Once again, Pfizer faces allegations of Neurontin mismarketing. Experts analyzed internal company documents made public in a lawsuit, concluding that Pfizer marketers put a positive spin on negative research. According to the documents, the company worked to suppress unflattering studies to make the drug look more effective for uses that hadn't yet been approved by the FDA. E-mails--some of them under the helpful subject line "Spinning Serpell," referring to a researcher on one unfavorable study--depict marketing managers casting about for ways to downplay any unflattering data and emphasize the positive instead.
All this, the lawsuit appears to show, after Pfizer took over Warner-Lambert, Neurontin's original maker. Settling off-label marketing claims in 2004, Pfizer maintained that any iffy marketing activities took place before it acquired the drug. Then--and again yesterday--Pfizer said that it set up procedures to prevent off-label promotion of the drug.
Dr. Kay Dickersin of the Johns Hopkins Bloomberg School of Public Health, one of the experts who reviewed court documents, told the New York Times that they spell out “a publication strategy meant to convince physicians of Neurontin’s effectiveness and misrepresent or suppress negative findings.” Pfizer disagrees: the company said in a statement that "study results are reported by Pfizer in an objective, accurate, balanced, and complete manner... regardless of the outcome of the study."
The Neurontin accusations emerge at a time when more and more drugmakers have come under scrutiny for their handling of research data. Consider allegations that Merck hired ghostwriters for journal articles in support of the painkiller Vioxx, then recruited big-name docs to add their bylines. Or concerns that Merck and Schering-Plough delayed the Enhance study, fearing it would cast doubt on the effectiveness of their cholesterol med Vytorin. Or a study that found a higher percentage of positive studies of antidepressants in publication, versus a much lower percentage of unfavorable data.