Originally Published IVD Technology January/February 2003
Regulations & Standards
Managing premarket approval for IVDsPart 2: Avoiding the pitfalls in PMA submissions
Thomas M. Tsakeris
|
Sidebar: |
FDA employs several methods for determining whether a proposed new IVD should be subject to review through the premarket approval (PMA) process. The first installment of this article (IVD Technology, November/December 2002) examined the criteria that the agency uses to make this determination and outlined a variety of ways that IVD manufacturers can determine in advance whether a proposed new IVD may be subject to the PMA process. New IVDs that cannot be associated with an identifiable predicate device and which FDA believes present an unusually high risk to the patient in the event of an erroneous test result are candidates for the PMA process. However, certain new IVDs without a clear identifiable predicate, but where FDA believes the patient risk is low, may be candidates for reclassification (see sidebar).
 |
| Thomas M. Tsakeris is a former director of FDA's Division of Clinical Laboratory Devices. He is now president of Devices and Diagnostics Consulting Group (Rockville, MD) and a member of the IVD Technology editorial advisory board. |
This installment examines how IVD manufacturers can best plan for and manage the process of submitting a PMA application, with special attention to facilitating the processing of an application and avoiding common pitfalls. Included are recommendations about alternatives within the PMA process, as well as preliminary notes on the role of FDA advisory panels in approving IVD PMAs.
Meeting with FDA
Companies should begin their PMA planning as early as possible—preferably as soon as it becomes apparent that FDA will require a PMA for the proposed IVD. Developing a PMA application can be very resource intensive, especially for small companies. To ensure the optimal involvement of key company personnel, companies should take a project management approach to the planning and execution of all PMA activities, including pre- and postsubmission interactions with FDA.
As a first step, companies should develop a plan for assembling all the information that will be required for the PMA application. As part of this process, the company should identify the individuals who will be responsible for developing the manufacturing, product description, product labeling, preclinical, and clinical sections of the application. It is highly advisable for this PMA team to include a biostatistician who will be responsible for developing the data analysis section of the PMA application and for responding to questions raised by FDA's biometrics staff. This biostatistician should be made available to participate in an FDA advisory panel hearing about the company's PMA should a panel be required.
Once the company's PMA team has been assembled and has developed a concrete plan for its application, it is a good idea for members of the team to meet with FDA reviewers to discuss the plan. FDA officials are usually agreeable to meeting on more than one occasion, but they are not always patient with meetings whose agenda is unfocused or open-ended. They prefer meetings in which substantive progress can be made toward resolving critical issues that have emerged during presubmission discussions.
In determination meetings, FDA provides the prospective PMA sponsor with a formal determination about the scope and type of studies and data that the agency will require in support of a proposed product's claims to safety and effectiveness. The product of a determination meeting is a written assessment of FDA's requirements, which the agency provides to the prospective sponsor within 30 days following the meeting.
The purpose of an agreement meeting is for FDA and the prospective PMA sponsor to forge an agreement regarding the salient parameters or critical elements of the investigational plan for a PMA device, including particularly the clinical study protocol. Critical elements to be agreed upon commonly include the study design, study objectives, clinical endpoints, and procedures for data collection and data analysis. An agreement reached in such a meeting is considered binding on both FDA and the sponsor, unless both later agree to modify it.
Companies should give careful consideration to their decision to request a collaborative meeting with FDA. Although both determination and agreement meetings are useful for restricting FDA's ability to change its mind about data or study requirements, they also have the same effect on the PMA sponsor. Experience suggests that it is usually not FDA that wants to tweak a product parameter or alter a critical element in a clinical protocol. Such requests are much more likely to come from manufacturers. Consequently, it is important for IVD PMA sponsors to exercise caution before requesting a collaborative meeting that will result in a binding agreement with FDA.
IVD sponsors may find it preferable to take advantage of nonbinding pre-IDE meetings, especially since there is little evidence to demonstrate that collaborative meetings produce better outcomes than nonbinding, informal meetings. Moreover, the latter can still enable sponsors to obtain valuable FDA feedback without the risks potentially associated with entering into binding agreements.
Traditional versus Modular PMA Submissions
Manufacturers have traditionally submitted their PMA applications as a single, complete set of documents for filing and review. In recent years, however, FDA has permitted sponsors to take a modular approach to submitting their applications.2
In the modular approach, the sponsor is permitted to separate the key sections of a traditional PMA (e.g., the sections containing preclinical data, clinical data, and manufacturing data), and to submit them at different times according to an agreed-upon schedule. As each module is submitted, it is then reviewed by appropriate FDA staff. In theory, when the final module is submitted, overall review of the PMA application should be at an advanced stage, since the preceding modules would have been reviewed already.
