As two retrospective analyses to be published in the Oct. 4, 2006, issue of the Journal of the American Medical Association tease apart the risks of different COX inhibitors and other NSAIDs for both heart and kidney, a third paper in the Proceedings of the National Academy of Sciences reported that targeting another step of the prostaglandin synthesis might be able to yield drugs with a better risk-reward profile.

Cyclooxgygenase-2, or COX-2, is a major enzyme in the pathway to make prostaglandins; two fats called PGE2 and prostacyclin are major members of the class that mediate pain and inflammation. Which is why drugs that specifically inhibit COX-2 were, once upon a time, blockbuster pain relievers: The thinking was that the drugs would be an improvement over nonselective COX inhibitors because they would no longer inhibit COX-1, which was responsible for gastrointestinal toxicity.

It was one case, however, where increased specificity came with its own price. COX-1 and COX-2 balance each other in their effects on blood vessels, and with COX-2 selectively inhibited, COX-1 was left unopposed to effect vasoconstriction and thrombosis.

In 2004, Whitehouse Station, N.J.-based Merck & Co. Inc. withdrew Vioxx (rofecoxib) after finding that the drug increased the risk of heart attacks and strokes. Vioxx has not returned to the market, despite a decision by the FDA that it could, with restrictions.

Merck is facing in excess of 10,000 lawsuits about Vioxx and has won four and lost three of the seven suits that have reached verdicts to date. Other COX-2 inhibitors have fared only marginally better. Sales of another COX-2 inhibitor, Bextra (valdecoxib), from New York-based Pfizer Inc., were suspended in 2005; Pfizer's Celebrex (celecoxib) still is available but has had black-box warnings added because it, too, raises the risk of cardiovascular problems.

In the first JAMA paper, researchers from Harvard Medical School and Harvard School of Public Health in Boston investigated the effects of COX-2 inhibitors on arrhythmias and kidney problems.

The researchers conducted a meta-analysis of 114 randomized double-blind trials with a total of more than 100,000 patients. While Vioxx increased the risk of both arrhythmias and kidney problems, other COX-2 inhibitors showed either no change in risk or a marginally decreased risk leading the authors to conclude that "a COX-2 inhibitor class effect was not evident."

In a second JAMA study, scientists from the University of Newcastle in New South Wales, Australia, conducted a meta-analysis of 23 observational studies to compare the risks of heart attacks and other "serious cardiovascular events" with both COX-2 selective and nonselective NSAIDs. Collectively, the studies included roughly 86,000 patients and half a million controls.

The researchers found that Vioxx, as well as indomethacin and the older drug diclofenac, increased the risk of adverse cardiovascular events. Celebrex increased risk at high doses.

Even good old ibuprofen is under something of a cloud: While it did not raise heart attack risk sufficiently to show statistical significance, an accompanying editorial calls the precise statistics of the finding "less than reassuring" and concludes that ibuprofen "probably" also increases cardiovascular risk.

Importantly, the study authors found the cardiovascular risk immediately, not after 18 months as Merck had initially claimed. They also found that Aleve (naproxen), while it did not raise risk, also had no cardioprotective effect, another suggestion initially made by Merck to explain the differences that were observed in a trial comparing Vioxx and Aleve users.

The authors summarized their findings thus: "This review confirms the findings from randomized trials regarding the risk of cardiovascular events with rofecoxib and suggests that celecoxib in commonly used doses may not increase the risk, contradicts claims of a protective effect of naproxen, and raises serious questions about the safety of diclofenac."

Hope springs eternal in drug development, though, and even as the JAMA studies are casting a pall over older drugs, a study slated for publication in the Sept. 26, 2006, issue of the Proceedings of the National Academy of Sciences, now available online, suggested that it is possible to develop NSAIDs that are heart-healthy.

The University of Pennsylvania team deleted the enzyme microsomal prostaglandin E synthase-1 in mice. mPGES-1 converts COX-2's chemical product into PGE2; previous research had suggested that inhibiting mPGES-1 might be able to combat pain without raising blood pressure or the risk of thrombosis; in the current study, the scientists showed that in mice susceptible to atherosclerosis, or hardening of the arteries, knocking out mPGES-1 slowed and stabilized disease progression. The mice appeared to increase the production of the heart-protecting fat prostacyclin.

Though the researchers caution that chemical inhibitors in humans are a far cry from gene knockouts in mice, their sanguine overall reading of the data is that "not only may reduce the likelihood of the hypertension and predisposition to thrombosis associated with COX-2 inhibitors, but also may retain their clinical efficacy and even confer cardiovascular benefit during sustained dosing."