Global Harmonization Seen at Risk

Originally Published MDDI January 2004

WASHINGTON WRAP-UP

More standards and requirements worldwide may impede the ultimate goal of harmonization and seamless borders.

James G. Dickinson

AdvaMed Opposes Bundling by Reprocessors [1] | FDA Tells the Top Device Offenses [2] | Counterfeit Ethicon Mesh Reported [3] | FDA Asks One in Three Staff to Retire [4] | AbTox Faces Jury in February [5] | LASIK Doctor to Pay $1.1 Million for Study Violations [6] | Roche's AmpliChip Isn't Exempt, FDA Says [7] | FDA Guidance on Modular PMAs [8] | Deaths Linked to Cordis Cypher Stent [9]

FDA's participation in the Global Harmonization Task Force (GHTF) is at a critical crossroads, CDRH director David Feigal told the Regulatory Affairs Professionals Society (RAPS) annual meeting in October. “I think the real challenge for global harmonization is to decide how far we want to take this,” he said. “At one level, we can certainly harmonize the application process even if we have different standards.

“For example, a pilot program [called STED, for Summary Technical Documentation for Demonstrating Conformity to the Essential Features of Safety and Performance of Medical Devices] is under way. The program allows a manufacturer to submit the same [device approved] application to multiple countries.”

He said that this, however, does not mean that the countries are “harmonized with how we look at things or what we require.

“What is really intriguing, and which caught my eye last year, was that Hong Kong said the way it was going to approve medical devices was that if the product was approved in any of the five original founding GHTF member countries, then it was automatically approved in Hong Kong,” Feigal said.

“I think we have to foster more of that because one of the things that has happened with global harmonization is that it [the situation] is getting worse before it is getting better. There are more countries coming on board that want to inspect, and they are requiring applications when they hadn't required applications before. And in the process of getting an international dialogue going about what the standards should be for devices, [we have] increased the requirements, when our intention was to move in the other direction.

“One of the more fundamental things we need to decide is whether we are going to continue with two basic systems,” Feigal continued. “The difference is that the U.S. system for Class II products does not require proof of conformance to quality systems at the time of submission, whereas [proof] is required in Canada, Europe, and Australia,” he said. 

The differing approaches have little to do with who is reviewing the applications (i.e., third parties, government regulatory agencies), but more with “the human philosophy of what should be in an application, which is very different. And we are seeing products being up-classified all the way to the highest classification in different parts of the world. In the United States, we only do that if we need to see clinical data. Other countries are doing it to get better manufacturing information.”

Feigal said European requirements for implantables are not about wanting long-term clinical trials. Rather, these requirements focus on a manufacturer's design controls. The standards also include requirements designed to ensure that these design controls are being inspected and reviewed by third parties.

“We are getting to the point where the details are starting to matter, and we need to pay attention to that,” Feigal said.

At FDA, recognized standards that can be cited in approval applications have just about “peaked out” at around 600, Feigal told RAPS. He said CDRH is interested in hearing from industry about which standards or types of standards are most important. He said work is mounting to keep the standards current by “reexamining the updates to standards and participating in that process.”

AdvaMed Opposes Bundling by Reprocessors

The Advanced Medical Technology Association (AdvaMed) reiterated in October its opposition to certain types of bundling. Specifically, the group opposed the suggestion from another trade organization that FDA permit reprocessors of single-use devices to bundle products made from the underlying devices of multiple original equipment manufacturers (OEMs) into a single 510(k) submission.

AdvaMed submitted its comments in response to FDA's bundling policy under the Medical Device User Fee and Modernization Act (MDUFMA), and challenged an August 8 request from the Association of Medical Device Reprocessors (AMDR). AdvaMed noted that it first responded in January 2003 to FDA's request for comments on the issue, and at that time, it recommended that reprocessed devices be handled in the same manner as original devices, with a caveat. It recommended that a single submission should not seek clearance for reprocessed devices manufactured from more than one OEM's devices.

“Prohibiting the bundling of devices made by different OEMs is necessary to ensure an accurate assessment of medical devices by FDA,” AdvaMed's latest letter said. “Even for devices within the same general type of device, each OEM may make its device using a unique design, materials, manufacturing processes, technology, and engineering processes. Each OEM is likely to use different patentable technologies to manufacture its devices, and, unless shown otherwise by the reprocessor, FDA should assume that these technological differences do not support bundling.” AdvaMed said it is significant that FDA states that bundling is not appropriate for OEMs when devices involve different engineering processes or technologies.

AdvaMed criticized an AMDR assertion to FDA that “since all devices are reprocessed in the same manner, they present the same regulatory and scientific issues in a submission.” AdvaMed said that different OEMs' device models may react differently to the agents used to reprocess them, thereby affecting the devices' performance. It also said a reprocessor's 510(k) is not merely an application to approve the cleaning of an OEM device, but rather is considered by FDA to be an application for a new device manufactured by the reprocessor.

