Influence Of Patient Baseline Characteristics And Concomitant Medication On Outcomes In Long-Term Trials Of Copd: Analysis Of The Uplift (R) Trial

Background: Several long-acting bronchodilators (BDs) were introduced recently as maintenance therapy for COPD. However, a direct comparison of the effects of these newer agents with existing therapy from long-term studies is lacking. Aims and objectives: To investigate if patient selection affects the magnitude of effects observed in clinical studies in COPD. Methods: Subgroup analysis of the UPLIFT® trial using in/exclusion criteria similar to clinical programs of other long-acting BDs (OLB), ie, exclusion of GOLD IV patients and concomitant long-acting β 2 -agonists. Spirometry and SGRQ: mixed-model repeated measures ANOVA with random subject effects; number of events: Poisson regression with correction for overdispersion. Results: Baseline patient characteristics are shown in the table. For UPLIFT total vs OLB subgroup, minimum, maximum improvements tiotropium versus placebo over 4 yrs were: trough FEV 1 , 0.09-0.10L vs 0.10-0.13L; SGRQ total score, -2.30 to -3.35 vs -2.14 to -4.14. Overall exacerbation rate (per pt-yr) in tiotropium/placebo groups differed between UPLIFT total (0.74/0.87) and OLB subgroup (0.43/0.48), rate ratios for tiotropium vs placebo were 0.86 and 0.89 respectively. Conclusion: Efficacy results in COPD placebo-controlled clinical trials are influenced by included patient population characteristics and/or allowed concomitant medication. Cross-trial efficacy comparisons should take this into account.