Glaxo Settles Paxil Lawsuit

GlaxoSmithKline PLC agreed Thursday to a schedule for releasing negative data on the safety and effectiveness of its drugs to settle a lawsuit by New York's attorney general that accused the pharmaceutical maker of misrepresenting data on prescribing its antidepressant drug Paxil to children.

The company will also continue to update the information as new data becomes available, and will keep it available for at least 10 years after the initial publication.

GlaxoSmithKline will put summaries of all its studies since December 2000 in a clinical trial registry on its Web site by the end of 2005. The London-based company is the first major drug maker to agree to disclose all its studies. The company also will pay $2.5 million to the state as part of the settlement.

GlaxoSmithKline spokeswoman Nancy Pekarek told CBS Radio News released negative data on the effects of Paxil on children on its Web site in June. The company that month also announced the launch of a clinical trial registry on all their marketed products.

She said the data Spitzer accused the company of withholding had already been released through other means that included posters, medical letters to healthcare professionals and during meetings.

"We are choosing to settle this basically to avoid the high cost and time of protracted litigation," she said.

New York Attorney General Eliot Spitzer sued the company for fraud in June, alleging the company withheld the negative information. The suit struck at drug companies' responsibility to show whether antidepressants increase suicidal tendencies in children and whether the companies skew information on their products by not publicizing all the studies conducted on medicines or editing information on published trials.

Pekarek said that the company still feels Spitzer's civil charge of fraud is unfounded and that the settlement accepts disclosure policies the company had been working on for months, even before Spitzer's suit was filed.

Test results from this week forward will be posted within 10 months of a drug's approval, she said.

Spitzer said GlaxoSmithKline's new standard will also drive the industry, in part because investors and funders of research will be drawn to a firm that is a leader in disclosure of product data.

"The immediate impact is sending a signal to the other pharmaceutical manufacturers that this is the new standard with regard to disclosure of clinical studies," said Joe Baker, Spitzer's health care bureau chief. Additional investigations of other drug firms continues, he said.

"We will continue to do that until we feel this industry as a whole has stopped this practice," Baker said.

Paxil has not been approved for use with children, but the company conducted research to see whether it might prove useful, Pekarek said.

Spitzer said Glaxo conducted at least five studies on the use of Paxil in children and adolescents, but released one that showed mixed results on its effectiveness. The Democrat accused the firm of suppressing the negative results that failed to show Paxil was effective and might increase the risk of suicidal thoughts and acts in some youths. Spitzer also accused the firm of omitting the negative data from "Medical Information Letters" to physicians, according to the settlement filed in U.S. District Court in Manhattan.

A key to Spitzer's case was an internal 1999 Glaxo document showed that the company intended to "manage the dissemination of data in order to minimize any potential negative commercial impact."

Only Prozac, made by Eli Lilly & Co., has been approved to treat depression in children. But doctors can prescribe drugs as they see fit and routinely recommend such medicines for children suffering from depression and other psychological disorders.

In May, a Journal of the American Medical Association article that reviewed 102 clinical trials found that 50 percent of efficacy outcomes and 65 percent of harm outcomes were incompletely reported. The article concluded that trial outcomes are frequently incomplete, biased and inconsistent with protocols.