The Evolving Landscape of the FDA
David E. Kandzari, MD: Hello. I am David Kandzari from the Piedmont Heart Institute in Atlanta, Georgia. Welcome to the American College of Cardiology (ACC) 2014 Scientific Sessions in Washington, DC. Washington is the home not only of our government, and this week the ACC, but also of the US Food and Drug Administration (FDA). This is a unique opportunity to take advantage of having Dr. Bill Maisel, who is the Deputy Center Director for Science for the Center for Devices and Radiological Health (CDRH) at the FDA. Bill, welcome.
William H. Maisel, MD, MPH: Thanks for having me.
Dr. Kandzari: You chaired the panel meeting for drug-eluting stents. As a cardiologist, you have had a long track record in clinical trials and regulation of these devices as well. We want to have a conversation about some evolution at the FDA and how we are seeing a changing landscape with the FDA as well.
Using the background of this ACC meeting as an example, clinical trials in renal denervation and in transcatheter aortic valves have been presented here. Some of it, we hope, will eventually come to the United States, and some therapies (eg, TAVR) are already commercially available in the United States. The big question is, from a device perspective: What is the primary role of the FDA? Is it safety? Is it efficacy? Is it both?
Dr. Maisel: The statutory language that defines the role of the FDA is that products need to demonstrate a reasonable assurance of safety and effectiveness. But, in reality, the FDA has a dual mission. Many people are familiar with the "protect" part of the FDA mission -- to protect patients. But there is also a "promote innovation" aspect to what the FDA does. We have started taking that much more seriously, and many of the efforts that we have going on are directed at bringing good products to patients more quickly.
When Are Advisory Panels Required?
Dr. Kandzari: We are going to talk about that more in a few moments. I mentioned the FDA panel meeting, and these are quite frequent for both drug and device therapies. That is an area of uncertainty for many practitioners, however. What is the role of the advisory panel meeting, and does the FDA always follow the panel's recommendation?
Dr. Maisel: We greatly value stakeholder input into the decisions that we make; and professional societies, healthcare practitioners, industry, and patients all contribute information and knowledge. There are several ways in which we get stakeholder input. Panel meetings are one of these ways. Advisory panels help the FDA review data and make recommendations for the FDA's decision. We are not obligated to follow the advice of a panel meeting, but we usually do.
Dr. Kandzari: Another question about panel meetings is whether they are required. When do you need to have a panel meeting for a device?
Dr. Maisel: There are certain circumstances under which an advisory panel is required, such as for a first-of-a-kind device. For other devices, we typically use our judgment about whether we need to use that mechanism -- for example, for a particularly challenging product, when we have a controversial dataset, or when we feel that stakeholder input would be of value.
Accelerating New Device Availability
Dr. Kandzari: You hinted at some programs in development to accelerate bringing technologies to patient care in the United States. Let's focus on that in more detail, starting with technologies -- let's use the example of next-generation transcatheter valves -- that might be commercially available elsewhere in the world but not yet in the United States. What are some of the mechanisms that the FDA is employing to bring that technology more quickly to the United States?
Dr. Maisel: First of all, we want to bring good technologies to patients quickly. Transcatheter valves are great examples of a game-changing technology. There are patients who are greatly helped by that technology. We are also very aware that there are other iterations, other generations of the device, that may reduce the risks for patients. So we are interested in bringing those technologies to the United States as quickly as possible. Not every device and not every iteration of a device requires a randomized trial to demonstrate that it's safe and effective. We are looking into many mechanisms, such as leveraging ACC registries like the Society of Thoracic Surgeons and ACC's National Coverage Determination Registry® for transcatheter valves, to try to understand the available information about some of the other generations of these devices.
Dr. Kandzari: One option would be to take existing registry data -- for example, from the United States -- to expand an indication for a device technology. Another would be perhaps to take existing data from other parts of the world and help support the indication or the application in the United States. How favorable is the FDA towards looking at data gathered in other studies elsewhere in the world?
Dr. Maisel: We already do that. The criteria are whether the quality of the data is sufficient and whether the manner in which the data were collected is applicable to a US population. Are the healthcare practices similar? For transcatheter valves, the techniques for implantation are very similar in other jurisdictions. We are very open to leveraging data, as long as they are good, from wherever in the world they are collected.
