There’s been a heated debate over the past year over whether legislative incentives are needed to encourage certain types of therapeutic innovation (http://wb.md/1OPWl0V). Much of this debate centered on the passage of the 21st Century Cures Act by the US House of Representatives (awaiting action by the Senate), which contained provisions intended to expedite availability of medical products such as new antibiotics and high-risk devices (http://1.usa.gov/1lTfwfU). Somewhat less prominent is legislation proposed in the Senate, the Combination Product Regulatory Fairness Act, which is intended to promote innovation among products that contain both a drug and device (http://1.usa.gov/1KFLZg9).
Aaron S. Kesselheim, MD, JD, MPH
Important drug-device combinations already on the market include transdermal patches, infusion pumps, and cardiovascular drug-eluting stents, with the growing global market for these products expected to reach $115 billion by 2019 (http://bit.ly/20teWVL). The Combination Product Regulatory Fairness Act was introduced to streamline the development of such products by changing the way they are reviewed by the US Food and Drug Administration (FDA).
Bo Wang, PharmD
Combination products—whether a drug and device, a drug and biological product, a biological product and device, or a product combining all 3—are currently overseen by the Office of Combination Products (OCP), which assigns investigational combinations to an FDA center based on the component producing the combination’s primary mode of action (http://1.usa.gov/20teYwP). Combinations deemed to be primarily drugs are assigned to the Center of Drug Evaluation and Research (CDER), biologics to the Center for Biologics Evaluation and Research (CBER), and devices to the Center for Devices and Radiological Health (CDRH). In particular, a combination product that has any chemical action within or on the human body is likely to be assigned as if it were a drug or biologic.
But how drug-device combinations are classified is controversial and may be challenged by the manufacturer.
One recent dispute involved the FDA and Halscion, a privately-held company based in Georgia. In 2008, Halscion acquired the license for an injectable hydrogel scaffold—a gel-like structural framework that purportedly promotes fibroblast migration and organization and that had been used for orthopedic applications—with the intent of developing it for improved wound healing of keloid scars (http://bit.ly/205wieB). The combination product initially had been regulated as a device under the jurisdiction of CDRH. But in 2012, the OCP determined that product’s primary mode of action for keloid scar healing was consistent with that of a biological product, placing it under the jurisdiction of CBER. The decision required the company to invest in larger-scale clinical trials to demonstrate adequate safety and substantial evidence of efficacy for this new indication to meet the review criteria for evaluating approvability of a new biological product.
By contrast, if the scaffold had remained under the jurisdiction of CDRH, it would likely have been cleared as a device via a 510(k) application. Unlike the process used for reviewing a drug or biologic, which is based on establishing a device’s safety and effectiveness via clinical trials, the more lenient 510(k) pathway requires only a demonstration that the product is “substantially equivalent” to an already-approved device (http://bit.ly/1nyJXd4). So Halscion appealed. In 2014, the ruling was changed and the product reclassified to CDRH.
The Combination Product Regulatory Fairness Act—introduced in July 2015—would broaden the ability for combination products to be classified as devices. The bill requires that jurisdiction as drugs or biologics not be assigned “solely because” of chemical action in the human body and orders the FDA to provide scientific support to the sponsor whenever the agency does not classify a combination product as a device.
In addition, the bill requires the FDA and sponsor to establish early in development a combination product review plan, an inflexible agreement on all aspects of clinical studies needed to establish its effectiveness and safety, including any postapproval studies. For example, if the FDA and sponsor agree on a clinical trial design that would test the product against placebo based on its effect on an intermediate outcome intended to predict clinical benefit or harm—a surrogate end point—the FDA would be highly restricted in its ability to later seek additional clinical data to inform its decision making.
Manufacturers often request their combination products be classified as devices because of the major financial implications that device classification can offer with respect to expedited market entry and reduced premarket testing (http://bit.ly/1SKDcB9). They further cite an unpredictable regulatory environment as a key reason for their holding back on research investments in products that might fall under the OCP’s jurisdiction (http://bit.ly/205xqii).
However, the bill’s aims also put it at odds with certain important public health interests. The classification changes it proposes would inevitably increase the number of combination products reviewed under the more lenient 510(k) pathway, even though in 2014, almost two-thirds of combination products submitted for approval were already classified as devices (http://1.usa.gov/1KbpOUa). The 510(k) process provides important flexibility in making small alterations to moderate risk devices, but it has been criticized for being a highly limited premarket screen for a product’s safety and effectiveness and for permitting products to evolve substantially over time without adequate oversight, based on numerous seemingly innocuous updates (http://bit.ly/1SjUH9W). In addition, unlike for drugs and biologics, the FDA does not have authority to require postmarket studies as a condition of clearing a 510(k) device.
The bill’s restrictive combination product review plan compounds these potential negative patient safety implications. This proposed attempt to restrict FDA flexibility is reminiscent of a similar provision in the 21st Century Cures bill that limits the FDA’s ability to alter study plans for drugs in the accelerated approval pathway. But the FDA does not capriciously change the goalposts for manufacturers with products under investigation; in fact, in 2015, FDA first-cycle approvals hit historically high levels with 95% of new products approved on their initial review (http://1.usa.gov/1MdLr0c). Rather, by binding the FDA to a study plan agreed upon during the initial stages of a combination product’s development—and then having that plan stretch out even as far as postmarket commitments—this clause prevents the agency from easily adapting to issues arising in the ongoing testing process.
Balancing faster patient access to innovative agents with sufficient proof of the therapeutic product’s clinical benefit continues to be a difficult and dynamic process, but the FDA should not have to forfeit its current authority and flexibility in overseeing the development of therapeutic products before they reach patients. And patients and the medical community gain little from forcing more combination products to be classified as devices so that they can be sped to market with limited evidence of their safety and effectiveness.
Corresponding Authors: Bo Wang, PharmD (email@example.com), and Aaron S. Kesselheim, MD, JD, MPH (firstname.lastname@example.org).
Published online: January 21, 2016, at http://newsatjama.jama.com/category/the-jama-forum/.
Disclaimer: Each entry in The JAMA Forum expresses the opinions of the author but does not necessarily reflect the views or opinions of JAMA, the editorial staff, or the American Medical Association.
Additional Information: Information about The JAMA Forum is available at http://newsatjama.jama.com/about/. Information about disclosures of potential conflicts of interest may be found at http://newsatjama.jama.com/jama-forum-disclosures/.
Bo Wang, Aaron S. Kesselheim. Promoting Therapeutic Innovation: What Do We Do About Drug-Device Combinations?. JAMA. 2016;315(9):857–858. doi:10.1001/jama.2016.1149