FDA retains trial protocols for atrial fibrillation devices
|
|
The FDA’s Circulatory System Devices Panel turned down proposed alternatives to the randomized clinical trial process for ablation devices for refractory atrial fibrillation.
“As we see the technology evolve for device-based therapy for atrial fibrillation and transition from medical therapies to interventional therapies, it is important that we discover and develop methods to adjudicate the techniques and the devices so that we can be certain of the risk-benefit ratios, and we can be certain of the real evidence of efficacy and safety,” Panel Chairman Clyde W. Yancy, MD, medical director of the Baylor Heart and Vascular Institute, said in an interview. “Our intent is to promote the development of intervention and promote the development of devices, but to do so responsibly and to do so in a manner that we can be reasonably confident that the benefits will be realized as they are intended.”
At the meeting, industry representatives expressed concern over the current – and mainstay – system and discussed alternatives to evaluate devices.
Industry problems with trials
One of the major concerns raised by industry representatives is the difficulty in randomizing and selecting patients for trials. An FDA executive summary noted that since the FDA requires that different types of AF be studied separately, patients screened for one form of AF may not have the specific form of AF required for the study. This leads to large numbers of patients screened for enrollment with few selected, thus greatly increasing the time and expense of conducting trials.
Another common problem is the difficulty in blinding studies. Patients who are enrolled in a control arm often drop out of studies when they feel there is a chance they may be randomized into an ineffective medical treatment.
“We respect the fact that device trials are not drug trials. The most difficult thing about device trials is that you can’t blind them,” Yancy said. “It is especially awkward in the setting of atrial fibrillation. The drugs have side effects, and so the control population has mediocre outcomes. It’s very difficult to feel at ease randomizing a patient to the control arm when you theorize on reasonable pilot data and reasonable experience that the intervention is likely to be better, and patients have even expressed the same disquiet with what is perceived to a loss of therapy.”
Panel keeps randomized trials
Despite drawbacks, the FDA panel decided to keep randomized clinical trials as the gold standard of clinical evidence for AF device approval. Their benefits, many members felt, outweigh their inconveniences.
“I’d be in favor of keeping the randomized clinical trials,” Douglas Zipes, MD, distinguished professor at the Indiana University School of Medicine, and Electrophysiology and Arrhythmia Disorders section editor of Cardiology Today’s Editorial Board, said in an interview. “It’s not as if patients are not having AF ablation. Obviously, these are off-label use ablations, so we’re not depriving them of a therapy that’s potentially beneficial. Unless we do prospective randomized clinical trials, we won’t have scientifically supportable conclusions.”
Although some possibilities and alternatives were discussed at length during the meeting, the attendees and panelists ultimately left in concurrence, according to Yancy.
“The FDA, industry and the professional organizations all suggested and made it clear that for the purposes of this sphere of atrial fibrillation, we really need to respect the process of the randomized control trial,” Yancy said.
“That doesn’t mean that there won’t be a scenario or a device that comes along that is so focused or has such a narrow window that we won’t have to be creative to work in concert to get that device approved, but I think it was a fundamental statement that the randomized trial really has authority in the hierarchy of evidence and we want to respect that,” he said.
Douglas Zipes, MD
Director of the Johns Hopkins Ciccarone Center
Print Version
Tell-a-Friend comments powered by Disqus
