Inclusion and Exclusion Criteria

Improvements in lung function with umeclidinium/vilanterol versus fluticasone propionate/salmeterol in patients with moderate-to-severe COPD and infrequent exacerbations

Authors:James F. Donohue, Sally Worsley, Chang-Qing Zhu, Liz Hardaker, Alison Church

Appendix B

Methods

Inclusion and exclusion criteria

For both studies, inclusion criteria were: males or females 40 years old with established COPD [1]; a post-albuterol (salbutamol) forced expiratory volume in 1s (FEV1) 30% and 70% predicted normal and a pre- and post-albuterol FEV1/forced vital capacity (FVC) ratio 0.70 [2,3]; a dyspnea score 2, modified Medical Research Council [mMRC] Dyspnea Scale); current or former (stopped smoking for 6months) cigarette smokers with a history of cigarette smoking of 10 pack-years. Key exclusion criteria were: asthma/other respiratory disorders; hospitalization for pneumonia within 12 weeks of screening; a documented history of 1 COPD exacerbation requiring oral corticosteroids, antibiotics and/or hospitalization in the year before screening.

Study visits and procedures

The study visits and procedures were the same for both studies. Patients attended up to six clinic visits (see Fig. 1 in the main paper). At the pre-screening visit (Visit 0), patients signed an informed consent form, and demographics, medical history (including COPD exacerbations in the prior 12 months) and concomitant medication were reviewed. Following this pre-screening visit, patients had a wash-out period for any medications as required (Table B.1).

At the screening visit (Visit 1; which could be combined with the pre-screening visit as appropriate), eligibility criteria were checked (including pre- and post-albuterol spirometry, mMRC score, vital signs, 12-lead electrocardiogram), and eligible patients then completed a 7- to 14-day run-in period. During this run-in periodbaseline albuterol use was assessed.

At Visit 2, patients were randomized (after baseline spirometry) to the double-blind study medication and commenced the 12-week treatment period. Each patient was given two inhalers for administration of one active treatment and one placebo treatment. All patients were provided with albuterol for use as needed throughout the run-in and study treatment periods. The Baseline Dyspnea Index was used to assess the severity of dyspnea at baseline. Disease specific health issues were evaluated using the St George’s Respiratory Questionnaire of COPD patients (SGRQ) and the COPD Assessment Test (CAT). Health outcome (EuroQol-5D [EQ-5D]) and vital signs were also evaluated.

Patients recorded (paper diary) the times of study medication taken the day before visits 3 (Day 28), 4 (Day 56) and 5 (Day 84). During these three visits, spirometry (including pre-dose trough spirometry), Transition Dyspnea Index, SGRQ, CAT (Visit 5 only) and EQ-5D (Visit 5 only) were assessed.

Approximately 7 days after the end of study treatment (i.e. after Visit 5 or early withdrawal), a safety follow-up assessment was conducted.

Outcome assessments

Efficacy (lung function) assessments: Spirometry was conducted at each visit in both studies. The primary endpoint, 0–24h wmFEV1 (Day 84), was calculated from pre-dose FEV1 and post-dose FEV1 evaluations at 5 and 15min and 1, 3, 6, 9, 12 (pre-evening dose), 13, 15, 18, 23 and 24h after the morning dose. The secondary endpoint, trough FEV1(Day 85), was the mean of the FEV1 values recorded 23h and 24h after morning dosing on Day 84. In a post-hoc analysis performed after the initial analyses had been completed, the proportion of patients achieving an increase from baseline in FEV1100mL on Day 1 was calculated for the following timepoints post-dosing: 5 and 15min, and 1, 3 and 6h.

Rescue use: Patients completed daily diaries, including rescue medication use and puffs/day; percentage of rescue-free days were calculated.

Statistical analyses

Sample size calculations: The sample size calculation was the same for both studies. Using an estimated residual standard deviation (SD) for wmFEV1 of 220mL (based on a mixed model for repeated measures [MMRM] analyses of previous studies in COPD patients [DB2113360/NCT01316900; DB2113374/NCT01316913; DB2113361/NCT01313637; DB2113373/NCT01313650])and a two-sided 5% significance level, 284 evaluable patients/group would detect a 60mL treatment difference in 0–24h wmFEV1 with 90% power. Approximately 355 patients per group were to be randomized to allow for an assumed 20% drop-out rate.

