Veil of secrecy to lift on drug tests

Firms are under pressure to release results of all their clinical trials, including the negative ones. The result could be a new level of industry openness.

September 13, 2004

Growing concerns over the safety and efficacy of anti- depressant drugs prescribed to children have caught the eye of Congress and the New York state attorney general. Now they're becoming the catalyst for calls to reform the way clinical trials of all drugs are reported.

Pressure is already causing some changes within the pharmaceutical industry. And it has put the US Food and Drug Administration (FDA), which approves new drugs, in the hot seat. If reforms are carried out, they could bring an unprecedented level of transparency to drug research.

The solution now under consideration: a public database, or registry, of drug trials, where companies would post the results of those trials.

In congressional testimony Thursday, a spokesman for the American Medical Association endorsed the registry and said it should include information on each trial's purpose and objective, its design, and the dates it begins and ends. If the trial is not completed, the registry should include an explanation.

While drug companies have been eager to make public any positive results of their trials, recent revelations suggest they've balked at divulging tests when the results are not what they'd hoped to see. The furor has centered around the use of anti- depressants on children.

Last month, for example, drug giant GlaxoSmithKline agreed to pay $2.5 million and post online summaries of all of its clinical trials to settle a lawsuit brought by Eliot Spitzer, New York's attorney general. The suit charged that Glaxo suppressed evidence that its antidepressant Paxil was ineffective in children and may lead them to have suicidal thoughts.

Then on Aug. 20, the FDA released a review of previous clinical research that confirmed a link between suicidal thoughts and antidepressants. It showed that children taking the drugs were nearly twice as likely to have suicidal thoughts or behavior as those receiving sugar pills. An FDA advisory committee meeting Monday and Tuesday is reviewing that evidence and is expected to determine soon if it is strong enough to warrant FDA action.

Meanwhile, in court, lawyers for a young defendant accused of murder are trying to draw a connection between an antidepressant and violent action. Christopher Pittman, who was taking the antidepressant Zoloft, is charged with killing his grandparents in 2001, when he was 12. According to The New York Times, Pfizer, the drug's manufacturer, is helping prosecutors with their case.

As allegations have piled up against the drug companies, the FDA also has come under scrutiny. In August 2003 the drug firm Wyeth asked the FDA for permission to place a stronger warning label on its antidepressant Effexor, saying it could worsen depression or suicidal thoughts. The FDA denied the request.

At the congressional hearings on Thursday the committee chairman, Rep. Joe Barton (R) of Texas, charged the FDA with "stonewalling, slow rolling, and plain incompetency" in its relations with the committee, which has been investigating the risks that mood-altering medications pose for children and teens.

The industry has begun to make some moves to address the concerns about drug trials. Drug companies have agreed to set up a voluntary system of posting their drug trials on the Internet. But that seems unlikely to satisfy some members of Congress, who are expected to introduce legislation to establish a mandatory drug registry.

Last week, editors of a dozen influential medical journals announced that they would begin requiring drug companies to post a drug trial in a public database prior to accepting an article about it. Doctors rely on these articles to make treatment choices. The editors hope that the registry will force unfavorable drug studies, before kept secret, into the open.

Medical journals already had been tightening up on the authorship of their articles, insisting that authors declare if they had any conflicts of interest, such as any financial or other ties to the drug company, says Daniel Callahan, a director at the Hastings Center, a nonprofit bioethics research institute in Garrison, N.Y.

"The most fundamental complaint [has been that pharmaceutical companies] function as a kind of 'black box,' " he says, because the results that were published could not be put in a context of other results, which were hidden from doctors, researchers, and journal editors. One benefit of a registry will be to head off "dead end" research, as other researchers are able to see what approaches have not worked, he adds.

Drug companies have an ethical obligation to report all their findings, says Arthur Caplan, director of the Center for Bioethics at the University of Pennsylvania in Philadelphia. When a company signs up subjects to take experimental drugs, it often promises to use the tests to generate knowledge, he says. But "if you don't publish and make public what you find, it's not honoring your contract."

Information from previously undisclosed clinical trials could lower prices, reduce the number of badly designed trials, and help doctors considering the use of a drug for a nonapproved purpose to know why it hasn't been approved for that use. "We give a lot of freedom to doctors to 'muck around,' " - try various remedies - "but they'd do a lot less of that if they knew that the drug companies had already tried to do it," Mr. Caplan says.

Antidepressant drugs "have some serious side effects ... that seem to be much more common than people realize ... much more common than you might think from seeing drug ads and from reports on drug studies," says Joel Gurin, executive vice president of Consumer Reports. His magazine just finished a survey of readers showing a "dramatic shift from talk therapy to drug therapy for mental health problems" during the past decade. In 1995, less than half of people getting mental health treatment - 40 percent - got drug therapy. Today 68 percent receive drug treatment, Mr. Gurin says.

Some studies coming to light show that antidepressants work no better than placebos. Even better than merely registering drug trials, Caplan suggests, would be to require that a new drug not only be "safe and do what it's supposed to do," but that it do it as well or better than other drugs already on the market. That, he says, would help push research into new areas and save money.