Editor, – It was disappointing to read that there are still people questioning the gastrointestinal safety and cost-effectiveness of the COX-2 inhibitors (Aust Prescr 2004; 27:2-3). It is even more disappointing when this opinion is referenced to a single non-systematic, heterogenous review article (that is, evidence level 5), which misrepresents the body of evidence in two important ways.
The review claims that non-steroidal anti-inflammatory drugs (NSAIDs) have minimal benefit against which to compare their adverse events. This is based on a very selective use of analgesic data from the literature (which still showed a significant difference to placebo). An alternative view is that NSAIDs are the mainstay of therapy worldwide for the symptomatic relief of arthritis and occupy the first five top rankings for analgesics on the Oxford pain relief table because of their clinical benefits.1 This is backed by clinical trials where both COX-2 inhibitors and traditional NSAIDs showed statistically and clinically different efficacy to placebo in arthritis.2,3,4,5
The article by Wright also states that there is no evidence for reduced gastrointestinal damage from COX-2 inhibitors. He bases this opinion on a single flawed study (CLASS) that had a statistical power of about 45% (that is, less than a 50% chance of detecting any real differences).6 He neglects to mention the wealth of other data from adequately powered studies that show a significant difference in safety and tolerability between celecoxib and the non-specific NSAIDs.7, 8, 9, 10,11,12,13
If the COX-2 inhibitors did not represent a cost-effective treatment then they would not be listed on the Pharmaceutical Benefits Scheme. The Pharmaceutical Benefits Advisory Committee makes this decision based on evidence, not opinion.
Dr Simon McErlane
Medical Director
Pfizer Global Pharmaceuticals
Pfizer Australia
Associate Professor J. Lexchin, the author of the editorial, comments:
Dr McErlane dismisses the results of the CLASS study on celecoxib by claiming that it was underpowered to find significant benefits. CLASS was funded by Pharmacia, the company that marketed celecoxib, and the corresponding author was a Pharmacia employee. Pharmacia is now owned by Pfizer. If there was a problem with the design of CLASS then Dr McErlane should look to his own house.
He criticises the article by Dr Jim Wright for ignoring seven articles showing the gastrointestinal benefits of COX-2 inhibitors. However, one was a poster presentation that was otherwise unpublished and two were published either just before or after Dr Wright's piece and would have been unavailable to him.
Dr McErlane has misread Wright's article. Wright does not say that COX-2 drugs have minimal benefits; what he does say is that the benefits need to be seen in the context of serious adverse events from these drugs. Serious adverse events include not only gastrointestinal problems but other adverse events. Wright combines all serious adverse events as reported in the CLASS study for celecoxib and for other NSAIDs and shows that there is no statistical difference in serious adverse events between celecoxib and the other NSAIDs. In other words, whatever reduction in gastrointestinal harms celecoxib produced was offset by a higher incidence of other serious adverse events.
Dr McErlane's letter provides a good lesson in why doctors should not rely solely on what companies have to say about their products.
