Defining disease stability in chronic obstructive pulmonary disease: evidence from phase 3 clinical trials

RATIONALE: Disease stability in chronic obstructive pulmonary disease (COPD) is a low disease activity state proposed as a treatment target. OBJECTIVES: Characterize COPD stability as a composite endpoint. METHODS: Post hoc analyses were carried out on IMPACT and FULFIL (fluticasone furoate/umeclidinium/vilanterol [FF/UMEC/VI] vs dual therapies) and MATINEE/METREX/METREO (mepolizumab added onto triple therapy). Stability was defined as "no moderate/severe exacerbations and no worsening from baseline in COPD Assessment Test (CAT) and FEV1." We assessed whether achieving stability at week 28 was prognostic of subsequent outcomes (time to first on-treatment moderate/severe exacerbation and time to all-cause mortality [ACM]) after week 28 (IMPACT). Bayesian joint modeling evaluated the probability of achieving stability (IMPACT). MEASUREMENTS AND MAIN RESULTS: Stability was achieved with FF/UMEC/VI by 22% of patients in IMPACT (week 52) and by 46% of patients in FULFIL (week 24); stability was achieved with mepolizumab + triple therapy by 18% of patients in MATINEE/METREX/METREO (week 52). Patients were more likely to achieve stability with triple vs dual therapy and with mepolizumab vs placebo. Many individuals achieving stability had clinically meaningful improvements in CAT and FEV1. Patients achieving stability at week 28 (vs patients who were not stable) had a 45.7% reduction in the risk of moderate/severe exacerbations and a 51.7% reduction in the risk of ACM after week 28. Bayesian joint modeling demonstrated a mean (SD) posterior probability of stability: 25.60% (0.66) at week 52 (FF/UMEC/VI). CONCLUSIONS: Our data provide further evidence for disease stability as an achievable COPD treatment target that predicts long-term clinical benefits in exacerbations and mortality.