A recent increase in medical device recalls has fueled scrutiny of current safety assessment methods, which has motivated reduced reliance on a presumption of safety based on predicate devices and increased data and device testing requirements.
The Pure Food and Drugs Act of 1906 marked the start of federal food and drug legislation designed to protect Americans against threats from harmful substances and deceptive practices. Medical devices were not specifically included in the Act. The first official recognition of the need for separate regulations for pharmaceutical drugs and medical devices was promulgated in the Federal Food, Drug, and Cosmetic Act (FD&C Act) of 1938. From 1938 until the early 1960s, medical devices were subject only to policing by the United States Food and Drug Administration (US FDA) and not preapproval. The official preapproval regulation giving US FDA the responsibility of ensuring the safety and efficacy of medical devices before they were legally allowed to be used in humans was codified in the Medical Device Amendments, which were enacted in 1976. This legislation required US FDA to classify devices according to the nature of their potential risks to patients (Classes I-III). Devices on the market prior to 1976 were grandfathered in. The Amendments describe the rigorous premarket approval (PMA) process for most Class III medical devices and those utilizing new technology. The PMA process described in the Amendments required clinical data in addition to preclinical safety assessment.
To reduce regulatory burden, Clause 510(k) of the Amendments stated that clearance for a new medical device could be obtained by demonstrating that the new device was substantially equivalent to devices on the market prior to the enactment of the Amendments. The 510(k) clearance process was based on the principle that pre-Amendment medical devices had already demonstrated an acceptable safety and efficacy profile, and, therefore, devices that are substantially equivalent to pre-Amendments devices could be cleared based on the presumption that they are similarly safe and efficacious. The regulations outlined in the Amendments were strengthened by subsequent legislation in the 1990s that allowed US FDA to require additional safety and efficacy data, including clinical studies, for all medical devices, including those cleared under the 510(k) process. The US FDA Modernization Act, enacted by Congress in 1997, directs US FDA to take a “least burdensome approach” to medical device premarket evaluation, which was defined as “the minimum amount of information necessary to adequately address a relevant regulatory question or issue through the most efficient manner at the right time” (US FDA, 2019).
Small changes to a medical device’s design or materials of construction may lead to unanticipated consequences.”
High-risk devices are generally subject to PMA application requirements, which may involve clinical trial evidence in addition to robust preclinical data packages. Most new medical devices comprise relatively minor modifications of existing products, such as a new shape or manufacturing process for an intravascular stent. For devices that pose lower potential risks to patients, the less rigorous 510(k) premarket evaluation process has been retained, although today, complete preclinical safety assessment data packages are required, and the definition of “substantially equivalent” has become more strict. Prior to 1997, any medical device that had no predicate with which to make a substantial equivalence comparison was automatically considered a Class III PMA device. The US FDA Modernization Act created a new approval route, called “De Novo Classification,” for medical devices with no predicate but which do not pose potential patient risks consistent with a Class III designation.
The 510(k) clearance process is controversial because of the limited testing requirements for devices that are considered substantially equivalent to a predicate device, especially in light of the fact that the safety and efficacy of many predicate devices have never been formally evaluated by US FDA. Over multiple generations of devices, this can lead to “layers” of “substantially equivalent” approvals, resulting in medical devices that are significantly different from the original predicate device on which the 501(k) approval was initially based. In addition, small changes to a medical device’s design or materials of construction may lead to unanticipated consequences. A well-known example of this involved the replacement of ceramic-polyethylene with cobalt-chromium alloy in artificial hip devices, a change considered “substantially equivalent” under the 510(k) program. The newly formulated artificial hips were later associated with a greater need for surgical revision by 7 years post-implantation, and in 2016, US FDA ordered manufacturers of these replacement joints to file for PMAs or face market removal (Darrow et al., 2021).
While new medical devices and biomedical technologies have resulted in considerable improvements to patient health, a recent spike in device recalls further calls the methods used to establish devices’ patient safety as well as the regulatory approval process into question. Between 2002 and 2016, US FDA initiated 487 Class I recalls (i.e., high-risk recalls, usually pertaining to defective products that can cause serious health problems or death). The vast majority of these recalls (405 out of 487) were for devices cleared via the 510(k) pathway (Darrow et al., 2021). In 2021, US FDA initiated 179 Class I recalls for a variety of devices, including ventilators, pacemakers, and insulin infusion pumps (US FDA, 2021). This recent increase in medical device recalls, combined with concerns about the rigor and protectiveness of the 510(k) clearance pathway, may be driving a change in the regulatory landscape towards data-driven decision making and away from presumptions of safety. Dr. Margaret Hamburg (the US FDA Commissioner from 2009 to 2015) was instrumental in moving US FDA’s process away from presumptions of safety and towards requiring that US FDA decisions be based on science and data. This trend towards heightened US FDA requirements for data to support biological safety and efficacy determinations for medical devices coincides with the increased sophistication of device technology and materials and testing processes.
