BACKGROUND: The drug development process can be streamlined by combining the traditionally separate stages of dose-finding (Phase IIb) and confirmation of efficacy and safety (Phase III) using an adaptive seamless design. This approach was used in a clinical study of indacaterol, a novel once-daily (od) inhaled long-acting beta(2)-adrenoreceptor agonist bronchodilator for the treatment of COPD (chronic obstructive pulmonary disease). METHODS: The study comprised a dose-finding stage with dose selection after 14 days of treatment, and a second stage evaluating efficacy and safety during 26 weeks of treatment. The dose-finding stage included seven randomized treatment arms: double-blind indacaterol 75 microg, 150 microg, 300 microg or 600 microg od, the beta(2)-adrenoceptor agonist formoterol 12 microg twice-daily or placebo, or the anticholinergic tiotropium 18 microg od open-label. An independent data monitoring committee selected two indacaterol doses based on unblinded results of an interim analysis performed by an independent statistician. The sponsor, investigators and patients remained blinded to the results. The indacaterol doses were selected using pre-set efficacy criteria for trough (24-h post-dose) and early (1-4 h post-dose) bronchodilator effect after 14 days, and all safety data. To qualify for selection, the doses had to exceed a threshold for clinical relevance or be superior to either tiotropium or formoterol, whichever was the highest value. Selected doses were continued into the second, 26-week stage. The two other indacaterol doses not selected, and formoterol, were discontinued following dose selection. RESULTS: 801 patients with moderate-to-severe COPD were evaluated. Indacaterol 150 microg was the lowest effective dose, exceeding criteria for trough FEV(1) (reference value 140 mL vs placebo) and FEV(1) AUC(1-4 h) (reference value 220 mL vs placebo). No safety signal was observed with any dose of indacaterol. Thus, indacaterol 150 and 300 microg were selected to continue into the second, 26-week stage. CONCLUSION: The adaptive seamless design is a novel and efficient way to combine dose selection with efficacy evaluation and safety confirmation in a single trial.