Device firms prepare for the unknown as the FDA gives its controversial 510(k) clearance process a much-needed facelift.
Managing Editor
Nothing stokes the flames of human emotion more than passion and anxiety. Both are extremely volatile sentiments, containing a complicated mix of combustible elements that quickly can ignite creativity or inflame truly destructive and painful digressions. They are the internal hurricanes of human nature—colorful yet dramatic, intimidating yet fascinating, seemingly harmless at first but eventually morphing into a menacing monster that meanders through the psyche, recklessly tossing about thoughts and rearranging the logic involved in making decisions.
These passion- and anxiety-spawned tempests usually cut very distinctive and very different swaths of destruction through the mind. Though these storms occasionally can result in detrimental behavior, they also can become a source of inspiration and innovation, particularly in the medical business world.
Take, for instance, researchers at the University of York in the United Kingdom. Driven by a passion for recycling and reducing e-waste, they have developed a method to turn discarded television screens into biomedical components. With an estimated 2.5 billion liquid crystal displays approaching their demise (and an untold amount maturing in smaller electronics such as cell phones, calculators and computer monitors), the York researchers have created an unlimited supply source for the biomedical industry.
Conversely, Anthony F. DiMarco, M.D., used anxiety as his muse in designing an implanted diaphragm pacing system that replaces the bulky, noisy venilators quadriplegics use to breathe. Ventilators often are a source of anxiety for quadriplegics who use motorized wheelchairs; many worry constantly about the hoses that attach to a hole in their necks becoming dislodged from the machine. DiMarco’s diaphragm pacing system resolves these patients’ anxieties by using electrodes that connect to phrenic nerves on the diaphragm. Wires from the electrodes connect to a control box that is the size of two decks of playing cards. When theelectrodes are stimulated by current, the diaphragm contracts and air is sucked into the lungs. When they are not stimulated, thediaphragm relaxes and air leaves the lungs.
Traditionally, physicians such as DiMarco looked to the U.S. Food and Drug Administration (FDA) for inspiration and guidance in designing medical devices and drafting a regulatory approval strategy that would bring new technologies to market in a timely manner. But over the last 18 months, the FDA has become a source of more anxiety than inspiration as it seeks to execute rules that some device executives claim have not been enforced for years. One of the largest thorns in the industry’s side for much of the last 12 months has been the FDA’s self-improvement and discovery process that has resulted in a re-examination and planned revamping of its much-maligned premarket review process, otherwise known as the 510(k) program.
Created in 1976, section 510(k) of the Federal Food, Drug and Cosmetic Act requires manufacturers of medical devices to notify the FDA of their intent to market a product. The notification must be made at least 90 days in advance of the product’s first planned sale, according to the statute.
This premarket notification process enables the FDA to determine whether a product is “substantially equivalent” to a device already on the market. Products that are significantly different— whether in design, material, chemical composition, energy source, manufacturing process, or intended use—are required to go through the agency’s more stringent premarket approval (PMA) system.
The 510(k) process has been used by manufacturers to clear about 90 percent of medical devices for the U.S. market since its inception 34 years ago. And, for the most part, the process has worked well: a study commissioned by the Advanced Medical Technology Association (AdvaMed) and conducted by Cambridge, Mass.-based research firm Battelle Memorial Institute concluded that less than 1 percent of all medical devices cleared through the FDA’s 510(k) review process have been involved in seriousproduct recalls over the last 12 years.
Battelle unearthed 77 Class I recalls (the most serious kind) among the roughly 47,000 medical devices cleared through the 510(k) process since 1998, giving the FDA a 0.16 percent recall rate for that time period. Even fewer devices (0.08 percent) were recalled for design reasons that may have been caught during the premarket review process. Twenty-nine percent of the recalls were traced to manufacturing deficiencies and 6 percent resulted from labeling mistakes, according to the report. Another 15 percent were recalled for “design or manufacturing” reasons (meaning data were not available to help officials pinpoint the cause of the recall).
Battelle’s report complements the results of two other studies released earlier this year that reaffirm the effectiveness of the FDA’s 510(k) review process. University of Minnesota professor Ralph Hall, Ph.D., determined the rate of Class I recalls for 510(k) products to be less than 0.2 percent while William H. Maisel, M.D., of Beth Israel Deaconess Medical Center in Boston, Mass., put the recall rate in the range of 1 to 1.5 percent.
