Digging for data from the COX-2 trials

We agree with Joel Lexchin's request for more access to the information from randomized controlled trials on new drugs.¹ To that end we felt CMAJ readers would find a brief synopsis of some of the findings from the latest cyclooxygenase-2 (COX-2) inhibitor trials valuable so that they could make informed decisions about the safety issues associated with these new agents compared with older NSAIDs.

Randomized controlled trials are the best way of determining whether a cause–effect relation exists between treatment and outcome.² They are also one of the best ways to determine if there are important clinical differences in efficacy or safety between therapies. In the last year or so, 2 trials looking at the safety of the COX-2 inhibitors have been published.^3,4 These results have been used to suggest that the COX-2 inhibitors are safer than older NSAIDs. However, the US Food and Drug Administration (FDA) has not put these agents in a class by themselves.⁵ Recently, Mucklow has suggested that journal reporting of clinical trial adverse events is inadequate.⁶ Since February 2001 it has been possible to review more complete trial results on the FDA Web site. Unfortunately, the data on this Web site are presented in numerous reports and it is difficult and time-consuming to get a complete picture of the overall differences in clinical end points. A synopsis of the information at this Web site follows. We chose to present primarily those end points that we felt readers would find interesesting, as well as results that were statistically different. Biochemical and laboratory test differences are not reported in this synopsis. The percentages presented are the crude incidence rates. Numerous subset analyses of the information are also presented at the FDA Web site, but a discussion of these data is beyond the scope of this letter. More detailed trial reports are available at www .fda .gov /ohrms /dockets/ac/01 /briefing/3677b1_03_med.doc and www.fda.gov/ohrms/dockets/ac/01/briefing/3677b2_03_med.doc.

Results of the CLASS trial as reported in JAMA are presented in Table 1.³ The CLASS trial was actually a planned pooled analysis of 2 trials, one comparing ibuprofen with celecoxib and the other comparing diclofenac with celecoxib. In the JAMA article the CLASS trial was presented as a 6-month trial (mean duration of exposure 4 months).³ However, the trial comparing ibuprofen with celecoxib was 15 months long (mean duration of exposure 7 months) and the trial comparing diclofenac and celecoxib trial was 12 months long (mean duration of exposure 6.5 months). A synopsis of the overall results of these 2 trials and a more complete presentation of the adverse events that occurred are presented in Table 2.

Table 1

Table 2

Results of the VIGOR trial are presented in Table 3. Results from the FDA Web site are virtually the same as those presented in the published version of this 9-month trial.⁴ The data presented in boldface are from the FDA Web site and were not presented in the N Engl J Med article.⁴

Table 3

The interpretation of the clinical importance of these results compared with the published data in journals is left to the reader.

James P. McCormack Associate Professor Faculty of Pharmaceutical Sciences University of British Columbia Vancouver, BC Robert Rangno Associate Professor Departments of Medicine and Pharmacology University of British Columbia

Vancouver, BC