Statistical methods

The primary and secondary endpoints were also descriptively summarized by GOLD II and III subgroups.

Peak FEV1 was analyzed using an MMRM analysis (covariates: baseline FEV1, smoking status, day, treatment, day by baseline interaction and day by treatment interaction, where day is nominal). The proportions of patients achieving various increases in FEV1 were analyzed using a logistic regression model with treatment as an explanatory variable and baseline FEV1 and smoking status as covariates. Inspiratory capacity (IC) (DB2114930 only) was analyzed by an analysis of covariance (ANCOVA) model with covariates of baseline IC, smoking status and treatment. FVC endpoints were analyzed using the same methods as the FEV1 endpoints. The mean number of puffs of rescue medication per day over the 12-week treatment period was analyzed using an ANCOVA model (covariates: baseline, smoking status and treatment). TDI focal score was analyzed using an MMRM analysis (covariates: baseline dyspnea index (BDI) focal score, smoking status, day, treatment, day by BDI focal score interaction and day by treatment interaction, where day is nominal).

The step-down, closed-testing procedure accounted for multiplicity across the primary and secondary endpoints: if 0–24h wmFEV1 (Day 84; primary endpoint) was statistically significant (5% level), then trough FEV1 (Day 85; secondary endpoint) was evaluated. If both endpoints were statistically significant (5% level) then inferences at the 5% significance level would be made for all other comparisons without adjustment for multiplicity.

Results

COPD medication taken pre-study and during the studies

Prior to the screening visit of both studies patients were washed out of COPD medicationsincluding ICS; wash-out times were 12 weeks for depot medication and 6 weeks for systemic corticosteroids. Thus, the medications listed below are not representative of what the patients were on at the time of providing consent to participate in the studies. Patients were permitted to take LABAs and LAMAs during the ICS wash-out period (provided that the wash-out period for these drugs was adhered to).

During the 30 days prior to the start of DB2114930, 81% of patients were taking COPD medication (not for an exacerbation). The most common classes of this medication were short-acting beta2-agonists (59% of patients), long-acting beta2-agonists (27%), short-acting anticholinergics (23%), and long-acting anticholinergics (20%).Overall, only 6% of patients in DB2114930 were taking an ICS within 30 days of screening (6% in the UMEC/VI group; 7% in the FP/SAL group). In DB2114951, 82% of patients were taking COPD medication (not for an exacerbation); the most common classes were short-acting beta2-agonists (70% of patients), short-acting anticholinergics (35%), long-acting anticholinergics (19%), and long-acting beta2-agonists (16%).Overall, only 4% of patients in DB2114951 were taking an ICS within 30 days of screening (4% in each group). COPD medication for an exacerbation was taken pre-study only by sevenpatients (twoin the UMEC/VI group; fivein the FP/SAL group) in DB2114930. In DB2114951, threepatients in the UMEC/VI group and 13 in the FP/SAL group took COPD medication for an exacerbation pre-study. Patients with an exacerbation during the run-in were not eligible for randomization.

During DB2114930, 3% of patients in each group took additional respiratory medication (not for an exacerbation), the most common of which was oxygen (1% patients in UMEC/VI; 2% in FP/SAL group). In the UMEC/VI and FP/SAL groups in DB2114951, 5% and 7% of patients, respectively, took COPD medication (not for an exacerbation) during the study; the most common classes of respiratory medication were mucolytics (2%) and oxygen (2%) in the UMEC/VI group, and mucolytics (3%) in the FP/SAL group.For COPD exacerbations during the treatment period, 2% of patients in each group took medication in DB2114930 (most common: hydrocortisone: 1% in UMEC/VI group; 1% in FP/SAL group). In DB2114951, 2% (UMEC/VI) and 3% (FP/SAL) of patients took medication for COPD exacerbations during the study.

