Summary Background Fluticasone furoate/vilanterol (FF/VI) is an inhaled corticosteroid/long-acting beta2-agonist (ICS/LABA), recently approved as once-daily maintenance therapy for COPD. We compared the lung function effects of FF/VI with those of twice-daily fluticasone propionate/salmeterol (FP/SAL). Methods Three 12 week studies comparing FF/VI and FP/SAL were conducted. Patients aged ≥40 years with moderate-to-very severe COPD were randomized to receive double-blind, double-dummy FF/VI 100/25 mcg once-daily, or FP/SAL 250/50 mcg twice-daily for 12 weeks following a 2 week placebo run-in period. The primary endpoint of each study was change from baseline trough in 0–24 h weighted mean FEV1 (wmFEV1) on Day 84. Safety was also assessed. Results In Study 1 (HZC113109) (intent-to-treat n: FF/VI = 260; FP/SAL = 259), the increase from baseline in 0–24 h wmFEV1 was significantly greater with FF/VI than FP/SAL (Δ80 mL, P < 0.001). In Study 2 (HZC112352) (intent-to-treat n: FF/VI = 259; FP/SAL = 252) and Study 3 (RLV116974) (intent-to-treat n: FF/VI = 412; FP/SAL = 416), the increase from baseline in 0–24 h wmFEV1 was not significantly greater with FF/VI than FP/SAL (Δ29 mL, P = 0.267; Δ25 mL, P = 0.137). The treatment difference was statistically but not clinically significant in a pooled analysis (Δ41 mL, P < 0.001). Pooled adverse events (FF/VI 27%; FP/SAL 28%) and serious adverse events (FF/VI 2%; FP/SAL 3%) were similar between treatments. Conclusions Our data suggest that once-daily FF/VI 100/25 mcg provides FEV1 improvement in COPD that is at least comparable with that conferred by twice-daily FP/SAL 250/50 mcg, although interpretation is limited by differences in individual study outcomes. The safety profiles of FF/VI 100/25 mcg and FP/SAL 250/50 mcg are similar. Clinical trial registration clinicaltrials.gov: NCT01323634; NCT01323621; NCT01706328. GlaxoSmithKline study codes: HZC113109; HZC112352; RLV116974.