Reengineering at CDRH: A Work in Progress

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  January 1998 Column

FDA REFORM

Of the nearly 50 initiatives under way at FDA's device center, three in particular have the potential to speed up device review.

The Center for Devices and Radiological Health (CDRH) devoted much time and effort in 1997 to a two-part program to reengineer the medical device regulatory process. One part involves improving the speed and efficiency of existing functions, such as premarket review. The second, more ambitious aspect of the program goes beyond simple streamlining to a complete reexamination of the entire medical device regulatory process. The ultimate goal of the reengineering program is to dramatically improve the device regulatory process by rethinking from top to bottom how FDA goes about regulating the medical device industry.

Figure 1. The new 510(k) paradigm (draft).

The reengineering program has grown out of effective but yet-unfinished efforts by CDRH management during the last four years to fix a broken premarket clearance process. The more-ambitious reengineering initiatives were prompted by the recognition that more work was needed to optimize not only premarket clearance, but also postmarket surveillance, compliance, and adverse event reporting. Two other factors also motivated the initiatives. First, if CDRH had not chosen to reengineer or reinvent itself, Congress would probably have produced legislation much more unacceptable to FDA than that recently passed. Second, CDRH recognized that in the face of inevitable budget reductions in coming years, it needed to change the process so that it could do the same job with fewer personnel.

All told, there are almost 50 specific reengineering initiatives under consideration at CDRH. Rather than seeking to address all of them, I will focus on the three leading product-clearance initiatives: the new 510(k) paradigm; the product development protocol process; and the streamlining of the premarket approval (PMA) process, including real-time supplement reviews.

THE NEW 510(k) PARADIGM

Even though the backlog of 510(k) filings has been largely eliminated and many Class I devices have been exempted from premarket review, obtaining 510(k) clearance can still be difficult. The average review time dropped from 216 days in fiscal year 1994 to 145 days in 1996, and to 97 days in 1997. However, many 510(k) submissions, especially those requiring clinical data, still take a long time to be cleared--sometimes as much as six months to one year. Moreover, under current FDA practices, the need to obtain FDA market clearance for any changes to an already marketed device that could significantly affect its safety and efficacy can delay the implementation of important technological modifications. To address these problems, CDRH has proposed the introduction of what it calls the Special 510(k) and the Abbreviated 510(k) (Figure 1).

The Special 510(k) for Device Modifications. Under current regulations, a new 510(k) notice must be filed when the safety and efficacy of a device could be significantly affected by a change or modification in design, material, chemical composition, energy source, or manufacturing process. The Office of Device Evaluation (ODE) and the medical device industry have been wrestling for years over the meaning of the phrase "could significantly affect safety and efficacy." To provide some direction to industry, ODE issued a guidance document on January 10, 1997, titled "Deciding When to Submit a 510(k) for a Change to an Existing Device."

THE SPECIAL 510(k) SUBMISSION

The Special 510(k) would include the following items:

The guidance has certainly helped industry in the decision-making process. Even so, 510(k) submissions for modifications are often subjected to one or more rounds of questions from ODE that seem to be related not to the changes but to the already cleared characteristics of the device. Therefore, industry has continued to push for a better way to deal with device modifications.

The new 510(k) paradigm, through what the agency calls the Special 510(k): Device Modification, seeks to address industry's concerns about what it sees as review overkill of modifications.¹ Under this approach, a manufacturer would first look to the January 1997 guidance and determine whether a new 510(k) is needed. If it is, and if the proposed modification does not affect the intended use or basic technology of the device, ODE will consider conformance with design controls sufficient to ensure that the device is substantially equivalent to its predicate.

Under the new quality system regulation (QSR), manufacturers must have a systematic set of requirements and activities for the management of product design and development, including documentation of design inputs, risk analysis, design output, test procedures, verification and validation procedures, and documentation of formal design reviews. This obligation applies not only to the original device design, but also to significant changes in the device. Consequently, the manufacturer must ensure that the design input requirements both for the original design and for the changes are appropriate, so that the device will meet its intended use and the needs of the user population. The manufacturer must also establish and maintain procedures for defining and documenting design output in terms that allow an adequate evaluation of conformance to design input requirements.

Under the proposed Special 510(k) procedure, a manufacturer who is intending to modify an already-cleared Class II device can conduct the verification and validation activities necessary under the QSR to demonstrate that the new design outputs meet the design input requirements. Upon completion of this design review process, a Special 510(k) can be submitted, through which the applicant would declare conformity with the design control requirements. FDA could then rely upon that declaration to clear the device. Of course, FDA could, and sometimes will, review the bases for such a declaration through its inspectional process. A false declaration could lead to FDA enforcement action, including, if warranted, criminal prosecution.

The Special 510(k) option is a welcome initiative. If a company can simply declare and certify that it has followed the required design control procedures and has not significantly changed the intended use or technology of the device, it might be possible to clear Special 510(k) submissions in a matter of days rather than months. Since companies should already be verifying and validating all significant device changes under the QSR design control procedures, it only makes sense to use that information for 510(k) clearance as well.

