition of the clinical and economic burden of AF/AFL has fueled research interest in management of this condition. Despite this growth in trial-level data, an analysis of the American Heart Association/American College of Cardiology/Heart Rhythm Society guidelines of AF/AFL management revealed that: 1) level of evidence (LOE) A accounted for a minority of all recommendations in the latest guidelines (7); and 2) the number of LOE A recommendations has not increased over a 13-year timespan (8).
There is limited understanding of the major factors that drive the lack of changes in these guidelines, including variability in study quality, inadequate blinding, lack of validated endpoints, and issues with publication and data transparency. Furthermore, whether similar adverse trends are apparent in the regulatory space with respect to new drugs and devices approved for clinical use is unclear. As such, we aimed to: 1) summarize contemporary trial-level patterns in AF/AFL by using publicly available data from the ClinicalTrials.gov registry; 2) systematically evaluate the methodological quality of these trials; and 3) identify AF/AFL drugs and devices recently approved by the U.S. Food and Drug Administration (FDA).
ClinicalTrials.gov search strategy
We identified uniquely registered AF/AFL studies updated prior to October 10, 2016 in the ClinicalTrials.gov database by using the following limits to the primary search strategy: interventional study design, closed enrollment, adult/senior population (≥18 years of age), and trial phases (II to IV). The terms “atrial fibrillation” and “atrial flutter” were searched as separate queries, and duplicate entries were subsequently deleted.
Published reports and data extraction
Full-text publications of these studies were then identified by searching the title, principle investigator, and ClinicalTrials.gov unique identifier in PubMed and MEDLINE in duplicate by 2 independent authors (R.V.V. and A.S.). Trials were included if enrollment criteria required previously diagnosed AF/AFL. When available, the following data were abstracted from the full text of published trials: number of patients per trial, number of sites per trial, region of enrollment, year of publication, AF category of study, type of primary endpoint, and method of AF/AFL detection (only if AF/AFL recurrence or AF/AFL burden was a primary endpoint). Data regarding study design (phase of trial, patient allocation, presence of parallel study arm, blinding strategy, Data and Safety Monitoring Board oversight) were initially abstracted from ClinicalTrials.gov and subsequently verified in all full-text publications. Primary endpoints were classified as clinical (mortality, hospitalization, stroke/systemic embolism, myocardial infarction), intermediate (functional status, quality of life [QOL], length of hospital stay), safety, or surrogate. In the case of co-primary endpoints, both endpoints were tabulated. In addition, information regarding study design was abstracted from ClinicalTrials.gov for all studies that were identified as completed but unpublished as of October 10, 2016 (date of ClinicalTrials.gov query).
Assessment of clinical trial quality
All published trials studying anticoagulation or pharmacological therapies were evaluated for methodological rigor by using the validated Jadad score, a 3-item scoring system that grades studies on the following characteristics: randomization, blinding strategy, and disclosure of patient disposition (e.g., patient withdrawals or dropouts) (9). Two investigators (R.B.P. and M.V.) independently graded each published trial, using the 3-item Jadad scale (range: 0 to 5, where 5 is highest quality). Given unique challenges to studying ablation approaches or device therapies, we did not use the composite Jadad scoring system, and instead detailed each trial element (randomization, choice of comparator, and blinding strategy) individually to assess clinical trial quality.
Identifying new FDA-approved drugs and devices
We used publicly available data from the FDA to identify drugs and devices approved for AF/AFL treatment over the past 5 years (2012 to 2016). We first queried all new drugs or therapeutic biological products approved by the FDA Center for Drug Evaluation and Research as new molecular entities (NMEs) or new biologics over the last 5 years. These applications often pertain to novel therapies containing entities that have not been previously evaluated by the FDA. We then searched original premarket approval(s) (PMA) for all class III medical devices with decision dates from January 1, 2012, to December 31, 2016. Two investigators (R.B.P. and M.V.) independently identified all NMEs for drugs and original PMAs for devices specific to AF/AFL.
Selection of primary endpoints and methods of AF/AFL detection were categorized separately for published trials focused on anticoagulation, antiarrhythmic/rate-control therapies, and ablation devices. Interrater reliability of Jadad scores was assessed using Lin’s concordance coefficient (ρc). Nonparametric tests of trend were used to examine the association between number of registered clinical trials and year of publication (in blocks of 4 years). Two-sided p values <0.05 were considered statistically significant. All statistics were performed using STATA version 14.0 software (StataCorp, College Station, Texas).
