The author says three factors are accelerating the need for change to the current FDA 510k process, and offers a proposed solution.
When a company’s marketing literature tells you how a new device is totally different from competitor products, and the 510k summary for the same new device states that it is “substantially equivalent,” you can understand why the FDA may not fully support the 510k process for innovative, medium-risk devices.
There are many things wrong with the FDA 510k process, but the concept of using a predicate device and demonstrating equivalence is inherently a wrong approach for innovative devices. A preliminary report about the 510k process was released in 2010 (http://bit.ly/510kreport). The report states, “While the concept of ‘substantial equivalence to a predicate’ is generally reasonable, CDRH’s application of this standard has, in certain cases, raised concerns.” Specifically, the use of predicate devices that were withdrawn from the market due to issues of safety or effectiveness, and the use of so-called “split predicates” may not ensure patient safety or device efficacy.
The FDA attempted to address issues identified in the report by issuing a draft guidance document for 510k submissions, but industry hated it. The FDA has also gradually requested more clinical study data to demonstrate that the new device is substantially equivalent to the predicate device. This practice results in unexpected delays and much higher costs of regulatory approval. The FDA’s published guidance for 510k content (http://bit.ly/510kContent) indicates that “Clinical data is not needed for most devices cleared by the 510(k) process.” However, more than 10% of 510k submissions now require clinical data because the 510k for predicate devices included clinical data in order to demonstrate safety and effectiveness.
More recently, the FDA issued a draft guidance document for the De Novo process (http://bit.ly/DeNovoGuidance). The De Novo process allows the FDA to reset the submission requirements for devices that are not substantially equivalent and specify new requirements. The process has been used most for new In Vitro Diagnostic (IVD) products, and it is quite similar to the concept of Common Technical Specifications (CTS) introduced for IVD products in Europe.
Why the FDA 510k Process Needs to Change Now
The simultaneous confluence of three factors is accelerating the need for change in the 510k process. First, the cost of healthcare is skyrocketing. Therefore, patients and healthcare providers are desperately searching for less expensive treatment solutions. Second, insurance providers are demanding clinical evidence that new products are more effective than existing products that cost less. Third, evidence-based medicine is becoming mainstream. Physicians are no longer accepting the word of salespeople and marketing literature. Instead, physicians demand clinical data demonstrating that products are safe and effective. Users also want detailed information regarding patient selection criteria.
The collision of these three factors has exponentially increased the value and importance of clinical data, but only 10%+ of the 510k cleared devices have clinical data at the time of product launch. Regulatory clearance to market a product is nearly useless if the product is not reimbursed, and users will not adopt its use. The modular Premarket Approval (PMA) process supports (http://bit.ly/PMAmethods) the need for preliminary safety data prior to clinical use, followed by a clinical study to demonstrate efficacy. However, 510k products are lower in risk and efficacy, and can often be demonstrated with simulated use, animal testing and cadavers.
As medical devices become more complex and innovative, bench-top testing and pre-clinical testing is not always adequate to demonstrate safety and efficacy for 510k products. Complex and innovative devices are extremely difficult to predict how the devices will interact with a broader population of users and patients, and it is difficult to predict the long-term effects of the devices—beyond the duration of a pre-market, clinical study.
The PMA process requires pre-market clinical data, but PMAs require exponentially greater amounts of data than a 510k submission, and the FDA requires supplemental approval of almost every minor change (e.g., – changing a component supplier, or changing a test method). If the PMA process is too burdensome for most devices, and the 510k process is not adequate, what is the right process for the next generation of devices?
The De Novo process offers one solution, but it is still a pre-market notification process. In order for the De Novo process to be effective for innovative devices, a Special Controls Guidance Document needs to include a requirement for both pre-market clinical studies and Post-Market Surveillance (PMS).
Pilot Parallel FDA-CMS Review Process
In 2011, the FDA initiated a pilot program to allow companies to have PMA and CMS (http://bit.ly/Medicare-Medicaid) review processes occur in parallel (http://bit.ly/ParallelReview). The concept behind this pilot program is that the same clinical data must be reviewed for PMA approval and CMS reimbursement. If the pilot program is effective, products will be approved and reimbursed at the same time. However, this pilot program is only applicable to PMA products at this time. This program could be expanded to 510k products, where clinical data is presented as part of the application, but most 510k products do not warrant a clinical study.
The best tool to measure the safety of a new device is a clinical study, but U.S. clinical studies focus on demonstrating efficacy. Therefore, the FDA should consider using smaller clinical studies, without comparisons to predicate devices, to demonstrate safety, rather than efficacy. This is the approach used by European Notified Bodies for medium and high-risk, innovative devices. This approach can also be integrated with a Special Controls Guidance Document for De Novo products.
For complex, innovative devices, the efficacy of the device is not reliably measured by clinical studies, because outcomes are highly dependent upon users. Pre-market clinical studies can only estimate efficacy due to the small number of users. The lack of accurate efficacy predictability is why the FDA requires PMS for many PMA products. The best tool to measure efficacy is Post-Market Clinical Follow-up (PMCF) studies—not the pre-market clinical studies the FDA uses to evaluate PMA applications. This is also the type of data that is required for CMS reimbursement and physician adoption.
Innovative devices are forcing regulations to evolve. The goal of regulators should be to produce a best-in-class method of evaluating which devices should be approved, reimbursed and adopted as the standard of care.