CHICAGO — Prolonged treatment with the COX-2 inhibitor etoricoxib carried no increased cardiovascular risk, compared with the traditional NSAID diclofenac in a pooled analysis of nearly 35,000 arthritis patients, reported Dr. Christopher Cannon at the annual scientific sessions of the American Heart Association.
For arthritis patients “who need to take one of these types of agents long term, either one will have the same risk profile for thrombotic events such as heart attack or stroke,” he said at a press conference.
The Multinational Etoricoxib and Diclofenac Arthritis Long-Term (MEDAL) program is a prespecified pooled analysis of three studies. The largest was MEDAL, performed in 23,504 patients during June 2002-May 2006 at 1,380 sites in 46 countries. This study randomized rheumatoid arthritis (RA) patients to treatment with 90 mg etoricoxib once a day or 75 mg diclofenac twice a day. Osteoarthritis (OA) patients were randomized to receive either 75 mg diclofenac twice daily or one of two dosages of etoricoxib, 90 mg or 60 mg.
The other two studies in the analysis were the Etoricoxib vs. Diclofenac Sodium Gastrointestinal Tolerability and Effectiveness (EDGE) trial, in which 7,111 patients received 90 mg etoricoxib once daily or 50 mg diclofenac three times daily, and EDGE II, in which 4,086 patients were treated with 90 mg etoricoxib once daily or 75 mg diclofenac twice daily.
In all three studies, patients were followed out to 3 years, with an average follow-up of 18 months. Patients had a broad range of cardiovascular risk. Use of aspirin and proton pump inhibitors was allowed in appropriate patients.
In the pooled analysis of the three studies, the primary end point was the first occurrence of any fatal or nonfatal venous or arterial thrombotic event, including MI, unstable angina, intracardiac thrombus, thrombotic stroke, and transient ischemic attack. The rates were 1.24 events/100 patient-years in the etoricoxib group and 1.30 events/100 person-years in the diclofenac group, a nonsignificant difference, reported Dr. Cannon, a cardiologist at Brigham and Women's Hospital, Boston.
Secondary end points were a subset of arterial thrombotic events only and the Antiplatelet Trialists' Collaboration (APTC) end point of MI, stroke, and vascular death. There were no significant differences in these end points.
The MEDAL program provides “more information that is very valuable, but it doesn't clarify the entire issue” of the thrombotic safety of COX-2 selective and nonselective NSAIDs, Dr. Roy Altman, professor of medicine at the University of California, Los Angeles, said in an interview.
The MEDAL program was the largest-ever trial in arthritis patients, and the first to demonstrate a significant reduction of GI events with a cyclooxygenase-2 (COX-2) inhibitor, compared with traditional NSAIDs. Rates of upper GI events clinical events were lower with etoricoxib than with diclofenac (0.67 and 0.97/100 patient-years, respectively). But there was no difference between groups in complicated upper GI events, such as perforated ulcers.
The nonsignificant difference in complicated upper GI events “is probably the most bothersome thing about the study,” because etoricoxib is supposed to be GI-protective and has even more COX-2 selectivity than some other COX-2 selective NSAIDs, such as rofecoxib, Dr. Altman said.
Whether these results can be applied to other traditional NSAIDs is questionable, however, because observational studies have shown diclofenac to have the highest relative risk for myocardial infarction, Dr. Luis Alberto Garcia Rodriguez and Dr. Paola Patrignani of the Spanish Centre for Pharmacoepidemiologic Research, Madrid, wrote in an editorial accompanying the published report (Lancet 2006; DOI:10.1016/S0140–6736[06] 69667–0). “The similar risk of thrombotic events between etoricoxib and diclofenac can be explained by the profound inhibition of COX-2 shared by the two agents.”
“The hypothesis that diclofenac is partially COX-2 selective I think is not supported” because only one article has ever said that diclofenac has some COX-2 selectivity, Dr. Altman said (N. Engl. J. Med. 2001;345:433–42).
But naproxen “clearly is the one that would have been more helpful to compare” with etoricoxib, he said, because naproxen was compared against rofecoxib in Merck's VIGOR (Vioxx Gastrointestinal Outcomes Research) trial and naproxen is more widely used (over-the-counter and prescription) in the United States than diclofenac.
Dr. Cannon told reporters that although observational studies have shown differences among COX-2 inhibitors in cardiovascular risks, randomized trial data have shown no such difference, with the exception of naproxen, which carries a lesser risk that may be explained by an antiplatelet effect at high doses.
“Our results today address one question: In these patients with arthritis, a selective agent didn't increase risk compared with other agents,” said Dr. Cannon. The message to clinicians and patients is that there are options to treat arthritis. “We are moving to an era where we do have choices, and we should be choosing different treatments for different patients.”