Studies registered with ClinicalTrials.gov
There was a total of 67,117 phase II to IV interventional studies with closed enrollment registered in the ClinicalTrials.gov database, of which we identified 465 uniquely registered studies related to AF/AFL (0.7% of all ClinicalTrials.gov trials with similar search limits). After initial screening, we excluded 117 (25%) trials because they did not include patients with a baseline diagnosis of AF/AFL. Of the remaining studies (n = 348), we found 75 (22%) had been terminated or withdrawn and 100 (29%) had not been published. Of the 100 unpublished studies, 70 trials were completed but unpublished at the time of data abstraction from ClinicalTrials.gov. Full-text published reports for 173 (50%) trials were available for final analyses (Figures 1 and 2), studying 167,803 patients enrolled from 12,744 sites, in total (Table 1).
Of the 173 published studies registered with ClinicalTrials.gov, there was a significant increase in volume over time (n = 12 prior to 2008 vs. n = 70 from 2014 to 2016; p for trend <0.001) (Table 1). Industry sources accounted for most funding (54%), whereas government agencies funded 9% of trials. The remainder of trials were funded by nonprofit organizations (23%), or were jointly funded (14%). Less than 25% of trials enrolled patients exclusively from North America, and more than half of all trials used open-label trial designs.
Overall, 173 trials evaluated 188 primary endpoints, of which 25 (13%) were clinical endpoints; 10 (5%) were measurements of functional status, QOL, or length of stay; and 28 (15%) were safety endpoints. The remaining endpoints were metrics of AF/AFL acute conversion, recurrence, or burden (n = 85 [45%]) or other surrogate endpoints (n = 40 [21%]).
Approximately one third of trials each studied anticoagulation strategies, pharmacological therapies, or ablation or device approaches, respectively. Of the 61 trials of pharmacotherapies, only 6 (10%) directly studied rate-controlling strategies. The target heart rate in these studies ranged from 80 to 110 beats/min at rest. A minority of all trials (3%) limited inclusion to patients with specific comorbidities (3 trials of heart failure, 1 trial of chronic kidney disease, 1 trial of hypertrophic cardiomyopathy). Primary clinical endpoints were uncommon across areas of study, representing 33% of anticoagulation studies and 7% of studies evaluating pharmacological therapies or ablation/devices (Table 2). Recurrence of AF/AFL was the most commonly selected endpoint for antiarrhythmic/rate control (41%) and ablation/device studies (56%), regardless of phase of study (15% of phase II, 52% of phase III, 41% of phase IV antiarrhythmic/rate-control trials; and 50% of phase II, 64% of phase III, 50% of phase IV ablation/device trials). Burden of AF/AFL was an uncommon primary endpoint (8% of antiarrhythmic/rate-control studies and 16% of ablation/device studies) and required long-term rhythm monitoring. Studies with a primary endpoint of burden of AF/AFL were commonly phase II trials (53%), industry-funded (47%), and increased over time (0% prior to 2008, 7% between 2008 to 2010, 93% between 2011 and 2016). The rhythm monitoring methods used to determine recurrence of AF/AFL varied considerably, ranging from electrocardiograms to admissions for telemetry monitoring (Table 2).
Completed but unpublished trials
There was a total of 70 completed but unpublished trials, which were completed a median of 59 months (25th to 75th percentile 24 to 96) prior to data abstraction from ClinicalTrials.gov and represented 29% of all completed trials (Table 1). The majority used open-label design (61%) and were phase III or IV (64%) and industry-funded (66%). Notably, none of these trials were government-sponsored, and a Data and Safety Monitoring Board was part of a minority of study designs (31%). Similar to published trials, a minority of studies used definitive clinical endpoints (4%) (Table 2). Only 30% of unpublished trials had reported limited results posted through either ClinicalTrials.gov or another platform (e.g., meeting presentation, FDA online resource, and others).
