Archived Content

The content on this page is provided for reference purposes only. This content has not been altered or updated since it was archived.

Medical Devices

Minutes from Negotiation Meeting on MDUFA III Reauthorization, March 30, 2011

 

FDA – Industry MDUFA III Reauthorization Meeting

March 30, 2011, 12:00 – 5:00 pm

FDA Switzer Building, Washington, DC

Room 3005

Purpose

To finish previous discussion on additional information requests in 510(k)s, discuss interactive review practices, and discuss the roadmap for future meetings.  

Participants

FDA

Malcolm Bertoni

Office of the Commissioner (OC)

Ashley Boam

Center for Devices and Radiological Health (CDRH)

Nathan Brown           

Office of Chief Counsel (OCC)

Kate Cook

Center for Biologics Evaluation and Research (CBER)

Natalia Comella

CDRH

Christy Foreman

CDRH

William Hubbard

FDA Consultant

Elizabeth Hillebrenner

CDRH

Thinh Nguyen

OC

Don St. Pierre

CDRH

Ruth Watson

Office of Legislation (OL)

Nicole Wolanski

CDRH

Barbara Zimmerman

CDRH

Industry                                               

Susan Alpert  

Medtronic (representing AdvaMed)

Hans Beinke

Siemens (representing MITA)

David Fisher

Medical Imaging Technology Alliance (MITA)

John Ford

Abbott Laboratories (representing AdvaMed)

Elisabeth George

Phillips (representing MITA)

Donald Horton

Laboratory Corporation of America Holdings (representing ACLA)

Mark Leahey

Medical Device Manufacturers Association (MDMA)

Joseph Levitt

Hogan Lovells US LLP (representing AdvaMed)

John Manthei

Latham and Watkins (representing (MDMA)

David Mongillo          

American Clinical Laboratories Association (ACLA)

Patricia Shrader

Becton Dickinson (representing AdvaMed)

Chuck Silverman

Quest Diagnostics (representing ACLA)

Janet Trunzo

Advanced Medical Technology Association (AdvaMed)

 

Meeting Start Time:  12:00 pm

Industry Perspectives on Additional Information Requests 

In follow-up to the previous meeting in which the 510(k) program was discussed, Industry described de-identified examples of additional information requests of concern to Industry.  The examples included cases described by Industry as late changes in FDA’s thinking, deviations from existing or recognized standards, changes in labeling requirements, unnecessary requirements, failure to communicate or clarify requirements, delays in holding meetings, inability to agree on a clinical protocol, and deviations from guidance documents.  Industry explained that the examples were not intended to be a statistically valid sampling.  Industry also noted the need to keep examples de-identified to protect Industry-reviewer relationships.  FDA noted that the lack of details made it difficult to understand the context for the requests and to determine whether the Agency’s requests were inappropriate. 

 

Industry questioned the threshold and processes by which FDA makes requests for additional information.  FDA explained that this includes clinical and scientific judgement.  In cases where different types of data could be used to answer a question, the Agency requests what they believe is least burdensome.  With high turnover rates and high reviewer to supervisor ratios, the Agency acknowledged that inconsistencies in requests may occur.  Industry requested that when the Agency asks for something new, they also provide clear guidance and the scientific justification for the request.  FDA replied that staff should be doing this now, as they are trained to write deficiencies in “four part harmony,” which includes a statement of what was provided, why the information provided is not sufficient, what additional information is needed, and why the requested additional information is needed to answer a regulatory or scientific question.  In such cases where the sponsor does not understand the basis for the request, FDA suggested that they contact the reviewer or branch chief to discuss it.

 

Industry expressed concerns regarding experiences that they believe reflect changing regulatory requirements.  They assert that sometimes when guidance documents are followed, FDA still requests additional information.  Industry requested that when new information is needed for scientific reasons, such changes be reflected in updated guidance documents rather than one 510(k) at a time.  FDA explained that good guidance practices are dictated by law.  The guidance development process is time consuming and cannot always keep pace with evolving technology and new information about safety and effectiveness; as a result,  when devices change, additional information may be needed.  FDA also suggested that if a published guidance document is several years old, sponsors could call the Branch Chief to find out if any information should be provided in addition to that outlined in the guidance. 

