Naproxen Label Shows Limitations Of Class Effect Thinking – Harvard’s Avorn

The inclusion of naproxen in NSAID cardiovascular warning label requirements is an overextension of the use of "class effect" labeling, pharmacoepidemiologist Jerry Avorn, Harvard University, said.

"I worry, especially as we talk about the [COX-2 inhibitors] or about NSAIDs in general, that we may have pushed" the concept of class labeling "beyond its usefulness," Avorn remarked. "It's a great approximation, but there are subtle differences."

"The most important concept is that it may not be the greatest predictor of real events in real patients."

Avorn spoke Aug. 23 at the International Conference on Pharmacoepidemiology & Therapeutic Risk Management in Nashville, Tenn.

"If you look at the molecular structure of a lot of non-steroidals, it is striking how unlike each other they look as compounds," he noted.

Avorn indicated that class labeling should not have been applied to naproxen because studies suggest that the agent is cardioprotective.

"If we look at the naproxen data... we are beginning to see an erosion of this notion of class effect," Avorn asserted. "The general pattern that seems to emerge is that naproxen is kind of good for your heart."

"Yet we have a class label now in the U.S. that all non-steroidals, including naproxen... can be associated with increased risk of cardiovascular disease," he said. "That is the best example that I can think of that maybe we shouldn't think of these drugs as [having] a class effect."

Naproxen products such as Bayer's Aleve , along with products containing ibuprofen or ketoprofen were recently required to amend labeling to include a statement that "long term continuous use may increase risk of heart attack or stroke" (¹ (Also see "NSAID Labeling Revisions Revised By FDA; Compliance Date Extended" - Pink Sheet, 25 Jul, 2005.), p. 3).

FDA determined in April that naproxen would be included among OTC NSAIDs that would carry class labeling following a February joint advisory panel meeting on COX-2 and other NSAID safety.

However, many panelists and other experts read the body of naproxen data to suggest it is in fact cardioprotective or neutral in its effect.

FDA should be less reliant on potential class similarities in evaluating safety and efficacy, Avorn said.

"We all need to grow up pharmacologically and epidemiologically and clinically and declare a moratorium on the concept of the class effect because it is sloppy thinking."

"We should stop talking about class effects for anything, either for benefits or for risks," he added. Avorn recently authored Powerful Medicines: The Benefits, Risks, and Costs of Prescription Drugs.

Class effects should be considered "hypothesis-generating, rather than as something which is the final word," he said.

"Once one has a hypothesis of a class effect... that really gives you a great question. It doesn't give you a great answer," he explained. "The next step is to do clinical trials or observational trials or, preferably, both to get the right answer."

Avorn suggested that it may not always be constructive for FDA to rely on class similarities in making regulatory decisions.

"There is another virtue to class effects if you work for a government agency, which is that it makes it possible to make sweeping statements about lots and lots and lots of drugs even if you don't have the same quality of data about all of them," he said.

"That can be appealing in some ways, but it may also be counterproductive."

FDA Office of Medical Policy Director Robert Temple recently said that pharmacologic similarity gives the agency greater latitude in making regulatory decisions.

While Avorn expressed particular concern about the class labeling of NSAIDs for cardiovascular risk, FDA also has proposed class labeling for non-steroidal anti-inflammatory drugs, antidepressants and antihypertensives in recent months.