“These studies reinforce the fact that the basic structure of the 510(k) process is sound,” AdvaMed President and CEO Stephen J. Ubl said.
The basic structure of the program may indeed be sound, but there is no denying its many shortcomings. Even Ubl, who is one of the industry’s biggest supporters of the 510(k) process, admits that the program needs some “tweaking.”
To many in the device industry, though, the 510(k) system needs more than a tweak. OEMs are clamoring for a quicker, more efficient review process, claiming the slowdown in product clearances increases costs and delays commercialization. With the FDA’s more critical eye of late, it’s little wonder that new devices have become mired in regulatory red tape. Average decision times, according to AdvaMed statistics, have jumped 20 percent between 2002 and 2008, going from 97 days in 2002 to 116 days in 2008. The number of days applicants spend gathering more information for the FDA also has risen, nearly tripling from 19 days in 2002 to 51 days in 2008. Consequently, these delays have prompted an increasing number of companies to abandon their 510(k) applications: AdvaMed data shows that 510(k) withdrawals skyrocketed 89 percent in just five years, going from 9 percent in 2004 to 17 percent in 2009.
“Over the last 18 months, the FDA has stepped up their review by applying more scrutiny and being more demanding. I don’t mean that in a negative way because you should always look with a more critical eye at these [510(k)] clearances,” said Rene van de Zande, CEO of Emergo Group Inc., a medical and in-vitro diagnostic device consulting firm with offices worldwide. “What is different now is the data that needs to be submitted with 510(k) applications to support safety, efficacy and equivalence. There are many more questions and demands for information as well as clinical data. All this has increased and we’re seeing more back and forth between the FDA and companies that can result in delays and the ability to get a product to market quickly.”
While making the 510(k) clearance process more efficient would satisfy most device firms (particularly smaller firms, which find it more difficult to cover unexpected expenses), startups are more concerned with the application process itself. They want more guidance on the type of data the FDA needs, as well as the kinds of devices that would qualify for the PMA system versus a 510(k). While small and midsize firms share much of the same concerns as startups and OEMs, their chief complaint is the turnover of FDA reviewers. Any change in the 510(k) review team, executives claim, can jeopardize a company’s application, costing them time, money and in some cases, the ability to commercialize a product. (Editor’s note: For an exclusive in-depth look at one company’s experience with the 510(k) application and clearance process, please visit our website, mpo-mag.com.)
Complaints about the 510(k) process, of course, are not limited to clearance delays, hard-to-understand application requirements and high turnover levels for FDA reviewers. Each company, regardless of its size, could probably find fault with the 510(k) system. More disturbing is the charge that the 510(k) process no longer lives up to its original goal of providing safe and effective medical devices to consumers, and that slower processes could harm innovation in the device industry.
Evidence that the FDA lost sight of its original goals for the 510(k) program became apparent early last year after a group of scientists with the FDA’s Center for Devices and Radiological Health (CDRH) charged management with staff intimidation and illegal application of device regulations. Since then, leaders within the CDRH’s Office of Device Evaluation have admitted that a portion of reviewers were not trained as well as they should have been on federal device statutes. Public health advocates and lawmakers also have added to the criticism, calling the overall device approval process “irredeemingly broken” and the 510(k) clearance system “distorted.”
Last year’s criticism about the 510(k) program helped convince the FDA to thoroughly reassess the program. It set up two panels to review the system and recommend changes, and commissioned the Institute of Medicine (IOM) to conduct its own independent study of the 510(k) process. The IOM’s report (as well as any recommended changes) is due next summer.
The FDA’s internal review panels—the 510(k) Working Group and the Task Force on the Utilization of Science in Regulatory Decision Making—released their recommendations for improving the system in early August.
Some of the key reforms suggested by the FDA panels include major changes to the de novo device classification process, which is reserved for products that cannot be cleared through the 510(k) system because they lacka clear predicate, but do not need a PMA level of review. The FDA wants to streamline this process and clarify the CDRH’s expectations for submissions that undergo this kind of review.