Grading quality of published clinical trials of pharmacological approaches
The mean Jadad scores for the 112 trials evaluating anticoagulation strategies and other pharmacological therapies rated by 2 independent investigators were high (3.9 ± 1.4 and 4.0 ± 1.3, respectively [range: 0 to 5, where 5 is the highest quality]). Trial quality was judged with high concordance between the 2 evaluators (ρc = 0.96; 95% confidence interval: 0.95 to 0.97). There were 16 trials (14%) that received a Jadad score of 2 or less by at least 1 investigator. In contrast, there were 62 trials (55%) that were identified by at least 1 investigator as Jadad 5, which represented double-blind, randomized controlled trials with adequate reported detail regarding patient disposition. These select “high-quality” trials were modest in size (median, 428 patients per trial; 25th to 75th percentile: 164 to 899), more often phase III or IV (68%), and commonly funded by industry (77%). Primary endpoints of trials identified with a Jadad score of 5 were commonly recurrence of AF/AFL, burden, or acute conversion to sinus rhythm (55%), whereas clinical endpoints were less frequent (18%).
Assessing methodological quality of trials of ablation and device therapies
Of the 61 AF/AFL trials involving ablation or device therapies, 69% (n = 42) included randomized designs, and 72% (n = 44) included 2 or more treatment arms (the remaining 17 trials were single-armed). Among studies with a comparator arm, an alternative ablation strategy was most common (n = 28), followed by drug therapy comparator (n = 16). No trials used a sham procedure as a comparator. Blinding strategies were highly variable among ablation trials. Relatively few ablation trials included patient blinding (16% [n = 10]) or proceduralist blinding (7% [n = 4]). However, event ascertainment blinding was used in nearly half of the ablation/device trials (44% [n = 27]). No trials used a blinding index to verify the effectiveness of blinding.
FDA-Approved drug and device therapies for AF/AFL 2012 to 2016
From 2012 to 2016, there was a total of 174 approved unique NMEs (drugs/biological products), of which 23 were indicated for the treatment of cardiometabolic disease or its risk factors (i.e., hyperlipidemia, diabetes mellitus, obesity, metabolic syndrome) (10). Only 3 therapies were approved for indications related to AF/AFL during this timeframe: apixaban, edoxaban, and idarucizumab (Table 3); all were anticoagulation-focused, and 2 of 3 drugs were approved based on results from double-blind randomized clinical trials with clinical primary endpoints. Of the 167 approved devices (original PMAs) from January 1, 2012, to December 31, 2016, 59 were approved through the cardiovascular advisory committee (11). Only 5 were indicated for the treatment of AF/AFL, of which 4 were new ablation catheters or systems approved on the basis of clinical trials demonstrating safety, procedural success, or reduction in recurrent AF/AFL alone (Table 3).
We systematically characterized 348 trials of AF/AFL registered with ClinicalTrials.gov, independently evaluated methodologic quality of the 173 published studies, and reviewed contemporary drugs and devices approved for AF/AFL over the past 5 years. Our analysis highlights several trial-level patterns: 1) even among high-quality and late-phase trials, recurrent AF/AFL is a commonly used endpoint, and the method of its detection lacks clear standardization; 2) more than 25% of the completed trials of AF/AFL remain unpublished, despite full data acquisition; 3) methodological quality of published trials of AF/AFL pharmacotherapies appears robust; and 4) trials of AF/AFL ablation and devices variably use randomization and blinding. In aggregate, our findings may offer insight into both the limited number and the overall lack of changes in guideline LOE A recommendations for AF/AFL management and the relatively low number of recent drug and device approvals for AF/AFL by the FDA.
Despite the overall global burden of AF/AFL, LOE A recommendations accounted for a minority of all recommendations for the management of AF/AFL in the latest guidelines (7); mortality remains substantial even in well-treated patients (4,12); and significant variations exist in treatment strategies across practices and providers (13,14). As such, there is an unmet need for novel drug and device therapies to improve cardiovascular and noncardiovascular outcomes in this population. Although we recognize that a number of these patterns in the quality of evidence base appear common across therapeutic areas in cardiovascular medicine (15,16), we highlight important opportunities to promote application of more rigorous and homogenous trial methodology specific to AF/AFL in the future.
The ClinicalTrials.gov registry, a Web-based platform maintained by the National Institutes of Health, provides a unique opportunity to systematically study the current research pipeline of select AF/AFL clinical trials (17). Although recent regulatory changes have clarified and expanded the specific requirements of clinical trial reporting (18), compliance with the ClinicalTrials.gov registry remains low across disciplines (19). As such, we linked the ClinicalTrials.gov database to a separate PubMed/MEDLINE query to clarify publication status of each identified trial. Although this approach was not intended to be exhaustive, this analysis represents a first step in characterizing the footprint of the AF/AFL clinical trial enterprise and we believe the observed patterns may be informative in future trial design.
