International Herald Tribune, 24/3/2008

Britain plans to force pharmaceutical companies to share more information with regulators about clinical trials after an investigation recently concluded that GlaxoSmithKline PLC deliberately withheld information about an antidepressant.

The four-year probe by the Medicines and Healthcare products Regulatory Agency, completed earlier this month, said the British company should have revealed more quickly that Seroxat sometimes increased the suicide risk in teenagers — by more than six times.

But without stronger legislation in place, the MHRA admitted there is no chance of prosecuting the company for what the agency perceives as an ethical lapse.

"I remain concerned that GSK could and should have reported this information earlier than they did," MHRA chief executive Kent Woods said in a statement.

GlaxoSmithKline rejected the suggestion that it withheld information.

"We firmly believe we acted properly and responsibly," said Dr. Alastair Benbow, the company's European medical director.

British legislation only obliges companies to report side effects in patients for which drugs are officially recommended.

Because Seroxat was only recommended for adults, GlaxoSmithKline was not required to report on any dangerous side effects it found in adolescents.

But Seroxat can still be given to adolescents if prescribed by a doctor. About half of psychiatric drugs are prescribed "off-label," meaning that doctors give them to patients for whom the drug is not strictly intended.

The MHRA said it sifted through more than 1 million pages of evidence after requesting details of clinical trials held between 1994 and 2002.

In response, Britain's government declared that by the end of the year, it will tighten laws forcing companies to share all their relevant safety research with regulators.

"Companies that conduct clinical trials should not compromise people's health by withholding information," public health minister Dawn Primarolo said. The new laws will require companies to disclose a drug's side effects in all patients tested, Primarolo said.

The GlaxoSmithKline case underlines a growing concern among many health experts that drug companies' tendency to hide damaging data could have disastrous consequences.

"If we make the wrong licensing decisions for whatever reasons, then patients may be put at risk," said Dr. Hans-Georg Eichler, senior medical officer at the European Medicines Agency, which licenses drugs across Europe.

Others warn that without more complete information from drug companies, consumers might see similar unpleasant surprises in the future.

"It would be unwise to assume that this particular case was anomalous," said Dr. Ike Iheanacho, editor of the Drug and Therapeutics Bulletin, an independent publication of the British Medical Journal that tracks drug treatments.

"It's perfectly reasonable for people to be anxious about what other drugs on the market might be questionable," he said.

Patients could be taking drugs they do not need or do not work, said Irving Kirsch, a professor of psychology at Britain's University of Hull and lead author of a study that concluded antidepressants like Prozac are mostly ineffective.

Kirsch and colleagues analyzed data about popular antidepressants after submitting a Freedom of Information Act request to the U.S. Food and Drug Administration. They found that depressed patients taking drugs did not actually do much better than those taking fake pills.

GlaxoSmithKline, makers of numerous antidepressants, responded in a statement that Kirsch's conclusions were incorrect because they evaluated only "a small subset of the data available."

But Kirsch and colleagues looked at nearly all the research submitted to the FDA. If that was only a small subset, experts wonder what other data might be out there.

"If we don't know what companies are doing, how do we know what's being reported?" said Davina Ghersi, coordinator of the World Health Organization's clinical trials registry platform.

Britain's National Institute for Clinical Excellence, the agency responsible for deciding what the health system pays for, issued guidelines on antidepressants in 2004. At the time, it did not have access to all the trial data.

"I shared with them the studies I had, but they did not have sufficient information to do a complete review," said Kirsch, then a consultant to the agency.

In recent years, once-blockbuster drugs, such as the painkiller Vioxx and the diabetes pill Avandia, have revealed worrying side effects, including significantly boosting the risk of a heart attack, years after being on the market.

"If we had seen all of the data in the first place, people might have realized that the claims being made for the drug were overblown and coy to the point of being fraudulent," said Dr. David Healy, a professor of psychiatry at Cardiff University in Wales.

But scapegoating of the drug companies may be too simplistic.

"A drug may turn out to be unsafe not because anyone has done anything wrong, but because new knowledge surfaces," Eichler said. Until a drug has been used by millions of people, the rarest and deadliest side effects may remain unknown.

Licenses are granted for drugs based on limited testing, Eichler said. Post-license monitoring should help sort out which drugs may be particularly risky, he said.

Without more data, people should be wary of what medications they take, experts say.

"If we had an overall picture, we might have a very different view about how good certain drugs are," Healy said. "There's an awful lot of people taking drugs that we don't know everything about."