Originally Published MDDI July 2001

Washington Wrap-Up

A medical device company's stock was sent reeling after FDA reacted adversely—and publicly—to its press release. How much attention does the agency pay to Wall Street?

James G. Dickinson

How much can a company say about clinical trial results with an investigational medical device? Not as much as Guidant Corp. did in a May 2, 2001, news release about its Contak CD/ Easytrak system, a combination cardiac resynchronization therapy (CRT) and implantable cardioverter defibrillator (ICD) system.

FDA's first reaction to the press release was condemnatory and word of it flew quickly through the Internet to Wall Street, sending the company's stock to the lowest point it has seen in 30 months. Guidant's indignant protest prompted a second reaction by FDA that amounted to a rare public admission of error, sending the stock's price back up.

That embarrassing agency flip-flop may have given a false impression, however, to observers who don't spend a lot of time focusing on the nuances of FDA reversals. Guidant's four-page press release broke the rules—as do many other releases that FDA doesn't see—by overstating the clinical trial's positives and understating (or, actually, not stating) its key negatives. Some enterprising media reporters saw something in its buoyant language that prompted them to ask FDA's press office about it. They got the response they probably expected from veteran FDA press officer Sharon Snider.

According to news reports, Snider told reporters: "It appears from the press release as if Guidant is making claims of safety and effectiveness for an unapproved medical device." Reuters quoted her as adding: "It is a violation of the law to make such claims for a product that is in clinical studies and has not been approved."

In these days of up-to-the-minute Internet news, Snider's comments almost instantly sent Guidant's shares tumbling 12.7%, or $4.80, to a 30-month low of $33.00 on the New York Stock Exchange.

While watching Wall Street has never been part of FDA's mission, these days due note is made of such financial events when they are brought to the agency's attention. When Guidant protested to Snider's "new guard" superiors at the agency, it caused officials to closely examine the company's release. Within hours, associate commissioner for public affairs Larry Bachorik was able to tell reporters that the agency felt there may not be a great deal wrong with it after all.

This difference of opinion derives from a subtle shift in FDA enforcement philosophy, coerced by budget restrictions that many "old guard" FDAers saw as punitive. The old FDA that trained Snider was more likely than the new, post–user fees, post–Modernization Act FDA to take a strict view of its regulations and to fault companies for not being properly familiar with them. A softer, more tolerant attitude seems to have been emerging lately. Another new factor is that the news media are more likely to name their FDA press office informants, ditching an old, unwritten rule that a source be referred to only as "an FDA spokesperson."

CDRH Office of Compliance acting director Larry Spears told Reuters that although he had not yet seen the release, part of it had been read to him over the telephone. Based on that section, he found nothing in the release to be "alarming." He promised to study it more over the weekend.

On the following Monday, Spears stated that after analyzing the release he thought Snider had gone "too far" in her comments to the media and that at the time in question (Friday), "it was an inappropriate suggestion to say that this may have violated the law. We weren't there yet. Yes, we had a few concerns about it, but we were really not in an appropriate position to characterize that as a violation based on just looking at a press release and having no interaction with the company."

In such cases, FDA's new-guard approach is first to engage in dialogue with the company in such cases, rather than simply apply to the offending document what an experienced official sees as the plain language of the regulations. Generally, dialogue with the company is the normal procedure, and by policy it is a necessary first step before issuance of a warning letter. In this case, a warning letter is not a likely result, according to Spears. Snider had told a reporter on Friday that FDA typically issues warning letters in cases like this.

Apparently, that is no longer true. At least in Guidant's case, there was no track record of the company ignoring prior FDA cautions about its language. Spears did acknowledge, however, that FDA did have "concerns" about a buoyant subheading in the release that said the unapproved device "Saves Lives, Improves Quality of Life." In addition, the news release neglected to mention that the reported clinical trial failed to meet its primary end point, but instead quoted Guidant president of cardiac rhythm management Fred McCoy as saying he was "looking forward to making available to patients the first and only heart failure therapy device with the potential both to enhance patients' lives and to protect against sudden cardiac death." FDA's old guard viewed this as an objectionable statement—a public presumption that the product would be approved.

Within just four days, the agency's remedial public statements substantially repaired the damage done to Guidant's stock prices.

To the fleeting glances of those with other things on their minds, it may have seemed that FDA goofed, admitted it, and everything was fine again. But that will be true only if Guidant and other companies have learned from the episode to be more circumspect when describing clinical trial results.

FDA to Be Even Less 'Burdensome'

In its regulatory activities with the device industry, FDA says it will apply a FDAMA provision on the so-called least burdensome concept more broadly than the act originally intended. A newly released guidance document, The Least Burdensome Provisions of the FDA Modernization Act of 1997: Concept and Principles; Draft Guidance for FDA and Industry, says the provision "could affect almost all premarket regulatory activities, including presubmission meetings with industry, premarket submissions, and the development of guidance documents and regulations."

In the guidance, FDA defines least burdensome as a "successful means of addressing a premarket issue that involves the most appropriate investment of time, effort, and resources on the part of industry and FDA." According to the draft guidance, the concept should be applied consistently to all premarket activities, including the following:

For PMAs where clinical outcome can be predicted from nonclinical data, well-designed bench or animal testing can support approval. "Conditions where nonclinical data may meet the threshold for approval are typically those devices or modifications of approved devices for which information is available in the public domain," the document says. If clinical data are needed, alternatives to randomized, controlled clinical trials should be considered.