The Abbreviated 510(k). The second initiative under the new 510(k) paradigm grows out of the concept of special controls as authorized by the Safe Medical Devices Act of 1990 (SMDA).² Under the SMDA special controls provisions, guidance documents can be drafted that, if followed, can be the basis for clearance of a Class II device. These guidance documents, referred to by FDA as special control guidance documents, are generically applicable to groups of Class II devices. Under the Abbreviated 510(k) process, a company could certify that it is in compliance with an applicable special control guidance document, and this could form the basis for a finding of substantial equivalence.

THE ABBREVIATED 510(k) SUBMISSION

In the Abbreviated 510(k), a manufacturer could use a third party to assess conformance with the recognized standard. The marketing application would then include both a statement of conformity from the third party as well as the summary information and declaration of conformity signed by the manufacturer.

Alternatively, if no such document yet exists, but FDA has determined and announced that a relevant voluntary standard is applicable, the Abbreviated 510(k) process would allow a declaration of conformity to this specific individual consensus standard. For example, compliance with IEC 60601, which has broad applicability to many electromedical devices, could be the basis for approval for certain types of devices. The consensus standard itself would therefore legally be considered a special control under SMDA.

Again, the Abbreviated 510(k) approach is a welcome addition to the premarket clearance options available to a manufacturer. The concept is a part of the movement toward harmonization with the European approach; namely, to rely upon conformity with standards as the basis for marketing clearances. If a special control guidance document exists for a specific device or FDA does accept a relevant standard, the 510(k) notices for the applicable devices could be cleared within a matter of days or weeks.

The only concern with respect to the Abbreviated 510(k) process is whether FDA has the resources or the inclination to draft numerous guidance documents. The consensus-standards alternative has great promise if FDA can overcome its past reluctance to accept consensus standards drafted by others as the basis for U.S. market clearance. Congress provided some impetus in this regard by statutorily recognizing in Section 204 of the FDA Modernization Act of 1997 that FDA can recognize a national or international standard as the basis for meeting a premarket clearance requirement.

THE PRODUCT DEVELOPMENT PROTOCOL

One of the most highly touted reengineering initiatives is the attempt to revive the product development protocol (PDP). In the past, many called the PDP the stillborn child of the Medical Device Amendments of 1976. Congress believed that the PDP would encourage the rapid development of innovative devices because it would be less expensive than the conventional two-step investigational device exemption (IDE) and PMA procedures. But the promise of the PDP has never been realized. Not one has ever been approved by FDA since 1976 as the basis for marketing clearance of a Class III device.

Why the renewed interest in the PDP if it never succeeded before? The answer lies in the belief of CDRH management that for some devices the PDP may yet prove to be faster than the standard IDE/ PMA approval processes.

Under the PDP process, any company seeking marketing approval of a Class III device could submit a summary outline of its protocol. FDA would respond within 30 days, possibly meeting with the applicant to discuss the eligibility of the device for the PDP process. Once the agency decided that the process applied to the specific device, the applicant would submit the PDP itself.

On October 1, 1997, FDA issued its revised draft of the PDP outline. This comprehensive outline includes a detailed description of the contents of a PDP, including device description, indications for use, and GMP/QSR information. There are also sections dealing with preclinical testing, the proposed clinical plan, and modifications made during the course of the product's development under the PDP. FDA has 120 days to conduct a review to determine whether the PDP should be filed. During that time, the agency can convene an advisory panel to evaluate the protocol and help decide whether it should be approved. After PDP approval, the sponsor would obtain institutional review board (IRB) approval and then commence the trial.

Upon completion of the protocol, the sponsor would submit a notice of completion delineating how the protocol was completed and specifying the study results. The comprehensive outline includes a detailed description of the contents of the notice as well as suggested formats for the clinical reports to be included in it. The notice of completion looks very much like a PMA application, and includes a summary of safety and effectiveness data, labeling and indications for use, a description of the study results and adverse effects, and the conclusions drawn from the study. Again as with PMA applications, follow-up notice-of-completion reports will have to be filed, and modifications to the PDP-approved device may require supplemental PDP filings.

FDA has been urging many companies to pursue the PDP process. However, no company should do so until the following concerns are addressed. First, is the PDP process realistic for novel Class III devices, which do not have a long history of regulatory approvals and recognized study plans and end points? And second, would the PDP provide the certainty and feedback on clinical trial design that companies seek, given how FDA has tended to raise its expectations for study end points and clinical trial requirements under the traditional PMA process?

