A model-based longitudinal meta-analysis of FEV1 in randomized COPD trials.

A model-based, longitudinal meta-analysis of the efficacy on morning trough forced expiratory volume in 1 second (FEV1 ) in chronic obstructive pulmonary disease (COPD) is presented. Literature data from 142 randomized maintenance trials were included, comprising 106,422 patients who received 19 compounds. 1982 morning trough FEV1 observations were available, each representing the mean FEV1 for a study arm at a specific timepoint. The final model for absolute FEV1 included baseline, disease progression, placebo effect, and drug effect estimates for all compounds, with interstudy variability on all model components and additional interarm variability on baseline. A dose-response relationship was identifiable for 10 of the 19 compounds. Drug-drug interactions among direct bronchodilators and the effect of concomitant background COPD treatment were included. Covariates were identified on baseline. Disease progression was proportional to the baseline FEV1 , and a mean baseline of <1.2 L resulted in a lower efficacy, in particular for antiinflammatory treatments.