Introduction

A COX-2 inhibitor remains the preferred option for patients who need antiinflammatory therapy and are at risk for gastrointestinal (GI) complications, Lee S. Simon, MD, and Vibeke Strand, MD, stated in a commentary published in the November issue of the Cleveland Clinic Journal of Medicine.¹ Before introduction of this drug class, an estimated 7500 to 16,500 excess deaths occurred annually due to toxicity of nonselective nonsteroidal anti-inflammatory drugs (NSAIDs).^2,3

"What has been forgotten is the benefit of COX-2 inhibitors," Dr. Simon, a rheumatologist and associate clinical professor of medicine at Harvard Medical School in Boston, Massachusetts, told CIAOMed. "People were dying because of this [NSAID toxicity] problem," he emphasized, adding that the number of deaths fell by about two-thirds with the introduction of COX-2s.²

GI Complications: a More Restrictive Role for COX-2s?

Yet, other clinicians are now taking a more restrictive view of the role of COX-2 inhibitors for patients at risk for GI complications. According to Leslie Crofford, MD, a professor of internal medicine and chief of rheumatology at the University of Kentucky College of Medicine in Lexington, Kentucky, her first choice for most patients at GI risk who are not taking aspirin would be a traditional NSAID plus a proton pump inhibitor (PPI). However, Dr. Crofford, a member of the editorial board of CIAOMed, noted that she would consider using celecoxib in "healthy people with GI intolerance to traditional NSAIDs (due to GI symptoms) or at risk due to anticoagulant use."

Claire Bombardier, MD, a CIAOMed board member and professor of medicine at the University of Toronto, said that based on available data, she now restricts COX-2s to patients with a previous history of a bleed or more than 1 of the remaining 5 risk factors for GI complications (Table 1). Previously, a single risk factor would have been sufficient to consider choosing a COX-2, she said.

Table 1.

Evaluating GI Risk

Six risk factors for GI complications¹

• Age >60 years
• History of a complication with or without NSAID use (eg, history of a GI bleed)
• Concomitant glucocorticoid use
• Use of >1 NSAID (eg, aspirin use plus an NSAID)
• Higher doses of NSAIDs
• Severe illness

One risk factor for bleeding

• Concomitant use of an anticoagulant

Patient reluctance also limits her use of the COX-2 class, but Dr. Bombardier said that when a COX-2 inhibitor is indicated, celecoxib is the first choice because "for the moment, [it] has no signal of any strength" for elevated CV risk.

Given that the APPROVe trial (Adenomatous Polyp Prevention on Vioxx) showed rofecoxib's elevated CV risk at 18 months, Dr. Bombardier said that she would await review of the results of long-term celecoxib safety data from ongoing trials of patients with colon polyps and Alzheimer's disease before making an overall determination. Some of these studies are not sponsored by Pfizer, she noted, and one is "very similar to APPROVe." (Click here for more details about trials of celecoxib sponsored by the US National Cancer Institute. http://www.cancer.gov/Templates/doc.aspx?viewid=ae3edaf3-74b9-4aa7-9a63-e979e69a4ab1)

The Fate of Valdecoxib

All the rheumatologists interviewed expressed reservations about using valdecoxib (Bextra^ÂR). "There is a cloud over Bextra given the CABG study," Dr. Bombardier said.

Pfizer notified health care professionals in October 2004 that valdecoxib was associated with an increase in CV events in patients having undergone recent coronary artery bypass graft surgery. Patients received valdecoxib alone or in conjunction with parecoxib (Dynastat^ÂR), the intravenous prodrug of valdecoxib not approved in the US.⁴

These findings contrasted with a pooled analysis published in the summer of 2004, which revealed no increased risk of CV thrombotic events with valdecoxib compared to traditional NSAIDs or placebo in OA and RA patients during what investigators termed "short and intermediate-term use" (6 to 52 weeks).⁵

"There are not enough data" about valdecoxib, Steven B. Abramson, MD, rheumatologist and professor in the department of rheumatology/medicine at New York University School of Medicine, New York, told CIAOMed. "I would not recommend valdecoxib in a population at CV risk. I also would be careful about blood pressure monitoring and watch for a dose-dependent rise in blood pressure."

Dr. Bombardier said that she considers prescribing the agent only in patients who meet her more restrictive criteria for COX-2 therapy and who fail to respond adequately to celecoxib. Even then, she said, "I might use a traditional NSAID plus a PPI" before trying valdecoxib.

PPIs and the Lower GI Tract

A PPI plus a traditional NSAID is an option for patients who cannot tolerate a COX-2 because of, for example, sulfa allergy. "It's not as safe," Dr. Simon told CIAOMed, but it is the best available option. One weakness of the PPI-plus-traditional-NSAID regimen is that PPIs do not protect the lower GI tract. "Nonselective NSAIDs can cause mucosal ulcers throughout the GI tract," according to the Strand/Simon commentary published in the Cleveland Clinic Journal of Medicine.¹ Dr. Simon remarked that this is one reason why he prefers celecoxib to that combination.

Dr. Bombardier agreed with that assessment "that PPIs do not protect the lower GI tract" but said that she would still opt for an NSAID plus a PPI before using celecoxib in patients with only one GI risk factor and in those who fail or refuse therapy with celecoxib.