First-in-Human Trials Coming Home?
Dr. Kandzari: Transcatheter aortic valve replacement is an example of a technology that already exists in Europe and Asia-Pacific. Let's talk about devices that might be at the phase of first-in-human study. For years, as interventional cardiologists, we have seen a lot of device technologies being studied in Latin America, in Asia-Pacific, or in Europe and then years later coming to the United States. Is there an endeavor in the FDA to bring back first-in-human trials to the United States?
Dr. Maisel: We recognize the importance of having clinical trials conducted in the United States, not just because we think it's a nice idea but because the sooner the trials are in the United States, the sooner our patients will have access to the devices. It's very important for healthcare providers and clinicians to get their hands on the devices during the investigational phase so that when the devices are rolled out and being used in patients, clinicians are already familiar with them, and we have a group of physicians who are knowledgeable about their use.
We have been focusing on trying to bring some first-in-human early feasibility studies back to the United States. A couple of years ago, we published guidance about how we can facilitate the use of those devices; for example, allowing companies to modify their devices during the clinical trial without having to run back to the FDA every time for an okay, as long as they are working in a prespecified area. We have started to see benefits of that. We are working with several companies to bring early feasibility studies to the United States, but there is still a lot of work to do.
Dr. Kandzari: That is an exciting development and probably one that a lot of practitioners have not heard about -- the fact that you are bringing first-in-human early-phase preliminary feasibility studies to the United States and that technologies can be iteratively developed during the trial process as well. Is that fair?
Dr. Maisel: Yes, that's right.
More Efficient "Next-Generation" Approvals
Dr. Kandzari: We are going to bring this full circle here. We have talked about devices in use in other parts of the world and bringing them to the United States from the ground up by doing early-phase feasibility studies. Let's talk about iterative changes to existing technologies, whether it's a drug-eluting stent, a transcatheter valve, or a closure device of some sort. For existing technologies (and let's use the drug-eluting stent as an example), we have often seen the requirement for yet another large randomized trial that is going to take more money, more energy, and more time before we see the device coming back to clinical practice. Are there efficiencies that you are finding within the FDA -- you mentioned registries as an example of something to leverage -- or other methods that can accelerate or be used to more efficiently facilitate the trials and make them less costly?
Dr. Maisel: In essence, you are asking a fundamental benefit-risk question. How much information do we need to know before we let the next iteration of a product go on the market? Unlike drugs, as you well know, devices change and are modified very quickly. If we don't have a facile way of evaluating the products, we are going to be left generations behind. So we are very interested in quickly evaluating products.
For the example you cited, we are looking into the idea that we could better evaluate drug-eluting stents through a registry infrastructure, for example, and monitor their performance and be able to say, "You know what, that stent is performing very similarly to an earlier generation," or maybe it's better, or maybe it's worse. But not every trial needs to be a randomized trial.
Dr. Kandzari: That would probably require collaboration from industry as well, to pool their data so that you have an objective performance criterion.
Dr. Maisel: You always want to know what you are comparing something with, and we have used different models. When we believe that there is a class effect, we can pool data and set a performance expectation for restenosis, for example, or the rate of myocardial infarction after stent placement. In other cases, there are enough differences between the devices that you need to do a more formal comparison with an individual product.
Postmarket Surveillance: Leveraging the EHR
Dr. Kandzari: The other part of your role is that you oversee postmarket surveillance, and you are by training an electrophysiologist. There is a lot of postmarket surveillance in that domain as well. Tell us about the role of postmarket surveillance. Has it changed in the past few years, or is it going to change as well?
Dr. Maisel: Postmarket surveillance is having a vital infrastructure for the collection of data, and the timely recognition of important public health signals is critical to the other half of the equation, which is having the confidence that we can let products on the market. It's easier to let a product on the market when you know that you are going to see a signal if there is one to capture. So we don't think that asking individual companies to conduct large postmarket surveillance studies every time a device is approved is a very efficient or economical way to do it.
In 2013, we released a vision for what we thought postmarket surveillance could look like. There are 2 pillars or foundational components to that vision. One is the use of registries, both international and national. The transcatheter valve registry is a great example of that -- for selected products, not for every device that comes on the market, but in targeted areas.
