Medical Devices: Lost in Regulation
The implanted medical device industry was founded in the United States and has been a major economic success and the source of numerous life-saving and life-improving technologies. In the 1950s and 1960s, technological innovations such as the cardiac pacemaker and prosthetic heart valve meant that thousands of suffering Americans had access to treatment options where none had existed before. And because so many breakthrough devices were developed in the United States, the nation’s citizens usually had timely access to the latest technological advances. In addition, U.S. physicians were at the forefront of new and improved treatments because they were working alongside industry in the highly dynamic innovation process. In fact, they rose to worldwide preeminence because of their pioneering work on a progression of breakthrough medical therapies.
But that was then. Although the United States is still home to numerous medical device companies, these companies no longer bring cutting-edge innovations to U.S. patients first. And U.S. clinical researchers now often find themselves merely validating the pioneering work that is increasingly being done in Europe and elsewhere in the world. Worse still, seriously ill patients in the United States are now among the last in the world to receive medical innovations that have secured regulatory approval and clinical acceptance elsewhere in the developed world.
What’s behind this erosion of leadership and late access to innovations? Simply stated, an overreaching, overly burdensome, and sometimes irrelevant Food and Drug Administration (FDA) regulatory process for the most sophisticated new medical devices. To be fair, occasional device recalls have caused great political pressure to be placed on the FDA for somehow “allowing” defective products to harm patients. The agency’s response to political pressure has been to add additional requirements and to ratchet up its tough-cop posture in order to assuage concerns that it is not fulfilling its responsibility to the public. It is presumed, incorrectly, that a lax approval process is responsible. In most instances, however, the actual cause of a recall is outside the scope of the approval process. The most frequent causes of recalls are isolated lot-related subcomponent failure; manufacturing issues such as operator error, processing error, or in-process contamination; latent hardware or software issues; and packaging or labeling issues. In addition, company communications that describe incorrect and potentially dangerous procedures used by some medical personnel are also considered a recall, even though the device is not faulty. Face-saving implementation of new and more burdensome clinical trial requirements, often called added rigor by the FDA, is an ineffective and wrong answer to such problems.
Excessive approval burdens have caused a once-vibrant medical innovation engine to become sluggish. Using the FDA’s statistics, we learn that applications for breakthrough approvals are near an all-time low. It is not that companies have run out of good ideas, but the regulatory risks have made it impractical to invest in the truly big ideas. A slow but inexorable process of added regulatory requirements superimposed on existing requirements has driven up complexity and cost and has extended the time required to obtain device approval to levels that often make such investments unattractive. It must be noted that the market for many medical devices is relatively small. If the cost in time and resources of navigating the regulatory process is high relative to the anticipated economic return, the project is likely to be shelved. The result is that companies will instead shift resources toward making improvements in existing products, which can receive relatively rapid supplemental approval and start generating revenue much sooner. Some patients will benefit from these updated devices, but the benefits are likely to be much less impressive than those that would result from a major innovation.
Perhaps the best measure of the FDA’s stultifying effect on medical device innovation is the delay, often of several years, between device approval in Europe (designated by the granting of the CE mark) and approval in the United States. The Europeans require that so-called Class III medical devices (products such as implanted defibrillators, heart valves, and brain stimulators) must undergo clinical trials to prove safety and functionality as well as compliance with other directives that relate to product safety, design, and manufacturing standards. In addition, the European approach relies on decentralized “notified bodies,” which are independent commercial organizations vetted by the member states of the European Union for their competence to assess and control medical device conformance to approval requirements. The primary difference in the U.S. system is a requirement for more and larger clinical trials, which can be extremely time-consuming and difficult to assemble. Ultimately, the European approach places more responsibility on physicians and their clinical judgment rather than on government officials who may have little appreciation of or experience with the exigencies of the clinical circumstance.
These Class III devices are complex and can pose a risk of significant harm to patients if they are unsafe or ineffective. It is for this reason that the FDA’s pre-market approval (PMA) pathway for these products is arduous and rigorous. It should be. Rigor, however, must be tempered with expert judgment that compares the demonstrable benefits with the possible risks to patients. And in setting requirements for evidence, regulators must distinguish between data that are essential for determining device safety and effectiveness and data that are nice to have.
Not to be lost in the FDA’s quest to avoid possible patient harm, however, is the reality that PMA devices offer the greatest potential for patient benefit. Delays in the approval of effective devices do result in harm to patients who need them. If we examine the date of approval for the identical device in Europe and the United States, we see that most devices are approved much later in the United States. Three examples illustrate this point. Deep brain stimulation for ineffectively managed symptoms of tremors and Parkinson’s disease was approved for use in the United States 44 months after European approval. A novel left ventricular assist device that permitted patients with severe heart failure to receive critical circulatory support outside the hospital was approved 29 months later. A pacemaker-like device that resynchronized the contraction sequence of heart muscle for patients suffering from moderate to severe heart failure was approved 30 months after it became available for patients in Europe.
These examples are drawn from experiences over the past 20 years. Each has matured into a treatment of choice. Table 1, which is based on data from the first 10 months of 2010, shows that delays continue to be long. Of the 11 new devices approved in this reporting period, 9 received the CE mark between 29 and 137 months earlier. It is not known whether the sponsor of the other two devices applied for a CE mark. In the case of an intraocular lens listed in the table, the FDA noted that more than 100,000 patients had already received the implant overseas. This level of utilization is significant by medical device standards and suggests strongly that its attributes have made it part of routine clinical practice. Yet U.S. patients had to wait more than five years for it to be available.
