This multinational study confirmed findings from a replicate study [9
] that a 90-mg daily dose of the COX-2 selective inhibitor, etoricoxib, was more effective than placebo for treating patients with RA. The treatment effects of etoricoxib occurred by the first assessment (at 2 weeks) and were sustained throughout the 12 weeks of the study.
Etoricoxib showed similar efficacy to a high dose of the non-selective NSAID naproxen, supporting results from studies with other selective COX-2 inhibitors in RA which also showed similar efficacy to non-selective NSAIDs [7
]. The results differed from those in the replicate study of etoricoxib in RA [9
] which found that etoricoxib was more effective than naproxen. The reason for the difference in results between the two studies is unclear since the studies had identical designs, including entry criteria, doses of study medication, and outcome measures. However, although patient enrollment criteria were identical, more patients in the present study were using concomitant corticosteroids (approximately 57% versus 32%) and disease modifying antirheumatic drugs, including methotrexate (approximately 82% versus 68%). It is conceivable that the increased concomitant RA medication use may have obscured the ability to detect small, but perhaps meaningful differences between active treatments. Interestingly, in this study as compared to the replicate etoricoxib RA study [9
], the placebo response rate was higher (ACR20-responders were 40.9% in the current study versus 20.8% in the replicate study) and the discontinuations due to lack of efficacy were lower in the placebo group (25.2% in the current study vs 54.5% in the replicate study), and this may also have influenced the results. Although unproven, other potential explanations may relate to the fact that the present study was conducted largely outside the United States and the previous study inside the United States [9
]. Therefore, differences in underlying disease characteristics and/or cultural differences in perceptions of efficacy may have also contributed to the difference in the results. Others have reported different findings for RA studies with similar designs which were conducted in the United States versus other countries. Serum C-reactive protein levels were noted to be elevated compared to placebo in the etoricoxib group. This was not observed in other etoricoxib RA studies [19
]; in one study no difference was seen from placebo [19
], and in the other study a lowering in serum C-reactive protein was observed [9
]. It is generally not believed that changes in serum C-reactive protein are substantially influenced by NSAID treatment. Therefore, this was not felt to represent a clinically meaningful finding.
Etoricoxib was generally well tolerated by the patients in this study, consistent with results from prior clinical trials of etoricoxib for RA and other indications [9
]. There was a significantly higher number of patients with drug-related clinical adverse events for etoricoxib versus placebo. However, the difference was small and should be interpreted with caution since more patients on placebo discontinued early (primarily due to lack of efficacy) and therefore had less chance of experiencing an adverse event. The findings may therefore overestimate adverse event rates for the active treatments compared with placebo.
Particular attention was paid to the typical NSAID-related renal effects of edema and hypertension since data with both selective COX-2 inhibitors and non-selective NSAIDs have suggested that they have an effect on renal physiology [14
]. Etoricoxib and naproxen showed a small increase in hypertension adverse events compared with placebo. Mean changes in blood pressure among the treatment groups was small, and both etoricoxib and naproxen treatment groups showed only small increases in mean systolic blood pressure compared to baseline. Among patients who had hypertension adverse events on etoricoxib, no patient discontinued from the study. The incidences of lower extremity edema adverse events for etoricoxib were similar to placebo and no patient treated with etoricoxib discontinued as a result.
The main proposed advantage for selective COX-2 inhibitors is reduced gastrointestinal toxicity. However, a thorough and adequate assessment of this can only be made either in very large long-term trials or pooled analyses because of the relatively low incidence of clinically significant gastrointestinal PUBs [6
]. In fact, only one confirmed PUB was reported in the present study.
It has also been suggested [6
] that the use of selective COX-2 agents may be associated with a higher incidence of cardiovascular thrombotic events than naproxen (a potent and sustained inhibitor of platelet aggregation at therapeutic doses). As with upper GI clinical events (PUBs), these events are rare and conclusive data can only be adequately amassed in large data sets. The incidence of confirmed cardiovascular thrombotic events in the present study was low (3 events), therefore no meaningful conclusions about the overall cardiovascular safety of etoricoxib can be determined from this single study.