Companies still hiding drug trial data, medical journals say.

Three major pharmaceutical companies are not disclosing enough information about their clinical trials and risk having those trials barred from publication, editors of several prominent medical journals said in a joint editorial in June. (See Catherine D. DeAngelis et al., Is This Clinical Trial Fully Registered? A Statement From the International Committee of Medical Journal Editors, 352 New Eng. J. Med. 2436 (2005).)

In the wake of claims that the pharmaceutical industry suppressed negative research on drugs such as Vioxx and SSRI antidepressants, many consumer advocates and public health researchers have called for mandatory registrations of clinical trials. If all trials are registered at the start, they say, researchers can track them, making it harder for companies to halt trials or hide negative results.

"The industry has developed a talent for submerging or suppressing uncomfortable data, and only if we know what trials are being done and when we can expect to see the results, can that kind of suppression be prevented," said Peter Lurie, deputy director of the Public Citizen Health Research Group in Washington, D.C.

Leonard Glantz, associate dean and professor of health law at Boston University's School of Public Health, agreed, noting that when a trial is halted before completion, it does not show up in the medical literature, but researchers still need to know about it.

"If you see one successful trial and six that have been withdrawn before they were completed, it will raise your level of suspicion about what that one that was completed actually demonstrated," he said.

Drugmakers are required to report the results of clinical trials to the FDA, but that information is kept confidential. The National Institutes of Health (NIH) has established a database at www.clinicaltrials.gov, but reporting is not mandatory. A bill that would require researchers to report more data was introduced in Congress in February, but at press time no hearings on it had been scheduled.

Last September, the International Committee of Medical Journal Editors (ICMJE) took action on its own, announcing in a joint editorial that its member journals--including The New England Journal of Medicine, The Journal of the American Medical Association, and The Lancet--would no longer accept for publication any article on a clinical trial that was not registered before participants were enrolled. The trials must be registered in a database that is electronically searchable and accessible at no charge to the public, such as www.clinicaltrials.gov.

The new policy, meant to encourage drugmakers to register all clinical trials at their outset, applies to trials begun on or after July 1, 2005. All so-called Phase III trials (those that compare the effectiveness of a new treatment with standard therapies or no treatment at all) must be registered by September 13, 2005. Phase I trials, which are designed to assess toxicity or biological activities, are excluded from the requirement. The editors said articles on Phase II trials--those designed to test the effectiveness of a treatment and which may provide preliminary data leading to a larger Phase III trial--will be considered on a case-by-case basis but encouraged researchers to register them.

In June, the ICMJE said in a second joint editorial that so far some drugmakers are not providing adequate information when they register their trials. A May review of the NIH database by its director, Deborah Zarin, found that several companies use a "meaningless phrase" (such as "investigational drug") in place of the name of the drug being studied. Without the specific name, researchers cannot accurately track all the trials on a given drug.

Jeffrey Drazen, editor of The New England Journal of Medicine and a member of the ICMJE, told the media that three companies were "making a mockery" of the database and efforts to make clinical trials more transparent. He said Zarin found that Merck did not provide drug names for 90 percent of its 123 trials, GlaxoSmithKline did not provide them for 53 percent of its 55 trials, and Pfizer omitted the names for 36 percent of its 75 trials.

"What the journals are trying to do is pretty much a serf-regulation system, in which these private entities try to work with other private entities to try to make the system better. But it appears not to be working," said Glantz.

To clarify what information it expects drug companies to provide, the ICMJE said it has adopted the "minimal registration" standard developed by the World Health Organization. A table showing the standard's 20 data points and further explanation accompanied the editorial.

However, Glantz and Lurie both said they do not expect comprehensive registration to happen without government intervention. "The pharmaceutical industry has had its chance to do this voluntarily--there may be a question of whether we want to give it one more chance--but ultimately, my guess is, if the FDA doesn't require this, it won't happen," Glantz said.

San Francisco attorney Don Arbitblit, who represents plaintiffs in drug cases, said more is needed to protect consumers. "What is needed is an agency that truly represents the interests of the public," he said. "Regulations requiring transparency in clinical trials would help, but much more is necessary, including stringent review of drug safety before and after marketing by officials without financial ties to those they are supposed to regulate."