Congress is currently taking action to reauthorize the user fee programs for devices and prescription drugs, which are currently set to expire on 30 September 2012, and to establish user fee programs for generic drugs and biosimilars. On 24 May 2012, the Senate passed the Food and Drug Administration Safety and Innovation Act.1 A similar bill, the Food and Drug Administration Reform Act,2 was passed by the House on 30 May. The two bills are now in conference committee. Congressional leadership would like to see the combined bill passed before the 4 July recess. In addition to reauthorizing the user fee programs, these bills include a range of other provisions seeking to reform the U.S. Food and Drug Adminstration's (FDA) regulation of medical devices, including a provision that seeks to clarify the definition of “custom devices.”
The custom device exemption set forth in Section 520(b) of the Federal Food, Drug, and Cosmetic Act (FDCA) allows a device firm to craft a unique medical device to meet the specific needs of an individual patient or physician without the need to obtain premarket clearance or approval from the FDA in limited circumstances. Although the definition of a custom device is set forth in the FDCA and FDA’s implementing regulations, device manufacturers have often interpreted these provisions somewhat differently than FDA. Indeed, several manufacturers who have called their devices “custom devices” have received warning letters indicating their products do not meet the definition of a custom device.
Current definition of custom devices
The FDCA prohibits the introduction of any adulterated device into interstate commerce with few exceptions. One such exception, the custom device exemption defined in the FDCA § 520(b) (codified at 21 U.S.C. 360j), exempts custom devices from performance standards, premarket clearance and approval, and investigational device exemption (IDE) requirements, but not from FDA’s good manufacturing practices, which are codified in the quality system regulation (QSR) (21 C.F.R. Part 820). In order to qualify as a custom device, a medical device must meet all five of the following requirements, as defined in 21 C.F.R. 812.3(b):
- The device necessarily deviates from devices generally available or from an applicable performance or premarket approval (PMA) requirement in order to comply with the order of an individual physician or dentist;
- The device is not generally available to, or generally used by, other physicians or dentists;
- The device is not generally available in finished form for purchase or for dispensing upon prescription;
- The device is not offered for commercial distribution through labeling or advertising; and
- The device is intended for use by an individual patient named in the order of a physician or dentist, and is to be made in a specific form for that patient, or is intended to meet the special needs of the physician or dentist in the course of a professional practice.
Although the custom device exemption is clearly defined in FDA law and regulation, interpretations of this definition vary, and are often a source of confusion for medical device manufacturers. FDA’s interpretation of each prong of the definition of a custom device, outlined above, is discussed in the paragraphs that follow.
Prong 1: The device necessarily deviates from devices generally available or from an applicable performance or premarket approval requirement in order to comply with the order of an individual physician or dentist. FDA’s current interpretation of the “necessarily deviates” requirement for a custom device requires that the device be “sufficiently unique that clinical investigations would be impracticable.”3 In keeping with this interpretation, FDA and the courts have agreed that contact lenses ordered specifically for a named patient on the basis of a prescription do not satisfy the definition of a custom device because, among other things, clinical investigations of contact lenses are practicable.4 In other words, any device with a standard template that is only modified by customizable options is considered by FDA to be a “customized” device subject to premarket notification or approval requirements, not a custom device. Examples of customized devices include Align Technology, Inc.’s Invisalign (K981095) and Polyclinic Medical Center’s hard tissue replacement – patient match implant (K924935).