In addition, the 172-page report issued by the FDA over the summer calls for better visibility into the 510(k) clearance process, the addition of a “Class IIb” for moderate-risk devices, and the creation of a guidance document that clarifies when a device should not be used as a predicate (such as when devices are taken off the market due to safety concerns). The report also recommends that the CDRH consider issuing a regulation that would clarify the circumstances under which the center could exercise its authority to rescind a 510(k) clearance to remove an unsafe device from the market and preclude its use as a predicate.
Though they applaud the FDA for embarking on a self-improvement plan, device manufacturers claim the report and recommendations for improving the 510(k) system create more questions than they answer. Anxiety over the proposed changes, in fact, is worse now than prior to the report’s release.
“The FDA is being very transparent, and very open to the comments it has received from industry. They truly are listening to what is going on. But we are in a state of unknown and that’s making everybody a little edgy right now,” said Nikki Willett, vice president, product marketing and regulatory affairs, at Pilgrim Software Inc., a Tampa, Fla.-based provider of enterprise risk, compliance and quality management software solutions. “Are we going back to normal? We don’t know. Are some things going to change? We don’t know. Will I have to do a 510(k) or a PMA? I don’t know. The FDA is going to have to think a lot about the impact to this industry. What are they trying to accomplish? What is the end goal? We all know it’s about patient safety, but if the process was working before, what was the tipping point that led the FDA to take a look at this system and determine that they wanted to reassess it?”
Most likely, it wasn’t simply one “tipping point” that triggered the FDA’s 510(k) overhaul.Rather, it was a confluence of factors, including criticism from its own workers and lawmakers, a growing frustration among device manufacturers with the 510(k) clearance system, the need to modernize the regulations, and, perhaps most importantly, the FDA’s realization that the program in its current state does more harm than good to device development and innovation.
The need for an image makeover almost certainly played a role as well. The FDA has long been hammered by companies and the public alike—among companies for its aloofness and lack of visibility into key decisions; and among the public for the perception that it blindly “rubber stamps” products regardless of their safety and effectiveness. FDA officials have tried their best to dispel such a fallacy but internal criticisms of the process (such as those levied by CDRH scientists last winter), skepticism among lawmakers and controversies such as the one surrounding approval of a surgical mesh by ReGen Biologics Inc. of Hackensack, N.J., only feed the public’s apparent mistrust of FDA decisions.
The public swallowed another dose of that mistrust in mid-October when the FDA revoked approval of ReGen’s surgical mesh, which is designed to stimulate growth of new meniscal tissue in damaged knees. While it is not unheard of for the FDA to revoke approval for a device (usually it is based on new safety data), it is rare for the agency to consider the entire approval process flawed and re-evaluate a product.
Public health advocates claim the ReGen decision demonstrates the FDA’s militant outlook on patient safety as well as its readiness to admit a mistake. But for those wary of the FDA and its judgement calls, even the slightest display of vulnerability by the agency easily can reinforce negative perceptions.
“I often hear the argument that ‘we have to change the 510(k) process because it’s a rubber stamp.’ It’s not a rubber stamp,” insisted Daniel R. Matlis, president of Axendia Inc., a Yardley, Pa.-based analyst firm that provides advisory services to life science executives on business, technology and regulatory issues. “I’ve heard FDA officials say that one of the reasons we need to change the 510(k) process is the amount of time it takes to get a medical device to market compared to a pharmaceutical product. A typical medical device takes about 12 to 18 months to go through the 510k process and get to market.By contrast a new pharmaceutical compound can take 12 to 15 years. There’s a false perception that the reason it takes less time for medical devices to get to market is because a less rigorous process is involved. That’s not the case. The primary reason 510(k) is a shorter review process is that when dealing with medical devices you’re generally evaluating physical and mechanical properties that are very well understood.By contrast, when dealing new pharmaceutical compounds, we must evaluate chemical and biological processes, and our level of understanding of those are not at the same depth that we understand physical and mechanical properties. As a result, the approval process requires more evidence, and time, to evaluate a new drug’s safety and efficacy profile.”
While Matlis’ explanation may help justify the relatively rapid path to market in the medical device industry, it doesn’t shed much light on the kinds of demands being placed on 510(k) applicants. Certainly, FDA reviewers understand the physical and mechanical properties of medical devices (though some in the industry would disagree) but that knowledge is not necessarily distributed equally within the agency.
