Drug Maker Withheld Paxil Study Data
ABC News Uncovers Documents Unknown to Regulators and Many DoctorsDec 9, 2004 | ABC NEWS
New documents uncovered by ABC News suggest GlaxoSmithKline, the maker of the popular antidepressant Paxil, failed to disclose important information about the possibility of an increased risk of suicidal behavior in some children taking the drug, as well as serious withdrawal symptoms when some patients stop taking Paxil.
The new documents obtained by the ABC News program "Primetime Live" have never before been made public. Earlier this year, the Food and Drug Administration ordered manufacturers to place a warning in bold print on antidepressants, alerting consumers that the drugs can cause suicidal tendencies in children and teenagers.
The FDA's move raised the issue of how much antidepressant manufacturers knew about this increased risk and when.
While Congress and the FDA have held hearings questioning GlaxoSmithKline and other drug manufacturers about suppressing studies that cast their products in a negative light, "Primetime Live" has obtained actual copies of GlaxoSmithKline reports from its own internal studies. In these studies, some children showed the same types of suicidal thoughts and behaviors that parents had for years claimed their children were exhibiting.
According to these documents, internal studies by GlaxoSmithKline concluded that Paxil had little or no effect in treating depression in children and adolescents. And as far back as 1997, the company was aware of studies reporting suicide-related behaviors in young patients taking the drug.
In spite of this information, GlaxoSmithKline distributed a memo to its sales force in 2001 touting the drug's "remarkable efficacy and safety in the treatment of adolescent depression."
See the memo.
Additionally, the internal documents reveal GlaxoSmithKline had scientific evidence Paxil could cause mild to severe withdrawal symptoms in some patients when they abruptly stopped taking the drug. A document the company distributed to sales staff referred to "withdrawal symptoms" and said they included problems like dizziness, nausea, fatigue, insomnia and nightmares, but advised the sales representatives to use the less-alarming term "discontinuation symptoms." The documents suggest GlaxoSmithKline's sales strategy was to downplay this information in order to maintain its share of the billion-dollar antidepressant market.
"This is about money," said Rep. Henry Waxman, D-Calif., who is on a congressional committee investigating antidepressant manufacturers. "This is not about science, because what they're doing is withholding the scientific information, suppressing the studies that could have a negative impact on their sales and their profits."
Drug Maker Under Attack
In addition to the investigation by Congress, the New York attorney general's office in June of this year leveled charges of consumer fraud against GlaxoSmithKline. The lawsuit alleged that the drug giant repeatedly concealed information about Paxil from doctors.
The attorney general's lawsuit noted that GlaxoSmithKline conducted at least five studies of Paxil's use in children and adolescents, but the company only published one of these studies a decision at the heart of the controversy.
Central to the attorney general's allegations were two Paxil studies for depression in children and adolescents. ABC News has obtained original documents from those studies.
In one study, referred to by the company as Study 377, researchers noted some children exhibiting suicidal behaviors and attempts to kill themselves. Referring to that study and another, Study 329, an internal GlaxoSmithKline memo acknowledged that Paxil "failed to demonstrate a statistically significant difference from placebo on the primary efficacy measures."
The same internal memo went on to identify a single goal: "to effectively manage the dissemination of these data in order to minimize any potential negative commercial impact."
The document recommends that the company not seek government approval for Paxil's efficacy or safety in treating depression in adolescents at that time because they believed that they would not get that approval and "it would be commercially unacceptable to include a statement that efficacy had not been demonstrated, as this would undermine the profile of [Paxil]." Finally, the memo concludes, "Positive data from Study 329 will be published in abstract and a full manuscript of the 329 data will be progressed."
The results from Study 329 were ultimately published in the Journal of American Child Adolescent Psychiatry in 2001 with the conclusion "Paxil is generally well tolerated and effective for major depression in adolescents."
The New York attorney general criticized the publication of this article, saying, "Although it allowed the data from study 329 to be published, GlaxoSmithKline concealed and suppressed studies 377 and 701, which failed to show that [Paxil] was more effective than placebo in treating [depression] in children and adolescents." Nonetheless, a 2001 internal memo to the company's sales reps touted the study as showing Paxil is "truly a remarkable product that continues to demonstrate efficacy" in adolescents.
See the memo.
GlaxoSmithKline says its reps were not supposed to discuss the memo with doctors and notes it was stamped "For representatives' information only," but the New York attorney general's complaint charged that the company would only have sent the memo to its reps "to falsely characterize Study 329 in their communications with physicians." GlaxoSmithKline later settled the lawsuit, without an admission of wrongdoing, but paid a $2.5 million fine and promised to make all of its research available online.
GlaxoSmithKline defends its public disclosures regarding suicide risks, saying that in each individual pediatric study the numbers were too small to be statistically significant. GlaxoSmithKline also claims that it wasn't until data were pooled from all antidepressant studies that the increased risk of suicide became clear. It and other drug companies also say there's no way to prove that any individual suicide-related behavior was caused by the drug and not the patient's underlying depression.