Prong 2: The device is not generally available to, or generally used by, other physicians or dentists. FDA and the courts have interpreted this prong to mean that, if the features of a device fall within a range of features that are commercially available to a physician, regardless of the manufacturer, the device cannot be sold as a custom device. In keeping with the view that a customized device is not a custom device, for example, prescription contact lenses are considered to be “generally available” if they provide merely a variation within a range of powers and surface contours that physicians can purchase for their patient.5
Prong 3: The device is not generally available in finished form for purchase or for dispensing upon prescription. If a would-be custom device is sold in sufficient quantities that it becomes commercialized, FDA will consider the device to be “generally available,” and, thus, ineligible for the custom device exemption. For example, if multiple shipments of devices of the same size are shipped to the requesting physician, these devices likely will be considered by FDA to be commercialized, and therefore “generally available in finished form.”6 In addition, if a commercially available product from another manufacturer could meet the patient’s or physician’s needs, FDA is likely to consider the subject device to be “generally available.” Along these lines, FDA has previously indicated that a company may make as many as 10 devices per year without the fear of an FDA enforcement action.7
Prong 4: The device is not offered for commercial distribution through labeling or advertising. Any device that is advertised, promoted or otherwise “offered for commercial distribution through labeling” cannot be sold as a custom device. Under Section 201(m) of the FDCA, “labeling” is a broad term that encompasses device labels as well as “all other written, printed, or graphic matter (1) upon any article or any of its containers or wrappers, or (2) accompanying such article.”8 The term “accompanying” is broadly interpreted by FDA and the courts to include material that is not only physically provided along with the device, but also material that does not physically accompany the device, but that is considered part of an integrated sales transaction, or that is offered in conjunction with the device for the purpose of providing an explanation of the device. Among the materials that have been construed as labeling are clinical study summaries, white papers, trade show exhibits, mailings, case studies, press kits, flyers, booklets, price lists, and training. Accordingly, if a device is featured or discussed in any of the above or similar materials, the FDA is likely to take the position that the device is offered for commercial distribution through labeling and does not meet the definition of a custom device.
Prong 5: The device is intended for use by an individual patient named in the order of a physician or dentist, and is to be made in a specific form for that patient, or is intended to meet the special needs of the physician or dentist in the course of professional practice. FDA has broad discretion in interpreting the terms “specific form” and “special needs.” As a result, the FDA often bases its determination that a device is not a custom device solely on the patient-specific form/physician special needs criterion. Would-be custom devices will fail to meet this prong if a similar product is commercially available,9 or if the device meets a need that is shared by others in the field, and is not unique to the patient or physician as individuals.10
Even for manufacturers who are fully apprised of FDA’s historical interpretation of the custom device definition, uncertainty over whether a given device meets this definition remains. For one thing, FDA has never provided guidance to assist firms in determining whether two devices are the “same” for the purposes assessing the devices under the custom device exemption. Thus, it is not clear when two devices with minimal changes in specifications will be treated as the same device and when they will be treated as unique. At present, device firms must rely on principles set forth in case law and warning letters to guide these assessments.
Furthermore, FDA has never defined a threshold number of devices that may be produced annually, above which the devices could not meet the definition of a custom device. As a result, device firms have struggled over the years to determine the maximum number of units of a would-be custom device that may be sold without exceeding the limitations of the custom device exemption. Over the years, FDA has made public statements that up to five or 10 units of a given “custom device” may be sold per year without fear of FDA enforcement action. In recent months, however, FDA has narrowly interpreted the custom device exemption to be limited to devices that are virtually one-of-a-kind — an interpretation that many in industry find overly restrictive.
A clarified definition?
The pending FDA reform legislation would bring added clarity to the definition of custom devices in two ways: (1) it will codify certain aspects of FDA’s and the courts’ interpretation of the existing definition of a custom device, and (2) it will introduce new limitations to the applicability of the custom device exemption.
The Food and Drug Administration Safety and Innovation Act,11 which passed the Senate on 24 May, proposes the following amendment to replace Section 520(b) of the FDCA (21 U.S.C. § 360j(b)):
- Custom devices —
- IN GENERAL — The requirements of sections 514 and 515 shall not apply to a device that
- is created or modified in order to comply with the order of an individual physician or dentist (or any other specially qualified person designated under regulations promulgated by the Secretary after an opportunity for an oral hearing);
- in order to comply with an order described in subparagraph (A), necessarily deviates from an otherwise applicable performance standard under section 514 or requirement under section 515;
- is not generally available in the United States in finished form through labeling or advertising by the manufacturer, importer, or distributor for commercial distribution;
- is designed to treat a unique pathology or physiological condition that no other device is domestically available to treat;
- (i) is intended to meet the special needs of such physician or dentist (or other specially qualified person so designated) in the course of the professional practice of such physician or dentist (or other specially qualified person so designated); or
- is intended for use by an individual patient named in such order of such physician or dentist (or other specially qualified person so designated);
- is assembled from components or manufactured and finished on a case-by-case basis to accommodate the unique needs described in clause (i) or (ii) of subparagraph (E); and
- may have common, standardized design characteristics, chemical and material compositions, and manufacturing processes as commercially distributed devices.