 

Without discussing specific examples, Industry also indicated they have seen cases in which the Agency raised new questions in the second additional information request that the Agency had not raised in their first letter, deficiencies which raised concerns with previously cleared devices and hypothetical concerns (i.e., deficiencies asking for supporting data when there is no evidence of a scientific issue).  FDA suggested that sponsors contact the Branch Chief when a request does not make sense to them.  FDA also explained that they sometimes learn information through review of a submission which may not be disclosed to other sponsors.  For example, if review of clinical data in a marketing application shows an adverse event that was not anticipated, the next sponsor who submits an IDE for the same type of device will be asked to look for this adverse event in the clinical study, possibly as a secondary endpoint.  Rather than holding every subsequent study to the same standards as the first for the device type, FDA works with sponsors to improve the next study design to better inform the labeling and physicians who will use the device.

 

Industry asked about supervisors' roles in reviewing proposed requests for additional information for consistency and appropriateness. FDA explained that team meetings are held.  so that all experts can weigh in if criteria are being changed; therefore, the outcome is not driven by a single reviewer’s opinion.  For example, many Divisions hold “Clinical Rounds” in which multiple medical officers and statisticians are brought together to obtain consensus on challenging clinical trial design and interpretation issues.  Additionally, the Agency is considering adding a Center Science Council as a mechanism to escalate such cases to a higher level and ensure staff are applying criteria consistently. 

 

Industry asserted that with recent reviewer turnover, they have experienced an increase in “nice to know” questions rather than “need to know” driven by appropriate  regulatory or scientific concerns; Industry further indicated that these questions are often asked without providing the basis for such requests.  The Agency replied that Branch Chiefs typically ask reviewers: what will you do with the answer to this deficiency?  If the response will influence the overall decision on the submission, how to label the device, or indications for use, then it is considered an appropriate question.  Otherwise, the Branch Chief may determine the question is unnecessary and remove it.  Industry agreed with the process described by FDA but expressed concern that it is not imposed consistently.  FDA replied that it is looking at potential new management structures and supervisor-to-reviewer ratios to ensure sufficient time and oversight for screening questions before they are submitted.  Industry asked if the issue of new questions raised for the first time in the second letter should be flagged by Branch Chiefs.  FDA replied that this is within the scope of their review, but they are currently limited in their ability to catch all issues consistently given the high number of reviewers (and consequently submissions) each individual Branch Chief oversees.  Furthermore, the Agency does not have the capacity for further quality control over Additional Information letters past Branch Chief review given the 510(k) submission volume.  In contrast, PMA major deficiency letters are reviewed by the Branch Chief, Division Director (or deputy), PMA staff, and in the case of Expedited PMAs, the Office Director (or deputy).

 

FDA Perspectives on Interactive Review 

FDA first defined the scope of Interactive Review (IR), assessed based on the applicable guidance document.  FDA’s presentation focused on requests for minor deficiencies noted during the review of a submission.  FDA engages in a broad spectrum of interactions with Industry such as informal requests for clarification not associated with a submission, the pre-IDE process, and regular check-ins on long-term projects; however, these are not covered in the guidance document and are not formally considered IR. 

 

Although various forms and degrees of IR had been employed by the Agency prior to MDUFA, it was formalized as part of MDUFA II.  Specifically, the commitment letter states that, “The [A]gency will continue to incorporate an interactive review process to provide for, and encourage, informal communication between FDA and sponsors to facilitate timely completion of the review process based on accurate and complete information. Interactive review entails responsibilities for both FDA and sponsors.”  The shared objectives were to prevent unnecessary delays in the completion of the review, avoid surprises to the sponsor at the end of the review process, minimize the number of review cycles and extent of review questions conveyed through formal requests for additional information, and ensure timely responses from sponsors.  The commitment letter further outlined the following provisions:  “There should be regular, informal communication from FDA to seek clarification on issues that can be resolved without substantive review or analysis. When appropriate, FDA will also informally communicate substantive review issues if FDA determines that it will facilitate a timely and efficient review process…  Whenever FDA informally requests additional information, the sponsor and FDA will determine an acceptable timeframe for submission.  If the information is not received within the agreed upon timeframe or the information is incomplete, the application will be placed on hold…until the information is received.”

 

Industry asked if it is standard procedure to alert sponsors that they will be getting an NSE letter.  The Agency explained that they do not provide an explicit warning that an NSE letter is on its way because negative decisions must be disclosed to the SEC; however, the Agency typically sends at least one AI letter prior to NSE, which indicates that pathway is likely.  The Agency does not use IR to suggest an NSE is likely because IR requests do not require management sign-off. 