Safety and tolerability

In DB2114930, the overall incidence of AEs was 26% (UMEC/VI) and 27% (FP/SAL). One patient died in the FP/SALgroup (myocardial infarction secondary to a pulmonary embolism; neither of these causes was considered by the investigator to be related to the study drug). Seven (2%, UMEC/VI) and10 (3%, FP/SAL) patients withdrew from DB2114930 due to AEs. The incidence of drug-related AEs was very low (2% in both groups), and only onepatient (UMEC/VI group) had a drug-related serious AE (obstructive airways disorder). Twelve (3%) patients treated with UMEC/VI and 11 (3%) treated with FP/SAL experienced COPD exacerbations during DB2114930; all but one(FP/SAL group) of these patients withdrew from the study due to these exacerbations.

In DB2114951, the overall incidence of AEs was 30% (UMEC/VI) and 31% (FP/SAL). Five patients (UMEC/VI: two;FP/SAL: three) died in this study. Only one of these deaths was considered related to the study drug:onepatient on FP/SAL died of pneumonia. The other deaths were not related to study treatment and the causes were: myocardial infarction and cardiac failure (both in the UMEC/VI group), and small-cell lung carcinoma and myocardial infarction (both in the FP/SAL group). Nine (3%, UMEC/VI) and14 (4%, FP/SAL) patients withdrew from DB2114951 due to AEs. The incidence of drug-related AEs was low (UMEC/VI: 2%; FP/SAL: 6%) and only onepatient (FP/SAL group) had a drug-related serious AE. Nine (3%) patients treated with UMEC/VI and 11 (3%) treated with FP/SAL experienced COPD exacerbations during DB2114951; all but four(twoin each group) of these patients withdrew from the study due to these exacerbations.

No clinically significant changes from baseline in vital signs at the end of the study were observed in either treatment group in either study.

References

1. Celli BR, MacNee W; ATS/ERS Task Force. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper. Eur Respir J. 2004;23:932-946.
2. Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med.1999;159:179-187.
3. Hankinson JL, Kawut SM, Shahar E, Smith LJ, Stukovsky KH, Barr RG. Performance of American Thoracic Society-recommended spirometry reference values in a multiethnic sample of adults: the multi-ethnic study of atherosclerosis (MESA) lung study. Chest. 2010;137:138-145.

1

Table B.1 List ofexcluded medications and time intervals prior to screening for DB2114930 and DB2114951.

Medication / Time Interval
Depot corticosteroids / 12 weeks
Systemic, oral or parenteral corticosteroidsa / 6 weeks
Antibiotics (for lower respiratory tract infection)b / 6 weeks
Cytochrome P450 3A4 strong inhibitors / 6 weeks
Herbal medications potentially containing oral or systemicsteroidsc / 6 weeks
ICS / 30 days
LABA/ICS combination products / 30 days
PDE4 inhibitors (e.g., roflumilast) / 14 days
Olodaterol and indacaterol / 10 days
Inhaled long-acting anticholinergics (tiotropium, aclidinium,glycopyrronium) / 7 days
Xanthinesd / 48 hours
Oral leukotriene inhibitors (zafirlukast, montelukast, zileuton) / 48 hours
Oral beta2-agonists
Long-acting
Short-acting / 48 hours
12 hours
Salmeterol and formoterol / 48 hours
Inhaled sodium cromoglycate or nedocromil sodium / 24 hours
Inhaled short-acting beta2-agonistse / 4 hours
Inhaled short-acting anticholinergics / 4 hours
Inhaled short-acting anticholinergic/short-acting beta2-agonist
combination products / 4 hours
Any other investigational medication for COPD / 30 days or within 5 drug half-lives (whichever is longer)

aIntra-aricular corticosteroid and epidural injections are permitted. These medications given for a COPDexacerbation within 12 months prior to Visit 1 were exclusionary

bAntibiotics given for a COPD exacerbation within 12 months prior to Visit 1 were exclusionary

cThese medications were classified into ‘Herbal medicine for respiratory diseases’

dIncludes methylated xanthenes (e.g. theobromine and theophylline)

fUse of study provided albuterol was permitted (as needed) during the study except in the 4-hour period prior to spirometry testing

COPD, chronic obstructive pulmonary disease; ICS, inhaled corticosteroid; LABA, long-acting beta2-agonist; PDE4, phosphodiesterase 4.

