In 1996, due to reports of negative side-effects with the use of GSK's newly approved salmeterol, and asthma-related deaths in patients taking the same class of drugs, the company initiated the SMART study--a randomised trial comparing salmeterol against a placebo. The results have never been published in a peer-reviewed journal. On July 13, 2005, salmeterol came before the FDA's Pulmonary Drugs Advisory Committee. Peter Lurie and Sidney Wolfe (Public Citizen's Health Research Group, Washington D.C., USA) examined the committee's briefing materials on the drug. The authors state that GSK submitted 6 months of post-trial follow-up data on the drug merged with the study's actual per-protocol 28 weeks of data. This produced lower risk values when compared with the original per-protocol data analysis. The authors claim that GSK did not clearly state to the FDA which dataset it had submitted. Lurie and Wolfe believe that the briefing documents provide insight into the manner in which GSK manipulated the data it sent to the FDA in an attempt to convince the agency that the drug's risks were smaller.
In a reply to the criticisms published alongside Lurie and Wolfe's letter, GSK states: "The protocol specified collection of reported events during the 28-week study period and a 6-month follow-up, since patients might have had access to, and used, study medication during this period. Study data were collected and analysed by a clinical research organisation. The only dataset analysed for the interim analysis included events from the 28-week plus 6-month period and were provided to an independent DSMB [Data Safety Monitoring Board], not to GSK. The interim analysis suggested worse outcomes in salmeterol-treated patients, mainly African-Americans…After discussions with the DSMB, GSK was provided with unmasked results of the interim analysis and subsequently terminated the study due to difficulties in enrolment and findings in African-Americans…The FDA requested that GSK further refine the analysis to include data from the 28-week treatment period. These data were reported to the FDA in May 2004. There were no substantive differences between the final data for SMART and those for the interim analysis…In collaboration with the FDA, numerous activities were undertaken to immediately communicate SMART results to health-care professionals. On the day SMART was stopped (Jan 23, 2003), a Dear Healthcare Professional letter was delivered by overnight mail to all US health-care professionals who had prescribed salmeterol or salmeterol-containing products within the previous year. Simultaneously, notices on both the FDA and GSK websites were posted. Results were presented at The American College of Chest Physicians in Oct 03. In addition, SMART study results, containing both analyses, have been posted on the GSK Clinical Trial Registry. The SMART paper submitted in November, 2004, to Chest, has been accepted for publication."
Contact: Dr Peter Lurie, Deputy Director, Public Citizen's Health Research Group, 1600 20th Street NW, Washington DC 20009-1001, USA. T) +1 202 588 7781 plurie@citizen.org
Marian Tadman, GSK, Director, Global Public Relations, Global Commercial Strategy, Respiratory. T) +44 (0) 20 7711 8831 or +44 (0) 20 8966 8831/ + 44 (0) 7749 822 616 marian.x.tadman@gsk.com
Journal
The Lancet