Forbes is running a piece called "Open Source Drug Safety?" that takes a look at a recent spate of independent researchers who publicize drug safety issues that pharmaceutical companies keep quiet. The lead example is the recent case of Avandia, a diabetes drug manufactured by GlaxoSmithKline. When Glaxo examined 42 of its own studies on the drug, they determined that a 30% increase in heart attack was likely; this result was communicated to the FDA, with an argument that "the risk only occurred in patients who already had serious heart problems, and that it didn't show up in long-term clinical trials." The FDA sat on this information.
Meanwhile, Steve Nissen, chairman of cardiology at the Cleveland Clinic, found a web site with the results of those 42 studies and did his own analysis, coming to a slightly different conclusion: "The patients who got the medicine were 43% more likely to have a heart attack, and 64% more likely to die." Instead of going to the FDA, however, Nissen went to the New England Journal of Medicine, which immediately published his results. And now the FDA is being spurred to pull an investigative panel together; in the meantime, Avandia sales are likely to be hurt by negative publicity.
Avandia is not the first drug that's been "outed" in this fashion; the article describes several similar cases (including Vioxx) as it builds its argument that what we're seeing is "open source" safety monitoring. This is resulting in a Congressional effort to compel pharmaceutical companies to more openly publish their clinical trial results on the web. Although some criticize the notion that we should rely on the interpretation of independent researchers, claiming for instance that "drug safety experts are too eager to use studies to answer questions they were not designed to answer," others see this as a necessary piece of the puzzle given the circumstances:
But drug safety experts say that the problem is that this is the best data available--and that the FDA, because of lack of authority to force drug firms to do anything, has not been able to mandate large studies that would answer safety questions. "This is the best data we have right now," says Bruce Psaty, a drug safety expert at the University of Washington.
The FDA's ability to track drug side effects is "a joke," says David Nathan, a diabetes drug expert at Harvard Medical School who reviewed Nissen's paper. " What Steve Nissen did was appropriate and clever." Adds Curt Furberg, from Wake Forest University: "In the U.S., we don't take action. Drugs are approved and then we discover problems down the road."
Certainly increased scrutiny over trial results is a good thing, but it seems like simply asking companies to publish on the web and hoping someone out there will take an interest in reviewing the data is shifting the responsibility. Senator Charles Grassley attempted to include a provision in the Senate bill that would set up a separate FDA division just to monitor side effects, but that provision got nuked. Maybe if enough of these cases accumulate, Congress will reconsider - hopefully a few more of these cases will occur before the drug is actually on the market.