 

To implement the commitment described above, FDA published a guidance document on IR in December 2007 (with minor updates in February 2008), added an IR log to the document tracking database, and trained all staff in November 2007.  A recording of the November 2007 training has recently been re-posted to the internal FDA website to serve as a resource to staff.  The Agency agreed to consider updating training following MDUFA III discussions.  In response to Industry’s question, FDA confirmed that IR is an element of the core competencies for reviewers.  CDRH interviewed ODE and OIVD staff on how they use IR, documentation of IR, and when they were more or less likely to use it.  This feedback, plus the experience of Center experts, reveals a number of challenges with the program as designed and as implemented.

 

Data regarding use of IR are limited as the IR log is a voluntary field in the submission tracking system.  Interviews revealed that reviewers are not consistent in documenting their use of IR, generally because it takes additional time.  Some log each interaction, others log a summary of interactions, and some only document IR in their review memos—which are not currently searchable.  This results in uncertainty in FDA’s assessment of how often IR is used across submissions as well as how intensively it is used within each submission.  Nevertheless, FDA presented available data showing that the number of 510(k)s with at least one IR logged in the tracking system has decreased from 47% in 2008 to 36% in 2010.  Although the data reflect under-reporting, FDA suggested that the trend is likely correct.  The Agency believes the observed decrease in IR is due to tightening capacity.  As reviewer workload approaches or exceeds the maximum volume they can handle while meeting goals, completion of each submission is occurring closer to the goal, reducing available time in which to use IR.  Industry asked what percent of reviewers use IR; FDA could not provide a definitive response due to the limitations of under-reporting described above.

 

Questions regarding use of IR were included in the 2010 ODE/OIVD staff perception survey discussed at the previous meeting.  Since 2005, 38% of staff indicate their use of IR has increased, 43% indicated it remained the same, and 9% indicated it decreased.  Since 2005, 20% of staff indicate the instances in which IR benefited the overall review have increased, 64% indicated it remained the same, and 10% indicated it decreased; 7% of staff indicated that IR does not work at all.  Staff were also asked to select all reasons why they may choose not to use IR.  Results show that 32% of staff have not used IR on occasions where they were planning to send a letter anyway (for more significant issues), 32% have encountered scenarios where there was simply not enough time left on the clock, and 31% have decided not to use IR because the data they were requesting required new testing which would take a long time to conduct.  Other reasons had smaller percentages.  Of note, 52% of staff indicated that this survey question was not applicable as they use IR.  Industry suggested that in the 31% of cases where IR was not used because the reviewer thought it would take too long to collect the necessary data, the reviewer should have first asked the sponsor if they have such data readily available.  Often sponsors do have additional data generated for other global regulatory purposes available to send to FDA immediately if asked. 

 

The Agency explained that the reviewer’s decision to use IR is multifactorial and depends on the nature of the questions (e.g., are they minor, not complex, not requesting new data or significant reanalyses?), the status of the submission with respect to the review clock (i.e., the likelihood of receiving the response in time to complete the review, the ability to place the submission on hold quickly if the sponsor does not respond on time, and the ability of the reviewer to review the response prior to the goal while balancing other workload and potentially involving a consulting reviewer), whether there are major questions that will require sending a letter or otherwise placing the submission on hold, and historical responsiveness of the sponsor (i.e., does the sponsor typically provide complete and timely responses to other requests without negotiating the standard of scientific evidence.)  The decision to use IR or not is often resource-driven. 

 

There are advantages and disadvantages to IR from the review clock perspective.  IR can help staff meet goals by avoiding the need for a formal response in hard copy (which adds processing time) and by minimizing the delay between initial submission review and review of the response (which eliminates the need for reviewers to re-familiarize themselves with the device/submission.)  IR can make it more difficult for staff to meet the goal if the reviewer/consultant(s) does not have the capacity to review the response in time.  In cases where sponsors do not respond in the allotted time and a formal letter must be sent, the current IR process penalizes FDA because the time spent waiting for an IR response counts on the FDA review clock.  In some cases, FDA chooses to miss performance goals when the Agency believes that interacting past the target goal will bring a safe and effective device to market more quickly than putting the submission on hold and awaiting a formal response.