The FDA hopes to correct this imbalance going forward by developing a network of external experts through Web-based social media. Such a network, the agency notes, can help CDRH staff to more efficiently and effectively leverage outside knowledge and answer questions that may arise during the 510(k) clearance process. These experts would not serve in an advisory capacity.
The creation of an outside expert network is perhaps one of the few recommendations in the FDA’s report that has received widespread support in the industry. Many, such as the agency’s proposal to create a new subset of devices to improve the predictability of necessary 510(k) data, have been panned by most device firms. Currently, devices regulated by the FDA fall into three categories:
• Class I: Devices that do not require premarket approval or clearance but must follow general controls. Dental floss, for example, is a Class I device.
• Class II: Devices cleared under the 510(k) process. Diagnostic tests, cardiac catheters, and amalgam alloys used to fill cavities all are Class II devices. The “predicate device” in question often is quite different, and this process largely is used to clear devices for marketing which do not meet the criteria to be considered Class III.
• Class III: Devices approved by the PMA process. These products usually are implantable or life-sustaining devices; an artificial heart meets both criteria. The most commonly recognized Class III device is an automated external defibrillator. Devices that do not meet either criterion generally are cleared as class II devices.
In its report, the FDA proposes adding a Class IIb subset of devices to its product classification system and requiring clinical or manufacturing data to support a substantial equivalence determination. The agency proposes to develop a guidance document to clarify the type of data applicants should include in their 510(k) documents. FDA officials claim the creation of this new class of devices would help them get the experience they need to make “reliable, well-supported decisions.”
Opposition to the creation of a Class IIb division practically has been universal. Johnson & Johnson, for instance, contends the new device class will further confuse applicants who are unfamiliar with the 510(k) system; the Medical Device Manufacturers Association believes the FDA lacks the statutory authority to create a new device class; and Boston Scientific Corp. fears that products in the new subclass will be treated differently than others.
“The increased clarity and predictability at the heart of this recommendation can be achieved if CDRH makes public a list of device types for which clinical information has been routinely required along with the associated rationale,” Natick, Mass.-based Boston Scientific wrote in comments it submitted to the FDA on Oct. 4. “Defining a new subclass implies that products in this subclass will be regulated differently. Creating a new subclass may also make it difficult to reduce the requirements on device types once sufficient information is known about the device type to no longer warrant enhanced data requirements.”
Protecting the public health is the impetus behind both the FDA’s overhaul of the 510(k) process and the industry’s push for a clearance system that balances the regulatory oversight necessary to encourage product innovation. In their year-long dialogue with the FDA, companies, device executives and financial consultants have warned the FDA against making too many changes to the 510(k) process at once. They have encouraged the agency to ask itself whether many of the recommendations in its report really are necessary.
“We are a very inventive industry,” said Susan Alpert, M.D., Ph.D., senior vice president and chief regulatory officer for Medtronic Inc. “Medical devices have developed from simple handheld tools to very sophisticated products. Many of the proposed changes sound reasonable but they can have a very big impact on us. Any recommendation that narrows what can go through the 510(k) process can have an impact on our industry. We should look at why some of these changes are being proposed. Just making a change for change sake is something that all of us in the industry are concerned about. Change can be good, as long as there is a benefit to making that change.”
Exclusive Online Sidebar
The 510(k) Follies: One Firm’s Foray into the Fray
Any medical device executive who has worked in the industry long enough has a war story to tell about the U.S. Food and Drug Administration (FDA). Some are truly alarming; others are more sanguine. The yarn Edward R. Teitel, M.D., is able to tell after roughly a decade in the sector falls somewhere in between.
Teitel’s tale begins in 2006 at a pre-investigational device exemption (IDE) meeting he set up with an FDA review team. Not unlike other small business owners who meet with FDA officials, Teitel was somewhat nervous and a little intimidated by the gathering. But those feelings quickly dissipated as he thought about the journey in which he was about to embark to bring an intriguing new technology to market.