Table B.2Descriptive summary of change from baseline in 0–24h wmFEV1 on Day 84 and in trough FEV1 on Day 85 by GOLD subgroup (ITT population).

DB2114930 / DB2114951
UMEC/VI
62.5/25mcg
(N=353) / FP/SAL
250/50mcg
(N=353) / UMEC/VI
62.5/25mcg
(N=349) / FP/SAL
250/50mcg
(N=348)
0–24h wmFEV1 on Day 84, L, change from baseline
GOLD stage IIa
n / 153 / 157 / 160 / 154
mean (SD) / 0.157 (0.2463) / 0.089 (0.2422) / 0.195 (0.2378) / 0.146 (0.2646)
GOLD stage IIIb
n / 162 / 154 / 162 / 157
mean (SD) / 0.174 (0.2419) / 0.092 (0.1891) / 0.230 (0.2793) / 0.080 (0.2048)
Trough FEV1 on Day 85, L, change from baseline
GOLD stage IIa
n / 154 / 158 / 161 / 154
mean (SD) / 0.143 (0.2613) / 0.064 (0.2714) / 0.170 (0.2407) / 0.120 (0.2673)
GOLD stage IIIb
n / 163 / 154 / 162 / 157
mean (SD) / 0.167 (0.2412) / 0.084 (0.1943) / 0.199 (0.2884) / 0.064 (0.2172)

aFEV150% to 80% predicted

bFEV130% to 50% predicted

FEV1, forced expiratory volume in 1second; FP/SAL, fluticasone propionate/salmeterol; ITT, intent-to-treat; SD, standard deviation; UMEC, umeclidinium; VI, vilanterol; wm, weighted mean.

1

Table B.3 Proportion of patients achieving an increase FEV1100mL above baseline at various times post-dose on Day 1 (ITT population;
post-hoc analyses).

DB2114930 / DB2114951
Endpoint / UMEC/VI 62.5/25mcg
(N=353) / FP/SAL 250/50mcg
(N=353) / UMEC/VI 62.5/25mcg
(N=349) / FP/SAL 250/50mcg
(N=348)
15min post-dose
n / 353 / 350 / 344 / 346
Increase, n (%) / 187 (53) / 120 (34) / 214 (62) / 157 (45)
No increase, n (%) / 166 (47) / 230 (66) / 130 (38) / 189 (55)
Odds ratio (95% CI) / 2.15 (1.59-2.92)
p 0.001
/ 1.98 (1.46-2.69)
p 0.001
1h post-dose
n / 352 / 349 / 345 / 347
Increase, n (%) / 229 (65) / 148 (42) / 251 (73) / 195 (56)
No increase, n (%) / 123 (35) / 201 (58) / 94 (27) / 152 (44)
Odds ratio (95% CI) / 2.53 (1.86-3.43)
p 0.001 / 2.08 (1.51-2.85)
p 0.001
3h post-dose
n / 353 / 351 / 345 / 347
Increase, n (%) / 249 (71) / 197 (56) / 268 (78) / 231 (67)
No increase, n (%) / 104 (29) / 154 (44) / 77 (22) / 116 (33)
Odds ratio (95% CI) / 1.87 (1.37-2.56)
p 0.001 / 1.74 (1.24-2.44)
p =0.001
6h post-dose
n / 353 / 351 / 346 / 347
Increase, n (%) / 247 (70) / 166 (47) / 263 (76) / 209 (60)
No increase, n (%) / 106 (30) / 185 (53) / 83 (24) / 138 (40)
Odds ratio (95% CI) / 2.59 (1.90-3.53)
p 0.001 / 2.09 (1.50-2.89)
p 0.001

CI, confidence interval; FP/SAL, fluticasone propionate/salmeterol; ITT, intent-to-treat; UMEC, umeclidinium; VI, vilanterol.

Table B.4 Results from the analyses of FVCother endpoints (selected) (ITT population).

