This article is reprinted from "The Gray Sheet" – January 7, 2008
The violation rate for clinical trial sponsors shot up in fiscal year 2007, hitting a 10-year high, with FDA finding many of the problems during inspections triggered by complaints, according to CDRH.
Of the 40 trial sponsors inspected by the device center's division of bioresearch monitoring (BIMO) last year, 33% were classified by FDA as "official action indicated" (OAI), denoting that significant regulatory violations were found. That compares with 11% of the 53 sponsors inspected in fiscal 2006 that were significantly out of compliance, the best industry performance in 10 years.
Device companies were the target of eight BIMO warning letters in 2007, five more than in 2006.
The BIMO division is responsible for checking on the quality and integrity of clinical trial data, and ensuring that trial subjects are protected from undue hazards and risks. The unit oversees inspections of sponsors, clinical investigators, institutional review boards and bench testing labs.
Of 323 total BIMO inspections in fiscal 2007, 14% were tagged OAI, up from 11% the year before. Twenty-nine resulted in warning letters, up from 24.
IRB noncompliance bumped up one percentage point to 9% in fiscal 2007. Clinical investigator noncompliance rose to 13% from 11% in 2006.
"For-Cause" Audits Spark Violation Increase
Insufficient monitoring of clinical sites was the most frequent problem for sponsors in 2007, according to Michael Marcarelli, director of CDRH's BIMO division. Other problems included failing to submit progress reports or ensure investigator compliance and, less often, hiring unqualified monitors or failing to receive FDA or IRB approval before starting a trial.
In a Dec. 18 interview with "The Gray Sheet," Marcarelli stressed that the 200% jump in sponsor noncompliance was largely due to "for-cause" inspections, which occur in response to complaints, not as part of routine inspections such as those performed during PMA reviews.
"As we look at our total program - our surveillance program during the early intervention phase, our PMA program - we find that, overall, clinical investigators and sponsors do a good job," Marcarelli said. "But some individuals and some companies, whatever the motivation is - money, lack of time, sloppiness, lack of supervision - are going to skew the data a little bit."
For-cause inspections are conducted by BIMO's special investigations branch, formed in 2004 to respond to the growing number of complaints received by CDRH (1"The Gray Sheet" Dec. 6, 2004, p. 6).
Complaints of deviations from device research regulations come from industry competitors, clinical trial subjects, clinical investigators, company employees and CDRH reviewers, among others.
In 2007, 19 of the 40 sponsor inspections were for-cause. Marcarelli noted that while the center is conducting roughly the same number of for-cause audits from year to year, these inspections are resulting in violations more often than before.
Removing for-cause inspections from the picture, 12% of sponsors were found to have significant violations in 2007, compared to 10% in 2006, Marcarelli said.
Supporting his point that overall industry compliance is not as bad as the top line numbers suggest, Marcarelli pointed out that BIMO placed only one "integrity hold" on a premarket submission last year, and that there were no new application integrity policies, which block a company from making any data submissions.
"We had very few integrity matters," he said. "So, although they were violative inspections, they were ones that we thought were remedial in nature and could be corrected once presented to the sites."
FDA Expects Tighter Quality Control - Attorney
Janice Hogan, a partner at the Washington, D.C., law firm Hogan & Hartson who represents device companies on clinical trial issues, agrees that it is important to note who the BIMO audits targeted before drawing sweeping conclusions from the high noncompliance rates. Nonetheless, she believes the 2007 numbers reflect higher FDA expectations.
"They expect sponsors to increase the level of process control that they use in clinical studies," Hogan said in an interview. "Now they're taking much more of a quality system approach to the conduct of clinical studies, so they expect sponsors to have very comprehensive procedures in place."
Marcarelli introduced the concept of a quality system for clinical trial oversight in 2005 after observing record noncompliance in 2004 (2"The Gray Sheet" Dec. 5, 2005, p. 3). The idea is for device firms to incorporate a corrective and preventive action program and management controls into their oversight of clinical trials, in much the same way they structure their manufacturing operations under FDA's Quality System Regulation.
Marcarelli reports positive responses from companies that have adopted this new strategy (3"The Gray Sheet" Dec. 4, 2006, p. 3), but he acknowledges that the formalized practices and procedures that companies use to monitor manufacturing quality are not familiar processes for doctors and researchers.
"We don't oftentimes see these really well structured training programs or standard operating procedures at clinical sites," Marcarelli noted. "We may see a protocol, but I'm talking about quality systems standard operation procedure."
That's a key factor for firms to consider when they pick clinical sites to work with, he said.
Warning Letters Cite Trial Monitoring As An Issue
"While many companies certainly have a system of procedures, like anything else, when this approach is new it takes some time to make sure that you're doing everything that the FDA expects you to do," attorney Hogan said.