In my view, it is unlikely that the PDP process would be a significant advance over the existing IDE/PMA process, especially for novel Class III devices. In a July 1997 advisory panel meeting for obstetrics and gynecology devices, members of the panel expressed some concern over whether the panel would be able to provide adequate feedback for a novel device. Panel member Michael Diamond of Wayne State University School of Medicine was quoted as saying, "As I understand it, both the company and the panel will be asked to review basically a concept and come up with a clinical study design, including inclusions/exclusions, before any animal data or even any in vitro data is available." Diamond and other panel members expressed some skepticism about whether the panel could offer educated advice on a study protocol where there is a paucity of preclinical, clinical, or other data.³

These concerns are well taken. It is probably unrealistic to believe that FDA and a panel can make decisions about clinical study design for a novel Class III device, which would then be written in concrete in a PDP, without any preclinical data or other information regarding the safety and efficacy profile of the device. Consequently, the PDP process would appear better suited to devices with a long history of PMA clearances, such as monofocal intraocular lenses, renal lithotripters, or extended-wear contact lenses. Similarly, where there are extensive guidance documents that set forth detailed requirements for the clinical study of Class III devices, such as those written for excimer lasers for photorefractive keratectomy, PDPs would more likely be of benefit. Well-understood guidelines, a history of agency review, and accepted end points are the key elements to consider in deciding whether to pursue the PDP path.

FDA very much wants the PDP process to be a usable alternative to the IDE/PMA process. The first companies through the process will likely be given much leeway and deference. On the other hand, FDA is inventing the PDP process as it goes along, and thus any novel issues could lead to much debate and delay. It remains to be seen whether the stillborn child of the 1976 device amendments can be brought to life.

REENGINEERING PMAs

The 120-Day Review. One of the reengineering proposals being considered at FDA would establish a separate streamlined review track for well-characterized devices. In this model, PMAs could be submitted in a user-friendly format and receive quick review (under 120 days). This new concept shares some characteristics with the real-time PMA supplement review program (described below).

The 120-day PMA, like the PDP, would be most appropriate for devices with a long history of FDA review, a well-known safety and efficacy profile, and accepted FDA guidance documents. It has been reported that the pilot for this program will initially involve PMAs for in vitro diagnostics (IVDs) reviewed by the Division of Clinical Laboratory Devices.⁴ Examples of IVDs that might be eligible for the pilot include hepatitis A and human papillomavirus tests. The 120-day PMA review concept is still in its infancy, but the use of fast-track review for me-too Class III devices seems entirely appropriate and should be easily implemented.

Real-Time PMA Supplements. This program began in April 1996 as a pilot and now is being extended to all PMA supplements in all ODE divisions. The purpose of the program is to speed up clearance of PMA supplements for minor modifications. Under FDA regulations, any change to a Class III PMA-approved device requires a supplemental approval before marketing if the change affects safety or effectiveness.

FDA has estimated that about half the 500 or so PMA supplements filed each year could be given real-time reviews. Since the program began, approximately 73 real-time reviews have been conducted.⁵

In real-time review, several weeks before filing a PMA supplement, the applicant is directed to set up a meeting, videoconference, or telephone conference with FDA to discuss the reasons for the submission, the time frame for the filing, and the reasons why the supplement qualifies for real-time review. On April 30, 1997, FDA issued a document titled "Real-Time Review Program for Premarket Approval Supplements" (http://www.fda. gov/cdrh/ode/realtim2.html). This document describes in detail the kinds of devices and changes that might qualify for real-time review in each ODE division. For example, for the Intraocular and Corneal Implant Devices Branch of the Ophthalmic Devices Division, qualifying changes are a change from one well-known sterilization method to another, minor design changes, minor material modifications, minor labeling changes, and changes similar to those in other previously approved supplements.

The results of the real-time review program so far are very good. The average time from the company meeting or conference with FDA to the sign-off on the supplement has averaged 5 days. The average approval time from initial contact to approval so far is 27 days.

The industry should recognize that the real-time PMA supplement review program is fairly narrow and is limited typically to minor labeling, sterilization, or expiration-dating types of changes. Nevertheless, by using presubmission discussions, possibly electronic submissions, and appropriate conferences or meetings, the PMA supplement process can be speeded up. The average review time for PMA supplements in fiscal year 1996 was 216 days of total elapsed time. Certainly the real-time review process will be able to improve upon that performance, and if successful it could be expanded to more significant types of modifications.

CONCLUSION

The foregoing reengineering initiatives are just a few of the many important changes now under review at FDA. CDRH is to be commended for its efforts in rethinking a program that has been the subject of extensive congressional and industry criticism over the last seven years. The initiatives are in harmony with the directions and goals of the FDA Modernization Act. It is to be hoped that the initiatives discussed above, as well as the others under consideration, will lead to a more streamlined, efficient, and user-friendly system. The end result should be a more-rapid introduction into the marketplace of important devices. The medical community and the patients who rely upon medical technology will be the ultimate beneficiaries of these initiatives.

REFERENCES

1. Federal Register, 62 FR:49247 (September 19, 1997); http://www.fda.gov/cdrh/ode/parad510.html.

2. 21 USC 513(a)(1)(B).

3. M-D-D-I Reports ("The Gray Sheet"), July 21, 1997.

4. M-D-D-I Reports ("The Gray Sheet"), November 3, 1997.

5. Burlington DB, letter to the medical device community, September 8, 1997.

Jonathan S. Kahan is a partner in the law firm of Hogan & Hartson (Washington, DC) and a contributing editor and editorial advisory board member for MD&DI.

Copyright ©1998 Medical Device & Diagnostic Industry