Seroxat, marketed in the US as Paxil, an SSRI anti-depressant whose active ingredient is paroxetine, was a big success for GlaxoSmithKline (GSK) from its release in the 1990s, earning billions of dollars worldwide for the drug giant. With a strategy of creating new markets for the drug, in the mid-1990s GSK began to explore the possibility of obtaining a licence for its use in children (under-18s), and conducted clinical trials on groups of children to provide data proving its effectiveness which would support its application.

However, the trials were a bust, in particular Study 329, the biggest of a series carried out over a period of years in the US. The results of Study 329 demonstrated that Seroxat had no measurable benefit to child patients. As is usual in the so-called gold standard clinical trial approach, the drug was tested on one group while another was administered a placebo, a sugar pill with no active ingredient. Neither group, nor the administrators of the pills, knew which pill they were getting.

The group given the placebo performed just as well, or as badly, as those given Seroxat in terms of the trial designers' criteria for improvement in their depression.

But the children who were given Seroxat suffered adverse reactions ranging from mild to serious. The serious adverse reactions included self-harm — mostly cutting; aggression — violence to others; and suicidality — suicidal thoughts or actions, meaning actual suicide attempts.

Seven out of the 93 children in the Seroxat group had to be hospitalised as a result. That's almost eight per cent. To independent researchers, that would be a wildly unacceptable proportion of a treatment population.

In terms of risk, for example, a prevalence of an adverse drug reaction of one in one hundred, one per cent, or more is defined as ‘frequent' and seen as unacceptable.

Yet Seroxat had put eight in one hundred in hospital in this clinical trial.

"This was the point at which GSK should have begun warning physicians who were prescribing the drug to children because not only is it not effective, it's not safe," says Karen Barth Menzies, the California lawyer heading the team pursuing a giant class action suit against GSK in America. Menzies possesses previously secret internal memos and e-mails from GSK, obtained as part of the disclosure process for the upcoming trial, which show what actually happened next.

The company dropped the plan to apply for a licence for the use of Seroxat in children. Its executives immediately recognised that the results of Study 329, which demonstrated no efficacy in treating adolescent depression, would ensure that no licence would be granted. In an internal company memo dated 14 October 1998, they concluded that the drug does not work and that a licence application would be refused: ‘The results of the study were disappointing. The possibility of obtaining a safety statement was considered, but rejected.' The best they felt they could achieve was a ‘ statement that although safety data was reassuring, efficacy had not been demonstrated'.

‘Efficacy had not been demonstrated' is a fancy way of saying that the drug did not work, and they knew it. ‘The possibility of obtaining a safety statement was considered, but rejected' means they realised that the extent of adverse reactions in the trial, at 8%, meant the drug was not safe.

GSK marketing teams commented that ‘the data would be unacceptable commercially'. Panorama's reporter Shelley Jofre translates this as ‘publishing the data in full would put profits at risk'.

Enter the spin doctors.

So the self-harm, aggression and suicide attempts were kept quiet, the lack of effectiveness filed away. PR people took over to implement a new plan: promote Seroxat to doctors as a treatment for under-18s.

How could this be? If it's not been granted a licence, how can a drug be prescribed? Unfortunately, as Panorama explained, even if a drug is not licensed for a particular group doctors can still prescribe it if they think it will benefit their patient — so-called ‘off-label' prescribing.

So GSK's plan was simple — persuade doctors that Seroxat was indeed suitable for their child patients. Which they duly put into effect, spinning the exact opposite of what the studies had shown: Seroxat, they said, was ‘remarkably effective and safe for children'.

Enter the academic psychiatrists, notably Professor Martin Keller of America's Brown University and child psychiatrist Dr Neal Ryan of the University of Pittsburgh, both co-authors of Study 329.

Panorama questions whether Keller is independent in any sense. The programme pointed out that in one year he earned $500,000 in fees from drug companies including GSK. Worse, it demonstrates that his paper on Study 329, published in the Journal of the American Academy of Child and Adolescent Psychiatry, which says it ranks as number one in child mental health in the world, ‘was written by a ghost writer who worked for a PR company, a PR company hired by GSK'. And it quoted an e-mail from Keller to the ghost writer, thanking him for his work: ‘You did a superb job with this. Thank you very much. It is excellent. Enclosed are some rather minor changes from me…'

The ghost had even penned Keller's covering letter submitting the paper to medical journals.

Child psychiatrist Neal Ryan was paid by GSK as a co-author of Study 329. In 2002 he gave a talk on childhood depression at a medical conference in Toronto sponsored by GSK, attended by 15,000 psychiatrists. He said that Seroxat could be a suitable treatment for children and later told Shelley Jofre that it probably lowered rather than raised suicide rates.

And among the archive of documents in the lawyers' possession, Jofre found the evidence that Ryan had forwarded to GSK her own e-mails to him from 2002 asking questions about the safety of Seroxat. His covering e-mail asked the company for advice on how to respond to her queries.

Nonetheless, the paper, of which Keller and Ryan claimed authorship, was published — but not before one journal turned it down after its peer reviewers found serious flaws, notably the downplaying of adverse reactions to Seroxat among the children trialed.

