New findings have implications for chronic use of COX-2 Inhibitors in pre-menopausal women

Heart disease is less pronounced in women than in men as humans age, but this difference narrows after menopause. Some studies have shown that estrogen slows heart disease in mouse models, but the mechanism is largely unknown.

Now scientists from the University of Pennsylvania School of Medicine show for the first time that in female mice protection from hardening of the arteries purported to come from higher levels of estrogen acts predominately through cyclooxygenase (COX)-2.

Garret FitzGerald, MD, Chairman of the Department of Pharmacology, and colleagues found that estrogen binds to a cell receptor that activates COX-2, which in turn ramps up the production of the prostacyclin PGI2. This biochemical provides protective benefits both by inhibiting platelet activation and by reducing oxidative stress in the circulatory system by increasing expression of an antioxidant enzyme. Earlier experiments in mice by the FitzGerald lab and others have shown that platelet activation and oxidative stress can independently hasten hardening of the arteries. The most recent findings appear in the November 18 issue of Science.

This study shows for the first time that prostacyclin can modulate gender differences in atherosclerosis and that estrogen increases prostacyclin in an animal model. In addition, this research also demonstrates that estrogen upregulates COX-2-dependent prostacyclin and that prostacyclin contributes to the atheroprotective effect of estrogen.

Disabling the prostacyclin receptor in female mice whose ovaries have been removed took away the atheroprotective effect of estrogen. By taking away the ovaries, the investigators can pinpoint the direct effects of estrogen. In mice treated this way, estrogen, as expected, slows hardening of the arteries. Taking away the receptor for PGI2 in those animals largely undermines this protection, which was based on measuring the extent of atherosclerosis. Increased platelet activation was demonstrated by increased levels of the chemical thromboxane, and increased oxidative stress was measured by increases of isoprostanes in the urine.

Because of the direct links among estrogen, COX-2 pathways, and atheroprotection in female mice, this study raises concern about the use of COX-2 inhibitors in premenopausal women. These studies also raise the possibility of an interaction between hormone replacement therapy and drugs which inhibit COX-2, including traditional NSAIDs. Of particular concern for selective inhibitors of COX-2 would be for patients with juvenile arthritis, which involves mostly long-term drug use in young, premenopausal women, says FitzGerald, also Director of the Institute for Translational Medicine and Therapeutics.

Although researchers extrapolate results from mice to humans with extreme caution, recent studies linking COX-2 inhibitors with cardiovascular risk have focused attention on the possibility of slowly evolving cardiovascular risk during chronic treatment with selective COX-2 inhibitors. This study provides insight into how this risk might occur and identifies potential biomarkers of this evolving risk. The work was funded by the National Heart, Lung, and Blood Institute of the National Institutes of Health. Other Penn researchers on this paper were Karine M. Egan, John A. Lawson, Susanne Fries, Daniel J. Rader, and Emer M. Smyth, along with Beverley Koller, University of North Carolina.