The upper gastrointestinal safety of rofecoxib vs. NSAIDs: an updated combined analysis

Background: Nonsteroidal anti-inflammatory drugs (NSAIDs) are nonspecific cyclo-oxygenase (COX-1/COX-2) inhibitors and are associated with gastrointestinal (GI) toxicity attributable to COX-1 inhibition. Rofecoxib, a COX-2 specific inhibitor, was developed to provide similar efficacy and less GI toxicity than NSAIDs. Objective: To update the results of a previously performed analysis of the incidence of upper GI perforations, symptomatic gastroduodenal ulcers, and upper GI bleeding (PUBs) with rofecoxib compared with non-selective NSAIDs. Research design and methods: We compared the incidence of PUBs in a combined analysis of 20 randomized, double-blind, clinical trials of rofecoxib versus NSAIDs. Men and women (N = 17 072) from multinational trial sites with osteoarthritis or rheumatoid arthritis were studied. There was no upper age limit in any of the trials. Investigator-reported PUBs were reviewed by a blinded, external adjudication committee using pre-specified criteria. The incidence of confirmed PUBs, the main outcome measure, among patients treated with rofecoxib 12.5 mg, 25 mg, or 50 mg (combined, N = 10 026) was compared to that among patients treated with ibuprofen, diclofenac, nabumetone, or naproxen (combined, N = 7046). Results: The incidence of PUBs over 24.8 months was significantly lower with rofecoxib vs. NSAIDs (cumulative incidence 1.6% vs. 3.1 %, p < 0.001; rate/100 patient-years 0.74 vs. 1.87; relative risk 0.36, 95% CI 0.24, 0.54). Results of subgroup analyses and comparisons of rofecoxib with individual NSAID comparators were consistent with the primary result, as was an analysis in patients with no PUB risk factors. Discussion: The analysis demonstrated a consistently lower incidence of confirmed PUBs with rofecoxib than with NSAIDs over 24.8 months. These results confirm those of a previous smaller combined analysis of clinical trials with rofecoxib vs. non-selective NSAIDs in OA patients only, in which the risk reduction for confirmed PUBs was approximately 50%. In addition, this analysis demonstrated risk reductions with rofecoxib vs. NSAIDs in risk subgroups and in patients who did not have any known risk factors for PUBs consistent with the primary result. Some of the studies in this analysis required scheduled endoscopies. Asymptomatic upper GI ulcers or bleeding diagnosed during scheduled procedures were not included in the primary endpoint, which may have caused a bias against rofecoxib. Conclusions: Treatment with rofecoxib was associated with a statistically significantly (p < 0.001) lower incidence of PUBs than was treatment with NSAIDs. The difference was maintained in subgroups of patients with risk factors, as well as in those with no risk factors, for PUBs.