© 2005 American Thoracic Society
Poor Choice of Primary Outcome in a Clinical Trial of Pirfenidone in Patients with IPFTo the Editor:We were surprised that Azuma and colleagues chose to use the amplitude of oxyhemoglobin desaturation during a 6-min walk test (6MWT) as their primary endpoint in a recent placebo-controlled trial of pirfenidone in patients with IPF (1). This outcome has not been validated as an endpoint for therapeutic efficacy and has several additional limitations. First, use of this endpoint required patients to complete a nonstandardized variant of the 6MWT on a treadmill set at a predetermined and individualized pace. Second, one might have predicted that it would be very difficult to detect worsening in the amplitude of desaturation during the course of the trial. A small but sustained drop in SpO2 from the mean baseline value of 87 to 84% would be enough to prevent completion of the test, because the protocol specified that the test be stopped when SpO2 fell below 80% or remained between 80 and 85% for 30 s. In fact, the authors found that 25% of their study population was unable to complete the follow-up 6MWT. The decision to analyze separately patients that completed the 6MWT subverts the process of randomization, and the statistically significant differences they found do not seem to be clinically important. Third, in the section describing their power calculations, the authors do not tell us what effect size they were hoping to detect, so we are left to wonder what constitutes a clinically important improvement in fall of SpO2. Given that pre-existing fibrosis is unlikely to respond to therapy and, as these authors found, anticipated efficacy will be in slowing, not reversing deterioration, it is hard to imagine a clinically relevant efficacy that would reliably be detected by the inherent limitations of this endpoint. Finally, Eaton and coworkers recently showed that amplitude of desaturation during 6MWT had poor reproducibility (SD/mean value = 42.5%) and correlated poorly with 6MWT distance (2).
Other outcomes in this study are more informative. The finding of a reduction in acute exacerbations is intriguing; and while in our opinion this result did not warrant early cessation of the trial, it is consistent with results of a published trial of
a Stanford University Medical Center, Stanford, California FOOTNOTES Conflict of Interest Statement: Neither of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
interferon in IPF (