By Ewen Callaway
Journal articles on hormone replacement therapy (HRT) ghostwritten by medical writers employed by the pharmaceutical industry serially understated the treatment's risks and promoted unapproved uses, according to an analysis of industry documents.
The analysis, published September 7 in the journal PLoS Medicine, is based on some 1,500 e-mails, contracts and other documents made public in July 2009, after The New York Times and PLoS Medicine successfully argued that their release would be in the public interest. Many thousands more papers remain sealed as part of ongoing lawsuits brought by more than 14,000 women against the drug maker Wyeth, which was bought last year by the pharmaceutical giant Pfizer, based in New York.
"People needed to know about what was going on behind closed doors in terms of these articles," says Adriane Fugh-Berman, a researcher at Georgetown University Medical Center in Washington DC, who conducted the analysis. She is a paid expert witness for plaintiffs against Wyeth and Pfizer.
Millions of women took a mix of the hormones oestrogen and progesterone called Prempro. HRT was intended to ease the symptoms of menopause, but many studies suggested that it also helped to protect against heart disease, osteoporosis and other conditions.
But a landmark 2002 study of tens of thousands of women enrolled in the Women's Health Initiative clinical trial found that HRT did not prevent heart disease, and increased the risk of breast cancer and stroke. Breast cancer rates in the United States have since fallen as many women abandoned HRT, although it is still prescribed to combat severe symptoms of menopause such as hot flushes.
Fugh-Berman's study focuses on Wyeth's previously disclosed use of ghostwriters on papers discussing Prempro. The documents detail arrangements made between Wyeth, DesignWrite--the company it employed to pen articles on HRT, based in Princeton, New Jersey--and the academic authors named on papers.
Between 1997 and 2003, DesignWrite oversaw the publication of dozens of peer-reviewed articles, conference abstracts and posters on HRT, receiving up to $25,000 per project. The documents suggest a deep involvement in writing, editing and overseeing the publication of the papers, with sometimes only minimal involvement from the named authors.
"The beauty of this process is that we become your postdocs," reads one 2001 e-mail from a DesignWrite employee to an academic author. "We provide you with an outline that you review and suggest changes to. We then develop a draft from the final outline. You have complete editorial control of the paper but we provide you with the materials to review/critique."
More troubling, says Fugh-Berman, is that the ghostwritten articles downplayed the risk that HRT might promote breast cancer by disputing epidemiological data showing such a link.
At the same time the papers supported unvetted 'off-label' uses for HRT that were not approved by drug regulators, including healthier skin, protection against Alzheimer's and Parkinson's diseases, and a generally higher quality of life. Drug companies are not allowed to make such claims in their advertising. "They have to be channelled through the mouths of academics," says Fugh-Berman.
Ghostwritten articles also questioned other therapies and the effectiveness of generic hormone treatments, she says.
In a statement provided by spokesman Christopher Loder, Pfizer disputes Fugh-Berman's conclusions. "This article completely--and conveniently--ignores the fact that the published manuscripts were subjected to rigorous peer review by outside experts on behalf of the medical journals that published them."
The company also notes that it has clamped down on ghostwriting by requiring that academic authors are involved in papers throughout their writing, and that the contributions of authors who did not write the paper are explained.
"This behaviour has happened, but arguably not often, and probably not recently," says Thomas Stossel, director of the Translational Medicine Division at the Brigham and Women's Hospital in Boston, Massachusetts. Fugh-Berman's analysis conflates legitimate services offered by professional medical writers and ghostwriting, he says. "The fact is that such writers provide assistance to academics as well as commercial entities, just as speechwriters abet politicians."
But Leemon McHenry, a medical ethicist at California State University in Northridge who has investigated the use of ghostwriters on articles about other drugs, thinks that the practice stretches beyond these high-profile cases.
"How many other drugs have been promoted in the same way, but you never find out about them because nobody's suffered heart attacks?" he says. "Nobody finds out about this at all until there's been some major damage and the lawsuits get filed."