- LIMITATIONS — Paragraph (1) shall apply to a device only if
- such device is for the purpose of treating a sufficiently rare condition, such that conducting clinical investigations on such device would be impractical;
- production of such device under paragraph (1) is limited to no more than five units per year of a particular device type, provided that such replication otherwise complies with this section; and
- the manufacturer of such device created or modified as described in paragraph (1) notifies the Secretary on an annual basis, in a manner prescribed by the Secretary, of the manufacture of such device.
- EXCEPTION — Paragraph (1) shall not apply to oral facial devices.
- GUIDANCE — Not later than two years after the date of enactment of this section, the Secretary shall issue final guidance on replication of multiple devices described in paragraph (2)(B).
The House bill, the Food and Drug Administration Reform Act,12 includes identical language, except that it does not categorically exclude oral facial devices from the definition of a custom device.
The definition of a custom device set forth in the reform legislation was taken largely from the existing statutory language (21 U.S.C. §360j(b)) and FDA’s implementing regulation (21 C.F.R. 812.3(b)) defining custom devices. The new language also incorporates previous interpretations of those provisions by FDA and the courts, as outlined above. Accordingly, the major effect of the proposed statutory language is to clarify and freeze the current FDA interpretation of custom devices in place, rather than to introduce a new standard for custom devices. The reform legislation does, however, also set forth specific limitations on what may be considered a custom device. In particular, the bills establish a new annual quantitative limit for custom devices: no more than five units of a given custom device may be produced in a year. The bills also mandate the issuance of a final guidance on this limitation within two years of the date of enactment of the legislation and introduce an annual reporting requirement for manufacturers of custom devices.
As indicated above, the Senate bill, but not the House bill, excludes all “oral facial devices” from the custom device exemption. This difference between the bills was present at the time of their introduction and is currently the only difference in language between the House and Senate versions of the custom devices provision. It remains to be seen whether the oral facial device prohibition will survive in the final version of the bill.
The bill also notes that custom devices “may have common, standardized design characteristics, chemical and material compositions, and manufacturing processes as commercially distributed devices.” This provision codifies the common understanding that a custom device may be derived from other devices. While it does not lighten the burden on manufacturers for demonstrating that a device is a custom device, this provision does provide a safeguard against a device being disqualified from the custom device exemption solely because the device shares a common composition or manufacturing process with commercially available devices.
Impact on Industry
While helpful in codifying FDA and judicial interpretation of the definition of a custom device, the pending FDA reform legislation falls short of addressing the major points of uncertainty surrounding the regulation of custom devices. Although the new five-unit limitation on the number of times a given custom device may be produced annually provides a clear threshold for when a device has been produced in sufficient quantities to be considered “generally available” for sale and, thus, ineligible for the custom device exemption, the legislation does nothing to provide additional insights into what makes a device the “same” as one that was previously sold. Thus, it may still be difficult for companies to determine whether the threshold has been met. For example, do the specifications of the units have to be exactly the same to be the “same” device, or are some differences so minimal that they do not actually amount to a “different” device? It is also unclear, for example, whether there is a material difference between sizes 0.23 and 0.231. There is no easy answer to these questions, except to say that the determination will depend on the nature of the devices, their similarities and differences, and the level of regulatory risk that a company is willing to accept.
Moreover, device firms should be cognizant of the fact that producing no more than five units of a given device is a necessary, but not a sufficient condition of the custom device exemption. In other words, the mere fact that a device is produced fewer than five times per year does not automatically mean it will qualify for the custom device exemption. To appropriately be considered a custom device, all prongs of the statutory definition will need to be met. The new quantitative limitation does, however, make it clear that all devices produced more than five times per year will not qualify for the custom device exemption, and will require premarket clearance or approval to exceed this limit.
On a more positive note, with a congressional mandate for FDA guidance on custom devices within two years of the enactment of this legislation, additional FDA guidance on these issues — which FDA has been promising since early 2000 — should be forthcoming.