Can you share your perspective on the recent FDA panel meeting?
The panel convened by the FDA, held June 19-20th, was tasked with providing the FDA answers to specifically focused questions.1,2 These questions focused around the recent meta-analysis3 of pivotal studies performed as randomized clinical trials that revealed the possibility of a long-term mortality signal in the combined data. There was a large amount of information provided to the panel by the FDA itself4, a number of industry representatives, and a large number of public entities and societies hoping to provide the panel with insight into the clinical arena and the studies performed both before and after the analysis. The significant questions posed focused around the possible presence of a signal and questions regarding possible mechanism. As part of the data presented, the FDA presented a more complete meta-analysis4 (section 3.2) it performed of updated data provided by industry. In this analysis, the FDA noted there is still a hazard ratio indicative of increased mortality in the patients treated with medicated devices. Additionally, Dr. Katsanos presented an update to his previous meta-analysis that continues to show increased mortality. Industry representatives provided enhanced information on all trials that revealed a narrowing of the differences between the groups, and in all instances, no specific difference between the medicated devices and non-medicated devices. The potential signal was only revealed when compiling the data. Industry representatives pointed out the problems with combining these trials with heterogeneous populations, small numbers, and varied definitions of even basic endpoints.
Information regarding very large data sets including the Vascular Quality Initiative (VQI), Medicare, Optum, and others were presented to provide a picture of real-world effects of these devices. For these studies, hazard ratios were centered around unity, implying no effect of paclitaxel-eluting agents. The panel members pointed out the issues with bias that can be present in large registry data sets. Overall there was good discussion around the patient and physician concerns as well. The panel concluded there was a signal, although the data was not strong. They also could not identify a mechanism and were not convinced this was a class effect.
The panel heard from representatives of all major societal database representatives, who informed the panel that database utilization may be a way forward and would allow data collection in a prospective fashion, but ultimately a new trial may be needed. Trial size for this type of project could vary from 3,000 patients to over 40,000 patients, depending upon trial design and assumptions regarding the strength of the signal. The panel appeared to be convinced of the benefits of paclitaxel-eluting devices and most panel members wanted these devices available for clinicians in a manner that would to allow them to educate and offer options to patients. It would seem from the panel deliberations and advice provided to the FDA that the FDA will likely need to work with industry to continue to collect information on completed trials, arrange better follow-up in trials already underway and extend those trials out to 5 years, and perhaps work with societies to devise a plan for collecting prospective data through registries to gain insight into the problem. Industry representatives asked the FDA to provide reassurance to clinicians regarding the safety and efficacy of these devices.
How will these issues and concerns affect your practice?
Medicated devices have become my own standard of care for patients with symptomatic peripheral arterial disease (PAD) of the lower extremities. I am thankfully practicing in a health system that has worked with clinicians to allow them to use these devices where we feel they are clinically indicated. Since a large portion of my practice consists of patients with complex lesions, I have continued to use these devices extensively and will plan to continue using them as I currently do. For most clinicians, I believe the information presented to the panel should provide reassurance. In large data sets from the Medicare fee-for-service population and in Optum’s managed care data of extremely large data sets, with over 200,000 individuals, and with and without propensity matching, there was no difference in the mortality outcomes of intervention with medicated devices. These devices appear to be working safely. Katsanos pointed out in his study that patients with few to no risk factors appear to be the group potentially affected by the long-term mortality he noted, but in reality, this is a small minority of the patients treated with these problems.
What questions remain to be answered?
There is primarily one question that needs to be answered and it is truly the question that initiated this entire process. Is there really a signal in the data or is this mostly “noise”? After the immense amount of information presented, the panel members could only conclude that there was a “signal” of increased mortality and could not conclude there was any strength to that signal. Moving forward, this will need to be fully refuted in order to bring this to rest. I think we are all hopeful that this would not require an extensive repeat trial, but data from ongoing trials, and perhaps a look at subsets within Medicare data could be helpful. I do think there is likely a role for societal registries and perhaps a consortium of industry, societies with databases, and clinicians could come together to provide a way to get an answer in a relatively short period of time.
Are there other experiences (outside of the FDA panel meeting) that influenced your thinking on this issue?
Clinicians who have used these medicated devices understand the value they bring in reducing the rates of recurrent stenosis. It is also clear that the risk profile for the majority of symptomatic peripheral arterial disease patients is very high. In light of these issues, I have not felt that I am putting patients at significantly increased risk by offering them a valuable adjunct to care for their lower-extremity disease. The large dataset analyses, including assessments from VQI and Medicare, make it clear that if there is increased risk, it is in a very small minority of patients and for most patients, these devices are safe and very effective.
Disclosure: Dr. Daniel Clair reports that he serves as an advisor for Boston Scientific and Medtronic, and a consultant and DSMB member for BD. Compensation for these services is paid to the medical group for which he works.
Dr. Daniel Clair can be contacted at firstname.lastname@example.org.
Reprinted with permission from Vascular Disease Management June 2019;16(6). Available online at vasculardiseasemanagement.com. Copyright 2019 HMP Global.
- FDA Panel Questions- Day 1. FDA Circulatory System Devices Panel Advisory Committee Meeting on Paclitaxel-Coated DCB and DES Late Mortality. June 19, 2019. Available online at https://www.fda.gov/media/127988/download. Accessed June 20, 2019.
- FDA Panel Questions (cont.)-Day 2. FDA Circulatory System Devices Panel Advisory Committee Meeting on Paclitaxel-Coated DCB and DES Late Mortality. June 20, 2019. Available online at https://www.fda.gov/media/127989/download. Accessed June 20, 2019.
- Katsanos K, Spiliopoulos S, Kitrou P, Krokidis M, Karnabatidis D. Risk of death following application of paclitaxel-coated balloons and stents in the femoropopliteal artery of the leg: a systematic review and meta-analysis of randomized controlled trials. J Am Heart Assoc. 2018 Dec 18; 7(24): e011245.
- FDA Executive Summary. Circulatory System Devices Panel Meeting. June 19 and 20, 2019. Available online at https://www.fda.gov/media/127698/download. Accessed June 20, 2019.