- Email: [email protected]
The Efficacy and Safety of the Novel Long-Acting β2 Agonist Vilanterol in Patients With COPD
CHEST
Original Research COPD
The Efficacy and Safety of the Novel Long-Acting b2 Agonist Vilanterol in Patients With COPD A Randomized Placebo-Controlled Trial Nicola A. Hanania, MD, FCCP; Gregory Feldman, MD; Wolfgang Zachgo, MD; Jae-Jeong Shim, MD, PhD; Courtney Crim, MD, FCCP; Lisa Sanford, MSc; Sally Lettis, PhD; Frank Barnhart, DVM; and Brett Haumann, MD
Background: Vilanterol (GW642444M) (VI) is a novel, inhaled, long-acting b2 agonist with inherent 24-h activity under development as a once-daily combination therapy with an inhaled corticosteroid for COPD and asthma. This study assessed the dose response, efficacy, and safety of VI at doses of 3 to 50 mg in patients with moderate to severe COPD. Methods: Six hundred two patients (intent-to-treat) were randomized (double-blind) to VI 3, 6.25, 12.5, 25, or 50 mg or placebo once daily for 28 days. The primary end point was change from baseline in trough FEV1 at the end of the 28-day treatment period. Secondary end points included 0- to 24-h weighted mean FEV1 on days 1 and 28 and time to increases of ⱖ 100 mL or ⱖ 12% from baseline FEV1 on day 1. Safety assessments included adverse events, vital signs, ECG assessment, and clinical laboratory tests. Results: VI once daily for 28 days significantly improved trough FEV1 in a dose-dependent manner vs placebo. Clinically relevant treatment differences of ⱖ 130 mL in trough and 0- to 24-h weighted mean FEV1 were observed with VI 25- and 50-mg doses vs placebo. All doses of VI were associated with a low incidence of treatment-related adverse events/serious adverse events, with no suggestion of effects on BP, pulse rate, QT intervals corrected for heart rate calculated by Fridericia formula, or blood glucose and potassium levels. Conclusions: VI 25 and 50 mg once daily provided both statistically and clinically relevant 24-h improvements in lung function in patients with COPD compared with placebo. All doses of VI had a safety and tolerability profile similar to placebo. Trial registry: ClinicalTrials.gov; No.: NCT00606684; URL: www.clinicaltrials.gov CHEST 2012; 142(1):119–127 Abbreviations: AE 5 adverse event; ANCOVA 5 analysis of covariance; ICS 5 inhaled corticosteroid; ITT 5 intentto-treat; LABA 5 long-acting b2 agonist; LOCF 5 last observation carried forward; QTcF 5 Fridericia formula; SAE 5 serious adverse event; VI 5 vilanterol (GW642444M)
inhaled bronchodilators, comprising Long-acting long-acting b agonists (LABAs) and long-acting 2
muscarinic antagonists, are recommended for symptomatic management of moderate to very severe COPD because they are more efficacious and convenient than short-acting bronchodilators.1 In patients with severe to very severe COPD, combining a longacting bronchodilator with an inhaled glucocorticosteroid can improve health status by reducing exacerbation frequency.1
journal.publications.chestnet.org
DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
Adherence to and compliance with LABA treatment could be improved by once-daily dosing, for which extended duration of activity is needed.2 Vilanterol (GW642444M) (VI) is a novel LABA with inherent 24-h activity currently in development for use as a once-daily treatment combined with an inhaled corticosteroid (ICS) for asthma and COPD and as a oncedaily treatment in combination with a long-acting muscarinic antagonist for COPD. Preclinical models with VI show that this LABA has a significantly faster CHEST / 142 / 1 / JULY 2012
119