Meanwhile, the reprocessing issue may be losing some of its heat. Even as AdvaMed was filing its comments, CDRH director David Feigal was telling the RAPS annual meeting in Baltimore that most hospitals are no longer reprocessing single-use devices because of FDA's efforts under MDUFMA. He said that the hospitals have either decided to use commercial reprocessors or have decided not to use reprocessed devices. FDA in April 2003 published a list of single-use devices requiring a 510(k) from entities reprocessing them.

FDA Tells the Top Device Offenses

Manufacturers' procedures for implementing a corrective and preventive action (CAPA) system were cited by FDA device investigators in FY 2003 more frequently than any other system, Baltimore District investigator Lori S. Lawless told the RAPS annual conference held in October. 

“Often, procedures are not established, and this means they are neither defined nor implemented,” she told the audience. One of the biggest problems with CAPA procedures is that they “lack true root-cause analysis . . . basically, the failure investigations are inadequate.”

Lawless said that often a device company attributes the root cause of a failure to an employee error. “I'll tell you, if I start seeing ‘employee error,' forget it, because that's an issue I look for. Everything is not [the fault of] the employees. I blame management. If employees were not doing what they were told, then it is up to [quality assurance] management to catch that.”

Failure investigations conducted under complaint handling is another top citation in device inspections, Lawless told RAPS. Frequently, a company will argue that a complaint is just one instance from one lot, she said. “When I look at the data I say, ‘Well, yes, it's with that lot, but it is also in lot B, C, D, and E over the past 12 months.' This shows me that the failure investigation is too specific and does not look at the overall system.”
Lawless also identified several other areas where inspectors found manufacturer offenses. They are:

• Medical device reporting.
• CAPA activities not documented.
• Validation.
• Frequency of quality audits.
• Management with executive responsibility.
• Procedures for quality audits.
• Procedures for design changes.
• Procedures for documenting CAPA. 

Counterfeit Ethicon Mesh Reported

CDRH says Ethicon Inc. has reported the existence in the marketplace of counterfeit Prolene polypropylene mesh products. According to the center, Prolene is a nonabsorbable mesh used to repair hernias and other fascial deficiencies.

Healthcare professionals are being advised to carefully examine all Prolene flat-mesh products and avoid use of any material suspected to be counterfeit. The suspect products have been identified as bearing lot numbers RBE609 (expiration date 1/07) and RJJ130 (expiration date 7/07).

FDA Asks One in Three Staff to Retire

One-third of FDA's 9000-plus workforce is being offered $25,000 payouts to retire early. Homeland security legislation passed after 9/11 provided authority to all federal agencies to offer early-out and incentive retirement packages to their employees. The purpose is to save money and reduce the size of government.

According to Kimberly Holden, interim director of FDA's Office of Human Resources and Management Services, three different packages are being offered to 3200 FDA support services employees.

Holden said in November that the main purpose of these packages is to meet the program's goal of a 15% reduction in the numbers of these types of employees at FDA during fiscal years 2004 and 2005. This is part of the President's Management Agenda, Human Capital Management Initiative. The first separation notice offering the $25,000 buyout to eligible employees who resign or take optional retirement closed October 31 and drew about 120 applicants. Two subsequent notices closed on November 28.

FDA's offers do not include personnel in the 696 investigator series, the 1320 chemist series, or in the 600 series (medical reviewers).

The majority of those affected are in the support series, which includes program managers, secretaries, and administrative officers—positions that, according to Holden, do not require an advanced education.

AbTox Faces Jury in February

FDA's criminal mail and wire fraud case against four former officers of the now-defunct medical device company AbTox Inc. of Mundelein, IL, has been set for February 9 in Chicago federal court. Court records show that AbTox president and CEO Ross Caputo and three former colleagues in the company will have a jury trial.

With Caputo, individual defendants named in the case are former vice president for regulatory affairs Robert M. Riley, marketing director Mark E. Schmitt, and clinical services director Marilyn M. Lynch. FDA alleges the defendants participated in the sale to hospitals of unapproved sterilizers that deposited toxic copper and zinc salts on medical devices.

In October, Judge Ruben Castillo denied the defendants' motion to dismiss the indictment.

LASIK Doctor to Pay $1.1 Million for Study Violations

FDA announced in November that Louisiana ophthalmologist Leon C. LaHaye, MD, and his LaHaye Center for Advanced Eye Care in Lafayette had agreed to pay the federal government a total of $1.1 million in civil money penalties. The penalties are for violating federal laws related to the conduct of clinical studies. The violations involved studies of a laser system built by LaHaye for LASIK treatment of nearsightedness.

An FDA news release said LaHaye would pay $150,000 and the clinic $950,000 for study violations that occurred on at least 175 occasions. In June 2002, FDA disqualified LaHaye from conducting any further clinical studies.

The civil money penalties complaint filed by FDA alleged that LaHaye and his center:

• Used an unapproved laser on patients before the study began. 
• Treated more subjects than allowed under the study plan that was approved by FDA. 
• Ignored parameters of the study by treating nearsightedness beyond the permitted range and by treating astigmatism and both eyes of some patients. 
• Failed to submit complete, accurate, and timely reports to FDA about the ongoing study. 
• Misrepresented to FDA that LaHaye was using an FDA-approved laser to treat patients when, in fact, the procedures were performed with an unapproved, experimental laser. 

