Trial safety lacking, says GAO

The system for protecting the safety of people who participate in clinical trials is in shambles and needs a major overhaul, according to the conclusions of a two-year undercover US government investigation. The undercover investigation, which caught a commercial institutional review board approving a fake study made up by investigators at the Government Accountability Office (GAO), is "absolutely" a positive step, said Adil Shamoo, a professor of biochemistry and molecular biology at the University of Maryland School of Medicine. "The current system is broken and it will require reform," Shamoo, who has worked extensively on research subject protection issues, told The Scientist. He noted that there are about 10,000 IRBs operating in the US. "So if there is 10% bad IRBs, you're talking about a million human subjects put at risk." At a Congressional hearing today, legislators lambasted a commercial institutional review board, linkurl:Coast IRB,;http://www.coastirb.com/ for giving a green light to the fake study. As part of the investigation, which started in 2007 at the behest of the House Committee on Energy and Commerce's Subcommittee on Oversight and Investigations, the GAO created a false company called linkurl:Device Med-Systems,;http://devicemedsystems.com/ and submitted a made-up study protocol, signed by a made-up doctor, to Coast and two other IRBs. The protocol of the fake surgical product, Adhesiabloc, was based on a real product that the Food and Drug Administration (FDA) had withdrawn from the market because of adverse events. Coast approved the study unanimously. "Nobody at Coast ever reviewed the data in the document to support those claims [the product was effective]-- if they had they would have seen they didn't exist," Bart Stupak (D-MI), who heads the Subcommittee, said at the hearing. The other two IRBs (which the hearing did not name) dismissed the protocol immediately, with one calling it "a terrible risk for the patients," Stupak added. Earlier this month, when Coast IRB was asked to appear before today's Congressional hearing to discuss the Device Med-Systems trial, it began to go through the data and discovered that both the company and its product was bogus. In response, Coast issued a series of press releases saying it had been unfairly framed by the government, and noting that the IRB had no reason to be on the lookout for such blatant fakery, since it never happens. "I find it to be the most pathetic example of trying to spin your way out of taking responsibility... that I've ever seen," said Greg Walden (R-Ore) a member of the Subcommittee, at the hearing. Independent IRBs require no accreditation to function, though most are accredited through an independent body called the linkurl:Association for the Accreditation of Human Research Protection Programs.;http://www.aahrpp.org/www.aspx Both the FDA and the Office of Human Research Protection within the Department of Health and Human Services (HHS) have regulations about the kinds of documentation IRBs need to obtain from trial sponsors, and the oversight IRBs should provide, but little power or resources to enforce them. "I think this [investigation] is a wonderful thing," said Paul Goebel, who served as associate director of the FDA's human protection program in 1997-99, and later served as vice president of linkurl:Chesapeake Research Review,;http://www.chesapeakeirb.com/us/default.aspx one of the larger independent IRBs in the country. "What we've got in the IRB community, particularly the independent IRBs, are a small number of highly reputable IRBs that are very careful and do a good job of vetting the studies, and then some that do a quick turnaround," Goebel said. Those "just take at face value of what the applicant tells them." Some independent IRBs have been quick to distance themselves from Coast's practices. (Last March, Coast IRB received a linkurl:warning letter;http://www.fda.gov/cder/warn/2008/08-HFD-45-1101.pdf from the FDA for breaching the agency's rules on expedited review of trials.) "The concern I'd like to speak to is the potential view of Coast as being a typical independent review board -- which is not valid," Currien MacDonald, medical director of linkurl:Aspire IRB,;http://www.aspire-irb.com/ an independent IRB in San Diego, told The Scientist. "It's completely ridiculous that they didn't catch this early on." He dismissed Coast's explanation that there was no way for them to expect the submitted study was a fake. Indeed, said MacDonald, IRBs must always be on the lookout for "feeder" studies -- essentially marketing ploys by drug companies whose sole purpose is not to examine a scientifically relevant question, but to expose physicians to their product. "Part of your regulatory practices would catch it, regardless of whether you're looking for it or not," he said. Meanwhile, the linkurl:Consortium of Independent Review Boards,;http://www.consortiumofirb.org/ a professional society of IRBs not affiliated with any university, released a statement affirming the importance of independent IRBs in the clinical trial process, noting that 75% of all human studies are done outside of an academic setting. "We think it's a responsible community that has worked hard to protect the public health and forward research," Gary Yingling, spokesperson for the group, told The Scientist. IRBs are not the only problem, the Subcommittee stressed. Procedures at the FDA and the HHS are also in dire need of reform. As a second part of the investigation, GAO also concocted a fictitious IRB called linkurl:Maryland Hause,;http://marylandhauseirb.com/ which it registered through an online form with the HHS, as the agency requires. Although the entry contained information that might have been a tip-off that the IRB was a fake, "it gets entered [into the database of IRBs] even if the town it's registered in is Cheatsville, Arizona, for which I assume there is no zip code," said Walden at the hearing. The outcome will depend on what the government does with the information it gleans from the investigation, said Goebel. "I mean, can Congress screw it up? Yeah," he said. "We don't need another overlay of laws. There are plenty of laws on the books. If they were enforced, it would be good." Editor's note (March 26, 2009): This story has been updated from a previous version to correct the spelling of the IRB invented by the GAO, Maryland Hause. **__Related stories:__***linkurl:Subject death halts clinical trial;http://www.the-scientist.com/blog/display/54825/ [8th July 2008]*linkurl:Paying for patients;http://www.the-scientist.com/article/display/54196/ [February 2008]*linkurl:Gene therapy's trials and tribulations;http://www.the-scientist.com/article/display/11688/ [7th February 2000]

Trial safety lacking, says GAO

Interested in reading more?

Become a Member of