onset of action than salmeterol3 and a longer duration of action than either salmeterol or formoterol.3,4 In addition, VI also has significantly greater selectivity for the b2 adrenoceptor than either salmeterol or formoterol.5 Results presented in congress abstracts from studies in healthy volunteers6 and patients with asthma7 or COPD8 have demonstrated that VI is well tolerated with a favorable safety profile. Furthermore, single doses of VI produce a rapid (5 min) increase in FEV1, an effect that is maintained over 24 h.8 The primary objectives of this study were to evaluate the dose response, efficacy, and safety of five doses of VI (3, 6.25, 12.5, 25, and 50 mg) in patients with moderate to severe COPD to determine the optimal dose(s) for further clinical development. Preliminary results have been presented in abstract form.9,10 Materials and Methods Study Population Men and women aged 40 to 80 years with a history of COPD and ⱖ 10 pack-years of smoking were enrolled. At screening, eligible patients had a measured post-albuterol/salbutamol FEV1/FVC ratio of ⱕ 0.70 and an FEV1 of ⱖ 35% and ⱕ 70% of predicted. Medications not permitted during the study are detailed in e-Table 1. Reasons for exclusion included current diagnosis of asthma, a1-antitrypsin deficiency, and receipt of long-term oxygen therapy. Complete inclusion/exclusion criteria are detailed in e-Appendix 1. All participants gave written informed consent. The study was approved by local ethics review committees (e-Table 2) and conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Study Design and Treatments This was a phase IIb, multicenter, randomized, placebo-controlled, double-blind, parallel-group study (GlaxoSmithKline study number: B2C111045; www.clinicaltrials.gov registration number: NCT00606684) conducted at 89 centers from February 2008 to October 2008. Following screening (visit 1) and a 2-week singleManuscript received September 2, 2011; revision accepted December 9, 2011. Affiliations: From the Section of Pulmonary and Critical Care (Dr Hanania), Baylor College of Medicine, Houston, TX; S. Carolina Pharmaceutical Research (Dr Feldman), Spartanburg, SC; Pneumologisches Forschungsinstitut Hohegeest (Dr Zachgo), Geesthacht, Germany; the Department of Pulmonology, Allergy, and Critical Care Medicine (Dr Shim), Guro Hospital, Seoul, South Korea; the Respiratory and Medicines Development Center (Drs Crim and Barnhart), GlaxoSmithKline, Research Triangle Park, NC; and the Respiratory and Medicines Development Centre (Ms Sanford and Drs Lettis and Haumann), GlaxoSmithKline, Uxbridge, England. Funding/Support: This study was funded by GlaxoSmithKline. Correspondence to: Nicola A. Hanania, MD, FCCP, Section of Pulmonary and Critical Care, Baylor College of Medicine, 1504 Taub Loop, Houston, TX 77030; e-mail: [email protected] © 2012 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.11-2231 120 Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
blind placebo run-in period, eligible patients entered a 28-day treatment period. On day 1/visit 2, patients were randomized to one of six treatments (VI 3, 6.25, 12.5, 25, and 50 mg or placebo) given once daily in the morning via a novel dual-strip dry powder inhaler, with predose and 0- to 12-h postdose serial spirometry performed. The dual-strip inhaler contained VI in one strip and placebo in the second, or both strips contained placebo. Patients returned to the clinic the following morning (day 2/visit 3) to capture trough (predose) FEV1. On day 7/visit 4, patients were monitored for compliance, adverse events (AEs), and other safety measurements. On day 14/visit 5, spirometry was performed predose and 0 to 4 h postdose. Predose and 0- to 12-h postdose spirometric evaluations were performed on treatment day 28/visit 6, with trough FEV1 measured the following morning (day 29/visit 7). AEs and concomitant medication were monitored at each visit. BP and pulse rate were assessed predose at each visit and 10 min, 2 h, and 4 h postdose on days 1, 14, and 28. The randomization schedule was generated by the sponsor using a validated computerized system (RandAll; block size 12). Patients were randomized (ratio 1:1:1:1:1:1) using Registration and Medication Ordering System (RAMOS). Further details of the randomization method used are described in e-Appendix 2. At randomization, patients were stratified by reversibility to albuterol/salbutamol (FEV1 ⱖ 12% and ⱖ 200 mL). Investigators and study participants were blinded to study medication by use of identical inhaler devices. Compliance was assessed by reviewing the dose counter on the inhaler. It was possible, however, for the patient to advance the counter without actually inhaling a dose. This was not checked and the device was not opened to measure any unused drug. Outcome Measurements The primary efficacy end point was change from baseline in clinic visit trough (prebronchodilator and predose) FEV1 at the end of the 28-day treatment period (trough FEV1 on day 29). Trough FEV1 was defined as the mean of the FEV1 values obtained 23 and 24 h postdose. Baseline FEV1 was defined as the mean of two assessments made 30 min and immediately predose on day 1. Secondary efficacy end points were 0- to 24-h weighted mean FEV1 on days 1 and 28, time to an increase of ⱖ 100 mL above baseline in FEV1 on day 1 (0-4 h), and time to an increase of ⱖ 12% above baseline in FEV1 on day 1 (0-4 h). Other end points included serial FEV1 on days 1 and 28 (0-24 h); serial FVC on days 1, 14, and 28 (0-24 h); percentage of rescue-free 24-h periods; and mean rescue use during the entire 28-day treatment period (see e-Appendix 3 for further detail and all other end points). Safety Evaluation Safety was evaluated based on AEs and serious AEs (SAEs) (coded using Medical Dictionary for Regulatory Activities), vital signs (including pulse rate and BP), ECG assessment, blood glucose and potassium levels, clinical laboratory tests, and incidence of COPD exacerbations. QT intervals corrected for heart rate were calculated by Fridericia formula (QTcF) and Basset formula. Statistical Analysis This study was designed to have 90% power to detect a treatment difference of 130 mL in change from baseline in trough FEV1 between a VI dose and placebo. Assuming a standard deviation of 250 mL and significance at the two-sided 5% level, a sample size of at least 80 evaluable patients per arm was required (it was planned that 480 patients would be randomized in total). All primary analyses were performed on the intent-to-treat (ITT) population, which comprised all patients randomized to treatment and who had received at least one dose of study medication. Original Research
The primary analysis of the primary end point was performed using an analysis of covariance (ANCOVA) model with covariates of baseline trough FEV1, reversibility stratum, sex, age, smoking status at screening, and study treatment. Missing data were imputed (where possible) using last observation carried forward (LOCF). LOCF trough FEV1 on day 29 was also assessed using a Bayesian analysis with a noninformative prior distribution to estimate the posterior probability distributions for the adjusted mean treatment differences against placebo. Details of other statistical analyses used are in e-Appendix 4.
Results Study Population A total of 605 patients were randomized; the ITT population comprised 602 patients, and 539 (90%)
completed the study (Fig 1). The primary reasons for withdrawing early from the study were protocol deviation (n 5 20), AE (n 5 12), investigator discretion (n 5 12), or reaching protocol-defined withdrawal criteria (n 5 10). Baseline demographic and clinical characteristics were similar across treatment groups (Table 1). The majority of patients in each treatment group were male, white, and graded 2 or 3 on the Modified Medical Research Council dyspnea scale (for scale definition, see Table 1 footnote). Mean treatment compliance during the study was ⱖ 99% in each treatment group, with at least 50% of patients in each treatment group being 100% compliant. Eight patients (1%) were , 80% complaint and 20 patients (3%) were ⱖ 110% compliant.
