Evolution & Revolution: Device Policy Priorities at FDA in 2019
This post is the first in a series of three in which we recap the Food and Drug Administration’s somewhat difficult year, having spent the majority of it without a permanent Commissioner and facing a slew of political and practical challenges. Today, our topic is “medical devices” writ large, which is a product class that is becoming more complex by the day as it grows to encompass software, diagnostics, laboratory tests, and new medical technologies that resemble nothing that Congress had in mind when it first gave FDA authority to regulate medical devices in 1976. In the coming days, we’ll also take a broad look at FDA’s year from the standpoint of therapeutic medical products, including drugs and biologics, as well as the more commonly used category of consumer products.
In part due to the leadership and political challenges faced by the agency, the year 2019 saw the agency’s medical device policies continue to advance, though not as quickly as some in the device ecosystem may like. Questions still linger about consequential proposals like how laboratory developed tests (LDTs) and in vitro diagnostic products (IVDs) (proposed to be collectively referred to as IVCTs) will be regulated, whether the digital health program will proceed with its reinvention as proposed by FDA, and other changes related to device safety and oversight.
FDA’s device center director Dr. Jeff Shuren has posited that the FDA’s approaches to device regulation and oversight have not kept pace with scientific and technological changes. Despite Congress authorizing or mandating new programs and policies every few years, Dr. Shuren contends the device regulatory program needs fresh ideas. In other words: think outside the box.
A prime example of creative thinking is the agency’s proposal to certify software developers who exhibit a culture of quality and organizational excellence. FDA’s Pre-Certification proposal (see our prior post here) would allow software developers to enjoy a streamlined review (in some cases, no review) following an intensive appraisal of their software development process and succeeded by a commitment to robustly monitor their software once marketed to quickly identify and fix safety and other problems. This is a major shift away from the existing model used for all other medical products FDA regulates, in which the agency reviews the safety and effectiveness of the actual product rather than the developer.
This proposal has been stuck in second gear since its announcement nearly two years ago, in part because FDA lacks statutory authority to advance it beyond a pilot phase in which the agency is reviewing software submissions as required by current law in parallel with a review based on the proposed program (see our prior post here). Meanwhile, members of Congress are not lining up to authorize Pre-Certification and, indeed, big names in the Senate, including one presidential hopeful, have raised significant questions about the program’s design and operation in two letters sent to the agency in October 2018 and October 2019, respectively. Members of the House Energy & Commerce Committee have also expressed concern with the concept of pre-certification because it is seen by some as FDA giving up oversight rather than holding products to the high standard of reasonable assurance of safety and effectiveness.
FDA did take steps this year to clarify software policies through multiple guidance documents issued in October 2019. However, questions remain about the agency’s policy on Clinical Decision Support software (CDS), which industry and others have been asking FDA to clarify ever since certain types of CDS were exempted from FDA oversight in the 2016 21st Century Cures Act (Cures Act).
In the revised CDS draft guidance released in October 2019, FDA clarified some aspects of its policy but at the same time created some additional ambiguities (see our prior post here). In many cases, the revisions to the draft guidance were relatively straightforward and expected based on the public comments to the initial draft. For example, FDA refined its interpretations of the Cures Act exemption criteria for CDS in the draft guidance, providing clearer guidelines for determining whether a CDS product is a medical device, and described how decision support software functions intended for use by patients and non-professional caregivers fit within the agency’s CDS policy. But, in an unexpected change, FDA proposed using the International Medical Device Regulators Forum (IMDRF) software device risk categorization framework as a way to determine whether or not a CDS device should be under enforcement discretion. This approach has generated some controversy because FDA’s interpretation and application of the risk factors with respect to CDS devices appears inconsistent with the general IMDRF framework. The timeline for finalizing the CDS guidance is unknown, but it is quite possible that FDA will issue the final guidance in 2020 or 2021.
Clinical Laboratory Tests
Another area where FDA is seeking to implement a creative, potentially revolutionary regulatory approach is in its oversight of lab tests. The proposals being discussed on Capitol Hill for inclusion in the soon-to-be-introduced VALID Act have been widely documented, including by us, and 2019 was supposed to have shown progress on pre-certification (the same concept being proposed for software), modifications to tests, and preventing overlap between FDA and CLIA oversight, among others. Looking back on the past year, however, little appears to have been successfully completed.
Progress on this topic overall has been slow: the agency’s announcement of its proposal to phase in oversight of lab tests—and the resulting backlash—occurred more than five years ago, and Congress has been unable to enact promised legislation resolving many of the concerns raised by industry, the FDA, and other stakeholders. A new version of the VALID Act is expected by the end of this calendar year, but will face strong headwinds in the second session of the 116th Congress as legislators also try to resolve public health concerns related to vaping, drug pricing, and surprise billing, among other topics that are expected to be given higher priority. The 2020 election will also limit what Congress can reasonably achieve during the time it is in session.
510(k) & PMA
FDA did make progress during 2019 towards implementing changes to the 510(k) program with a final Safety and Performance Based Pathway guidance and draft Safer Technologies Program guidance, both issued in September (see our prior post here). These policies are outgrowths of the FDA’s April 2018 Medical Device Safety Action Plan, in which the agency committed to spur innovation towards safer medical devices.
The Federal Food, Drug, and Cosmetic Act requires substantial equivalence to be based on comparison of a new device to a predicate device’s intended use and technological characteristics. However, FDA says the Act does not preclude the use of performance criteria to facilitate that comparison. Under the agency’s Safety and Performance Based Pathway, if a predicate meets certain criteria and a new device meets those same criteria, direct comparison between the predicate and the new device is not required to demonstrate that the new device is substantially equivalent, if the devices share an intended use. FDA has not made suggestions to this effect, but it is possible the Safety and Performance Based Pathway could be laying the groundwork for a predicate-free 510(k) process. (In other words, if you can rely on performance criteria to assure a device is safe and effective, why do you need a predicate at all?)
The Safer Technologies Program (STeP) mirrors the breakthrough devices program that FDA created via guidance years ago and which Congress codified into law in the 21st Century Cures Act passed in December 2016. FDA created STeP because it says that the agency sees a lot of devices that are not substantially different from an already-marketed device. While that’s kind of the point of the current 510(k) model—new devices need to follow a predicate—it means new devices may not be coming to market with more modern safety mitigations than what the predicate has. Like the breakthrough devices program, STeP is aimed at enabling more interactions with FDA during the device design and development process if the device has more modern safety mitigations than its closest predicate. Unlike the breakthrough programs, eligible devices for STeP are those devices for treating non-life threatening or reversible diseases or conditions (breakthrough is limited to devices to treat life-threatening or irreversibly debilitating diseases or conditions).
FDA is also making waves with a proposal to allow progressive (also called conditional) approval of PMAs if a device is intended to treat life-threatening or irreversibly debilitating diseases or conditions. Progressive approval would permit the marketing of a PMA device based on a demonstration of reasonable assurance of safety followed by a demonstration of reasonable assurance of effectiveness once marketed. If effectiveness cannot be demonstrated, the device’s approval would expire after a set amount of time. Last year, Congress authorized conditional approval for certain animal drugs in limited circumstances but was assured by then-Commissioner Scott Gottlieb that FDA was not planning to seek authority for progressive approval for human medical products. Some in Congress are upset that the agency now appears to be withdrawing from that commitment; this particular proposal is therefore not expected to advance quickly.
In closing, a myriad of proposed changes to the FDA’s premarket device program have kept the agency busy in 2019. We look forward to seeing what progress can be made on these and other regulatory policy topics in 2020.