Biocompatibility subject matter experts are therefore tasked with two distinct (but not mutually exclusive) goals: (1) establish whether a medical device in its final, finished form is safe for clinical use, and (2) develop a data package demonstrating the device’s safety that will pass regulatory scrutiny. The first step towards both goals in any premarket assessment involves establishing the device’s use category and biological safety endpoints for evaluation. The International Organization for Standardization (ISO) 10993 series of standards detail a framework for evaluating biological safety based on the nature and duration of the device’s contact with the human body. For example, an endovascular stent graft permanently implanted inside the aorta may require rigorous test data to establish safety prior to market approval. In contrast, an intact skin-contacting electrode would have different testing requirements, focusing mainly on local effects (skin irritation, sensitization, etc.). Furthermore, a wound-healing device that would come into contact with compromised skin may require additional safety considerations, including risk assessment for genotoxicity and carcinogenicity endpoints.
Once the medical device’s use category is established, known and potential risks associated with the device are typically documented in a biological evaluation plan. This data-gathering exercise leverages available information on the clinical use history of the device, as well as its materials of construction, manufacturing process, and packaging materials. Additional testing may be necessary to fill data gaps identified during this early evaluation stage, typically including biocompatibility tests for local effects (irritation, sensitization [discussed in more detail earlier in this issue], cytotoxicity, etc.). As noted earlier in this issue, systemic toxicity, genotoxicity, carcinogenicity, and developmental toxicity endpoints may be addressed via toxicological risk assessment of the chemical components of the device, including an evaluation of extractable and leachable (E&L) chemical constituents, as long as the E&L compounds are adequately identified and the toxicological risk assessment process determines that there is limited toxicological risk posed by these compounds. Achieving these two criteria has been made more complicated by the implementation of ISO 10993- 18:2020, which requires that analytical methods detect, identify, and quantify all compounds that may migrate from a medical device during its use in quantities exceeding safe limits of exposure. These limits are based on the toxicological threshold of concern (TTC) daily exposure limits from ISO/TS 21726:2019 or International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) M7(R1). For devices composed of nondegradable materials, these methods are designed to identify and assess risk for those compounds that may migrate from the device materials in quantities exceeding the TTC. Special consideration for devices made from absorbable or degradable materials are also required to account for systemic exposure to degradation byproducts. Toxicological risk assessment of E&L compounds can only be used to address systemic toxicity, genotoxicity, carcinogenicity, and reproductive/developmental toxicity. Specific biological assays assessing local tissue contact or implantation site effects are required to address all potential adverse effect endpoints named in ISO 109931:2018.
The nature and quantity of data required for establishing “proof” of a medical device’s biological safety, in order to meet the expectations of consumers and regulators, is constantly increasing. As biomedical technology continues to advance, systematic and harmonized approaches for establishing medical device safety will be critical for protecting patient health and regaining public confidence in medical technology and the regulatory approval process.
Bill Wustenberg is a long-time consultant in the biomedical field and the owner of Mycroft Medical, LLC, a company that provides expert comprehensive biological safety, biomaterials toxicology, and preclinical regulatory services to medical device clients. He can be reached at firstname.lastname@example.org.
Darrow, JJ; Avorn, J; Kesselheim, AS. 2021. “FDA regulation and approval of medical devices: 1976-2020.” JAMA 326(5):420-432. doi: 10.1001/jama.2021.11171.
United States Food and Drug Administration (US FDA). 2019. “The Least Burdensome Provisions: Concept and Principles – Guidance for Industry and Food and Drug Administration Staff.” 24p., February 5. Accessed at https://www.fda.gov/media/73188/download.
United States Food and Drug Administration (US FDA). 2021. “2021 Medical Device Recalls.” December 16. Accessed at https://www.fda.gov/ medical-devices/medical-device-recalls/2021-medical-device-recalls.