Is it the disease or the design?
AF/AFL has proven to be a difficult disease to study on the trial level. AF/AFL represents a heterogenous clinical syndrome with varying natural history based on overall arrhythmia burden, presence of symptoms, and certain comorbidities. The disease may be sporadic and difficult to capture with short-term rhythm monitoring. Endpoints are varied and span domains of QOL and stroke/systemic embolism. Although AF/AFL has benefited from significant innovation and progress in the development of various ablation and device therapies, trials evaluating technical success, safety, durability, and clinical outcomes of these approaches have varied in scientific rigor.
Variable trial quality
Trial quality is highly variable across contemporary AF/AFL trials. Average methodological quality was high for trials of pharmacological approaches with more than half of studies using double-blind, randomized designs. For instance, mean Jadad scores for these trials were comparable or even higher than that observed in other therapeutic areas (20,21), even when restricted to those studies published in high-tier journals (22). However, even among rigorously designed trials, overall sample sizes were modest, and only a minority of primary endpoints were clinical. In contrast to trials of pharmacological approaches, more than a quarter of ablation or device studies included single-treatment arms, and a minority of ablation or device studies incorporated elements of patient or proceduralist blinding. Indeed, of the trials supporting ablation or device therapies recently approved for use by the FDA, 1 was nonrandomized and 2 were unblinded.
Publication of completed trials
As in other therapeutic areas (23-25), the registration, prompt reporting, and disclosure of trial results remain suboptimal in AF/AFL. We estimate that, although registration of trials has gradually increased in recent years, more than a quarter of completed AF/AFL studies registered with ClinicalTrials.gov remain unpublished, of which, relatively few have posted limited results. Similar nonpublication rates of completed studies have been reported across fields of medicine (∼30% to 35%) (23–25). Concerted, broad-scale efforts from members of the academic community, study sponsors, and regulatory bodies will be necessary to bring acquired data quickly into published and accessible form.
Establishing high-quality AF/AFL clinical trial data: Next steps
The goal of AF/AFL therapeutic programs is to place better quality information in the public domain to inform guideline recommendations, regulation, and clinical decision-making. New strategies and efforts are required to refine the design and conduct of emerging AF/AFL trials with respect to power, endpoint selection, and blinding, as outlined in a recent expert consensus statement by Calkins et al. (26) and previous FDA regulatory guidance (27).
Power and sample size
We observed that, on average, published trials of AF/AFL enroll ∼200 patients from 15 sites, likely reflecting sample sizes sufficient to detect differences in endpoints of arrhythmia recurrence. The emergence of multicenter trials supporting the AF/AFL trial enterprise is encouraging. Larger, multicenter trials with longer follow-up times should continue to be designed to improve the overall robustness and generalizability of trial results. However, given rapid evolution in AF/AFL ablation technology and approaches, appropriate trial sizes should be planned to facilitate timely completion such that background therapies and comparator arms selected at the start of a trial remain relevant at trial completion.
Due to the relatively low event rates in AF/AFL (∼5% annualized mortality) (4), trials designed with solely clinical endpoints require both considerable size and follow-up duration. For instance, the CABANA (Catheter Ablation vs Anti-arrhythmic Drug Therapy for Atrial Fibrillation Trial; NCT00911508) trial with enrollment of more than 2,200 patients was initiated in August 2009 and was presented in May 2018. Such an endeavor by the CABANA investigators has proved challenging, and thus, the current trial infrastructure may not represent a sustainable model for future AF/AFL trials designed to examine clinical endpoints. Future trials with primary clinical endpoints will require not only broad commitment from study participants, investigators, and coordinators but also dedicated support over a potentially lengthy trial duration from funding bodies. AF/AFL trials with clinical primary endpoints may be conducted along shorter timelines in enriched higher risk subsets of patients, such as patients with prevalent comorbid heart failure, as studied in CASTLE-AF (Catheter Ablation vs. Standard Conventional Treatment in Patients with LV Dysfunction and AF) trial (28). Overall, such contemporary examples of trials driven by clinical endpoints offer promise for future AF/AFL trials.
