Spirometry testing was performed in accordance with American Thoracic Society standards. Trough FEV1 defined as the mean of two measurements at 23 hours 15 minutes and 23 hour 45 minutes post dosing. Baseline FEV1 was defined as the average of the -45 minutes and -15 minutes FEV1 values taken on Day 1. An analysis-of-covariance (ANCOVA) for repeated measurements, also known as mixed model for repeated measures (MMRM), was performed for the change from baseline of trough FEV1 at Week 12. The model included treatment, COPD severity, baseline smoking status, baseline ICS use, region, and visit (Day 1, and Weeks 12 and 26) as factors and baseline FEV1 as a covariate.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and week 12 were included in this analysis.

The standardized Area Under the Curve (AUC) for Forced Expiratory Volume in one second (FEV1) is assessed for different time spans (0-12h, 0-24h, 12-24h) within the overall serial measurement post dosing at week 12 of treatment. Baseline FEV1 was defined as the average of the -45 minutes and -15 minutes FEV1 values taken on Day 1.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at baseline and week 12 in different time spans were included in this analysis.

The standardized Area Under the Curve (AUC) for Forced Expiratory Volume in one second (FEV1) is assessed for 0-12 hour, post dosing at Day 1 of treatment. Baseline FEV1 was defined as the average of the -45 minutes and -15 minutes FEV1 values taken on Day 1.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and Day 1 were included in this analysis.

The standardized Area Under the Curve (AUC) for Forced Expiratory Volume in one second (FEV1) is assessed for different time spans (0-12h, 0-24h, 12-24h) within the overall serial measurement post dosing at week 26 of treatment. Baseline FEV1 was defined as the average of the -45 minutes and -15 minutes FEV1 values taken on Day 1.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and week 26 in different time spans were included in this analysis.

The health status, as reported by the patients, is assessed using the St. George's Respiratory Questionnaire (SGRQ). The SGRQ contains 50 items divided into 2 parts covering 3 aspects of health related to COPD:

• Part I covers "Symptoms" and is concerned with respiratory symptoms, their frequency and severity
• Part II covers "Activity" and is concerned with activities that caused or are limited by breathlessness
• Part II is also concerned with "Impacts", which covers a range of aspects concerned with social functioning and psychological disturbances resulting from airways disease.

A score was calculated for each of these three subscales and the total score was calculated. In each case, the lowest possible value was 0 and the highest 100. Higher values corresponded to greater impairment of health status.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and post-baseline time points were included in this analysis.

The health status, as reported by the patients, is assessed using the St. George's Respiratory Questionnaire (SGRQ).

The SGRQ contains 50 items divided into 2 parts covering 3 aspects of health related to COPD:

• Part I covers "Symptoms" and is concerned with respiratory symptoms, their frequency and severity
• Part II covers "Activity" and is concerned with activities that caused or are limited by breathlessness
• Part II is also concerned with "Impacts", which covers a range of aspects concerned with social functioning and psychological disturbances resulting from airways disease.

A score was calculated for each of these three subscales and the total score was calculated. In each case, the lowest possible value was 0 and the highest 100.

A clinically significant improvement is defined as ≥ 4 unit improvement from baseline score (a decrease of ≥ 4).

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. n= the number of patients with a SGRQ total score at both baseline and indicated post-baseline time point

Breathlessness at Week 12 and Week 26 is measured using the Transition Dyspnea Index (TDI). On day 1, breathlessness is assessed by the Baseline Dyspnea Index (BDI). Baseline Dyspnea Index (BDI)/Transition Dyspnea Index (TDI) focal score is based on three domains: functional impairment, magnitude of task and magnitude of effort and captures changes from baseline. BDI was measured at day 1 prior to the first dose with domain scores ranging from 0=very severe to 4=no impairment and a total score ranging from 0 to 12(best). TDI captures changes from baseline. Each domain is scored from -3=major deterioration to 3=major improvement to give an overall TDI focal score of -9 to 9. Higher numbers indicate a better score.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and post-baseline time points were included in this analysis.

