The New York Times
May 12, 2006
By BENEDICT CAREY and GARDINER HARRIS
After analyzing data from clinical trials, GlaxoSmithKline has sent letters to doctors warning that its antidepressant drug Paxil appears to increase the risk of suicide attempts in some young adults.
The company said it had changed the labeling on the drug to reflect the finding of the study, which analyzed clinical trial data involving some 15,000 people. The study found that reported suicide attempts were rare but significantly more common in adults who took the drug for depression than in those who received placebo pills.
The Glaxo researchers reported only one suicide in the trials, a number so small it says nothing about the drug's risk, experts said.
In October 2004, the Food and Drug Administration ordered drug companies to place a strong warning on antidepressant labels after studies suggested that some drugs increased suicidal thinking or behavior in children and adolescents. But the Glaxo study — the first by a drug company to find a link between antidepressants and suicidal behavior in adults, experts say — is likely to persuade some skeptics that the risk is real and not confined to minors.
The studies of children and adolescents found mainly evidence of suicidal thinking and agitation. There were no completed suicides reported.
In a statement issued this week, the F.D.A. said that though it was still evaluating the data, "we are recommending that consumers and prescribers follow current advice to carefully observe adults being treated with antidepressants for worsening of depression and for increased suicidal thinking and behavior." The statement said, "It is essential that patients taking Paxil do not suddenly stop taking their medication."
Last year, the agency asked psychiatric drug makers to review all their data on side effects in adults after a prolonged international debate over whether antidepressant drugs increase the risk of suicide in some children. Other companies have not yet reported their findings.
"This is the first analysis to show a relationship between suicide attempts and one of the antidepressants" since the F.D.A. required the warning label for children and adolescents, said Kelly Posner, an assistant professor in the department of child psychiatry at Columbia, who has helped the agency interpret bad reactions to antidepressants.
Dr. Posner said the Glaxo findings should be treated with caution, because the antidepressant trials done to date were not designed to evaluate suicide risk.
"It's not clear that the drug caused the behavior," she said.
Glaxo sent out the warnings voluntarily, and its data still show that the drug's benefits outweigh the risks for people with depression, said Mary Anne Rhyne, a company spokeswoman.
"We are now advising doctors to monitor all patients to make sure their symptoms don't worsen" in the full course of treatment, Ms. Rhyne said.
Previous research has suggested that the risk of suicidal thinking or behavior was highest in the first few weeks of treatment, or when people went off the medication. One large review of antidepressant trials, published last year in BMJ, a British medical journal, found that people taking Paxil and similar drugs like Prozac reported suicide attempts more often.
But experts have debated the interpretation and value of these findings. In the Glaxo analysis, the researchers analyzed trials that included 8,958 people who took Paxil and 5,953 who received placebo pills. The study participants ranged in age from 18 to 64 years old and were taking the medication for depression or other disorders, like panic attacks and obsessive compulsive disorder.
The analysis found that 11 of 3,455 people who were taking Paxil for depression reported an attempted suicide, compared with 1 in 1,978 taking placebo in the trials. Most were among adults ages 18 to 30, the company said.
Over all, the analysis found no increased risk of suicidal behavior in adults over 30.
"The new findings are not going to change my practice a lot, but I say, yes, they provide a reason to do even more careful monitoring of people on the medication," said Dr. George Simpson, a professor in psychiatry and behavioral sciences at the Keck School of Medicine at the University of Southern California.
Dr. Simpson said the warning underscored the need for more careful tracking of side effects once drugs went on the market.
"The current system of postmarketing surveillance is lousy," he said.