Teitel had discovered the technology the previous year while working for MicroMed Technology Inc. (now MicroMed Cardiovascular Inc.). The technology, licensed and patented by a team of Utah-based scientists, enables doctors to measure the stickiness level of platelets and their responsiveness to drugs such as Plavix and aspirin. The concept fascinated Teitel, a vascular surgeon-turned-financial advisor-turned lawyer who was building an impressive resume at MicroMed overseeing clinical trials, regulatory matters and fundraising activities. He also was building a proficiency in platelet function and anti-platelet therapy while working at the firm, which indubitably helped spark his interest in the scientists’ invention.
Intent on commercializing the technology, Teitel teamed up with the scientists to form a company called ThromboVision Inc. By the middle of 2006, the startup firm had moved into a 3,000-square-foot manufacturing and headquarters facility near downton Houston, Texas, so it could begin producing the platelet therapy monitoring system for clinical trials. The team even had a name for the system—the ThromboGuide platelet function monitor.
Teitel knew that the clinical trials would play a key role in ThromboVision’s 510(k) application with the FDA. He also knew that he couldn’t afford to botch the application process—the company did not have deep pockets (a trait that most startups and small device companies share) and its access to funding was limited. Any hiccup in the 510(k) process would jeopardize the much-anticipated commercialization of the ThromboGuide.
“If you’re a Roush or J&J [Life Sciences] and something happens, you have numerous other revenue streams so it’s not a big deal if one of your 510(k)s goes belly-up,” Teitel said. “But for early-stage entrepreneurial companies, it’s hard to raise money. If you run up against a brick wall with the FDA, there’s usually no way to raise the next round of funding.”
Using that do-or-die scenario as inspiration, Teitel worked with the FDA at the initial pre-IDE meeting to anticipate most of the issues that could arise during the design, development, manufacture and distribution of the ThromboGuide. The two sides also drafted strategies for overcoming those issues, a move that Teitel believed would surely boost his chances for a hiccup-free product approval process. As an added bonus, one of the FDA review team members was a hematologist. He understood platelet stickiness and anti-platelet therapies well.
Still, there were hiccups. The FDA and ThromboVision executives engaged in a question-and-answer exchange that lasted 18 months (which is about average for the industry). During that time frame, however, the FDA took liberties with response time deadlines, Teitel claimed. ThromboVision was now forced to play The Waiting Game.
Then the FDA threw a wrench into the works. As ThromboVision was conducting its clinical trials on the ThromboGuide, the agency changed the composition of the original review team. The hematologist who understood the clinical technology behind the device was no longer on the team. Other members changed as well and their cardiology background appeared limited. To make matters worse, the new review team was not bound by any agreements made by the original group. All bets were off.
“One of the big problems with the 510(k) review process is that the FDA is not bound by what they agree to at the pre-IDE meeting,” Teitel told Medical Product Outsourcing. “I don’t agree with that. I think the FDA should be bound to what is agreed to at the pre-IDE meeting unless the science is out of date and no longer applicable. I’ve heard the same complaint from other colleagues. They thought they had an agreement on how to move forward on their technology and regulatory pathway only to have some of the things they had already agreed to not honored anymore. It’s exceedingly frustrating.”
Teitel’s worst fears were realized. ThromboVision simply did not have the money to switch gears in the middle of the game, and its prospects for raising more funding were bleak. Bankruptcy was Teitel’s only option.
Though he still harbors some indignation at the 510(k) approval process, Teitel admits that he is partially to blame for the failed application. In retrospect, he acknowledges that ThromboVision did not have the “wherewithal or the resources” to negotiate some of the issues and may not have engaged in a strong enough dialogue with the FDA. He also believes that an FDA review team with a background in cardiology would have helped the company’s cause.
A stronger cardiology foundation also would have helped the final FDA review team better understand the technology behind the ThromboGuide, Teitel said. “We had a good product that worked. I don’t think they [FDA] ever really understood how our product would be used and applied by physicians,” he charged.
Teitel has intentionally stayed out of the recent fray caused by the FDA’s decision to re-examine its 510(k) review process. He has not submitted any public comments on the system to the agency, nor has he publicly discussed the FDA’s own recommendations for improving the oversight of the device industry.
He does, however, have two suggestions for the agency:
“The first thing is there should be input at the beginning as to what kind of members should be on the [FDA] review team,” he said. “The second thing is that the pre-IDE meeting roadmap that you get should be close to binding unless there’s a major reason to change it. The roadmap really should be less arbitrary.”