 

The Agency acknowledged that its use of IR is inconsistent.  Based on recent staff interviews, there appear to be different interpretations of the guidance among staff.  Some review teams will not use IR until the major issues are resolved, while others will use IR to reduce the number of issues prior to sending a hold/major deficiency letter.  While sponsors appreciate resolution of minor issues via IR even when the submission will be placed on hold anyway, this process results in more time on FDA’s clock, can set an expectation that this should occur for all submissions, and can be misinterpreted to suggest that because FDA is using IR there are no major/significant deficiencies.  Some reviewers even use IR to request more than “minor” information such as test data or reanalyses, which reflects more substantive information than was intended in the guidance.  Industry asserted that the guidance did intend for larger questions to be asked via IR, only such responses were to be developed “off the clock.”  FDA refers to this form of IR as “alerts” to sponsors about major (PMA) or significant (510(k)) deficiencies prior to sending a formal letter.  This requires appropriate management review and concurrence prior to communication with the sponsor to ensure that the issues are not mitigated by data being reviewed by another team member, consistent with Branch, Division, or Office policy, and appropriately worded to clearly outline the deficiency (i.e., per “four part harmony”).  This means that management will have to review the information and question twice – once during IR and again during review of the deficiency letter.  In some cases, providing the alert has led to the sponsor seeking to negotiate the scientific review standards or to submit a version of the response prior to issuance of a letter. 

 

In summary, FDA believes that appropriate use of IR can reduce unnecessary review cycles and therefore reduce total time to a final decision.  FDA believes that IR works best when it is initiated by FDA, initial submissions are complete, reviewers interact on minor issues when only minor issues are left, and sponsors provide a complete response in the time specified.  Unfortunately, the current process does not allow sufficient flexibility in certain circumstances, such as the scenarios in which FDA chooses to use IR to eliminate a cycle and achieve a shorter time to MDUFA decision at the cost of missing the goal and when the sponsor does not respond in a timely manner.  Although helpful to Industry, trying to provide an “alert” of upcoming major issues has been challenging from a resource perspective.

 

Industry Perspectives on Interactive Review 

Industry presented de-identified examples of both positive and negative experiences with IR.  FDA and Industry discussed their respective understandings of IR and Telephone Holds (THs).  FDA considers IR to refer only to “on the clock” phone, e-mail, or fax requests for additional information.  Industry understood IR to include this as well as THs.  FDA explained that a TH results in stopping the clock and is followed by a formal letter indicating the submission is on hold and responses should be sent within a given time to the Document Mail Center.  Industry explained that they are not focussed on “on the clock” vs “off the clock”, but rather on total time to market regardless of the review clock time. 

 

FDA noted that the program analysis to date has revealed this fundamental concern about the MDUFA program:  meeting the MDUFA goals does not necessarily result in the desired outcome of timely access to safe and effective devices, as evidenced by increasing number of review cycles per 510(k) and increasing average total time to final decision.  Industry also noted that predictability is another key concern, given that industry needs to establish plans for the product life cycle, and if FDA changes the requirements during the course of the product development and regulatory review process, that disrupts their business model.

 

Both FDA and Industry agreed to continue thinking about and discussing potential improvements to the program structure to address this.  It is a challenge to develop a program with goals that FDA can manage that also encourage speed to market, which involves shared FDA and Industry responsibility.  FDA and Industry share responsibility for developing a program that results in timely access to safe and effective devices.


Industry also presented concerns that IR is not used by all reviewers and when it is used, it is inconsistent.  Industry pointed out one purpose of IR in the commitment letter was to improve interactions between FDA and Industry, and that the IR guidance puts little emphasis on goals.  The scope of IR outlined in the guidance includes many of the same deficiencies identified in FDA’s previous analysis of AI letters, which points to the variation in use of IR.  Industry also noted that the “least burdensome” principle is an important part of the IR guidance.  Predictability is of paramount importance to Industry, which identified the need for a well defined process that is clearly understood by both parties and tracked.  FDA agreed to consider improvements to the IR process under MDUFA III, and further suggested the group carefully consider the type of data that should be collected to monitor its success.  

 

Roadmap for Future Meetings 

FDA reiterated that the Agency and Industry were responsible to transmit a package to Congress by January 2012.  Congress has recently emphasized to FDA the importance of receiving this package on time.  When considering the need for HHS and OMB review and publication of an FR Notice and public meeting, FDA suggested that the group has three months to develop a package of recommendations.  The Agency plans to present a comprehensive set of proposals on April 13, 2011, developed based on concerns heard from Industry and the analyses of program data presented to date.  FDA suggested holding working group teleconferences between this meeting and the May 4, 2011 negotiation meeting to clarify the details for various aspects of the proposal.

 

Next Meeting 

All agreed to meet from 10:00am – 4:30pm on April 13, 2011.  The agenda will include a brief discussion of IDEs, in addition to an overview of FDA’s proposed solution.

 

Meeting End Time:  5:00 pm

Page Last Updated: 07/16/2014
Note: If you need help accessing information in different file formats, see Instructions for Downloading Viewers and Players.