Neuromodulation: Regulatory Considerations

Brief History of Medical Device Regulation

It’s easy to forget that the FDA has been regulating medical devices only since 1976 with the passage of the Medical Device Regulation Act, a congressional reaction to the 1973 spate of lawsuits over the Dalkon Shield intrauterine device in which over 200,000 women claimed injury following use of the implant ( ). This is a relatively short history of regulatory oversight, in stark contrast to the century-long regulation the agency has provided over biologics (1902), food (1906), and drugs (1906). The laws crafted in the 1970s for oversight of medical devices were intended to provide a modern regulatory framework that acknowledged the significant differences between engineered medical devices, as compared to drugs and biologics, while allowing for a similar risk-benefit approach. Subsequent amendments to the law, including the Food and Drug Modernization Act (1997), the Medical Device User Fee and Modernization Act (2002, with its reauthorizations in 2007 and 2012), and the FDA Innovation and Safety Act (2012), strengthened the FDA’s postmarket regulatory authorities, introduced user fees while specifying statutory review times, and promoted innovation. The introduction of user fees has done the most to shape the inner workings of the FDA, bringing tighter control over review times and resulting in a more predictable experience for industry. The evolving history of FDA’s regulatory framework for medical devices incorporates some of the most sophisticated thinking about product regulation than any other regulatory body, with legal authorities extending from before a company begins studying a device in humans through market approval and including postmarket requirements for adverse event (AE) reporting, product recalls, advertising, and promotion.

Time and Cost of Device Development

With the introduction of medical device laws, the previously unfettered pursuit of technological solutions to medical problems in the 1960s and 1970s gave way to increasing regulatory oversight in the 1980s until today. A number of sophisticated neuromodulation technologies used in cardiac electrophysiology were already part of standard care at the time FDA began regulating medical devices, including pacemakers and external defibrillators (1950s). Methods of neural recording were also clinically in use prior to the introduction of FDA regulations: electrocardiography (1930s); electroencephalograpy (1940s); and electromyography (1940s). The first three decades of FDA authority over medical devices saw the introduction of a number of innovative concepts in neuromodulation, including the approval of the first spinal cord stimulator for pain relief (1978), the first bone growth stimulator (1979), the first cochlear implant (1985), and the first neuroprosthetic to restore function (1997).

Investigators and small start-up companies pursuing new medical technologies often found the FDA requirements overwhelming, particularly at the earliest stages of development. Recognizing this, Makower, et al. conducted a survey of over 200 medical device companies to calculate the time and cost associated with bringing medical technologies to market, including those activities tied directly to FDA’s requirements. The resulting 2010 report, “FDA Impact on Medical Technology Innovation” ( ) showed that the cost to bring a Class II (defined in Table 142.3 ) product to market averaged $31 million, including $24 million in FDA-dependent, or FDA-related activities. The cost to bring a Class III (defined in Table 142.3 ) product to market averaged $94 million with $75 million directed toward FDA-related activities. The report further underscored the fallacy of “statutory review times,” pointing out that the total time—from FDA’s receipt of an application to its final approval—often exceeded the statutory time by considerable margins. The statutory review time for a Class II device is 90 days, but the effective total time was found to be 10 months. Likewise, for Class III devices, the agency reported average total interaction times of 9 months, whereas survey respondents reported effective total time of 54 months.

Similar criticisms leveled by congress ( ), and demands for FDA to abandon sections of its laws ( ) resulted in the creation of several internal programs designed to improve transparency, facilitate the early stages of translation, and promote innovation. The 2011 Innovation Initiative saw the start of new efforts, including the Innovation Pathway program, which this author directed ( ) and later evolved into the Expedited Access Pathway ( ); the Early Feasibility Investigational Device Exemption (IDE) Pilot ( ); and critical improvements in the De Novo 510(k) pathway ( ), all of which are discussed in this chapter.

Investigational Device Exemption (IDE)

FDA’s premarket regulatory authority restricts medical device companies from introducing their products into interstate commerce without prior approval from the FDA. This restriction is waived in the case of companies wishing to study an investigational device, and hence the word “exemption” in the term IDE. An approved IDE permits the holder to manufacture, ship, and sell medical devices under controlled conditions, to collect evidence to support either a marketing application or a research objective, and without complying with other requirements necessary for devices in commercial distribution.

The IDE regulations referenced in Title 21 Code of Federal Regulations, Section 812, only specify what companies should submit; the regulations do not provide a framework for review. This is in contrast to the framework for review established for premarket applications (PMAs) (“reasonable assurance of safety and effectiveness”) and 510(k)’s (“substantially equivalent to a predicate”). However, an implicit, but unstated, framework for IDEs has evolved over time to include two main ideas. First, an investigational device must be safe enough to begin use in humans. This concept underscores FDA’s emphasis on safety over effectiveness, although proof-of-principle may need to be established. Basic elements of “safety” include a device’s biocompatibility, sterility, electrical safety, mechanical reliability, software performance, and electromagnetic compatibility. Second, the investigational plan itself must be designed to generate evidence to support a marketing application. In the past, this resulted in IDE disapprovals that were based on study design, hypotheses, or statistical analysis plans, even though the device itself was safe enough to begin use in humans. The FDA modified its approach to IDE decision-making, allowing FDA reviewers to separate their review of safety from their review of the study design. At the same time, the agency developed new internal processes to facilitate the clinical trials enterprise, following criticism that the high regulatory burden in the US was causing companies to abandon US studies in favor of clinical trials in Europe, Asia, and elsewhere. The result is that from 2011 to 2014, the median number of days to full IDE approval decreased from 442 to 101 days ( ).

One way this was achieved was to correct a misperception about the meaning of FDA’s decisions for IDE’s. “Approved” and “Approved with Conditions” both legally permit the start of the clinical trial. However, in the past, many Institutional Review Boards (IRBs) would not honor the conditional approval status, sometimes re-reviewing FDA’s concerns and conducting their own assessment, further delaying the start of the study. FDA corrected this in 2014, making it clear that even conditionally approved studies could begin enrolling patients ( ).