Journals Insist Drug Manufacturers Register All Trials
Editors Say That, Otherwise, Studies Will Not Be Published; Goal Is to Ferret Out Suppressed Data

By Shankar Vedantam
Washington Post Staff Writer
Thursday, September 9, 2004; Page A02

A dozen editors of prestigious medical journals jointly announced yesterday they will refuse to publish drug research sponsored by pharmaceutical companies unless the studies are registered in a public database from the outset -- a step designed to ferret out unpublished studies that find medications to be ineffective or dangerous.

The initiative creates a potent incentive for companies to register their drug trials and is expected to give physicians and the public a window on unfavorable studies that companies routinely suppress.

"When a pharmaceutical company sponsors a clinical trial and the results turn out not to be in the best financial interests of the company, it has been our experience these results are never made public," said Gregory D. Curfman, executive editor of the New England Journal of Medicine. "They are buried away."

More than two-thirds of studies of antidepressants given to depressed children, for instance, found the medications were no better than sugar pills, but companies published only the positive trials. If all the studies had been registered from the start, physicians would have learned that the positive data were only a fraction of the total.

The new requirement calls on companies to register their trials well before anyone knows whether a study will turn out positive.

The Journal of the American Medical Association, the Annals of Internal Medicine, the Lancet, the New England Journal of Medicine and several other international publications have signed on to the initiative, and their editors hope that more will join in.

Jeff Trewhitt of the Pharmaceutical Research and Manufacturers Association of America said several companies are already registering clinical trials on a voluntary basis and that a lot of information is already being made public. Individual companies, he said in a statement, "may have reservations about divulging proprietary information in clinical tests that are in very early phases. . . . [I]n the end it is their decision."

Responding to the growing outcry over suppressed medical studies, the House Energy and Commerce subcommittee for oversight and investigations plans today to ask executives of seven leading drug manufacturers whether they withheld information about the potential risks of suicidal behavior in children taking their popular antidepressant medications.

While companies could forgo publication in the journals and thereby avoid registration, the editors hope that preserving the opportunity to publish a positive study in a prominent journal, which can greatly boost the visibility of a new drug, will be a juicy enough carrot.

The medical editors' initiative also asks companies to report the studies' results in the database, but the editors acknowledged that if a trial turns out to be unfavorable, a company could decide not to. Still, they said, just knowing of the existence of the trial through the registration requirement would be telling.

"Registration in and of itself is a very big step," Curfman said. "Even if there aren't results out there, the very fact it is known this trial was conducted is a very big step forward -- it becomes much harder to bury a trial away."

Christine Laine, senior deputy editor of the Annals of Internal Medicine, said the requirement would apply to any trial begun after July 1, 2005. For trials already in progress, she said, companies would have to register them before seeking publication.

A number of initiatives are also underway in Congress to enforce registration of trials. The federal Food and Drug Administration Modernization Act of 1997 established a database called ClinicalTrials.gov and required companies to register trials, but many companies have evaded the requirement or filed incomplete information. FDA officials said there has been no enforcement, in part because Congress was vague about who would wield the stick.

Democrats in the House and Senate are expected to introduce legislation soon that would impose penalties on companies that do not register their trials with the government-run database before recruiting patients. Unlike the 1997 law, proposals would require registrants to list not only when a trial began, how many patients were recruited and what yardsticks of performance were used, but also what the study found.

"I understand the concern of some companies that if they report an adverse result their stock might suffer," said Rep. Edward J. Markey (D-Mass.), who said he and Rep. Henry A. Waxman (D-Calif.) would introduce a bill in the next two weeks. "But consider the alternative -- patients suffer as doctors prescribe in the dark."

The ClinicalTrials.gov database, which was a party to the journal editors' new initiative, recently expanded its format to include the primary and secondary outcome measures of a trial and details of when trials end. While journal editors will encourage the use of ClinicalTrials.gov, Laine said other databases could also be used, so long as they are maintained by nonprofit entities and abide by the rules laid down by the editors.

In today's congressional hearing, executives from GlaxoSmithKline, Wyeth Pharmaceuticals, Forest Laboratories Inc., Eli Lilly & Co., Bristol-Myers Squibb Co., Organon USA and Pfizer Inc. are expected to be asked about the controversy over suppressed studies information and the use of antidepressants among children.

Last year, British authorities warned doctors not to prescribe a number of antidepressants to children, citing an increased risk of suicidal behavior. The FDA has declined to take a similar step, citing the need for more careful evaluations. Today's hearing will also question FDA officials. In February, an internal agency report found that the medications were associated with an increased risk of suicidal behavior. Top FDA officials played down the report at the time and refused to make it public until recently.

A second internal agency analysis, undertaken after researchers from Columbia University reclassified suicidal cases reported in company trials, recently produced statistical findings that appeared to confirm the February assessment.

Last week, the author of the new report, FDA scientist Tarek Hammad, told investigators working for Senate Finance Committee Chairman Charles E. Grassley (R-Iowa) that his analysis agreed with the earlier finding -- and that he favored stronger warning labels on the drugs. The first internal report had recommended that the agency discourage the use of antidepressants other than Prozac for depressed children.

Grassley's investigators said Hammad went further this week and told them, "I can no longer say that Prozac is okay for children."

© 2004 The Washington Post Company