The other pillar is the use of unique device identifiers, which are unique numbers that can be assigned to devices and that will then be incorporated into electronic health information. For the first time, we will be able to leverage existing electronic health information in a way that would allow us to conduct postmarket surveillance that we can't do right now because those data fields are unavailable.
Dr. Kandzari: This is very interesting. Are there any examples of that strategy that we are able to demonstrate now? Or is that still a work in progress?
Dr. Maisel: Last fall, we released the final regulation for the unique device identifier. By September 2014, companies need to have unique device identifiers for the class III products, which are the highest risk products. The following year, they need to have unique device identifiers for life-sustaining and implantable products, and then it goes on to the moderate-risk and the lower-risk devices. Over the next several years, all devices will ultimately have unique device identifiers.
Dr. Kandzari: That will be an exciting turnaround from conducting those standard, routine large postmarket surveillance studies.
Dr. Maisel: Right. It will be a matter of using existing resources. We can envision a day when we won't have to ask a company to do those studies because the information will be captured in an electronic health record, and we can query an electronic health record database.
FDA and CMS: Killing 2 Approvals in 1 Process
Dr. Kandzari: That's great. Speaking of health records reminds me of the issue of reimbursement. One of the challenges, when thinking about new device therapies, is that when we are doing a trial, we are conducting it within the mandates of the FDA in terms of demonstrating safety and effectiveness, and then we end up with a technology that might be successful, but we find out that we didn't meet the criteria for reimbursement through Centers for Medicare & Medicaid Services (CMS).
There has been a lot of discussion (and a lot of rumbling) about the possibilities of the FDA working together with CMS. How can we get uniformity between the agencies of the government to more efficiently design trials and more successfully design trials? Is anything afoot in that area?
Dr. Maisel: The FDA and CMS have developed a parallel review program. It's important to keep in mind that the FDA's standard for approval is different from the CMS standard for reimbursement. The FDA is looking for reasonable assurance of safety and effectiveness, and CMS is assessing whether it is reasonable and necessary for the Medicare population. So what we have done is allow companies to voluntarily sign up for the parallel review program. Then they can have conversations with the FDA and CMS together and understand what the expectations would be both for FDA approval and for CMS reimbursement.
Dr. Kandzari: FDA and CMS would probably encourage companies to get that process started as early as possible.
Dr. Maisel: Right. Certainly there is the greatest value for companies when they are planning their study and deciding what information they need to collect. The parallel review process officially starts with the submission of the application for device approval, which, in some respects, is a little later than companies need that information. But the bottom line is: We are open to helping coordinate and collaborate, understanding that the FDA's data expectations are not going to change just because CMS may need something extra that we don't require.
Getting Inside the FDA
Dr. Kandzari: Too often, we hear comments from clinicians and from industry trialists like this: "The FDA made us do this," or, "We wanted to do this, but we had to do something that the FDA wanted." More often than not, that probably represents a lack of understanding of what goes on at the FDA and how the FDA regulates clinical trials and health safety.
I did a medical device fellowship at the FDA, which was one of my best professional experiences ever. How can physicians become more engaged with the FDA? I suspect that there is a need for the FDA to learn more from practicing clinicians as well. How do we go about having greater transparency and communication?
Dr. Maisel: At the outset, we need stakeholder input, including the healthcare provider community and professional societies. We have established something we call the "network of experts," which is a relationship with more than 30 organizations -- professional societies and professional, scientific, and clinical organizations -- that allows us to reach out to those communities and obtain input and help on issues so that when we are making decisions, we are knowledgeable about the issues and understand how the clinical community feels.
There are certainly other ways for individual physicians to experience the FDA. We have a medical device fellowship program, in which a small number of physicians can spend some time at the FDA and learn the inner workings. We always try to get out to meetings and talk to people so that we can engage and understand the issues that are of importance to them.
Dr. Kandzari: Your being here at ACC is one of the best examples of that. Thank you for your time and your insight. It's a great opportunity for us to be here in DC and to have you here to educate us about the FDA. Thank you.
Dr. Maisel: It was a pleasure to be here. Thank you.