< Back to Front Page Text size +

Drugs approved near FDA deadlines more likely to have safety problems later, study finds

Posted by Elizabeth Cooney March 26, 2008 05:58 PM

Drugs that were approved by government regulators just before their deadlines were much more likely to have later safety problems than medications approved at other times, Harvard researchers report.

Writing in tomorrow's New England Journal of Medicine, Daniel Carpenter and colleagues say congressional mandates to speed up the US Food and Drug Administration review process for new drugs have been followed by spikes in the number of new medications approved in the weeks before deadlines.

"The ones that are piled up before deadlines are significantly more likely to experience safety problems," Carpenter said in an interview.

The FDA disagrees with Carpenter's analysis of approvals after the Prescription Drug User Fee Act was enacted.

"We have found at least two fundamental problems with the data," Clark Nardinelli, director of the economics staff at the FDA, said in an interview. "We don't think their conclusions hold up."

According to the journal article, drugs approved within two months of their deadlines are two to three times more likely to be pulled off the market, two to seven times more likely to add "black box" warnings, twice as likely to have changes in how they are manufactured, and two to seven times more likely to be withdrawn by their makers because of low demand, the authors found.

Nardinelli of the FDA said when staffers tried to replicate what Carpenter and his co-authors had done, they found different numbers of drugs approved early in the review process and that had "black box" warnings added after approval. Carpenter said he had used data from published studies for his journal article, but in response to an FDA inquiry about his study, he had come to the similar conclusions using data from a former FDA official.

"FDA won't approve a drug if we are not ready," Dr. Janet Woodcock, director of the FDA Center for Drug Evaluation and Research, said in an e-mailed statement . "The timeframes established by PDUFA are guidelines, and we have the flexibility to miss a PDUFA date if we're not certain about the safety of a drug and want more time to review it. And we have the option of denying approval altogether if there is any question about safety."

Carpenter said the study is the first to examine deadlines imposed as part of a 1992 law intended to get medications to patients more quickly. The law, which introduced user fees from drug manufacturers applying for approval, said the FDA must make a decision on 90 percent of proposed drugs within 12 months or lose funding. That window was narrowed to 10 months in 1997 but extended again in 2002. The authors looked at approval data from 1950 through 2004.

Carpenter cautioned that his work can only point to patterns in drug approvals and not single out the cause for later harm done by individual drugs.

"I can't say whether deadlines caused problems for a specific drug," he said. "Deadlines are a necessary, useful part of life. The question is, should we rely on them so heavily to accelerate this process?"

Carpenter said examples of drugs approved close to deadline whose problems drew regulators' attention after they were on the market include:

-- Baycol, a cholesterol drug that Bayer voluntarily withdrew after 31 deaths;

-- Vioxx, a pain reliever ordered off the shelves after a clinical trial showed triple the rate of heart deaths;

-- Rezulin, a diabetes drug that was withdrawn from the UK market for toxic effects on the liver.

  • CommentComment
  • EmailEmail
1 comments so far...
  1. Another flawed Harvard study. One of the primary efforts that academic health care leaders have undertaken is to seek more rapid approval of life saving drugs. A quick review of the "black box" drugs indicates that most of them are cancer drugs. These have side effects, often discovered with increased use, but the early approval based upon efficacy allows for significant positive effects on disease free and overall survival. If one looks at the data in the paper, delays of 12 -24 months in the pre-PDUFA days will translate into hundreds if not thousands of lives shortened or ended prematurely. The proper analysis would have been to compare the effect on event free survival (including both toxicity and anti-disease effect) and look at the difference in drugs approved timely and not.

    Posted by WPP March 27, 08 04:36 PM
add your comment
Required
Required (will not be published)

This blogger might want to review your comment before posting it.

about white coat notes We post updates every weekday about the region's hospitals, labs and medical schools – covering everything from the latest research findings to what's on the minds of the innovative doctors, nurses and scientists who work here. Send news items and tips to whitecoat@globe.com

Contributors

blogger

Elizabeth Cooney is a former health reporter for the Worcester Telegram & Gazette, where she also was a business reporter and an editor. Earlier in her career, she edited medical books and journals at Little, Brown, and worked for Boston magazine.

Boston Globe Health and Science staff:

archives