The New Nonsteroidal Antiinflammatory Drugs Are Coming - Are You Ready?

Peter Barland, M.D.

Educational Objectives

Upon completion of this Cyberounds®, the participant should be able to:

Describe the basic metabolism of prostaglandins
Discuss the differences between COX 1 and COX 2
Recognize some of the advantages of inhibiting COX 2 selectively.

Introduction

Until recently, the drugs used for the treatment of chronic inflammatory diseases, such as rheumatoid arthritis (RA), were introduced either through clinical observations or serendipity. Aspirin, for example, was found to be the active ingredient in willow bark and other plant extracts, observed, since antiquity, to have analgesic and antipyretic properties. Likewise, patients with RA were also treated with gold salts because it was believed that RA was caused by an organism similar to tuberculosis, and heavy metals, including gold, were used to treat TB.

As scientists unraveled the mysteries of the cellular and molecular mechanisms in the pathogenesis of chronic immunoinflammatory conditions, aspirin, indomethacin, gold salts, penicillamine and antimalarials, already in use for the treatment of RA, were then tested for their effect on these mechanisms. It was shown that gold salts stabilize lysosomal membranes, penicillamine interferes with T and B cell interactions, antimalarials inhibit antigen processing, aspirin uncouples oxidative phosphorylation and indomethacin alters T cell function.

I call this Cinderella pharmacology -- the clinical scientist has a shoe (effective drug) and proceeds to try it on every promising woman (cellular pathway) that comes along. Unfortunately, none of these observations has been observed in patients in vivo.

COX inhibition and prostaglandin production

A major advance came when Dr. John Vane and his colleagues showed that aspirin and other nonsteroidal antiinflammatory drugs (NSAIDs) inhibited the enzyme, cyclooxygenase (COX), which is the rate limiting enzyme in the production of prostaglandins from arachidonic acid that is found in all cell membranes. For a review of arachidonic acid and prostaglandin metabolism the reader is referred to the Terre Haute Center for Medical Education's Medical Biochemistry Page.

Vane demonstrated that there was a correlation between the cyclooxygenase inhibitory activities of the drugs and their antiinflammatory properties and that there was inhibition of prostaglandin production in vivo in patients receiving pharmacological doses of these drugs.

Prostaglandins are intercellular messengers that are found in high concentrations at sites of chronic inflammation. They are capable of causing vasodilatation, increasing vascular permeability and sensitizing pain receptors. More recently, certain prostaglandins have been shown to induce the synthesis of IL-11, a cytokine that can elicit an acute phase response.

As a consequence of Vane's research, pharmaceutical companies began screening compounds for their ability to inhibit cyclooxygenase in vitro on the assumption that they would thus be able to predict a compound's antiinflammatory action. This approach has spawned the introduction of a large number of NSAIDs. Unfortunately, all of these drugs produced gastritis in many patients -- especially the elderly, those patients with a prior history of peptic ulcer disease and patients on corticosteroids. The gastritis was caused by the inhibition of a gastric COX that regulates mucosal cell production of mucous. (The mucous acts as a barrier to the acid and pepsin present in gastric secretions.)

The NSAIDs also inhibited platelet aggregation meiated by thromboxanes released from platelets as a result of COX activity. In addition, some patients with impaired renal blood flow experience a rapid decrease in glomerular filtration following exposure to NSAIDs. This is believed to result from inhibition of renal COX which produces prostacyclins that augment renal blood flow.

There's not one but two

Between 1989 and 1992, a number of experimental observations suggested that there were two COX enzymes. These two enzymes are referred to as cyclooxygenase 1 (COX 1) and cyclooxygenase 2 (COX 2). While COX 1 and COX 2 are 70% homologous, they are the products of two different genes that reside on different chromosomes.

Cyclooxygenase 1 is a constitutive enzyme present in all cells, while COX 2 is inducible in macrophages, endothelial cells, synoviocytes, chondrocytes, ovarian granulosa cells and smooth muscle cells of the myometrium and arterial wall by a number of modulators. In addition to being inducible, COX 2 appears to be constitutive in the macula densa of the juxtaglomerular apparatus and in certain areas of the central nervous system. Using computerized molecular modeling and X-ray crystallography, it was found that the sites where these enzymes attach to their substrate, arachidonic acid, differed slightly, raising the possibility that specific inhibitors could be constructed.

The traditional NSAIDs were found to bind to the active sites of both COX 1 and COX 2, though some were slightly more specific for either COX 1 or COX 2. With this knowledge, molecular pharmacologists have been able to design drugs that are extremely selective for COX 2, both in vitro and in vivo in humans, using pharmacologically achievable and well tolerated doses.