Patient-reported outcomes, including health-related QOL, remain important ancillary measures of therapeutic success. However, patient-reported outcomes are especially prone to bias (related to placebo and nocebo effects) if proper blinding is not also undertaken. Factors beyond AF/AFL recurrence, including general mental and physical health, may be important determinants of arrhythmia-related symptoms in patients with AF/AFL referred for ablation (29). Indeed, patient-reported outcomes poorly correlate with physician-assessed AF/AFL outcomes after ablation, and symptom benefit may be observed even with ongoing documented AF/AFL burden (30). AF/AFL-specific QOL metrics have been developed but require further validation.
In our trial-level analysis we found that use of a single time point or periodic assessments of recurrence of AF/AFL as primary endpoints remain common, even in late-phase investigation. Recurrent AF/AFL is easily studied, may correlate with certain metrics of QOL, and facilitates trial efficiency. Indeed, the FDA has supported freedom from AF/AFL as an efficacy endpoint for trials of AF/AFL ablation therapy (27). However, the validity of this endpoint has been questioned, as isolated recurrence of AF/AFL does not capture true arrhythmia burden (31). In addition, recurrence of AF/AFL has not been rigorously validated as a true surrogate for more durable clinical endpoints (i.e., incident heart failure, cardiovascular hospitalization, cardiovascular mortality). Conversely, AF/AFL burden offers promise as a clinically meaningful endpoint but is met with its own challenges. It is intuitive that total AF/AFL burden would correlate with adverse long-term clinical outcomes, however this association has not been rigorously studied, as noted by the FDA in a recent advisory statement (27). A recent think tank report by Piccini et al. (32) supported the utility of implantable cardiac monitors in quantifying AF burden in ongoing and future trials; however, the authors cited concerns regarding excess costs and lack of understanding of the true background of arrhythmia in a healthy population. In addition to ongoing uncertainty regarding selection of arrhythmia endpoints, there is no current standardized modality for assessing AF recurrence. Although the FDA suggests periodic Holter monitoring may serve this purpose, other modalities, including transtelephonic monitors and serial electrocardiograms are also deemed adequate (27).
Blinding and feasibility of sham controls
The lack of adequate blinding or use of sham controls have made broad applications of emerging AF/AFL ablation techniques or devices challenging (33,34). Although blinding certainly adds cost, complexity, and effort to clinical trial design, it may be necessary to limit bias and misattribution of placebo or nocebo effects (35). For instance, the blinded SYMPLICITY HTN-3 (Renal Denervation in Patients With Uncontrolled Hypertension) trial (36) failed to demonstrate a reduction in systolic blood pressure in patients with resistant hypertension with renal artery sympathetic denervation compared with a sham control, despite previous nonrandomized and unblinded trial experiences suggesting its benefit. Sham-controlled designs have been used successfully in other cardiovascular domains including percutaneous coronary intervention for stable angina (37), pacemakers for vasovagal syncope (38), and patent foramen ovale closure for migraine headaches (39). Certain aspects of AF/AFL ablation, including the transseptal puncture, may introduce patient risks and raise specific ethical issues related to sham protocols. Recent trials, such as REDUCE LAP-HF I (Randomized Trial to Reduce Elevated Left Atrial Pressure in Heart Failure) (40), have navigated these issues and minimized patient risks by using transesophageal echocardiographic imaging of the atrial septum (rather than transseptal puncture) in the sham arm, in addition to the insertion of femoral venous sheaths and administration of conscious sedation. The lack of transseptal puncture and complexity regarding periprocedural anticoagulation management in the sham protocol designs of AF/AFL ablation trials may potentially introduce bias in the sham control group, especially in trials involving QOL endpoints. Sham controls in procedural trials within cardiology and electrophysiology continue to be of interest to the scientific community, but challenges remain in their optimal and safe application.
We recognize that sham controls may be challenging to operationalize, inappropriate, or non-ideal in certain scenarios. Regardless, well-executed controlled studies should use adequate blinding, especially given the high utilization of surrogate endpoints. Although patient- and operator-blinding may not be feasible from a cost or pragmatic perspective, blinding of event ascertainment in AF/AFL ablation/device studies should be undertaken (41). When applicable, a simple blinding index (36) may be helpful to confirm effectiveness of patient blinding.
