© 2008 American Thoracic Society
Treatment Step Down Should Be Avoided at Entry in COPD Therapeutic TrialsTo the Editor:Dr. Wedzicha and colleagues have published a study comparing the effect of salmeterol/fluticasone (SFC) and tiotropium on the prevention of exacerbations in chronic obstructive pulmonary disease (COPD) (1). They did not show a difference between the two treatment arms for the primary endpoint, but other outcome measures favored the SFC treatment. One of these was the withdrawal rate, which might suggest that SFC better prevented the clinical deterioration that often leads to withdrawal. About 50% of the patients were treated by inhaled corticosteroids (ICS) at entry. The authors state that withdrawing from the study after tiotropium was unrelated to the subject's prior use of ICS. The data presented, however, suggest the contrary; the drop-out rates were 113 of 325 and 166 of 340 in patients allocated to tiotropium without and with prior use of ICS, respectively (P < 0.001, chi-square test). Interestingly, the drop-out rate was similar in the two arms of the study in patients without prior use of ICS (106/339 vs. 113/325 in the SFC and tiotropium group, respectively; P = 0.38), while significantly higher in the tiotropium arm in patients with prior ICS use (126/319 vs. 166/340; P = 0.02). A similar analysis performed with data from the ISOLDE trial (2) leads to the same conclusion; the difference in withdrawal rate between treatment groups is driven by the patients who discontinued ICS at entry in the trial. Withdrawal of ICS therapy is associated with an increased risk of exacerbation in patients with COPD (3, 4). These exacerbations can be a trigger for withdrawal. It would be interesting to know how other endpoints of the study were influenced by prior ICS use. Particularly, data on exacerbation rate and quality of life in ICS-naive patients would be informative, although the prednisolone run-in might also have influenced these outcome measures. Therapeutic trials are difficult to perform in patients with COPD, and the authors are to be praised for their commitment in the field. Patients are not easy to recruit and maintain in long-term studies. Unfortunately, there are often issues in trial design since they can favor dropout, leading to difficulties in the interpretation of the results. Moreover, adaptation of COPD treatment is usually in a step-up and not a step-down direction, such as used with some patients in this and previous trials. These limitations make the results of such trials difficult to generalize since they do not match with real-life practice. Avoiding therapeutic step down in future studies would reinforce the validity of results from such trials.
Cliniques Universitaires UCL de Mont-Godinne FOOTNOTES Conflict of Interest Statement: E.M. has been reimbursed by AstraZeneca, GlaxoSmithKline (GSK), and Boehringer Ingelheim for attending conferences; he has participated as a chair in a scientific meeting sponsored by Boehringer Ingelheim; he has received fees as a consultant for GSK. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