Misoprostol

Another option for GI protection, misoprostol (Cytotec^ÂR), has the disadvantage of being "hard to use," Dr. Simon said. Dr. Abramson concurred, noting that many patients cannot tolerate the agent due to diarrhea and other side effects, and that it can require 3- to 4 times daily dosing. "Arthrotec^ÂR (a combination containing misoprostol 200 mcg plus diclofenac [Voltaren^ÂR] 50 mg or 75 mg) is a good drug," Dr. Abramson said. "However, the medication contains about 50% of the amount of misoprostol needed to protect against NSAID-induced ulcers, as was demonstrated in the MUCOSA trial. Therefore, we are not certain that the lower dose of misoprostol is gastroprotective."

Dr. Bombardier said she had stopped prescribing the misoprostol/diclofenac combination because many patients cannot tolerate the diarrhea associated with its use, but indicated that she is prescribing more of it now. "Most of my patients want to return to older NSAIDs," Dr. Bombardier noted, adding that she turns to the combination of misoprostol/diclofenac or recommends an older NSAID plus a PPI.

COX-2s Plus Aspirin: A Question of GI Protection

Dr. Crofford said she would not use celecoxib in patients taking aspirin partly because "you lose the GI benefit" if aspirin is on board. This finding comes from the Celecoxib Long-term Arthritis Safety Study (CLASS), in which celecoxib reduced rates of upper GI complications combined with symptomatic ulcers in the overall population, but not among those also taking low-dose aspirin for cardioprotection.⁶

"It's not worth giving a COX-2" in such cases, Dr. Bombardier concurred in a separate interview. "The cost is higher and aspirin may eliminate the benefit." It is possible that there is still some GI benefit when COX-2s are used with aspirin, but "we're not sure of it," she said.

Given this caveat to the use COX-2 inhibitors, how does Dr. Crofford manage patients with GI risk who need aspirin for cardioprotection? "They all go on a PPI" or misoprostol and a nonselective NSAID such as naproxen (Aleve^ÂR, Anaprox^ÂR, Naprelan^ÂR, Naprosyn^ÂR) or diclofenac (Voltaren^ÂR), she said.

If a person at elevated CV risk had to stop aspirin due to ulceration, some physicians might use celecoxib, Dr. Crofford noted, adding, "I would use acetaminophen in that patient if the aspirin was prescribed for cardioprotection." In patients who must receive aspirin or clopidogrel (Plavix^ÂR) due to high CV risk, Dr. Bombardier said she prescribes a traditional NSAID with a PPI.

Dr. Simon conceded that aspirin cancels out most of the benefit of COX-2s but said, "there is less [GI] risk than with the nonselective NSAIDs."

As to cardiovascular risk, Drs. Simon and Abramson agreed during interviews at the end of October that evidence did not demonstrate increased thromboembolic risk with celecoxib.¹ Additionally, Dr. Simon noted, "nonselective NSAIDs also have a signal for CV events." A recent observational study by the US Food and Drug Administration and Kaiser Permanente reported that naproxen and indomethacin each were associated with increased risk of thromboembolic cardiovascular events.⁷

Dr. Crofford is less convinced of the cardiovascular safety of celecoxib, citing her published case series of ischemic complications in celecoxib patients at high risk for thrombosis due to factors such as Raynaud's phenomenon and other connective tissue diseases.⁸ She did concede, however, that there does not appear to be a risk for cardiovascular complications in lower-risk subjects.

Overall, Dr. Bombardier said that she assesses both GI and CV risk in all candidates for COX-2s, a practice that predates the withdrawal of rofecoxib. That assessment is based on the US Preventive Task Force Guidelines from January 2002 for CV risk assessment in persons with no history of myocardial infarction, stroke, or coronary artery bypass graft surgery.⁹ "We practice defensively. There were signals [of CV risk] before the APPROVe study," she said.

The third and final installment focuses on the future of COX-2s, including those in Phase III of development.

References

1. Simon LS, Strand V. A world without Vioxx: to COX-2 or not to COX-2? Cleve Clin J Med. 2004;71:849-856.

2. Simon LS. Stemming the tide of rhetoric: the fate of COX-2s after Vioxx. Advanced Studies in Medicine. 2004;4:522-523.

3. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. N Engl J Med. 1999;340:1888-1899.

4. Pfizer Provides Information to Healthcare Professionals About Its Cox-2 Medicine Bextra^ÂR (Valdecoxib). Pfizer news release. October 15, 2004. Available at: http://www.pfizer.com/are/news_releases/2004pr/mn_2004_1015.html. Accessed November 1, 2004.
5. White WB, Strand V, Roberts R, Whelton A. Effects of the cyclooxygenase-2 specific inhibitor valdecoxib versus nonsteroidal antiinflammatory agents and placebo on cardiovascular thrombotic events in patients with arthritis. Am J Ther . 2004;11:244-250.

6. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS study: a randomized controlled trial. JAMA. 2000;1247-1255.

7. American College of Rheumatology. Cardiovascular complications related to COX-2 Inhibitors. ACR Hotline. September 30, 2004. Available at: http://www.rheumatology.org/publications/hotline/0904chfvioxx.asp. Accessed November 10, 2004.

8. Crofford LJ, Oates JC, McCune WJ, et al. Thrombosis in patients with connective tissue diseases treated with specific cyclooxygenase 2 inhibitors. A report of four cases. Arthritis Rheum. 2000;43:1891-1896.

9. U.S. Preventive Services Task Force. Aspirin for the Primary Prevention of Cardiovascular Events: Recommendations and Rationale. January 2002. Agency for Healthcare Research and Quality, Rockville, MD. Available at: http://www.ahrq.gov/clinic/3rduspstf/aspirin/asprr.htm. Accessed December 1, 2004.