FDA Performance Review Reveals Problems

Originally Published MDDI August 2001

Washington Wrap-Up

FDA Performance Review Reveals Problems

The CDRH director's year-end report admits to a marked slowdown in device review time.

James G. Dickinson

TMJ Patient Group Protests | Third-Party GMP Inspections | Unapproved IDE Changes Are OK | More on Reprocessed Devices

Stepping past the confines of FY 2000, Feigal—a former ombudsman in FDA's biologics center—acknowledged that device PMA review time has slowed by 28% this year, and it is taking CDRH an additional 100 days, on average, to issue an approval.

Eighty percent of the slowdown is on the manufacturers' side, the report says, accounting for 80 days of the additional approval time. Last year, CDRH was reviewing PMAs within 363 days, on average; the average approval time this year has jumped to 463 days. PMA supplement approval time has remained relatively flat, taking 122 days this year compared to 118 last year.

Feigal says 510(k) clearance is taking 102 days this year, the same amount of time it took in 2000. Device sponsors are also using more nontraditional review clearances, such as third-party reviews and special or abbreviated 510(k) processes. For example, in 2000, 62% more special 510(k)s were submitted than were submitted in 1999 (583 vs. 361).

The fact that the majority of CDRH dispute resolutions are being resolved in industry's favor raised more than a few eyebrows at an AdvaMed submission workshop in June. The trend prompted center regulations and policy deputy director Linda Kahan to suggest that device reviewers might need more training on the center's "least-burdensome" principles.

As if to confirm an internal deficiency, Weinstein shared with the AdvaMed attendees recent survey data on what he referred to as "early collaboration meetings" between industry and CDRH. Industry respondents to the survey said CDRH reviewers were not applying least-burdensome principles during agreement and determination meetings. For example, the majority of industry respondents said preclinical testing was not considered in lieu of clinical data, and reviewers did not consider previously collected non–U.S. data, literature, or registry data.

CDRH science and regulatory policy deputy director Phil Phillips told the workshop that one of the obstacles in integrating least-burdensome concepts into CDRH practices is that bureaucracies like FDA don't change easily. "Those responsible for overly burdensome regulation do not recognize the unnecessary burden that they impose," he said. Phillips added that change will require a commitment and will need to be managed from the top.

Preliminary training for CDRH staff and advisory panel members has been completed, according to Phillips. The least-burdensome draft guidance was open for public comment until August 1, 2001. Phillips encouraged the audience to submit feedback, especially if the document seemed unclear or was missing critical elements. Once the draft is finalized, FDA will conduct additional training of its personnel, he added.

TMJ Patient Group Protests

Ideological arguments (i.e., those stating that FDA should approve only inherently safe products) surfaced again in a May petition to FDA from the TMJ Association, the largest U.S. patient advocacy group for people with temporomandibular joint (TMJ) conditions. The association sought a public hearing on the agency's February 27 approval of the TMJ Fossa-Eminence, a partial prosthesis from Golden, CO–based TMJ Implants Inc.

The device was approved after ODE director Bernard Statland intervened in a review-staff dispute with the sponsor. He ordered full labeling disclosure of device and procedure risks, thereby sharing FDA's risk-benefit assessment role with practitioners and patients.

In a press release, the association called the approval—which capped 20 months of acrimonious review—"unconscionable and cowardly," because it "doesn't just put patients painfully suffering from TMJ at risk, it asks the patients to share the risk." The association added that the approval "contradicted the scientific and medical conclusions of two dental products panels" and FDA's "own internal scientific staff."

In response, TMJ Implants CEO Robert Christensen said the association's petition ignored 40 years of successful surgeries that placed his device in over 10,000 patients. "Over 90%, perhaps even 95%, don't progress to total-joint replacement after receiving the Fossa-Eminence," Christensen said.

Third-Party GMP Inspections

Speaking during a break at a June 14 AdvaMed device submissions workshop in Washington, DC, center regulations and policy deputy director Linda Kahan said that device manufacturer inspections had dropped to an alarming level over the past five years. As a result, she said, CDRH is considering a third-party inspection program to help FDA inspect device companies every two years. Third-party inspections would allow outside certified auditors to perform quality system/GMP inspections in lieu of FDA investigators.

Budget cuts by the Clinton administration led to a 50% decline in inspections— falling from 3602 in 1996 to 1841 in FY 2000. Of these, 710 were classified as quality system/ GMP audits, 247 were for bioresearch monitoring, 156 were preapproval inspections, and 609 were related to mammography quality standards.

Kahan told the AdvaMed audience that FDA's device program performance plan targeted 28% of high-risk domestic manufacturing facilities for inspection last year. Only 13% were inspected, however, due to resource constraints. Optimistic that relief is on the way in President Bush's FY 2002 budget proposal (which calls for a $2.8 million increase for device- related inspections), Kahan said CDRH is projecting a rise in inspections to 22% of high-risk firms.

Kahan also said that FDA district offices are asking for alternative inspection programs to allow for faster inspections. She emphasized that CDRH discussions on new inspection techniques are in the early stages and could require regulation amendments or new legislation.

Unapproved IDE Changes Are OK

FDA says IDE changes are permitted to a device, its protocol, monitoring procedures, and labeling without prior FDA approval if the changes do not affect either patient safety or the "scientific soundness" of the study.

According to a new final guidance, Changes or Modifications during the Conduct of a Clinical Investigation, it is the sponsor's responsibility to consider the effect that any change made to the investigational plan may have on a clinical investigation and its data. The guidance is intended to implement a FDAMA provision that establishes criteria permitting sponsors to make certain modifications to investigational devices, including manufacturing changes and clinical protocols.

"Any change to the basic principles of operation of a device is considered to be a significant change and, thus, requires prior FDA approval," the guidance says. "In assessing the effect of a device design or manufacturing change, a risk analysis and supporting credible information should help to identify those changes that represent a significant change."

To assist sponsors in determining if a change represents a significant change, the document provides a decision-tree flowchart. To follow the flowchart, FDA recommends that sponsors use data generated by design control procedures or other credible information to determine whether the change has a significant effect on the device design. The guidance states that "credible information may include data generated under the design control procedures of [CFR sec.] 820.30, preclinical or animal testing, peer- reviewed published literature, or other reliable information gathered during a trial or marketing."

The new guidance may be accessed at http://www.fda.gov/cdrh/ode/guidance/1337.pdf.

More on Reprocessed Devices

FDA says it has identified several "technical concerns" as it formulates a policy on reviewing device premarket submissions for reprocessed single-use devices.

Speaking at a device reuse workshop in Phoenix on May 30, CDRH Office of Surveillance and Biometrics consumer safety officer Lily Ng said these concerns include controlling raw materials; defining specifications; identifying changes to OEM devices; cleaning, disinfecting, and sterilizing procedures; the functionality of reprocessed devices; and the labeling of reprocessed devices. Ng also outlined FDA's enforcement strategy for reprocessors, adding that the agency is implementing an inspection program for hospital reprocessors as it continues to inspect all commercial reprocessors.

Copyright ©2001 Medical Device & Diagnostic Industry