Last year, Professor Keller admitted to lawyers suing GSK that he had not studied the results of Study 329 in full. "I reviewed data analytic tables. I don't recall how raw it [the data] was. [These are] huge printouts, you know, item by item numbers and variable numbers, and don't even have words on them. I tend not to look at those… I do better with words than I do with symbols," he recounted in a session recorded on video. In effect, he admitted that his ‘authorship' of the paper, apart from being written by a PR company's ghost, was based on at best an inadequate and incomplete scrutiny of the data — hardly the gold standard scientific review we are constantly told validates clinical trials.

Says Karen Barth Menzies: "These academics are truly spokespersons for the drug companies."

But the drug company in question, GlaxoSmithKline, buried the trial data, including data from another, later trial on children which found that the placebos given to the control group of depressed kids ‘worked' better than Seroxat. And later, forced by US medicines regulator the FDA to re-evaluate the raw data from Study 329, the company admitted to four further adverse reactions in which children became suicidal, raising the number suffering severe reactions to the drug from seven to 11 — a shocking 12 per cent of the total, and representing a 600% increase in events related to suicide.

Memos in the lawyers' boxes of GSK papers show that company executives clearly told the PR woman who designed their spin strategy that ‘Seroxat may have caused all of these'. Yet the final article, approved by the PR, stated: ‘Of the 11 patients, only headache (one patient) was considered to be related to the treatment' and concluded that the drug was ‘generally well tolerated and effective'. This was, finally, reviewed by and accepted for publication in the Journal of the American Academy of Child and Adolescent Psychiatry. And it was published relatively unchanged, even though the journal's peer reviewers objected in these terms: ‘Overall, results do not clearly indicate efficacy. Authors need to clearly note this.' ‘The relatively high rate of adverse effects was not addressed in the discussion (core portion of an academic paper)' ‘Given the high placebo response rate, are these drugs an acceptable first-line therapy for depressed teenagers?'

Though the article appeared, GSK abandoned a part of its PR plan which intended to use it extensively in publicising Seroxat's use with children. At company headquarters in Britain, a PR executive commented: ‘Originally we had planned to do extensive media relations surrounding this study until we actually viewed the results. Essentially, the study did not show [Seroxat] was effective in treating adolescent depression, which is not something we want to publicise.'

After publication, the third of GSK's trials reported that the placebo worked better than Seroxat. The company's public claims had been flatly contradicted by its own trial, but it maintained a deafening silence on this, telling doctors instead that Study 329 showed the drug ‘was superior to placebo by several assessment methods'. The two failed studies were not mentioned either when GSK informed its US sales force that Seroxat/Paxil ‘demonstrates remarkable efficacy and safety in the treatment of adolescent depression'.

"They didn't tell the regulators or physicians or parents about these risks or the lack of efficacy. Instead, they went out and promoted this specific study as remarkably effective and safe for kids," says Menzies. "That's a lie. A fraud."

And Panorama clearly demonstrated that GSK lied in implementing Plan B.

Here's what their marketing people concluded after analysing the unpublished trial data: "The results of the studies were disappointing. The possibility of obtaining a safety statement from this data was considered but rejected." And here's what GSK told their sales reps: "[Paxil/Seroxat] demonstrates remarkable efficacy and safety in the treatment of adolescent depression."

And GSK went ahead and promoted the drug for children.

Dr David Healy, Professor of Psychiatry at the University of North Wales, is angry that doctors have been deceived into prescribing a harmful drug. He says: "The experts all said that it was extremely safe and very effective. We were all hoodwinked. Misled. Duped. They [GSK] produced these clinical trials which appeared to be evidence. And they aren't: they're adverts."

In June 2003 Britain's Medicine and Healthcare Products Regulatory Authority (MHRA) advised that Seroxat should not be given to anyone under 18 because its analysis had shown that GSK's own studies demonstrated that the drug actually trebles the risk of suicidal thoughts and behaviour in depressed children.

But this came too late for Charise Gatchell, who watched the programme with her parents — who did not know she was secretly taking Seroxat again, having come off it following serious adverse effects, including cutting herself and suicidal thoughts, during a previous prescription. Short weeks later, her mother found the 16-year-old's warm body hanging on the upstairs landing when she and her husband returned from a weekend away. Says Stephanie Gatchell: "The decision-makers in that company should be brought to justice. They have a lot to explain. They just kept on denying and denying and denying, so a programme like Panorama virtually had to force them to admit… were they ever going to come out with the truth on their own? I don't think so."

Says David Healy: "People who are ill have been deeply betrayed by the whole process. They have been betrayed by the pharmaceutical companies and by all the experts who have been willing to lend their names to the process."

In the US, bereaved families have joined together to sue GSK. The Panorama programme closes with a sequence of some of the children who killed themselves after being prescribed Seroxat. Of the last child, Karen Barth Menzies says: "This is a little boy who was 11 years old. He started taking Seroxat and he ended up… he hanged himself in the closet… with the leash of his new pet puppy."

This isn't suicide; this is murder.

Dublin, Ireland: 1 February 2007