PhRMA urges more disclosure

Experts applaud new guidelines, but wonder about their effectiveness | By Paula Park

A new set of guidelines issued by the Pharmaceutical Research and Manufacturers of America (PhRMA) calling for the release of more data from clinical trials is being met with guarded skepticism by advocates who have been clamoring for increased transparency.

The guidelines, approved by PhRMA's executive committee on June 30, say the results of clinical trials for marketed drugs should be made public "whether they are favorable to the company sponsoring the trial or not." They also suggest that the clinician in charge of the trial be paid in currency, and not with stock options from the company commissioning the trial.

The voluntary guidelines have been in development since 2002 and mostly reflect protocols already in place, according to Alan Goldhammer, PhRMA's associate vice president for regulatory affairs. They have been updated in recent weeks to clarify the association's views on conflicts of interest and the publication of negative trial results, he added. The guidelines do not specify which studies will be made public.

"We've made it a little more clear that a clinical trial registry is compatible with PhRMA," Goldhammer told The Scientist. "What's most important is to focus on what prescribing physicians need to know."

PhRMA's release of the clarified guidelines follows a June 2 lawsuit by New York State Attorney General Eliot Spitzer, charging that GlaxoSmithKline engaged in "repeated and persistent fraud" by withholding negative information about Paxil and misrepresenting data concerning Paxil's safety and efficacy when prescribed for depression in children and adolescents. Spitzer's office would not comment on the new guidelines.

Two weeks later, the American Medical Association House of Delegates called for the Department of Health and Human Services to establish a comprehensive registry for all US clinical trials. On June 18, GlaxoSmithKline announced the creation of a register that would provide summaries of clinical trial results on the Web. The World Health Organization would like to establish a worldwide registry of all trials.

An American Medical Association (AMA) trustee welcomed the PhRMA guidelines, but questioned their potential effectiveness and said they do not go far enough. All clinical trials—not just those for marketed drugs—should be added to a public registry, said trustee Joseph Heyman, and institutional review boards should require registration as a condition of allowing the study to continue. Even trials on therapies that are later discarded provide significant information to scientists, Heyman added. For marketed drugs, it's important to release information on all studies, even the most exploratory ones.

"We would agree with measures to make studies public," Heyman told The Scientist. "But [PhRMA] does not commit to make them available at inception. From what I'm reading, if the drug never goes to market, nobody knows about it."

Jonathan Moreno, director of the Center for Biomedical Ethics at the University of Virginia, said that PhRMA's guidelines are "more interesting for what they don't say than what they say," and labeled the effort "damage control."

The guidelines do not provide enough information to ensure accountability, Moreno said. "This is for general consumption," he added. "How much they are doing with their members is hard to tell. But you don't get the sense of trying to push back the boundary of business ethics."

Sheldon Krimsky, Tufts University professor and author of the forthcoming Science and the Private Interest, also questioned the lack of detail. "These are all positive things, but one of the problems is when you get down to the details, are they talking about phase 3, phase 4, or are they talking about all trials, 1 to 3?" Krimsky asked. "How are things going to be reported?"

Goldhammer said that such details needed to be worked out among PhRMA, the Food and Drug Administration (FDA), and the AMA. "We are all in favor of transparency," he said.

American Medical News, the AMA's weekly newspaper, reported July 5 that there are already 300 trial registries in use. The largest is the National Institute's of Health ClinicalTrials.gov. A November 1997 federal law requires that companies report the results of phase 2, 3, and 4 trials for therapies treating "serious or life-threatening diseases" on that Web site.

But Theresa Toigo, an associate commissioner of the FDA, said that only about 48% of all cancer trials done in 2002 were reported. Toigo's study of reporting on cancer therapies was released last fall, and the full study results are expected later in the year. "We think more trials should be registered than actually are registered," Toigo told The Scientist.