DB2114930 / DB2114951
Endpoint / UMEC/VI 62.5/25mcg
(N=353) / FP/SAL 250/50mcg
(N=353) / UMEC/VI 62.5/25mcg
(N=349) / FP/SAL 250/50mcg
(N=348)
wm 0–24h FVC on Day 84, L
n / 315 / 310 / 322 / 311
LS mean (SE) / 3.018(0.0202) / 2.892(0.0204) / 3.122(0.0212) / 2.934(0.0215)
LS mean (SE) change from baseline / 0.203 (0.0202) / 0.077 (0.0204) / 0.293 (0.0212) / 0.105 (0.0215)
Treatment difference (95% CI) / 0.126 (0.070-0.183)
p 0.001
/ 0.188 (0.129-0.247)
p 0.001
Trough FVC on Day 85, L
na / 341 / 339 / 335 / 336
nb / 317 / 312 / 323 / 311
LS mean (SE) / 3.001 (0.0201) / 2.859 (0.0203) / 3.060 (0.0216) / 2.873 (0.0219)
LS mean (SE) change from baseline / 0.184 (0.0201) / 0.042 (0.0203) / 0.238 (0.0216) / 0.051 (0.0219)
Treatment difference (95% CI) / 0.142 (0.086-0.198)
p 0.001 / 0.187 (0.127-0.247)
p 0.001
wm 0–6h FVC, L
Day 1
na / 353 / 349 / 346 / 348
nb / 353 / 347 / 344 / 347
LS mean (SE) / 3.117 (0.0146) / 2.992 (0.0147) / 3.183 (0.0155) / 3.073 (0.0154)
LS mean (SE) change from baseline / 0.308 (0.0146) / 0.183 (0.0147) / 0.355 (0.0155) / 0.246 (0.0154)
Treatment difference (95% CI) / 0.125 (0.084-0.165)
p 0.001 / 0.110 (0.067-0.153)
p 0.001
Day 84
na / 353 / 349 / 346 / 348
nb / 317 / 313 / 324 / 312
LS mean (SE) / 3.104 (0.0207) / 2.954 (0.0208) / 3.232 (0.0256) / 3.023 (0.0259)
LS mean (SE) change from baseline / 0.296 (0.0207) / 0.146 (0.0208) / 0.404 (0.0256) / 0.196 (0.0259)
Treatment difference (95% CI) / 0.150 (0.093-0.208)
p 0.001 / 0.208 (0.137-0.280)
p 0.001

aNumber of patients with analyzable data for 1 or more time points

bNumber of patients with analyzable data at the current time point

CI, confidence interval; FVC, forced vital capacity; FP/SAL, fluticasone propionate/salmeterol; ITT, intent-to-treat; LS, least squares; SE, standard error; UMEC, umeclidinium;
VI, vilanterol; wm, weighted mean.

Table B.5TDI focal scores and SGRQTotal scores (ITT population).