Do selective COX 2 inhibitors produce ulcers?

The first of these selective COX 2 inhibitors, celocoxib (Celebrex®), has been approved by the FDA for the treatment of osteoarthritis and rheumatoid arthritis. In addition, the drug will probably be used for the treatment of arthritic pain and for the treatment of other forms of chronic arthritis, including psoriatic arthritis and reactive arthritis. A second selective COX 2 inhibitor, refocoxib (Vioxx®), will probably be approved later this year.

The primary advantage of these selective COX 2 inhibitors is the very low incidence of gastric ulcerations observed with doses that provide antiinflammatory and analgesic effects equivalent to those provided by the traditional nonselective COX inhibitors. In double blind prospective endoscopic studies, the prevalence of significant gastric ulcerations, after three and six months, was the same with celocoxib and refocoxib as it was with placebo (approximately 5%). This compares favorably with the 15-20% incidence seen in those patients who received traditional NSAIDs.

The incidence, however, of gastric ulceration in patients on concomitant steroids and patients with a recent history of upper GI bleeding from gastric or duodenal ulcerations has not been studied with the newer agents. Patients on low doses of aspirin for the prevention of atherosclerotic thrombi have been treated with celocoxib and have been shown to have a slight increase in the number of gastric ulcerations compared with placebo treated patients. The incidence of significant upper GI bleeding for several thousand patients treated for over six months with celocoxib and refocoxib has been very low, and statistically insignificant, but these patients have been followed for less than two years in most instances.

Other side effects

Both celocoxib and refocoxib have no effect on platelet aggregation or bleeding time. Therefore, these drugs do not have to be discontinued prior to elective surgery. Effects on renal blood flow and creatinine clearance have not been adequately studied to date. Since COX 2 is constitutively expressed in certain areas of the kidney, it may be that these drugs will cause some of the renal effects seen with the traditional nonselective COX inhibitors. Some patients, when given NSAIDs, develop urticaria, nasal polyposis and asthma (Samter's syndrome). Whether these patients will develop a similar syndrome, when given the selective COX 2 inhibitors, is not known.

The COX 2 enzyme is present in some malignant cells and has been demonstrated in the vascular tissue surrounding adenomatous premalignant polyps of the colon. It is, therefore, reasonable to expect that the selective COX 2 inhibitors will be equally effective in suppressing carcinomatous degeneration of colonic polyps as the traditional nonselective NSAIDs.

Celocoxib

Celocoxib is rapidly and completely absorbed following oral ingestion, reaching a peak blood level after three hours. The absorption is not affected significantly by the presence of food. After ten days, a steady blood level is achieved. The drug is metabolized by the liver through a P450C39 oxidative enzyme to inactive metabolites, which are excreted both in the gut and in the urine. Less than 3% of the active drug is excreted in the urine. The drug's metabolism is slowed by fluconazole, which is metabolized by the same oxidative enzyme. There is no interaction between celocoxib and coumarin or the oral hypoglycemic agents. Because celocoxib contains a sulfa side chain, it should be given with caution to patients with a history of sulfa allergy. The drug is not recommended during the third trimester of pregnancy because it may induce closure of the ductus arteriosus. The effect on early pregnancy has not been studied, but in animals the drug does not appear to be teratogenic. The effects of COX 2 inhibition on the maturation of the corpus luteum and on ovulation remain questions for further study.

Dosage

The dose of celocoxib recommended for osteoarthritis is 100-mg b.i.d. or 200-mg per day. In rheumatoid arthritis, the recommended dose is 100 to 200-mg b.i.d. Higher doses are not accompanied by greater antiinflammatory efficacy. The cost of these drugs is comparable to the cost of the newer NSAIDs. The most common side effect associated with celocoxib and refocoxib, so far, has been the development of dyspepsia. There is no correlation between the development of dyspepsia and the incidence of endoscopically observed gastric ulceration or significant blood loss.

Conclusion

While it is likely that several more selective inhibitors of COX 2 will be developed, some of which may be targeted to enzymes in specific cells, it is still apparent that the first generation of these drugs represents a significant step forward in antiinflammatory therapy.

References

1. Vane JR, Bakhle YS, Botting RM. Cyclooxygenases 1 and 2. Annu. Rev. Pharmacol. Toxicol. 1998,38:97-120.

2. Hawkey, CJ. COX-2 Inhibitors. Lancet 1999,353:307-314.

created 2/28/99; reviewed 3/2/99; modified 3/10/99; end date 3/10/01.