November 30, 2004

One in ten

David Graham's study will finally see the light of day Wednesday when the Journal of the American Medical Association publishes his analysis of the dangerous side effects of statins.

Graham, a top medical officer in the Food and Drug Administration's drug safety office, along with several colleagues reviewed a quarter million HMO records of patients, many elderly and diabetic, who took statins to lower cholesterol in the late 1990s through 2001. Besides Lipitor and other leading statins, the study looked at Bayer's Baycol (cerivastatin), which was withdrawn from the market in late 2001.

The shocking numbers: for every ten persons treated with Baycol and a fibrate (generics used to lower triglycerides, another lipid that in excess can cause heart problems), one came down with the debilitating muscle wasting disease, rhabdomyolysis.

The other leading statins were much less dangerous. If one looked at patients who only took statins, one in 20,000 patients taking the leading brands for a year would wind up in the hospital with rhabdomyolysis. But for those taking Baycol, it was 10 hospitalizations for every 20,000 users.

No wonder the FDA has been trying to shut him up and denigrated his study. With that kind of comparative risk profile, how did Baycol ever get approved? And what does it say about the state of post-marketing surveillance at the FDA if the agency allowed it to stay on the market for nearly two years?

As the accompanying commentary (which received widespread press coverage when it was released early) pointed out, putting companies in charge of collecting adverse events reports is a flagrant conflict of interest. The time has come for an adequately funded office of drug safety that is independent of the section of the agency that approves new drugs.

Meanwhile, a tip of the hat to Sen. Chuck Grassley (R-IA) for going to bat for Graham after he testified at the Vioxx hearing two weeks ago. Grassley wrote the FDA this week to warn its leaders not to punish Graham for going public.

Attorney Tom Devine of the the non-profit Government Accountability Project also has been doing yeomanlike work defending Graham. After Graham called him up seeking whistleblower protection, some higher-ups at FDA tried to scare Devine off the case by making anonymous phone calls and casting aspersions on Graham's work. Devine checked his caller i.d., saw the calls were coming from inside the FDA, and leaked the cad-like behavior to the Washington Post.

Posted by gooznews at November 30, 2004 11:04 PM
Comments

Although I share your concerns about the FDA not doing its job, it is not the FDA's job to study the combination of one drug with another. Data that seem obviously worrisome when based on a sample of more than 250,000 patients would not show up when a few thousand people are studied. When the FDA asks for studies of safety and effectiveness, the standard practice is to study one product at a time. Doctors need to be more aware of possible drug interactions, and they often are not -- even ignoring warning letters from the FDA about specific risks.

On the other hand, the FDA is remiss when it fails to regulate how medicines are used in the real world -- and especially when it fails to require long-term safety data for products that are taken for year after year. Even implants (heart valves, shunts, breast implants or whatever) are rarely studied for more than 2 years. I just wrote an article on that topic, published in several Knight Ridder newspapers:
http://www.center4research.org/medcabsafety.html

Diana Zuckerman, Ph.D.
President
National Research Center for Women & Families
1901 Pennsylvania Ave, NW, Suite 901
Washington, DC 20006

Posted by: Diana Zuckerman at December 2, 2004 09:05 AM