Antiviral drugs oseltamivir (Tamiflu) and zanamivir (Relenza) made big bucks for Roche and GSK respectively during the H1N1 influenza pandemic: Roche reportedly sold about $3 billion of Tamiflu in 2009. Although sales have dropped off precipitously, the drugs are still recommended by CDC for serious cases of seasonal influenza.
But do Relenza and Tamiflu even work for the purpose for which they're recommended -- to reduce complications from serious influenza? Leading academic investigators say not only is the answer "probably not," Tamiflu may actually suppress the body's natural production of antibody against influenza.
These answers would become more clear if Roche and GSK would release the data from their randomized controlled trials on the antiviral drugs. They won't, causing the Cochrane Collaboration to cry foul as it has sought the data to conduct its own updated analysis for the respected Cochrane Database of Systematic Reviews. (Cochrane's first analysis was in 2006, and recommended Tamiflu based on the incomplete data Roche gave without informing Cochrane there was other data being withheld.) The Atlantic Monthly then ran a piece called "The Truth About Tamiflu" detailing Roche's successful manipulation of the CDC and Cochrane, whose endorsement likely increased sales.
Tom Jefferson of Cochrane and Peter Doshi of Johns Hopkins have had enough: they went nuclear by writing an Op-Ed, "Drug Data Shouldn't Be Secret," in the New York Times this week. They call Roche's data secrecy "a disservice to those who volunteer their bodies for clinical trials, and dangerous to those being asked to swallow approved medicines."
Using the limited randomized trial data available in their new 2012 analysis, Tom Jefferson and Cochrane concluded that oseltamivir (Tamiflu) reduced symptom duration by 21 hours (from a median of 160 hours with placebo), but had no effect on hospitalizations. However, the strength of the conclusions even from this data is poor: only 7 randomized trials had analyzable data on hospitalizations, and authors felt there was a high likelihood their data were compromised by publication and reporting biases.
Jefferson and the Cochrane team didn't analyze the zanamivir (Relenza) data, since GSK hinted it would share its data from randomized trials (they're still waiting, according to their Op-Ed).
Jonathan Hsu, Nancy Santesso, and Holger Schunemann published a systematic review and meta-analysis of a mound of available observational data in the February Annals of Internal Medicine, to try to shed extra light on the picture from the real world use of antivirals outside of clinical trials. They did find the data supported a benefit of oseltamivir in reducing hospitalizations and death, but they found the data to be low-to-very low quality, and also felt it was probably rife with reporting and publication bias.
Tom Jefferson, Peter Doshi Op-Ed, "Drug Data Shouldn't Be Secret," The New York Times April 10, 2012.
Hsu J et al. Antivirals for Treatment of Influenza. A Systematic Review and Meta-analysis of Observational Studies. Ann Intern Med 2012;156:512-524.
Shannon Brownlee and Jeanne Lenzer, The Atlantic Monthly, "The Truth About Tamiflu."
Center for Medical Consumers, "Drugs to Treat the Flu."
Jefferson T et al, Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database Syst Rev 2012;Jan 18;1:CD008965.