Concerns about naproxen were probably overstated

Decision to halt ADAPT trial with naproxen and celecoxib was due to evidence that participants would not stay on study treatments.

Reports that naproxen may increase risk of stroke and myocardial infarction may have been overstated, the lead investigator of the Alzheimer’s Disease Anti-Inflammatory Prevention Trial told Cardiology Today.

“We did not suspend treatment with naproxen for straightforward safety reasons,” said John C. S. Breitner, MD, of the VA’s Puget Sound facility in Seattle. “The Treatment Effects Monitoring Committee had voted a week earlier to recommend continuation of the ADAPT protocol unaltered.”

ADAPT was designed to assess the potential benefit of long-term use of nonsteroidal anti-inflammatory drugs — naproxen (Aleve, Bayer) and celecoxib (Celebrex, Pfizer) — in decreasing the risk of Alzheimer’s disease. Researchers had enrolled 2,500 patients who were at least 70 and randomized them to naproxen (220 mg BID), celecoxib (200 mg BID) or placebo. Patients had no symptoms of Alzheimer’s, but they were considered at risk due to family history.

Breitner said the decision to halt the ADAPT trial was based on “operational difficulties” stemming from news that one trial with celecoxib had presented increased risk of heart disease and stroke. “We had rapidly accumulating evidence that participants would not stay on study treatments in ADAPT and would lose confidence in us if we asked them to stay in the celecoxib or placebo arms,” Breitner said. “With the preliminary evidence we had in ADAPT, it would have been disingenuous and misleading to have continued the naproxen arm and suspend the celecoxib arm.”

Safety concerns about naproxen that were reported in many news accounts resulted from “insistent press questions” at a news conference, Breitner said. “We cautioned that the figure quoted was a preliminary, global estimate of the apparent strength of the safety signal — substantially weaker than was reported from the celecoxib trial — but it was immediately reported as a hard number. We don’t know the real hard numbers yet. We are in the process now of cleaning up these data in the shortest possible time.”

Disappointing phase

Reports of safety concerns over naproxen were the latest in a series of negative reports about pain relief agents. Raymond Woosley, MD, PhD, director of The Center for Education and Research on Therapeutics, University of Arizona and section editor of Cardiology Today’s Cardiovascular Pharmacology section, called these developments “a very disappointing phase of medicine because of the way the new data on these drugs has been handled without in depth analysis of the data.”

In September 2004, Merck pulled rofecoxib (Vioxx, Merck) from the market after data from the APPROVE (Adenomatous Polyp Prevention on Vioxx) trial showed a 3.5% risk of cardiovascular events among patients assigned rofecoxib compared to 1.9% among patients assigned placebo.

Raymond Woosley

Immediately after Merck’s announcement, Pfizer released a statement reminding the medical community that celecoxib was considered safe. Three months later the company halted a cancer prevention trial after it was shown that patients assigned the 800-mg dose of celecoxib had a 3.4- fold greater risk of cardiovascular events compared to placebo; patients assigned the 400-mg dose had a 2.5-fold greater risk.

A. Mark Fendrick, MD, University of Michigan, said that new prescriptions for celecoxib and valdecoxib were down approximately 50% since that news was reported. Two Food and Drug Administration advisory panels will meet this month to review data on Cox-2 inhibitors and traditional NSAIDS.

For some experts, Pfizer’s recent decision to halt direct-to-consumer advertising of celecoxib is a positive development.

“I was not disappointed to see the direct-to-consumer advertising go away because I have always felt that it was just totally inappropriate,” Woosley said. “Direct-to-consumer advertising was intended to let people know that there was a treatment intended for their disease and they should ask their doctor about it. But that was never the case here. Direct-to-consumer advertising was hyping one drug over another just to expand their market.”

Steve Chen, PharmD, University of California School of Pharmacy, said Cox-2 inhibitors would not have been so popular without the “excellent marketing by the pharmaceutical companies.

Steve Chen

“The detailing sparked a lot of confusion about the belief that Cox-2 inhibitors were GI sparing, and only Vioxx ever showed that,” Chen said. “By the time the full data were released to the FDA, the [FDA] refused to give them a label that said [Celebrex] was GI sparing. But doctors never got that message.”

Douglas Jackson, MD, an orthopedic surgeon in Long Beach, California, told Cardiology Today that he had “never seen such an intense drug-marketing effort directed at orthopedic surgeons and their patients.

“I had never received so many invitations to attend drug company-sponsored dinners, ball games and concerts, or to become a consultant and meet at some resort,” Jackson said. “I’m a bit disappointed in myself for being so susceptible to the unprecedented marketing hype around Cox-2 inhibitors.”

There is still a role for Cox-2 inhibitors. Woosley said the drugs were intended for a “relatively smaller market” of people who are at risk for gastrointestinal bleeding.

“Use them for the way they were intended, which is for people at high risk for ulcer disease in need of pain and arthritis treatment. Don’t use it as a first-line treatment for someone not at risk, monitor its effects on blood pressure and do not use it beyond the dosages that are recommended,” Woosley said.

The FDA has issued a public health advisory recommending limited use of Cox-2 inhibitors.

Fendrick said a similar level of gastrointestinal risk reduction as demonstrated by Cox-2 inhibitors can be achieved by combining a traditional NSAID with a proton pump inhibitor like ezomeparazole (Nexium, AstraZeneca) or lansoprazole (Prevacid, Tap). Since the GI safety advantage of Cox-2 inhibitors is strongly diminished in the setting of aspirin, this PPI/NSAID combination should be considered for the millions of NSAID users who also take a daily aspirin to prevent MIs and strokes. “Why spend an extra $2 a day on a Cox-2 inhibitor when it provides neither an advantage in either pain relief or GI safety? For aspirin users, it would be preferable to spend the money on a traditional NSAID and proton pump inhibitor [to reduce risk in these at-risk patients],” he said.

Many questions remain

Controversy over Cox-2 inhibitors has left many questions unanswered and has “generated more heat than light,” said Franz Messerli, MD, editor of Cardiology Today’s Hypertension and Vascular Disease section and director of the hypertension program at St. Luke’s Roosevelt Hospital, New York.

“Nobody has thoroughly analyzed the reason why some of these Cox-2 inhibitors increase myocardial infarctions and cardiovascular events,” Messerli said.

In a report published in Archives of Internal Medicine, James Sowers, MD, University of Missouri-Columbia, and colleagues said that hypertension could be the causal link in diabetic hypertensive patients with osteoarthritis. After six weeks of treatment with rofecoxib, 24-hour systolic blood pressure increased from a mean of 130 mm Hg to 134 mm Hg, which was significantly greater than that seen with celecoxib or naproxen.

“The blood pressure increase seems fairly minimal, but you have to consider also that this was in diabetic patients and in diabetic patients we know that, if anything, an increase in blood pressure just about doubles or triples the risk of outcome events when compared to the nondiabetic,” Messerli said. “If blood pressure is destabilized by a Cox-2 inhibitor for a prolonged time period, the implications of this population-wise could be potentially devastating.” – by Jeremy Moore

Dr. Fendrick is a paid consultant for AstraZeneca, Merck, McNeil, TAP, Novartis, and Procter and Gamble, and he is on the speaker’s bureau for AstraZeneca, McNeil and TAP.