DB2114930 / DB2114951
Endpoint / UMEC/VI 62.5/25mcg
(N=353) / FP/SAL 250/50mcg
(N=353) / UMEC/VI 62.5/25mcg
(N=349) / FP/SAL 250/50mcg
(N=348)
TDI focal score, change from baseline
Day28
na / 334 / 335 / 334 / 331
nb / 334 / 335 / 333 / 330
LS mean (SE) / 2.0 (0.16) / 1.7 (0.16) / 2.2 (0.15) / 1.9 (0.16)
Treatment difference (95% CI) / 0.3 (–0.2-0.7)
p =0.246 / 0.4 (–0.1-0.8)
p =0.099
Day 56
na / 334 / 335 / 334 / 331
nb / 326 / 322 / 330 / 321
LS mean (SE) / 2.8 (0.15) / 2.5 (0.15) / 2.8 (0.15) / 2.2 (0.16)
Treatment difference (95% CI) / 0.2 (–0.2-0.7)
p =0.266 / 0.6 (0.1-1.0)
p =0.010
Day 84
na / 334 / 335 / 334 / 331
nb / 316 / 309 / 323 / 307
LS mean (SE) / 3.3 (0.16) / 3.0 (0.16) / 3.0 (0.16) / 2.6 (0.17)
Treatment difference (95% CI) / 0.3 (–0.2-0.7)
p =0.193 / 0.3 (–0.1-0.8)
p =0.170
SGRQ Total score, change from baseline
Day 28
na / 335 / 340 / 332 / 332
nb / 332 / 337 / 329 / 330
LS mean (SE) / 41.67 (0.566) / 41.56 (0.562) / 41.80 (0.618) / 43.75 (0.617)
LS mean change (SE) / –4.14 (0.566) / –4.25 (0.562) / –5.74 (0.618) / –3.79 (0.617)
Treatment difference (95% CI) / 0.10 (–1.46-1.67)
p =0.898 / –1.95 (–3.67-–0.23)
p =0.026
Day 84
na / 335 / 340 / 332 / 332
nb / 312 / 313 / 321 / 307
LS mean (SE) / 39.48 (0.658) / 39.02 (0.655) / 40.32 (0.742) / 41.87 (0.752)
LS mean change (SE) / –6.33 (0.658) / –6.79 (0.655) / –7.23 (0.742) / –5.67 (0.752)
Treatment difference (95% CI) / 0.47 (–1.36-2.29)
p =0.616 / –1.55 (–3.63-0.53)
p =0.143

aNumber of patients with analyzable data for 1 or more time points

bNumber of patients with analyzable data at the current time point

CI, confidence interval; FP/SAL, fluticasone propionate/salmeterol; ITT, intent-to-treat; LS, least squares; SE, standard error; SGRQ, St George’s Respiratory Questionnaire;
TDI, Transition Dyspnea Index; UMEC, umeclidinium; VI, vilanterol.

Table B.6Rescue albuteroluse (ITT population).

DB2114930 / DB2114951
Endpoint / UMEC/VI 62.5/25mcg
(N=353) / FP/SAL 250/50mcg
(N=353) / UMEC/VI 62.5/25mcg
(N=349) / FP/SAL 250/50mcg
(N=348)
Rescue albuterol use
Mean number of puffs/day, weeks 1–12
n / 332 / 334 / 328 / 329
LS mean (SE) / 1.4 (0.08) / 1.4 (0.08) / 1.4 (0.08) / 1.7 (0.08)
LS mean (SE) change from baseline / –1.4 (0.08) / –1.3 (0.08) / –1.6 (0.08) / –1.3 (0.08)
Treatment difference (95% CI) / 0.0 (–0.3-0.2)
p =0.689
/ –0.3 (–0.6-–0.1)
p =0.006
Percent rescue-free days during 12 weeks, change from baseline
n / 332 / 334 / 328 / 329
Mean (SD), % / 24.2 (36.87) / 22.4 (35.56) / 26.8 (34.04) / 24.2 (34.19)

CI, confidence interval; FP/SAL, fluticasone propionate/salmeterol; ITT, intent-to-treat; LS, least squares; SE, standard error; SD, standard deviation; UMEC, umeclidinium;
VI, vilanterol.

Table B.7EQ-5D Utility Scores and CAT Scores (ITT population).

DB2114930 / DB2114951
Endpoint / UMEC/VI 62.5/25mcg
(N=353) / FP/SAL 250/50mcg
(N=353) / UMEC/VI 62.5/25mcg
(N=349) / FP/SAL 250/50mcg
(N=348)
EQ-5D utility score on Day 84, change from baseline
n / 321 / 317 / 326 / 312
Mean (SD) / 0.04 (0.216) / 0.06 (0.235) / 0.06 (0.204) / 0.02 (0.200)
CAT score on Day 84, change from baseline
n / 322 / 317 / 326 / 312
Mean (SD) / –2.34 (6.317) / –3.21 (6.664) / –2.23 (7.393) / –2.05 (7.152)

aNumber of patients with analyzable data for 1 or more time points

bNumber of patients with analyzable data at the current time point

CAT, COPD Assessment Test; EQ-5D, EuroQoL-5D; FP/SAL, fluticasone propionate/salmeterol; ITT, intent-to-treat; SD, deviation; UMEC, umeclidinium; VI, vilanterol.

1