One advantage of the modular approach is that the sponsor can submit modules as their required information becomes available. For example, the manufacturer may choose to submit its preclinical and clinical data sections as soon as it has completed the necessary studies, leaving the submission of manufacturing data for later. In this way, the sponsor avoids delaying FDA's review of the all-important preclinical and clinical information, but gains the additional time necessary to bring its manufacturing facility into a state of compliance with the good manufacturing practices requirements of the agency's quality system regulation.
Sadly, the modular approach has not enabled FDA to achieve greater efficiency in the review of PMA submissions. To date, the processing times for traditional and modular PMAs are not appreciably different.3 Nevertheless, PMA sponsors may find that the flexibility of the modular approach is beneficial in permitting a more efficient utilization of staff resources.
Expedited Review
IVD manufacturers should consider requesting expedited review of their PMA submissions. When FDA grants expedited-review status, the application is moved to the front of the review queue, that is, reviewed before any other pending nonexpedited PMAs.
The agency uses several criteria to determine whether expedited-review status should be granted to a PMA application.4 Successful candidates are those that offer a clear, substantive diagnostic or therapeutic benefit in respect to a life-threatening or irreversible debilitating condition, or those whose use offers a significant advantage in safety and effectiveness over existing devices.
Notwithstanding a widespread industry misconception, expedited-review status does not necessarily translate into an accelerated review. Accelerated review is available only for devices that offer substantial diagnostic or therapeutic benefit in respect to life-threatening or irreversible debilitating conditions, and for which no alternative modalities exist. Most expedited reviews are not approved in much shorter time than nonexpedited reviews.4
When managing their public relations, PMA sponsors should carefully consider how best to present the fact that their PMA applications have been granted expedited review. Such an announcement can help to build stakeholder enthusiasm and can even result in increased stock prices for publicly traded companies. But such initial gains can just as easily be lost if expectations for a rapid approval are dashed by a protracted FDA review.
Preparing the PMA
Preparation of a PMA application is a complex topic that cannot be covered fully in the context of this overview. For detailed information about how to prepare PMA documents, IVD manufacturers should review the many PMA guidance documents available via the Web site of FDA's Center for Devices and Radiological Health (http://www. fda.gov/cdrh). Nevertheless, the following pointers about how to prepare the core PMA information may be useful.
When assessing the safety and effectiveness of a PMA device, FDA's primary concern is to determine whether the manufacturer's statement of indications for use—as supported by the established performance characteristics of the device and by data collected from a valid scientific investigation—represents a discernible clinical utility that significantly outweighs any known risks to patients.5 To ensure that their PMA applications will have a successful outcome, IVD manufacturers must therefore carefully consider two key factors: the exact wording of the intended-use statement; and the design, implementation, and outcome of the clinical investigation. In short, the design and execution of the clinical investigation should produce sufficient analyzable data to represent how the device will perform under intended-use conditions.
Intended Use and Indications for Use. The intended-use statement for an IVD PMA typically includes two components: a description of the functionality of the test (e.g., an immunoassay for the detection of analyte x in serum or plasma), and a statement of the clinical utility of the test (sometimes referred to as the indications for use).
The indications-for-use component can be thought of as the mission statement of the IVD PMA. The sponsor should choose each word of the statement carefully so as to communicate as precisely as possible the clinical application of the test (e.g., for screening, diagnosis, monitoring, risk assessment, or prognosis), the specific disease or condition of clinical interest that the test is intended to address, the intended target patient population, the intended user, and any conditions of use.
In their desire to attain broad market coverage, IVD sponsors too often present indications-for-use statements that are vague, ambiguous, or overly ambitious. Such statements frequently result in FDA finding the PMA application deficient, and requesting the sponsor to clarify or be more specific about some aspect of the application. If an IVD is intended to be used in risk assessment for a given disease or condition, for example, the sponsor may need to specify whether the test is to be used as an adjunct to or in conjunction with other risk tests or risk factors. Depending on the clinical situation, the terms an adjunct to and in conjunction with can suggest very different meanings. A test that is described as adjunctive may be assumed to have a relatively subordinate role, serving primarily to optimize or refine the information derived from other risk tests or risk factors. Alternatively, a test that is intended for use in conjunction with other tests may be assumed to offer greater independent value, often by providing additional risk information derived from the study of a unique clinical parameter, such as a novel analyte.
Clinical Study Design. Designing and executing a proper clinical study is often the most serious challenge for IVD PMA sponsors. This is particularly so when the IVD under study is intended to address such diseases or conditions as certain cancers or Alzheimer's disease, which develop slowly, have multiple stages, and may be marked by analytes that are also present in other related conditions. It can also be challenging to study IVDs for diseases or conditions whose incidence is relatively low, making it difficult to obtain patients or samples in a quantity sufficient for the analysis to achieve appropriate statistical power.
