Saturday, December 18, 2004

COX-2 Inhibitors and Cardiovascular Toxicity

A great deal of media attention was paid to Merck's pulling of Vioxx (rofecoxib) from the market, and now on the heels of that Pfizer announced some danger in high doses of Celebrex (celecoxib) the COX-2 inhibitors are getting a lot of media attention. While I think COX-2 inhibitors deserve some more examination, I also think the media is getting a tad hysterical about the whole thing.

My experiences in toxicology at this point are solely with liver and immune system toxicology, but we've discussed the COX-2 inhibitors this semester so I thought I might put out the ideas I've learned about them.

First, it is important to understand what COX enzymes are and why they're important in our biology. Cyclooxygenase (COX) comes in two forms, COX-1 which is expressed in nearly all cells continuously, and COX-2 which is inducible in response to injury or invading germs (it is also mostly found only in inflammatory cells of the immune system). COX is an enzyme that converts a fatty acid (arachidonic acid) from the cell membrane into compounds called prostaglandins. Prostaglandins come in a many forms, and depending on the mix of enzymes in a cell different ones will be produced, mediating differing effects. Prostagladins can be proinflammatory, antiinflammatory, and probably other effects that I didn't take the time to look up. The slowest step in the making of prostaglandins is the step that COX enzymes catalyze. Once made prostaglandins can also be made into thromboxanes, hormones important to blood clotting.

Pfizer and Merck were both involved in clinical studies where they were trying to find their COX-2 drugs may prevent or slow the growth of colorectal cancer (which causes 10% of all cancer deaths). It was a reasonable hypothesis to pursue, as some antiinflammatory drugs are useful in the treatment colorectal cancer. Researchers have established that a drug called sulindac, a Non-Steroidal Antiinflammatory Drug (NSAID) that inhibits both COX-1 and COX-2, is helpful in treating colorectal cancer. Often the undesirable effects of NSAIDs like aspirin are blamed on their inhibition of COX-1, as normal prostaglandin production in the stomach is necessary for making the mucous that protects the lining of the stomach from self-digesting. Researchers thought that it might be possible to get the beneficial effects of COX inhibition in the treatment of colorectal cancer without the undesirable side effects like stomach ulcers.

What researchers at the National Cancer Institute observed was that those people who took the highest doses, 400mg to 800mg, of celecoxib (Celebrex) had 2.5 times the risk of having a fatal or non-fatal cardiovascular event (things like heart attacks or strokes) when compared to those taking a placebo. This was in 1 of 2 studies Pfizer had going. The other study only had doses up to 400mg and revealed no increased risk.

It looks to me like this is a case of "Dosis facit venenum," or "the dose makes the poison." All this hand wringing and fear-mongering in the media, and we're talking about a small fraction of people. The recommended dose for celecoxib is 100mg-200mg for osteoarthritis and 200mg-400mg for rheumatoid arthritis. Osteoarthritis accounts for something like 80% of all arthritis, so most of the people taking celecoxib would not be likely to be at risk for "fatal or non-fatal cardiovascular event." What really bothers me is these drugs are being painted as risky, when they pose little risk for most people and offer excellent therapeutic help to people suffering with arthritis pain!

Being in pharmacology (not pharmacy) I'm of course interested in how this observed effect works mechanistically. In my pharmacology class we did get a plausible hypothesis, but it has yet to be proven.

Our bodies have many different kinds of cells, and they often respond differently to a given stimulus like a drug. I alluded to the idea that COX can be important to many processes in the body. It's not accurate to look at prostaglandins as all good or all bad, they must function in a balancing act. One of the things that they help to balance is the formation of clots in the body. COX-2 helps to produce a prostaglandin called prostacyclin, which normally dilates the blood vessel and inhibits clotting.

If we selectively inhibit COX-2 we'll reduce the amount of prostacyclin produced, and thus decrease blood vessel diameter and increase clotting. At the same time COX-1 is still functioning and producing prostaglandins and thromboxanes that can increase clotting. By inhibiting solely COX-2 and not all COX forms we may create a situation where the clotting stimulus is on, but the anticlotting signal is switched off, leading to clots that could float around until they reach a constricted point in the blood vessels, and then stick, reducing blood flow. Boom, stroke or heart attack.

The research that generated this hypothesis was done in mice, so it can't be said for certain that it holds true in humans. It does give a very serious hint at it though.


Post a Comment

<< Home