• 6/2/2005
  • Bethesda, MD
  • Adriana Albini, Douglas M. Noonan
  • Journal of the National Cancer Institute, Vol. 97, No. 11, 859-860, June 1, 2005

COX-2 inhibition has been a main target for new nonsteroidal anti-inflammatory drug (NSAID) development. The effectiveness of two COX-2–specific drugs, Vioxx (rofecoxib) from Merck and Celebrex (celecoxib) from Pfizer, in chronic inflammatory diseases such as arthritis, was so highly promoted that they were among the most prescribed drugs in the category; experimental studies also showed promise in other diseases, including cancer (1). Both drugs have recently met criticism. Vioxx was voluntarily pulled off the market in September 2004 by Merck and recently returned with cautions; a threefold higher relative risk of mortality from cardiovascular complications as compared with older non–COX-2–selective NSAIDs was reported (2,3). After optimistic publicity by Pfizer but controversial clinical trial results, the consumer advocate group Public Citizen requested that Celebrex and its close cousin Bextra (valdecoxib) be withdrawn. We suggest that clues to the origin of the cardiovascular side effects come from the area of cancer research, in which COX-2 inhibitors are being tested for chemoprevention (1).

Although similar, the two COX enzymes show functional differences. Aspirin, one of the first and most widely used COX inhibitors, largely targets COX-1, whose constitutive enzymatic activity is linked to renal function, gastric mucosal maintenance, stimulation of platelet aggregation, and vasoconstriction. Although occasionally causing gastrointestinal bleeding, aspirin is also associated with a reduced risk of heart attack and stroke; chronic low-dose aspirin use has become a preventive practice for cardiovascular and possibly oncologic diseases (1). In contrast, COX-2 is associated with inflammatory events, and inhibitors of this enzyme, in theory, provide beneficial anti-inflammatory effects without undesired gastrointestinal effects. COX-2 also mediates vasodilation and inhibits platelet aggregation, opposing the effects of COX-1; thus, it can be speculated that selective inhibition of COX-2 would be prothrombotic (4). Many drugs that inhibit both COX enzymes with different efficiencies are currently available and most have been found to be relatively safe after extensive use on the market. However, the Alzheimer disease trial that put Celebrex in question also indicated similar risks for naproxen, an older COX-2–preferential NSAID. So why have the more specific COX-2 inhibitors been associated with increased risk for thrombotic cardiovascular events while other NSAIDs have not?

An increasingly large body of evidence from cancer research supports the association between expression of COX-2 and vascular endothelial growth factor (VEGF), a key endothelial survival factor. Recent studies indicate that COX-2 is an upstream stimulator of VEGF production and is beneficial for endothelial cells, possibly by helping surrounding stromal tissues provide VEGF (5). Given the relationship between COX-2 and endothelial cell survival, COX-2 inhibitors may reduce the levels of VEGF available to endothelial cells, making them prone to apoptotic cell death. Survival factors could be critical in vessel areas exposed to flow stress and previous damage. After endothelial cell death, the corresponding “hole” of exposed basement membrane is rapidly plugged by platelet aggregation and thrombosis. Patients treated with COX-2 inhibitors may have increased endothelial cell death in critical areas, such as preexisting atherosclerotic plaques, due to reduced VEGF production. In these patients, thrombosis that is favored by COX-2 repression, if not balanced by a simultaneous COX-1 inhibition as an anticoagulant, could occur and rapidly propagate further endothelial cell death. In patients treated with older NSAIDs that also block COX-1, thrombosis might be avoided thanks to reduced coagulation.

This scenario of endothelial cell death susceptibility and risk for subsequent thrombotic events may not be limited to COX-2 inhibition. Treatment with Avastin (bevacizumab), a humanized anti-VEGF antibody recently approved for use in colon cancer therapy, is also associated with an increased risk of cardiovascular events (leading the manufacturer Genentech to recently modify its warning labels), further implicating VEGF in cardiovascular homeostasis.

Although specific COX-2 or VEGF inhibitors with little or no effect on COX-1 may endanger cardiovascular homeostasis, we should be careful not to abandon an entire sector of promising pharmaceuticals for what may be severe, but relatively easily counteracted, side effects. Because the advantages of selective COX-2 inhibitors would be lost if accompanied by COX-1 blockers, possible approaches are the administration of anticoagulants of a different nature, e.g., reviparin, a low-molecular-weight heparin found to be beneficial when given soon after an infarct (6), or of antiapoptotic chemoprevention agents, such as N-acetyl cysteine (7). We suggest that COX inhibitors can be rescued from the “waste bin,” where they risk ending up: Combining COX-2–specific inhibitors with anticoagulant, antithrombotic agents that lack gastric side effects might compensate for the increased risk of potentially fatal cardiovascular complications.

Authors’ Affiliations:
Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy (AA); Università degli studi dell’Insubria, Varese, Italy (DMN)

References:
(1) Brown JR, DuBois RN. COX-2: a molecular target for colorectal cancer prevention. J Clin Oncol 2005;12:2840–55.

(2) Graham DJ, Campen D, Hui R, Spence M, Cheetham C, Levy G, et al. Risk of acute myocardial infarction and sudden cardiac death in patients treated with cyclo-oxygenase 2 selective and non-selective non-steroidal anti-inflammatory drugs: nested case-control study. Lancet 2005;365:475–81.[CrossRef][ISI][Medline]

(3) Juni P, Nartey L, Reichenbach S, Sterchi R, Dieppe PA, Egger M. Risk of cardiovascular events and rofecoxib: cumulative meta-analysis. Lancet 2004;364:2021–9.[CrossRef][ISI][Medline]

(4) Taking stock of coxibs. Drug Ther Bull 2005;43:1–6.[Medline]

(5) Gately S, Li WW. Multiple roles of COX-2 in tumor angiogenesis: a target for antiangiogenic therapy. Semin Oncol 2004;31:2–11.

(6) Yusuf S, Mehta SR, Xie C, Ahmed RJ, Xavier D, Pais P, et al. Effects of reviparin, a low-molecular-weight heparin, on mortality, reinfarction, and strokes in patients with acute myocardial infarction presenting with ST-segment elevation. JAMA 2005;293:427–35.[Abstract/Free Full Text]

(7) Albini A, Morini M, D’Agostini F, Ferrari N, Campelli F, Arena G, et al. Inhibition of angiogenesis-driven Kaposi’s sarcoma tumor growth in nude mice by oral N-acetylcysteine. Cancer Res 2001;61:8171–8.[Abstract/Free Full Text]