Breathlessness at Week 12 and Week 26 is measured using the Transition Dyspnea Index (TDI). On day 1, breathlessness is assessed by the Baseline Dyspnea Index (BDI). Baseline Dyspnea Index (BDI)/Transition Dyspnea Index (TDI) focal score is based on three domains: functional impairment, magnitude of task and magnitude of effort and captures changes from baseline. BDI was measured at day 1 prior to the first dose with domain scores ranging from 0=very severe to 4=no impairment and a total score ranging from 0 to 12(best). TDI captures changes from baseline. Each domain is scored from -3=major deterioration to 3=major improvement to give an overall TDI focal score of -9 to 9. Higher numbers indicate a better score.

Clinically important improvement indicates ≥ 1 unit in the TDI focal score at Weeks 12 and 26 in comparison to BDI focal score (an increase of ≥ 1).

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. n= the number of patients with a TDI focal score

Spirometry testing was performed in accordance with American Thoracic Society standards. Trough FEV1 defined as the mean of two measurements at 23 hours 15 minutes and 23 hour 45 minutes post dosing. Baseline FEV1 was defined as the average of the -45 minutes and -15 minutes FEV1 values taken on Day 1. An analysis-of-covariance (ANCOVA) for repeated measurements, also known as mixed model for repeated measures (MMRM), was performed for the change from baseline of trough FEV1. The model included treatment, COPD severity, baseline smoking status, baseline ICS use, region, and visit (Day 1, and Weeks 12 and 26) as factors and baseline FEV1 as a covariate.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and post-baseline time points were included in this analysis.

Mixed model for repeated measures was used to analyze change from baseline in FVC. Baseline FVC is defined as the average of the -45 min and -15 min FVC values taken on Day 1 prior to first dose.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following intent-to-treat principle, patients in FAS were analyzed according to the treatment to which they were randomized. n = number of patients included in respective analysis (i.e. who had a FVC at this time point on at least one visit).

Mixed model for repeated measures was used to analyze change from baseline in IC.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following intent-to-treat principle, patients in FAS were analyzed according to the treatment to which they were randomized. n = number of patients included in respective analysis (i.e. who had a IC at this timepoint on at least one visit).

Mixed model for repeated measures was used to analyze change from baseline in FEV1. Baseline FEV1 is defined as the average of the -45 min and -15 min FEV1 values taken on Day 1 prior to first dose.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following intent-to-treat principle, patients in FAS were analyzed according to the treatment to which they were randomized. n = number of patients included in respective analysis (i.e. who had a FEV1 at this timepoint on at least one visit).

Patients report the number of puffs of rescue medication (salbutamol / albuterol) using an electronic diary. change from baseline in percentage of days without rescue medication usage over 26 weeks was analyzed.

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and week 26 were included in this analysis.

Patients reported symptoms by using an electronic diary. The mean daily total symptom score, the mean morning symptom score and the mean evening symptom score were calculated for each patient over 26 weeks. Each symptom measured in a numeric rating scale of 0-10; 0 indicates no symptom and 10 indicates severe symptom. 0 is no waking due to symptoms, 1 woke up once, 2 woke up more than once due to symptoms ; 10 was the worst score.The daily score for an individual symptom score was the worst of the morning and evening scores on a particular day. If either the morning or evening score was missing for a symptom then the non-missing value was taken as the worst. A negative change indicates improvement. Only the scores for the 6 COPD symptoms (respiratory symptoms, cough, wheeze, production of sputum, sputum color, and breathlessness) were used to derive the total symptom score

The Full Analysis Set (FAS) included all randomized patients who received at least one dose of study drug. Following the intent-to-treat principle, patients in the FAS were analyzed according to the treatment to which they were randomized. Patients with evaluable data at both baseline and week 26 were included in the analysis.

This endpoint reports patients affected by any adverse events (AE), serious adverse events (SAE) and death. Only treatment emergent AE, SAE, deaths are reported for this endpoint.

The Safety Set (SAF) included all patients who received at least one dose of study drug whether or not they were randomized. Patients were analyzed according to the treatment they received. If patients switched treatment during the study, they were to be analyzed according to the treatment to which they were randomized.