More often, FDA is letting companies and clinical investigators know through warning letters that it expects a higher level of compliance with good clinical practices (GCPs), she said. "But if that is the way we're going to go, then it would be helpful to have more interpretive guidance than just, 'You should follow the GCPs.'"
Of the 29 device BIMO warning letters in 2007, 16 targeted clinical investigators, five were sent to IRBs and eight were sent to sponsors. Investigators were the subject of most inspections by far - 183 compared to 92 IRB inspections and 40 sponsor inspections.
"The position is pretty harsh right now in terms of the level of perfection that the FDA expects," Hogan said. Well-done studies and subject protection are obviously important, she noted, but "to ask a sponsor to have 100 percent follow-up and 100 percent compliance? We all know how many of us have been unable to stay for a doctor's appointment. We can't remove ourselves from reality."
Several warning letters issued in 2007 criticized sponsors for failing to keep up with what is going on at sites that are performing a company-sponsored clinical trial.
In March, for instance, BSD Medical, which makes hyperthermia therapy products for cancer and other diseases, was cited for failing to secure the investigator's compliance after agency inspectors observed that a researcher was enrolling subjects who did not meet eligibility criteria and was not performing certain pre- and post-treatment evaluations.
In April, FDA reprimanded breast implant maker Silimed, now owned by Sientra, for submitting a detailed monitoring plan to FDA but not documenting whether it was carried out. Other letters censured firms for not defining their monitoring policies and procedures at all.
Hogan says FDA has recently made efforts to give industry more direction on monitoring, but not enough. Firms are looking to GCPs in addition to FDA guidance, but "it's not the same as FDA really giving people more details on what they expect," she said. "It's not desirable to have people learn by warning letter."
Marcarelli agrees that the rules on device monitoring are neither very specific nor "as prescriptive as our attorney would like. But that allows us some flexibility."
BIMO regulatory counsel Sonali P. Gunawardhana anticipates that more concrete regulations will be necessary in the future, as trials become more complex, but cautions: "That's a lengthy process."
Marcarelli says his office plans to release final guidance within the next few months on BIMO's role in premarket approval inspections, a draft of which came out in June 2006 (4"The Gray Sheet" June 26, 2006, p. 10).
But two planned 2007 BIMO documents - a long-awaited draft guidance defining sponsor responsibilities with regard to trial monitoring and another on Web-based training for clinical investigators - have been put on the back burner.
"The monitoring guidance document has been subsumed by the larger effort at the commissioner's level," Marcarelli said. "The agency is looking at developing an agency-wide guidance with respect to monitoring clinical trials."
At present, the online investigator training guidance is effectively dead, Marcarelli said, though clinical investigator training and certification and the idea of certifying trial sites will be explored at the agency-wide level.
In the absence of specific guidelines, Hogan advises companies to commit to at least quarterly monitoring, and suggests that sponsors provide basic training for clinical investigators on how to run an FDA study.
Charma Konnor, an executive director with consulting group Phoenix Regulatory Associates and former CDRH BIMO director, says companies need some "wiggle room" to determine on a case-by-case basis how often site monitoring is needed.
"If the sponsor is running their overall system of oversight and conduct of clinical trials ... correctly and compliantly, they can be the best judge of how frequently they need to monitor a site," she said.
Konnor does urge companies to self-audit their monitoring protocol or hire an outside auditor early in the study. "That's a good reality check and can be very helpful to getting records straightened and other things corrected and taken care of before BIMO comes in."
Research Oversight Priorities Go Agency-Wide
Marcarelli's group, along with BIMO divisions within the other FDA centers, is under increasing pressure to improve clinical trial oversight and strengthen compliance since the HHS Office of Inspector General criticized FDA's study oversight in a September report (5"The Gray Sheet" Oct. 8, 2007, p. 5).
Marcarelli says agency-wide committees are addressing some of the deficiencies outlined in the OIG report, and FDA's Good Clinical Practice Program, directed by former CDRH staffer Joanne Less, is spearheading many of the initiatives.
The shelved guidances on monitoring and Web-based training are among many BIMO activities being merged with broader FDA collaborations, or taking a backseat to higher priority Critical Path initiatives.
The latest collaboration to emerge from the agency's Critical Path program is the FDA/Duke public-private partnership to improve the efficiency of the clinical trial process (6"The Gray Sheet" Dec. 10, 2007, p. 7).
"We sit right at the table," Marcarelli said of the partnership. "It's going to focus primarily on the premarket aspect. I think they're going to take a look at things like monitoring of clinical trials, because that's a large expense with respect to costs associated with a clinical trial."
- Jessica Bylander
“The Gray Sheet” – Comprehensive news and analysis for medical device professionals. Click here for a free, 30-day trial.