The makers of Paxil, Effexor and Zoloft said they have not endorsed the use of their antidepressants in children or adolescents. They also point out that during their pediatric trials no one actually committed suicide.
Downplaying Withdrawal Symptoms
Paxil belongs to a class of antidepressants known as selective serotonin reuptake inhibitors, or SSRIs. Like other SSRIs, Paxil can cause withdrawal symptoms when patients stop using the drug. Until 2001 there was no clear warning or precaution in the prescribing information provided to doctors by GlaxoSmithKline.
In prescribing information up to December 2001, the company only mentioned "withdrawal syndrome" as a rare potential side effect associated with the nervous system, but without explaining what "withdrawal syndrome" is. GlaxoSmithKline also mentioned spontaneous reports of "discontinuation symptoms," but explained these reports "may have no causal relationship with the drug."
But "Primetime Live" discovered an internal safety review dating from 1997 that provided scientific evidence from GlaxoSmithKline studies of withdrawal effects caused by Paxil. There was never any mention of these studies and the specific results in any of the official prescribing information made available to American doctors and patients.
Starting in December 2001, GlaxoSmithKline included a precaution to make doctors and patients aware of some possible "discontinuation" side effects affecting "2 percent or greater" of patients based on studies.
According to the 1997 review, one study found that 25 percent of patients taking Paxil in one experienced discontinuation symptoms (vs. 5.9 percent for patients taking a placebo). In a study of patients with major depression, 42 percent of the patients taking Paxil experienced one discontinuation symptom.
GlaxoSmithKline told "Primetime Live" that "most of the individual studies have been made public either through publication in a peer-reviewed journal or presentation at a major medical meeting." But none of the medical experts contacted by "Primetime Live" knew about this data, and GlaxoSmithKline did not make those publications or presentations available, despite repeated requests.
Despite these studies, these documents reveal that GlaxoSmithKline instructed its sales force to downplay or minimize the negative effects of discontinuing Paxil from doctors
A 1998 "business plan guide" for sales reps instructed them to "minimize concerns surrounding discontinuation symptoms." Sales reps were also told to explain to doctors that the "discontinuation incident rate is two in 1,000 patients." This rate is markedly different from the results of the studies in the safety review published nine months earlier.
See the Business Plan Guide.
In a company-sponsored education program for sales representatives, "withdrawal syndrome" was clearly defined as "a class effect that can occur when an SSRI is stopped abruptly. Symptoms may include asthenia, flu-like symptoms, fatigue, dizziness, nausea, and sleep disturbances (insomnia, vivid dreams or nightmares)." However, the next direction recommended reps avoid using the term: "instead of 'withdrawal syndrome,' which implies addictive properties, try to refer to this phenomenon as 'discontinuation symptoms.'"
One vivid reminder of the importance of Paxil to GlaxoSmithKline is found on a May 1997 sales memo. Sales reps are reminded, "Let's face it in the end. [sic] The only thing the anxious and agitated patient will be saying is: 'Where's my Paxil!!!!!!'" The comment is illustrated by a cartoon of a screaming woman.
See the May 1997 sales memo.
As if to underscore the importance of Paxil to GlaxoSmithKline, another company sales document graphically explained why discontinuation is an issue. The answer: $1 billion, the total sales of Paxil and Seroxat (a related SSRI) as of September 1997. This statement is illustrated by a large, black money bag.
See the "money bag" memo.
A Publishing Coincidence
In a 1997 memo, a public relations firm working for GlaxoSmithKline drafted a letter on the topic of discontinuation symptoms that was apparently intended for publication by a company spokesperson or possibly a physician.
The memo explained that "complete duplication will look fishy if we decide to submit both. Are there other references we could draw on for the various drugs? At the very least, we can't have the references appear in the same order." The draft letter listed three physicians as the intended senders, including Dr. Bruce Pollock of the University of Pittsburgh.
A letter by Pollock appeared in the October 1998 issue of the Journal of Clinical Psychiatry. Although Pollock's letter was not the same as what the PR company drafted, it made nearly every point in the draft almost in the same sequence.
See the PR company's draft letter and Dr. Pollock's letter to the journal.
Pollock, who was on the company's advisory board at the time, told "Primetime Live" he wrote the published letter and stands by its accuracy. He did say, however, he "could imagine a scenario where a representative from the makers of Paxil said, 'Could you make this point?'" because it was well-known that he had studied Paxil in elderly patients. But he denied that he had any knowledge of intentions to use his letter in a marketing plan, or that he had any knowledge of the draft letter prepared by the PR firm that included his name.
This published letter resurfaced once again in the internal memorandum obtained by "Primetime Live." The company's 1998 business plan guide for sales reps cited Pollock's letter as "an effective tool for addressing discontinuation."
GlaxoSmithKline declined to comment on these documents, saying "it is difficult to respond to court documents taken out of context." In court, they have said they provided adequate and complete warnings to physicians about the risks of withdrawal. In a letter to "Primetime Live," they also said the majority of patients who experience any symptoms have mild to moderate symptoms that are usually self-limiting within two weeks.