COX-2 Inhibitors: Where Are We Now and Where Are We Going?

Nonsteroidal anti-inflammatory drugs (NSAIDs) have long been a mainstay in the management of pain and inflammatory conditions, such as osteoarthritis and rheumatoid arthritis. A major limitation, however, has been the high incidence of gastrointestinal (GI) toxicity, including major hemorrhage. During the past 20 years, research began to show that the anti-inflammatory and analgesic effects of the NSAIDs are mediated by the inhibition of one form of cyclooxygenase (COX), the enzyme responsible for the conversion of arachidonic acid into prostaglandins (COX-2), while the GI toxicity of these agents is mediated by the inhibition of another form of the enzyme (COX-1). Thus, compounds were synthesized that could selectively inhibit COX-2 and thereby provide the benefits of traditional NSAIDs without the associated GI side effects. The earliest of these agents, celecoxib (Celebrex) and rofecoxib (Vioxx), were approved by the FDA in 1998 and 1999, respectively.

Soon after the introduction of the COX-2 inhibitors, 2 large-scale trials, the Celecoxib Long-Term Arthritis Safety Study (CLASS)¹ and the Vioxx Gastrointestinal Outcomes Research (VIGOR)² study, were begun to demonstrate the improved GI toxicity profile of these new agents. The VIGOR study showed a higher incidence of myocardial infarction in the rofecoxib group, compared with the control group, which was treated with naproxen. The significance of this finding was unclear and was credited by some to a cardioprotective effect of naproxen.

Last year, however, the Adenomatous Polyp Prevention on Vioxx (APPROVe) trial, designed to show that COX-2 inhibitors could prevent recurrent colon polyps, was stopped early due to an increased risk of cardiovascular (CV) events.³ These data led to the withdrawal of rofecoxib from the market by Merck in September 2004. Simultaneously, 2 trials investigating rofecoxib in cancer patients were discontinued. The interim data from one of these trials, the Vioxx in Colorectal Therapy, definition of Optimal Regime (VICTOR) trial, showed an increased risk of thrombotic CV events.⁴ A similar trial, the Adenoma Prevention with Celecoxib (APC) study, also was stopped early when a retrospective review revealed an increased incidence of CV events.⁵ Concurrently, a trial investigating pain control in coronary artery bypass graft (CABG) patients using valdecoxib (Bextra) and its intravenous prodrug parecoxib also demonstrated an increased risk of CV events.⁶

In light of these results, the FDA assembled an advisory panel in February 2005 to review the available data regarding the safety of the COX-2 inhibitors available in the United States (ie, rofecoxib, celecoxib, and valdecoxib), as well as naproxen and other nonselective NSAIDs. A large body of data was presented at the conference, and a summary of the panel's findings, as they relate to CV risk, is found in the Table.^4,7

On April 7, 2005, the FDA announced a series of changes to the NSAID class.⁸ The overall risk-benefit profile of valdecoxib was deemed unacceptable. Agency officials requested that Pfizer remove it from the market?leaving celecoxib as the only COX-2 inhibitor available in the United States. This decision was based on documented risk in CABG patients, lack of long-term safety data, reports of serious skin reactions, and lack of a demonstrated advantage over other NSAIDs.

The FDA also announced that all prescription NSAIDs, including celecoxib, will carry a black-box warning regarding the risk of CV and GI events, as well as a contraindication in patients who have recently undergone CABG surgery. Agency officials announced that the data do not suggest an increased risk associated with the short-term use of low doses of the available OTC NSAIDs, although this is the first time that a product with a boxed warning on the prescription version would be available over the counter.⁹

Complicating the issue of COX-2 inhibitor safety is the fact that much of the data are retrospective or drawn from observational studies. The controlled trials available were often not intended, or powered, to detect CV events. Additionally, many of the controlled trials have utilized active comparison arms, and therefore few placebo-controlled trial results are available. For rofecoxib,^2,3,4,9,10 celecoxib,^4,5,9 and valdecoxib/parecoxib,^4,6,9,11 there are data that show a signal for increased CV risk, as well as data that show no increased risk. These data include unpublished data from the Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT), the Prevention of Spontaneous Adenomatous Polyps (PreSAP) trial, the VICTOR trial, and 2 long-term Alzheimer's disease prevention trials involving rofecoxib.

Similarly, there is little information available directly addressing the CV profile of the nonselective NSAIDs. FDA officials have recognized the limitations of the available data and have stated that further study is necessary to adequately characterize the safety profile of both selective and nonselective NSAIDs.

Based on this information, preliminary recommendations can be made while waiting for more definitive guidelines. First, it is important to note that increased CV risk may be a class effect of all NSAIDs (except aspirin), and they should be used only in patients where there is a clear indication and the benefits of therapy outweigh the potential risks. In patients deemed appropriate for NSAID therapy, it may be warranted to begin with an agent that seems to carry less risk, such as naproxen.^4,7 If a traditional NSAID is used, an acid-suppressing agent (eg, a histamine²-receptor antagonist or a proton pump inhibitor) could be added to improve GI safety.

COX-2 inhibitors have never been shown to be more effective than traditional NSAIDs in the treatment of pain. Therefore, they should be reserved for those patients who require NSAID therapy but are intolerant of or unresponsive to traditional agents, and these drugs should be avoided altogether in high-risk patients.

Additionally, it may be recommended that all NSAIDs be used at the lowest effective dose for the shortest possible duration. In patients in whom risk is deemed to be prohibitive, other agents such as tramadol or acetaminophen (with or without an opioid) may be utilized. Perhaps most important, the decision whether or not to use an NSAID must be individualized, based on each patient's particular risk factors, likely benefit, and informed choice.

The future of NSAIDs?and of the COX-2 inhibitors in particular?remains far from certain. More studies are needed to fully elucidate the CV risk of both drug classes, as well as of the individual agents. In the meantime, the FDA has chosen to allow celecoxib to remain available on the US market. In light of the current ambiguity, it is important that pharmacists stay abreast of developments regarding this important issue and take an active role in promoting the safe and effective use of these medications.

Dr. Wells is a clinical pharmacist with Cooper Green Hospital in Birmingham, Ala.

For a list of references, send a stamped, self-addressed envelope to: References Department, Attn. A. Stahl, Pharmacy Times, 241 Forsgate Drive, Jamesburg, NJ 08831; or send an e-mail request to: astahl@ascendmedia.com.