Roche's AmpliChip Isn't Exempt, FDA Says

CDRH says Roche Molecular Diagnostics' AmpliChip CYP450 ASR v1.0 microarray is not considered a Class I 510(k)-exempt analyte-specific reagent (ASR). The microarray cannot be commercially distributed without a premarket determination from the center. (See this column, October 2003.)

In a recent letter from the CDRH Office of In Vitro Diagnostic Device Evaluation and Safety, director Steven Gutman wrote that the office “believes that the intended use of the company's AmpliChip—to identify polymorphisms related to drug metabolism—is of substantial importance in preventing impairment of human health.” And, therefore, he recommended that Roche submit a 510(k) premarket notification.

Once a 510(k) is submitted, FDA will conduct a premarket review and, if the AmpliChip is found not substantially equivalent, Gutman said, Roche will need to “seek de novo classification under section 513(f) of the Federal Food, Drug, and Cosmetic Act (21 U.S.C. § 360c(f)).”

Gutman said his office based its decision on 510(l) of the act (21 U.S.C. § 360(l)), which says “a device that is within a type of device that has been classified into Class I is not exempt from the premarket notification requirement of Section 510(k) of the Act if it is ‘intended for a use which is of substantial importance in preventing impairment of human health' or if it ‘presents a potential unreasonable risk of illness or injury.'”

Even if the AmpliChip were to fall within the classification of an ASR, Gutman continues, the office “believes the technological characteristics of the AmpliChip would cause it to differ from existing or reasonably foreseeable ASRs such that the AmpliChip would not be exempt from premarket notification.”

In July, Gutman met with Roche officials over his objections to their marketing of AmpliChip CYP450 as “the world's first pharmacogenomic microarray for clinical applications” that include cardiovascular disease, high blood pressure, depression, and attention deficit hyperactivity disorder.

FDA Guidance on Modular PMAs

CDRH says its newly revised PMA modular review approach should allow applicants to resolve any FDA-identified deficiencies earlier in the review process than with the review of traditional PMA applications. According to a new guidance document, “Premarket Approval Application Modular Review,” qualified PMA applicants may submit “preclinical data and manufacturing information for review while still collecting, compiling, and analyzing the clinical data.”

The new guidance replaces a 1998 guidance and clarifies that PMA supplements are not eligible for modular review. The new guidance also modifies the procedures for preparing and filing a submission and explains the timing importance for submission of the modules. It revises the sample shell format and specifies the expected content of each module.

CDRH says in the guidance that it does not have the resources to extend the modular review process to supplemental PMAs. “In the agency's experience, reviews of applications that are supplements to an already approved device do not ordinarily lend themselves to a modular format because the changes do not usually involve multiple modules,” the guidance says.

CDRH recommends that applicants use the guidance's sample PMA shell “as a model when designing PMA shell proposals. In your PMA shell proposal, you should describe the contents of each module in sufficient detail to provide FDA with a complete understanding of the modules you plan to submit. The PMA shell proposal should address all elements required under 21 CFR 814.20, Application, and the timing of each modular submission.” 

If a PMA applicant's clinical trial period is expected to be lengthy, or if the product development timeline is long, the guidance suggests that applicants should carefully consider their schedule for submitting PMA modules when developing a PMA shell. “Premature submission of PMA modules (i.e., prior to finalizing your device design) could result in changes to the device that require you to submit additional data and FDA to reevaluate closed PMA modules,” CDRH says.

To view the guidance, visit www.fda.gov/ohrms/dockets/98fr/1998d-0896-gdl0002.doc [10].  

Deaths Linked to Cordis Cypher Stent

FDA says it has received more than 60 reports of subacute thrombosis (SAT) associated with patient deaths involving Cordis Cypher drug-coated coronary stents since that device's approval in April 2003. The device was MD&DI's cover feature in September. Overall, 290 reports of Cypher-related SAT have been submitted, with most events occurring between 24 hours and 30 days postprocedure, the agency says in a new public health notice.

“Also, we have received more than 50 reports, including some deaths, that Cordis considers possible hypersensitivity reactions,” FDA says. Symptoms reported include pain, rash, respiratory alterations, hives, itching, fever, and blood pressure changes. 

So far, FDA says there are insufficient data to establish rates for the reported adverse events, and it is still too early to tell whether adverse-event rates are different from those experienced with bare-metal stents. FDA notes that as of this month, more than 450,000 Cypher stents have been distributed worldwide. Additional information on adverse events is being requested from foreign regulatory bodies whose countries have also approved the device.

FDA says that while the etiology of the adverse events has not been identified, it is exploring a number of possibilities, including “product characteristics, patient characteristics (e.g., concomitant medications or illnesses), procedural factors, or a combination of these....”
The agency is encouraging health professionals to be vigilant for patient symptoms that may be related to hypersensitivity and to report these events as soon as possible. 

Copyright ©2004 Medical Device & Diagnostic Industry