Clinical Considerations: COX-2-Selective Inhibitors

Some patients with asthma experience respiratory reactions after ingesting aspirin or other NSAIDs. With the introduction of COX-2-selective inhibitors, the question was raised as to whether patients with aspirin-sensitive respiratory disease (ASRD) would tolerate these drugs. In a small double-blind, crossover study, 12 patients with ASRD received either an increasing dose of rofecoxib (1.5 to 25.0 mg over 5 days) or a placebo.(46) Patients then crossed over to the complementary arm. None of the patients receiving rofecoxib had dyspnea or decreases of >20% in forced expiratory volumes (FEV1). In a randomized, doubleblind, placebo-controlled study of 60 patients with confirmed ASRD, none of the patients receiving rofecoxib 12.5 or 25.0 mg over 48 hours had symptoms, declines in FEV1, or changes in nasal examination findings.(47) A study of 17 patients with asthma and aspirin intolerance did not have bronchoconstriction or extrapulmonary reactions after a graded challenge with celecoxib (10, 30, 100, and 200 mg).(48) Although based on these studies selective COX-2 inhibitors appear to be tolerated by patients with ASRD, product labeling for all available agents lists this as a contraindication to therapy. It should be emphasized that these observations apply only to aspirin-sensitive respiratory reactions, not urticaria or angioedema; these processes are likely mediated through different pathobiologic mechanisms. It is also important to note that urticaria, angioedema, and anaphylaxis have been reported with the currently available COX-2-selective agents. Up to one third of patients with NSAID-induced urticaria and angioedema have had reactions when challenged with COX-2-selective agents.(49-52)

Sulfonamide hypersensitivity

The presence of a sulfonamide group in the celecoxib molecule prompted concern that patients with sensitivity to sulfonamides may be reactive to celecoxib. Patients with hypersensitivity to sulfonamides were excluded from the largest outcomes study of celecoxib safety.(7) A meta-analysis of 14 double-blind trials of celecoxib in patients with arthritis found that the overall incidence of allergic reactions with celecoxib was not statistically different from that seen with placebo or active comparators. Although patients with a history of sulfonamide hypersensitivity had a 3- to 6- fold higher incidence of dermatologic reactions, the trend was consistent in all 3 groups (placebo, NSAIDs, and celecoxib).(53) The nature and description of these dermatologic reactions is not reported, making interpretation of these results difficult. Prospective trials are needed to confirm these findings. Pending these studies, celecoxib labeling contraindicates its use in patients with known allergic reactions to sulfonamides. Rofecoxib does not possess a sulfonamide moiety, and patients with sulfonamide sensitivity were not excluded from rofecoxib clinical trials. Of note, both valdecoxib and parecoxib have sulfonamide moieties in their structures, but patients with sulfonamide sensitivity have not been excluded from clinical trials with these agents. Whether dermatologic reactions will be increased in incidence has not yet been reported.

Further discussion

A deeper understanding of the physiologic roles of COX-1 and COX-2 will clarify the clinical implications of selective COX-2 inhibition. COX- 2 has a complex and uncharacterized role in normal physiology.(54) Experience with NSAIDs has verified the tolerability of COX-2 inhibition in the context of these nonselective drugs. It must be acknowledged, however, that biologic effects of prostaglandin production by unopposed COX-1 may differ from that of combined inhibition.(55) For example, COX-2-selective inhibitors decrease levels of the vasodilatory PGI2 while COX-1-derived platelet TXA2 production is unaffected. COX- 2-selective inhibitors, therefore, may possess less antithrombotic and cardioprotective properties than nonselective NSAIDs. Animal studies suggest a role for COX-2-derived prostacyclin in coronary circulation.(56) Another area deserving further investigation is the apparent increased risk of cardiovascular events that occur in RA patients and the implications of use of coxibs in this patient population.

In the kidney, both COX-1 and COX-2 are constitutively expressed, and it is unclear which enzyme is predominantly responsible for NSAID-induced renal toxicity. Nephrotoxicity induced by conventional nonselective NSAIDs is most commonly associated with reduced glomerular filtration rate (GFR); COX-2 appears to be most important in sodium retention without a decrease in GFR.(57) Published clinical experience shows that the incidence of renal adverse events with COX-2 inhibitors is similar to that of NSAIDs.(6,7) In patients with a high risk of renal side effects, COX-2-selective inhibitors should be approached with the same caution as other NSAIDs.(58)

The advent of new COX-2-selective agents and their ensuing clinical experience may shed greater understanding on these issues, leading to optimal management of COX-2-mediated inflammatory diseases.

Etoricoxib, an investigational COX-2-selective inhibitor, demonstrates a high degree of COX-2 specificity (106-fold in ex vivo human blood assays) and has a lower potency of COX-1 inhibition than other reported agents (Figure 2).(5)

Figure 2. Inhibition of COX-1 by COX-2-selective agents as determined with a sensitive microsomal assay. (Adapted from Riendeau et al with permission.)(5)

Parecoxib, the first COX-2-selective inhibitor formulated for parenteral use (intravenous or intramuscular), compares favorably with ketorolac. Parecoxib is not biologically active; it is a water-soluble prodrug that is rapidly hydrolyzed to valdecoxib.

Bingham CO III. Development and clinical application of COX-2-selective inhibitors for the treatment of osteoarthritis and rheumatoid arthritis. Cleve Clin J Med 2002;69:SI5-12.