Assessing methodological quality in future interventional AF/AFL trials
Given the evolution and complexity of contemporary AF/AFL procedural trials, traditional metrics to evaluate methodologic quality of trials (e.g., Jadad scale) may not fully capture the salient features of interventional AF/AFL trials. Methodologic quality evaluation of ablation and device trials may require a more nuanced and structured checklist, incorporating elements of randomization, blinding (patient, proceduralist, event ascertainment), use of a comparator group (including sham controls), and patient disposition (number of patients enrolled/consented, achieved access, completed procedure, and deployed device, if relevant).
This ClinicalTrials.gov-based analysis is subject to certain limitations. Compliance with registration and reporting of ongoing or completed cardiovascular clinical trials remain poor (19). Trials, especially those conducted outside the United States, may be registered with other data repositories. Data extracted from ClinicalTrials.gov and the primary publications rely on investigator- or sponsor-reported information. Given variable registration of phase 0 or 1 studies, we did not evaluate these trials in this analysis. It is plausible that our search strategy missed a portion of studies that were in fact related to AF/AFL. We did not count publications of clinical trials in abstract form alone, given variable abstract data quality and content, which may accentuate publication bias. Lastly, since the completion of our systematic search, several clinical outcomes trials of AF/AFL have been published or presented. We are hopeful that increased focus on clinical endpoints will be accompanied by ongoing improvements in other design elements and in overall clinical trial conduct.
We provide a contemporary, comprehensive survey of registered AF/AFL studies within the ClinicalTrials.gov database and recently FDA-approved AF/AFL therapies. Our analysis highlights the fact that AF/AFL trial quality is variable, which may be due to difficulty in studying AF/AFL, given its inherent heterogeneity and challenges related to the evaluation of ablation and device therapies in this space. In addition, we found that primary endpoints in AF/AFL trials are commonly recurrence of arrhythmia, which currently lacks a standardized modality for assessment. These trial-level patterns may account for the relative lack of high-level guideline-supported recommendations and the low number of FDA-approved drugs and devices. Although AF/AFL will remain a complex disease to study, rigorous study designs incorporating adequate blinding and sham controls, utilization of both standardized and validated endpoints specific to AF/AFL, and increased enforcement of trial transparency and data reporting offer promise for the AF/AFL clinical trial enterprise. Further, high-quality, appropriately-powered trials are necessary to define therapies that may improve quality and length of life in AF/AFL.
COMPETENCY IN MEDICAL KNOWLEDGE: Despite the growing burden of AF/AFL, level of evidence A recommendations account for a minority of all recommendations guiding its management. Only 3 new drugs and 5 devices have been approved by the U.S. FDA over the last 5 years for AF/AFL.
TRANSLATIONAL OUTLOOK: Acute conversion, recurrence, or burden of AF/AFL represents the primary endpoint in nearly half of all contemporary AF/AFL trials, including high-quality and late-phase investigations. Future studies should continue to incorporate clinically-meaningful endpoints and use standardized modalities of rhythm monitoring. Methodological quality was high for published trials of AF/AFL pharmacotherapies. Patient- and operator-blinding is infrequently applied in trials of AF/AFL ablation/device therapies.
