RADAR explores adverse drug event detection

It takes time for the Food and Drug Administration's Adverse Event Reporting System to detect problems with drugs on the market. But a group of physicians at Northwestern University's Feinberg School of Medicine in Chicago aren't waiting. They've taken matters into their own hands, and their efforts are paying dividends.

According to a study in the May 28 Archives of Internal Medicine, this independent drug safety project, which proactively seeks out data on suspected adverse events, is sometimes able to gather more information than the FDA on such incidents and disseminate alerts with greater detail to guide physician response. Those behind this initiative -- Research on Adverse Drug events And Reports, or RADAR -- are seeking more cooperation with the agency and drug companies.

"We're filling a void," said Charles Bennett, MD, PhD, the study's lead author and RADAR's principal investigator. "But we don't want to compete with the FDA. We want to collaborate."

Dr. Bennett, who is also the A.C. Buehler professor in economics and aging, established RADAR in 1998. He set out to develop strategies that would detect the kind of events that other drug safety monitoring bodies were missing.

The most common means for the FDA to detect these events are the hundreds of thousands of reports it receives annually, although an investigation also could be triggered by a report in the scientific literature or the mainstream media. RADAR, which is funded primarily by the National Institutes of Health, the American Cancer Society and the Dept. of Veterans Affairs, receives far fewer such reports and, therefore, focuses efforts on a handful of drugs at any one time. Sometimes theories are generated and pursued because of signals from basic research or known problems with drugs that have similar effects even though they might have a different mechanism of action.

"Our hypothesis is that the quality, not quantity, counts," Dr. Bennett said. "A small number of very good case reports is better than a large number of very bad case reports."

When project researchers suspect a problem, they actively ferret out more detailed information about the initial report as well as other incidents. For example, when an adverse event requiring plasmapheresis was suspected to be caused by the antiplatelet drug ticlopidine, RADAR contacted physicians who provided this treatment to find out more.

None of Dr. Bennett's research has triggered the removal of medicines. But it has changed how some are used. Notably, RADAR discovered that the antiretroviral nevirapine, when administered to health care workers for HIV prophylaxis, was linked to liver damage, and its use in these cases fell out of favor. Also, research on thalidomide, which revealed links to venous thromboembolism, led to this drug being prescribed more with a preventive for this condition such as low molecular weight heparin. The FDA praised the project.

"Bennett and coworkers have identified important, serious adverse drug reactions through the RADAR project," says a statement from the FDA's Office of Surveillance and Epidemiology. "The study shows the value of reports from physicians who are attentive to the possible side effects of medicine."

A complementary role

The FDA's ability to monitor drug safety has come under increasing fire after the sudden withdrawal of Vioxx (rofecoxib) in September 2004. Bills now working their way through Congress call for shoring up the agency's ability to detect drug problems after approval. The American Medical Association supports enhancing the agency's postmarketing surveillance ability, including the expanded use of targeted postapproval studies and the active tracking of data from clinical and drug utilization databases.

Those who work in the drug safety arena praised RADAR for providing an independent strategy for detecting adverse events related to real-world medication use.

"What they're doing is complementary to what the FDA is doing, and I think it's good," said Curt Furberg, MD, PhD, professor of public health sciences at Wake Forest University School of Medicine in Winston-Salem, N.C. He has served on FDA advisory panels. "The FDA has a passive monitoring system that is very good for detecting rare serious conditions such as liver failure and rhabdomyolysis and ineffective for common serious conditions such as heart attacks and stroke."

But many experts also expressed caution about the chance the drug safety system could become overly reliant on the small but growing number of projects like RADAR.

"This is a nice way to conduct some drug safety studies and make some important findings along the way," said Bruce Psaty, MD, PhD, professor of medicine, epidemiology and health services at the University of Washington, Seattle. "But it's not a comprehensive system in and of itself. We cannot have the drug safety system rely on investigator-initiated work in the long term."

There also are challenges associated with getting the word out about what RADAR researchers find. While this study found that data collected from this project were more complete, it also found that the FDA and drug firms were swifter at distributing information through "Dear Doctor" letters and package insert changes. RADAR depends on journal publication, which can be much slower.