O blog do GAT

Medical Journal:
Drug Studies Hide Key Data
By SUZANNE SATALINE
Staff Reporter of THE WALL STREET JOURNAL
December 29, 2005; Page B1

Several major pharmaceutical companies are withholding important details about clinical drug trials, despite urging from federal regulators and medical-journal editors to be more forthcoming, according to a study published in this week's New England Journal of Medicine.

The study says that companies including Merck & Co., Pfizer Inc., and GlaxoSmithKline PLC are obscuring basic information -- including the names of some drugs under study -- in reporting on trials of drugs to treat serious or life-threatening diseases. Some of the drugs involved are already on the market, and the companies are seeking approval for new uses of them. In an editorial, the medical journal calls for investigators and patients to avoid participating in drug trials where companies take a secretive approach.

The push for more information comes after safety concerns over big-selling drugs -- including GlaxoSmithKline's antidepressant, Paxil, and Merck's painkiller, Vioxx -- appeared well after the drugs were on the market. For years, some drug companies kept secret the results of studies that reflected badly on their products and published only positive results. Critics of this practice said it helped to mask safety problems.

Last year, in an effort to encourage more disclosure, the editors of several of the world's top medical journals said they would refuse to publish any study that wasn't first registered at the federal clinical trial Web site maintained by the National Institutes of Health or in a similar public database.

The federal Web site for reporting on clinical trials --http://www.clinicaltrials.gov/ -- was created in 2002 after Congress, in 1997, mandated a publicly accessible registry of clinical trials of drugs and devices designed to treat serious or life-threatening diseases and conditions. By law, companies are allowed to register trials only if they provide the required information. The reports on the site concern drugs not yet approved, as well as some that are being investigated for secondary uses.

Investigators had until Sept. 13 to register their trials if they wanted to publish the results in major journals. The editors' policy appears to have worked well in some respects. From May through October of this year, the number of clinical trials registered at the government site increased 73% to 22,714, and the journal study found that the vast majority of entries complied with the editors' new standards.

But several large companies are providing information that is incomplete or vague, according to the study. Most notably, some companies don't disclose the names of the drugs they are studying. In a data field that asks for the name of the drug, the companies instead use terms such as "investigational drug."

The study said that four companies -- including Merck, Pfizer and Glaxo SmithKline -- were responsible for all of the non-specific entries. Some drug companies also provided scant information about what results their clinical trials are measuring. Novartis AG, for example, reported details about what outcomes it was measuring in reports on only eight of 239 trials, according to the study. Merck reported outcome details in nine of 46 trials. The study found that the fourth company, Eli Lilly &Co., failed to list the name of the drug in three out of 96 trials.

In general, the study, written by staffers of the federal Web site, found that smaller companies and non commercial researchers were better at disclosure than large pharmaceutical concerns.

Several of the large drug companies said they were meeting all federal requirements for disclosing trial data, but that they were leery of giving out too much information because of competitive concerns.

"We're not ready to give to all our rivals what we're doing," said Stephen Lederer, a Pfizer spokesman. Pfizer, like other companies, says it reveals all the clinical trial details on its corporate Web sites when trials are completed. Trials can take years to complete. Detlef Niese, vice president of clinical development at Novartis, said that patients in trials are given all the details about the trials, but that the company is concerned about providing public information that competitors "may want to copy."

A GlaxoSmithKline spokesman said that 95% of the company's listed trials met the tougher new standard of the medical-journal editors. As for the other entries, he said, "we are looking into any omissions.'' A spokesman for Lilly said, "We've met all the reporting requirements for the registry."

Jeffrey M. Drazen, editor in chief of the New England Journal, called the companies' concerns about disclosing competitive information "a smoke screen." He noted that other drug makers provided full and detailed information on their clinical trials. He also said that by not requiring companies to name the drugs being tested, the federal requirements had a "loophole.'' As a result, he said, companies "are meeting the letter but not the spirit of the law.'' The medical journals' requirements for disclosure are more stringent, and do require the name of the drug or procedure being studied.

In an editorial accompanying the study, Dr. Drazen wrote that it "makes moral sense" to provide complete trial information. "When patients put themselves at risk to participate in clinical trials, they do so with tacit understanding that their risk is part of the public record, not merely the secret record of the sponsor," he wrote.

The study credits Merck with amending many of its filings over the summer, after questions were raised about its disclosures. The company provided more details on the names of drugs being studied, boosting its compliance to more than 99%. However, Merck was less forthcoming with details about the outcomes of studies. It detailed information on what outcomes were being measured in only 20% of the trials. In a statement, Merck said it was reluctant to provide details on outcome data in early stage research trials because that information is"highly proprietary."

Merck's painkiller Vioxx was pulled from the marketplace last year after safety concerns were raised by a study showing an increase in heart-attack risk after 18 months of use. Earlier this month, the New England Journal published an unusual expression of concern about a 2000 Vioxx study it had published; the journal said it had discovered that data about three Vioxx patients who suffered heart attacks had been excised from a crucial study sponsored by Merck. Merck has said that the heart attacks occurred "after a pre-specified cutoff date."

In 2004, Glaxo SmithKline agreed to settle a lawsuit over unpublished clinical trials for $2.5 million and to publish summaries of all of its drug trials in a registry. In the suit, New York State Attorney General Eliot Spitzer accused the company of "repeated and persistent fraud" for concealing problematic issues of efficacy and safety when children took Paxil for depression. On settling, Glaxo said that it believed the state's charges were "unfounded.''