Dr. Bhatt has served on advisory boards of Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, Regado Biosciences; and is on the board of directors of Boston VA Research Institute, Society of Cardiovascular Patient Care; Chair, American Heart Association Quality Oversight Committee; and is on the data monitoring committees of Cleveland Clinic, Duke Clinical Research Institute, Harvard Clinical Research Institute, Mayo Clinic, Population Health Research Institute; and has received honoraria from American College of Cardiology (Senior Associate Editor, Clinical Trials and News), Belvoir Publications (Editor in Chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Harvard Clinical Research Institute (clinical trial steering committee), HMP Communications (Editor in Chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (Guest Editor and Associate Editor), Population Health Research Institute (clinical trial steering committee), Slack Publications (Chief Medical Editor, Cardiology Today’s Intervention), Society of Cardiovascular Patient Care (Secretary/Treasurer), WebMD (CME steering committees), Clinical Cardiology (Deputy Editor), NCDR-ACTION Registry Steering Committee (Chair), VA CART Research and Publications Committee (Chair); and has received research funding from Abbott, Amarin, Amgen, AstraZeneca, Bristol-Myers Squibb, Chiesi, Eisai, Ethicon, Forest Laboratories, Ironwood, Ischemix, Lilly, Medtronic, Pfizer, Roche, Sanofi Aventis, The Medicines Company; and has received royalties from Elsevier (Editor, Cardiovascular Intervention: A Companion to Braunwald’s Heart Disease); and has served as site co-investigator for Biotronik, Boston Scientific, St. Jude Medical; and is a trustee of American College of Cardiology; and has received unfunded research support from FlowCo, PLx Pharma, and Takeda. Dr. Fonarow has relationships with, Novartis, Amgen, Bayer, Gambro, Medtronic, and Janssen; and holds the Eliot Corday Chair of Cardiovascular Medicine, UCLA, and is supported by the Ahmanson Foundation, Los Angeles, CA. Dr. Passman has received consulting/speaker fees for Medtronic and Biotronik; and royalties from UpToDate. Dr. Javed Butler has received research support from U.S. National Institutes of Health and European Union; and has consulted for Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, CVRx, Janssen, Luitpold Pharmaceuticals, Medtronic, Merck, Novartis, Relypsa, Vifor Pharma, and ZS Pharma. Dr. Vaduganathan is supported by the U.S. National Heart Lung Blood Institute T32 postdoctoral training grant T32HL007604. All other authors have reported that they have no other relationships with industry relevant to the contents of this paper to disclose.
All authors attest they are 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.
- Abbreviations and Acronyms
- atrial fibrillation/atrial flutter
- level of evidence
- quality of life
- new molecular entities
- Received January 1, 2018.
- Revision received April 17, 2018.
- Accepted April 19, 2018.
- 2018 American College of Cardiology Foundation
- Chugh S.S.,
- Havmoeller R.,
- Narayanan K.,
- et al.
- Gomez-Outes A.,
- Lagunar-Ruiz J.,
- Terleira-Fernandez A.I.,
- Calvo-Rojas G.,
- Suarez-Gea M.L.,
- Vargas-Castrillon E.
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Barnett A.S.,
- Lewis W.R.,
- Field M.E.,
- et al.
- ↵U.S. Food and Drug Administration. New Drugs at FDA: CDER’s New Molecular Entities and New Therapeutic Biological Products. Available at: https://www.fda.gov/Drugs/DevelopmentApprovalProcess/DrugInnovation/. Accessed April 16, 2018.
- ↵U.S. Food and Drug Administration. Premarket Approval (PMA). Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPMA/pma.cfm. Accessed April 16, 2018.
- Pokorney S.D.,
- Piccini J.P.,
- Stevens S.R.,
- et al.
- Hsu J.C.,
- Maddox T.M.,
- Kennedy K.F.,
- et al.
- Huisman M.V.,
- Rothman K.J.,
- Paquette M.,
- et al.
- Subherwal S.,
- Patel M.R.,
- Chiswell K.,
- et al.
- Gautier I.,
- Janiaud P.,
- Rollet N.,
- et al.
- Pica N.,
- Bourgeois F.
- Chen Y.P.,
- Liu X.,
- Lv J.W.,
- et al.
- Calkins H.,
- Hindricks G.,
- Cappato R.,
- et al.
- ↵Clinical Study Designs for Surgical Ablation Devices for Treatment of Atrial Fibrillation: Guidance for Industry and Food and Drug Administration Staff. February 15, 2013. U.S.D.H.H.S. FDA Center for Devices and Radiological Health Cardiac Electrophysiology Devices Branch Division of Cardiovascular Devices Office of Device Evaluation. Available at: https://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM182017.pdf. Accessed April 16, 2018.
- Marrouche N.F.,
- Brachmann J.,
- Andresen D.,
- et al.
- Charitakis E.,
- Barmano N.,
- Walfridsson U.,
- Walfridsson H.
- Björkenheim A.,
- Brandes A.,
- Magnuson A.,
- et al.
- Piccini J.P.,
- Clark R.L.,
- Kowey P.R.,
- et al.
- Ozeke O.,
- Cay S.,
- Ozcan F.,
- Baser K.,
- Topaloglu S.,
- Aras D.
- Olshansky B.
- Tobis J.M.,
- Charles A.,
- Silberstein S.D.,
- et al.
- Feldman T.,
- Mauri L.,
- Kahwash R.,
- et al.