Conducting a well-controlled clinical study can be time-consuming and expensive—but the alternatives can be even worse. Many prospective IVD PMA sponsors—particularly small start-ups—can be readily misled into embarking on a study that contains flawed elements. Such flawed studies almost always result in difficult and protracted FDA reviews. Some notable examples of such study design flaws include the following. FDA is familiar with such difficulties and is usually willing to help sponsors design an appropriate least-burdensome study. To ensure that critical study elements are clearly defined and achievable, however, it is also advisable for IVD sponsors to retain their own medical advisory board of experts.
FDA Review of PMA Applications
Although the administrative mechanics of FDA's review process for PMAs differ from those used for the premarket notification (510(k)) process, there are many basic similarities. Unlike the practice for 510(k)s, FDA performs a formal administrative filing review for each incoming PMA application. As a result of this review, the agency typically notifies the PMA sponsor within 45 days that the PMA is filed, filed with some cited minor deficiencies, or not filed because of major deficiencies.
Once the PMA application has been officially filed, it undergoes an in-depth scientific review not unlike a 510(k) review. If the agency review staff require additional information to complete their review, they may communicate with the sponsor via faxed or mailed correspondence, e-mails, or telephone calls. As in the case of the 510(k) review process, FDA usually communicates major technical deficiencies in writing. In that case, the FDA review clock stops when the correspondence is issued, and starts over at zero when the sponsor responds in writing.
FDA also refers many PMA applications to standing advisory panels of outside experts, whose task is to make a recommendation about the approvability of the applications. In recent years, FDA has become increasingly discriminating about when to convene a panel hearing. In general, PMAs for first-of-a-kind IVDs are referred for panel review and panel hearings. Also referred for additional panel review are IVD PMAs that have been significantly modified, where FDA believes that a new issue of safety and effectiveness may exist. Most such cases involve a new indication for use or a new target population for the IVD. FDA usually notifies the PMA sponsor that its application will be subject to a panel review at the same time it sends notification that the PMA has been officially filed.
Panel input into the PMA review process occurs during the postfiling phase. Typically, one or more panel members are selected as primary reviewers and are asked to critique the PMA. The panelists are usually chosen because of their focused expertise in the subject area covered by the PMA.
At the 100-day point in the review FDA must, upon written request by the PMA applicant, meet with the applicant to discuss the review status of the application. Prior to the meeting, FDA must inform the applicant in writing of any significant identified deficiencies and what information is required to correct those deficiencies.6
Assuming that no major deficiencies are found during this phase of the review, FDA will likely advise the sponsor that its PMA is scheduled for a full advisory panel hearing. The sponsor will then be asked to prepare and submit to FDA several premeeting documents intended both for panel review (a panel package) and for the public.7
Conclusion
Although the preparation of a PMA application is a complex undertaking, IVD manufacturers can streamline their processes and improve their chances of swift approval by taking a project management approach to the preparation of PMA documents. It is advisable for IVD PMA sponsors to consider all of their options as early as possible, including especially the potential utility of the de novo classification process.
The third and final installment of this article will examine the role of FDA advisory panels in the PMA process, and will offer recommendations about how IVD manufacturers can best prepare for a panel meeting.
References 1. Early Collaboration Meetings under the FDA Modernization Act (FDAMA); Final Guidance for Industry and FDA Staff, in CDRH Home Page [on-line] (Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, 1998 [cited 12 December 2002]); available from Internet: http://www.fda.gov/cdrh/ode/guidance/310.html.
2. Guidances for the Medical Device Industry on PMA Shell Development and Modular Review, in CDRH Home Page [on-line] (Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, 1998 [cited 12 December 2002]); available from Internet: http://www.fda.gov/cdrh/ode/pmashell.html.
3. Office of Device Evaluation Annual Report: Fiscal Year 2001, in CDRH Home Page [on-line] (Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, 2002 [cited 12 December 2002]); available from Internet: http://www.fda.gov/cdrh/annual/fy2001/ode/annualreport2001.html.
4. PMA/510(k) Expedited Review: Guidance for Industry and CDRH Staff, in CDRH Home Page [on-line] (Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, Program Operations Staff, 1998 [cited 12 December 2002]); available from Internet: http://www.fda.gov/cdrh/modact/expedite.pdf.
5. Code of Federal Regulations, 21 CFR 860.7.
6. Guidance on PMA Interactive Procedures for Day-100 Meetings and Subsequent Deficiencies for Use by CDRH and Industry, in CDRH Home Page [on-line] (Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, 1998 [cited 12 December 2002]); available from Internet: http://www.fda.gov/cdrh/modact/day100mt.html.
7. Guidance on Amended Procedures for Advisory Panel Meetings, in CDRH Home Page [on-line] (Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, 2000 [cited 12 December 2002]); available from Internet: http://www.fda.gov/cdrh/modact/amendpan.pdf. Thomas M. Tsakeris is a former director of FDA's Division of Clinical Laboratory Devices. He is now president of Devices and Diagnostics Consulting Group (Rockville, MD) and a member of the IVD Technology editorial advisory board.
Copyright ©2003 IVD Technology