A legitimate question is whether the hastier approval of Class III devices in Europe harms overseas patients. A study conducted by Ralph Jugo and published in the Journal of Medical Device Regulation in November 2008 examined 42 PMA applications that underwent review between late 2002 and 2007. Of the 42, 7 resulted in FDA disapproval, of which 5 had received prior CE mark approval. Reasons for disapproval were attributed to study design, failure to precisely meet primary study endpoints, and the quality of the collected data in the FDA’s opinion. In other words, the problem was that these devices failed to satisfy some part of the FDA protocol, not that the FDA found evidence that they were not safe. The majority (34 of 42) of applications garnered both European approval and a subsequent, but considerably later, PMA approval.
Examples of Class III devices that received the CE mark and were subsequently pulled from the market are few. In recent testimony before the health subcommittee of the Energy and Commerce Committee, the director of the FDA’s device branch cited two examples. One involved certain breast implants. The other was a surgical sealant. These events indicate that the European approval process is imperfect, but hardly one that has subjected its citizens to a large number of unsafe devices. It is simply unrealistic to expect an event-free performance history, given the complexities and dynamic nature of the device/patient interface and the incomplete knowledge that is available.
But what about the harm caused by delaying approval? Delay may not be of much consequence if the device in question serves a cosmetic purpose or if there are suitable treatment alternatives. Delay is of major significance if the device treats an otherwise progressive, debilitating, or life-threatening disease for which medical alternatives don’t exist or have only limited effects. Such afflicted patients can’t wait for what has become an inefficient process to run its course. The paradox is that the FDA’s current regulatory approach may be causing unnecessary patient suffering and death by virtue of the regulatory delay imposed by its requirements.
It is particularly frustrating that devices invented and developed domestically are unavailable here for significant periods of time whereas patients elsewhere receive tangible benefit. It is not unusual for second and third generations of some products to be available internationally before the now outdated device finally secures U.S. approval.
The example of a minimally invasive transcatheter heart valve for the treatment of inoperable aortic stenosis illustrates the implications of excessive delay on the well-being of ill patients. Patients suffering from severe aortic stenosis have an estimated 50% mortality within 2 years after symptom onset if they do not undergo open-heart surgery for valve repair or replacement. Quality of life is adversely affected because of shortness of breath, limited exercise capacity, chest pain, and fainting episodes. A definable subset of affected patients includes those who are too frail to undergo the rigors of open-heart corrective valve surgery. The transcatheter approach, whereby a new replacement valve is inserted via the vasculature, much the way in which coronary balloon angioplasty is done, offers a much less invasive and less traumatic therapeutic option for the frail patient. Even though the technology and procedure are still evolving, clinical results have been impressive, and thousands of patients have received it. In a recently published clinical study, one-year mortality has been reduced by 20 percentage points when compared to the mortality of patients in the standard medical care group. Quality-of-life measures also improved substantially. The transcatheter heart valve was approved in Europe in late 2007; it is still awaiting FDA approval. A transcatheter valve of different design was approved in Europe in March 2007 and has produced impressive results in high-risk patients. Over 12,000 patients in Europe and 40 other countries where approval has been granted have received this valve. It too is still not approved in the United States. In the case of a disease with a poor prognosis, years of delay do not serve the best interests of affected U.S. patients, especially if there is credible clinical evidence that a new intervention performs well.
A more subtle effect of over-regulation is the loss of a leadership position by U.S. physicians and clinical researchers. Whereas pioneering clinical trials used to be the province of U.S. physicians at major academic medical centers, today non-U.S. physicians and medical centers are conducting a substantial and growing number of safety and effectiveness trials. As a result, overall clinical expertise and identification of ways to further improve a new technology have shifted overseas. International physicians increasingly supplant U.S. clinical researchers as medical pioneers. The United States can no longer be assured that its physicians are the preeminent experts at the cutting edge or that U.S. patients are receiving world-class treatments.
The peer-reviewed medical literature serves as a good indicator of where innovation in clinical practice and technology is taking place. The role of journals is to publish findings that are new, true, and important. Reported findings inform the future course of medical practice. A review of the current medical literature concerning transcatheter heart valves, as an example, shows that non-U.S. investigators and centers dominate the field. Published reports not only document the initial clinical experience but also identify advances in technique, refine indications for use, and propose next-generational improvements. High-caliber clinical studies are, without question, being performed in the United States as part of the data package for the FDA, and they are producing valuable information. The point is that although they are adding layers of relevant confirmatory data, they are not driving the cutting edge of medical practice.
A rigorous approval process for medical devices is absolutely necessary. However, the process must be relevant for the safety and effectiveness questions that pertain to the product under review. The process must be efficient, streamlined, administratively consistent, predictable, and conducted with a sense of urgency. It must limit its scope of requirements to those data that are central to demonstrating safety and effectiveness. There are always more questions that could be asked of a new product. A patient-centered regulatory process prioritizes and limits questions to those that are essential to the demonstration of safety and effectiveness in the context of the disease. The FDA has a very legitimate role to play in ensuring that new technologies are sufficiently safe and effective for patient use. This is a relative, not absolute, standard. Benefits must be balanced against risk. As practiced today, the regulatory process is unbalanced at the expense of innovations that could help patients.
Current FDA processes for the approval of medical device innovations need to be reengineered to balance the quest for avoidance of possible harms with the potential for helping today’s seriously ill patients. The agency must also limit the scope of studies to address necessary questions rather than to aspire to scientific elegance and excessive statistical certainty. As Voltaire said, “The perfect is the enemy of the good.” The European experience demonstrates that it is possible to make safe and effective new medical devices available to patients much more quickly. Actual clinical experience demonstrates that an excessively cautious and slow regulatory process conflicts with the interests of patients suffering from serious and progressive diseases. They simply don’t have the luxury of time.