The use of scientifically valid surrogate end points and statistical methods, such as Bayesian analyses, should be considered when appropriate. The guidance says FDA and industry should rely on information available from earlier versions of the same device or from marketing experience with similar devices. "The role of postmarketing information should be considered in determining the appropriate type or amount of data that need to be collected in the premarket setting to support PMA approval." FDA feels the use of recognized standards can also streamline PMA submissions.

For 510(k)s, the guidance says information unrelated to the substantial equivalence decision should not be submitted to, or requested by, the agency. It adds that substantial equivalence will normally be determined by comparative device descriptions, including performance characteristics; performance testing should be submitted if there are important descriptive differences between devices of the same type or the descriptive characteristics for the new device are not precise enough to ensure comparability.

In these instances, the most appropriate bench or animal testing—or in the case of IVDs, analytical testing—to address the performance issue should be provided. Summary information about the testing should generally suffice, but the manufacturer should also provide the test protocol and a description of test methods, and cite any standards followed in conducting the testing. The guidance also says that in the rare cases where clinical data may be necessary for 510(k) clearance, the agency should adhere to the least burdensome standards as outlined for PMAs.

The new guidance may be accessed at http://www.fda.gov/cdrh/ode/guidance/1332.pdf.

Nine New PMA Consensus Standards

FDA has added nine new standards to its list of recognized consensus standards. Device manufacturers can now declare conformity to these standards in lieu of including data to satisfy portions of product premarket review submissions or other requirements. The newly recognized standards are as follows:

The standards may be accessed at http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfStandards/search.cfm.

Abbott on FDA Comeback Trail

In what may be a signal that Abbott Laboratories (Abbott Park, IL) is making significant progress in correcting GMP deficiencies that resulted in a $100 million consent decree in 1999, FDA says it has agreed to allow the company to again market its HIV test—the Abbott Murex single-use diagnostic system (SUDS) HIV-1 test, the only rapid HIV test currently licensed for the U.S. market.

The test was pulled off the market in October of 2000 because of manufacturing problems related to the company's failure to meet certain specifications. After reviewing lot-release data submitted by Abbott for five consecutive lots of the test kit, FDA has determined that the test may now be released for clinical use.

Two New FDA Guidances Coming

Two industry guidance documents, one on premarket submissions and the other on GMP/quality system regulation inspections, are now under development, according to CDRH Office of Surveillance and Biometrics director Larry Kessler. The transcript of an April 23 presentation by Kessler says that FDA is reevaluating its list of high-risk premarket-exempt products and whether the products should remain exempt.

In addition, Kessler says, the agency is still developing policies on "open but unused devices," on the labeling of reprocessed and OEM products, and requirements for healthcare facilities other than hospitals.

Kessler's presentation may be reviewed on-line, at http://www.fda.gov/cdrh/reuse/L_Kessler_NJ_042301.pdf.

FDA Cites GE France Facility

In an April 10 warning letter, FDA declared some of GE Medical Systems' responses to agency findings inadequate, and it is requesting more information. The findings were the result of an inspection last December at the company's Senographe 2000D full-field digital mammography system manufacturing facility in Buc Cedex, France. The inspection found the mammography system was adulterated because of the methods used, or because the facilities or controls used for manufacturing, packaging, storage, or installation didn't conform with the quality system regulation.

CDRH rejected GE's responses to four specific failures cited by the center: One, a failure to establish and maintain a procedure to ensure that design requirements are appropriate and address the device's intended use, including the needs of the user and the patient; two, a failure to establish and maintain adequate procedures for identification, documentation, validation or verification, review, and approval of design changes before implementation; three, a failure to establish and maintain adequate procedures for finished device acceptance to ensure that each production run, lot, or batch of finished devices meets acceptance criteria; and four, a failure to establish and maintain adequate acceptance procedures to ensure that specified requirements for in-process products are met. In each instance, FDA asked for specific information or documentation so the staff can assess the adequacy of the response.

The company's responses were deemed to be adequate in three other categories: a failure to follow established procedures for implementing corrective and preventive action and failure to document all activities and their results; a failure to assess and determine whether service reports may represent an event that must be reported to FDA; and a failure to ensure that device packaging and shipping containers are adequately designed and constructed to protect the device from adulteration or damage during processing, handling, storage, and distribution.

Device Firms Impressed with QSIT

The majority of medical device firms are pleased with FDA's new quality system inspection technique, according to a just-released survey of 559 firms. Fifty-two percent of the firms said the inspection process was better than that used previously, 40% said it was the same, and 8% said it was worse.

For firms that received an FDA-483 after the inspection, 95% said that all of the inspection observations were "understandable." Eighty-eight percent of firms inspected also said that FDA investigators reviewed observations noted on a daily basis during the inspection.

The surveys showed that domestic medical device firms received 5 days advance notice of an upcoming inspection, compared with 55 days for foreign manufacturers. Eighty-three percent of FDA's in-plant inspection time lasted 5 days or less and 14% lasted 6 days or longer.

James G. Dickinson is a veteran reporter on regulatory affairs in the medical device industry.

Copyright ©2001 Medical Device & Diagnostic Industry