Not So SMART: Public Citizen Charges GlaxoSmithKline Manipulated Data on Salmeterol

In a letter to Lancet [Lurie P, Wolfe SM. Misleading data analyses in salmeterol (SMART) study. Lancet 2005; 366: 1262-1262. Requires subscription for online access.], members of the US group Public Citizen charged that GlaxoSmithKline (GSK) manipulated the data from the Salmeterol Multicenter Asthma Research Trial (SMART). (Summaries in news media from Newsday and Bloomberg are available.) The main points are as follows (quoting from the letter, which may not be accessible to all readers.)

• SMART "was designed to randomise 60,000 patients to either salmeterol 42mcg twice daily or placebo. The study duration was 28 weeks.... Investigators were also asked to report (but not actively seek out) any serious adverse event that occurred within 6 months after the patient completed the trial. The primary outcome was combined respiratory-related deaths and life-threatening experinces...."
• "In September, 2002, the SMART Data Safety Monitoring Board (DSMB) met to review the data on the 25,858 patients who had been enrolled to date. Based on nearly signficant treatment-related increases in both primary outcome variable and asthma-related deaths, the DSMB offered two alternatives: either increase the target sample size by at least 10,000 persons or terminate the study 'with dissemination of findings as quickly as possible.'"
• "GSK presented the interim analysis of the SMART data at the 2003 meeting of the American College of Chest Physicians, claiming that 'The interim analysis was inconclusive.' Based on the the interim data, on Augu. 11, 2oo3, the FDA modified the labelling for both Serevent (salmeterol) and Advair (salmeterol plus flucatisone) to include a black-box warning of a 'small but signficant increase in asthma-related deaths.'"
• "In a significant deviation from the the original protocol, the [statistical methodology] appendix permitted including in the study dataset adverse events that were reported in the 6 months after the termination of the trial."
• "For the primary outcome variable, as well as for for each of the three major secondary outcome variables, the inclusion of the post trial data produces an apparent diminution of the dangers associated with salmeterol."
• "GSK did not specify to the FDA which dataset it had submitted. Only on April 26, 2004, after the FDA specifically enquired as to which dataset had been provided, did the agency learn that the company had provided only the dataset that included the 6-month data."

Lurie and Wolfe's comments were:

It is now approaching 3 years since the SMART study was terminated. The results have still not found their way into print. (GSK told the [FDA] Advisory Committee that they have been accepted for publication in Chest) and the drugs' labels have still not been finalised. Instead, the company, under seemingly little pressure from the FDA, has succeeded in drawing out the process and initiallymisleading the agency, physicians, and patients with not-per-protocol analyses that diminished the drugs apparent risks.

Kathy A Rickard, Vice-President, Clinical-North America, Respiratory Medicine Development Center for GSK, replied, "Contrary to the assertions posed by Peter Lurie and Sidney Wolfe, GSK has acted responsibly and transparently and communicated with a sense of urgency to ensure that health-care professionals and patients had access to the results of SMART." She declared, "the only dataset analyzed for the interim analysis included events from the 28-week plus 6-month period and were provided to an independent DSMB, not GSK." But she never explained why data from the six-month post-treatment period was included in the main GSK analysis, even though events from that period were less likely to have been related to the administration of salmeterol than those which occured during the time patients actually used the drug. Although she noted that after SMART was stopped, GSK sent two letters to US physicians, she did not explain why it has taken more than two years to get a manuscript in press in a medical journal.
This case provides yet another example of how pharmaceutical companies (and presumably, other organizations) may design studies, and analyze and disseminate their results so as to put their products and interests in a maximally favorable light. Health care professionals now must be very skeptical about the extent that published studies reflect completely and honestly what is known about products and services produced by large health care organizations.
But if the published research literature cannot be trusted, we are practicing in a fog.
I strongly suggest that health care professionals must become better watchdogs to keep the clinical literature honest. Furthermore, I'm afraid we need better government regulation of large health care organizations, starting with pharmaceutical companies, even though I fear that the heavy hand of government will not create its own problems.