Celecoxib is just as effective as naproxen and diclofenac for treating osteoarthritis, but it causes significantly fewer serious upper gastrointestinal events, compared with the other agents, according to data from a large international study.
The finding “shows conclusively that celecoxib does reduce the risk of upper GI complications, compared to conventional NSAIDs,” the study's lead author, Dr. Gurkirpal Singh, said. “Up until now managed care has been saying there is no evidence in a randomized, clinical trial that celecoxib is better than NSAIDs in reducing GI bleeding. But here it is; these are level 1 data that conclusively prove that.”
However, Dr. Brennan M.R. Spiegel noted that while the difference favoring celecoxib reached statistical significance in the study, the actual difference was a matter of 1 patient per 100 patient-years, which “is a tiny difference that to me is not enough to warrant spending as much money as we do on COX-2 inhibitors,” said Dr. Spiegel, of the division of digestive diseases at the University of California, Los Angeles.
He added that the study is “notable because it's very large, [but] I believed it before that GI events are less common with coxibs than with NSAIDs. I didn't need another study to demonstrate that.”
In a trial called the Successive Celecoxib Efficacy and Safety Study-1 (SUCCESS-1), Dr. Gurkirpal and his associates randomly assigned 13,194 osteoarthritis patients from 39 countries to double-blinded treatment with celecoxib 100 mg b.i.d., celecoxib 200 mg b.i.d., or nonselective NSAID therapy for 12 weeks. The NSAID therapy consisted of diclofenac 50 mg b.i.d. or naproxen 500 mg b.i.d. (Am. J. Med. 2006; 119:255–66).
Patients with a history of two or more episodes of active peptic ulceration were excluded from the study, as were those with gastrointestinal bleeding or recurrent gastric or duodenal ulcers and those with an esophageal, gastric, or duodenal ulcer within a month prior to randomization. Patients with active gastrointestinal disease or any condition that required NSAID therapy were also excluded from the study.
The mean age of study participants was 62 years, 76% were women, and 80% were white, reported Dr. Singh, of the division of gastroenterology and hepatology in the department of medicine at Stanford (Calif.) University. The mean duration of osteoarthritis was 8 years.
Instruments used to measure efficacy included the Patient's Assessment of Arthritis Pain-Visual Analog Scale, Patients' Global Assessment of Arthritis, and the Western Ontario and McMaster Universities (WOMAC) osteoarthritis index. Serious GI events were evaluated by two independent committees that were blinded to patient randomization.
The researchers reported that the primary efficacy measures for both doses of celecoxib were equally effective as the NSAIDs in treating osteoarthritis.
There were 37 confirmed upper GI events: 19 in the patients who took the NSAIDs and 18 in the patients who took celecoxib. That translated into a rate of 2.1 per 100 patient-years for patients who took the NSAIDs vs. a rate of 1.0 per 100 patient-years for those who took celecoxib. The difference was statistically significant with a P value of .023.
A key limitation of the study, Dr. Singh said, is the fact that it was not powered to detect differences in terms of cardiovascular adverse events. “So we can't make any conclusion about that,” he said. “We did not see any statistically significant differences [between treatment groups], but at the time the study was designed, [concerns about COX-2 inhibitors and risk of] myocardial infarction were not an issue, so that's not something we followed.”
To tease that information out, he added, “you'd need a larger study over a longer period of time, perhaps in a high-risk population that has had myocardial infarction. I'd probably want to do it for at least 2 years.”
Dr. Spiegel said that the current standard of care for older patients with osteoarthritis has “overtaken” the overall impact of the SUCCESS-1 study findings.
“The reality is that people are moving to adding a proton pump inhibitor to an NSAID when [osteoarthritis] patients exceed the age of 65 or if they're put on aspirin,” he said. “They're not moving to COX-2 inhibitor, and this study doesn't give me any more reason to put someone on a COX-2 inhibitor as opposed to just adding a PPI to an NSAID, which is overall a cheaper thing to do. [That approach] is more relevant because it's cheaper, it's cost effective, and there's no risk of MI as there is with COX-2 inhibitors. In fact, there's less overall dyspepsia in patients who are on an NSAID plus a PPI, compared to [those on] a COX-2 inhibitor.”