Figure 1. Patient disposition. *One patient not randomized but took study drug (considered as a withdrawal, not a run-in failure). †Three patients excluded from the ITT population due to not taking any doses of study medication. ITT 5 intent-to-treat; VI 5 vilanterol. journal.publications.chestnet.org
DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
CHEST / 142 / 1 / JULY 2012
121
Table 1—Patient Demographics and Baseline Clinical Characteristics (ITT Population) VI Demographic/Clinical Parameter Male sex Age, y BMI,a kg/m2 Race White Asian African American American Indian or Alaska Native Mixed Tobacco history Current smoker Former smoker Smoking pack-yc mMRC Scale graded 1 2 3 4 5 FEV1 % predicted FEV1/FVC, % Reversibility in FEV1, % Reversibility stratum,e Reversible ICS use at baseline
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
Total (N 5 602)
57 (56) 61.6 ⫾ 8.53 28.51 ⫾ 6.028
68 (69) 61.1 ⫾ 8.57 27.62 ⫾ 7.119
64 (63) 62.0 ⫾ 7.94 27.23 b⫾ 6.693
57 (56) 62.6 ⫾ 8.03 27.44 ⫾ 6.833
59 (58) 62.6 ⫾ 8.88 28.01 ⫾ 6.872
65 (66) 61.4 ⫾ 8.12 27.27 ⫾ 6.719
370 (61) 61.9 ⫾ 8.34 27.68 ⫾ 6.705
90 (89) 7 (7) 3 (3) 0
84 (85) 11 (11) 1 (1) 2 (2)
84 (83) 13 (13) 3 (3) 0
86 (85) 9 (9) 2 (2) 4 (4)
84 (83) 10 (10) 4 (4) 2 (2)
87 (88) 6 (6) 3 (3) 3 (3)
515 (86) 56 (9) 16 (3) 11 (2)
1 (, 1)
1 (1)
1 (, 1)
0
46 (46) 55 (54) 47.7 ⫾ 28.25
50 (51) 49 (49) 52.4 ⫾ 32.43
48 (48) 53 (52) 49.6 ⫾ 29.66
52 (51) 49 (49) 49.2 ⫾ 28.66
45 (45) 56 (55) 49.5 ⫾ 29.12
44 (44) 55 (56) 43.1 ⫾ 20.58
285 (47) 317 (53) 48.6 ⫾ 28.38
9 (9) 37 (37) 34 (34) 17 (17) 4 (4) 51.5 ⫾ 10.67 52.5 ⫾ 8.66 16.5 ⫾ 15.08
7 (7) 43 (43) 30 (30) 18 (18) 1 (1) 50.3 ⫾ 10.14 51.4 ⫾ 10.17 15.1 ⫾ 13.01
7 (7) 48 (48) 25 (25) 14 (14) 7 (7) 49.9 ⫾ 8.57 49.9 ⫾ 8.33 12.8 ⫾ 10.86
8 (8) 37 (37) 39 (39) 13 (13) 4 (4) 50.2 ⫾ 9.20 51.0 ⫾ 9.12 15.8 ⫾ 14.81
8 (8) 32 (32) 42 (42) 19 (19) 0 49.9 ⫾ 10.44 50.1 ⫾ 10.34 18.1 ⫾ 16.45
12 (12) 39 (39) 30 (30) 18 (18) 0 51.0 ⫾ 9.64 52.6 ⫾ 9.58 13.3 ⫾ 14.78
51 (8) 236 (39) 200 (33) 99 (16) 16 (3) 50.5 ⫾ 9.78 51.2 ⫾ 9.41 15.3 ⫾ 14.34
39 (39) 33 (33)
36 (36) 33 (33)
34 (34) 30 (30)
37 (37) 23 (23)
34 (34) 30 (30)
34 (34) 36 (36)
1 (, 1)
0
4 (, 1)
214 (36) 185 (31)
Data are presented as No. (%) or mean ⫾ SD. ICS 5 inhaled corticosteroid; ITT 5 intent-to-treat; mMRC 5 Modified Medical Research Council Dyspnea Scale; VI 5 vilanterol. aBMI categories: underweight (, 18.5 kg/m2), normal (18.5 kg/m2 to , 25 kg/m2), overweight (25 kg/m2 to , 30 kg/m2), and obese (ⱖ 30 kg/m2). bn 5 100. cSmoking pack-years 5 (number of cigarettes smoked per day/20) 3 number of years smoked. dmMRC scale grade: 1 5 not troubled with breathlessness except with strenuous exercise; 2 5 troubled by shortness of breath when hurrying on level ground/walking up slight hill; 3 5 walks slower than others of same age on level ground because of breathlessness; 4 5 stops to breathe after walking about 100 yards/meters, or after few minutes on level ground; 5 5 too breathless to leave house or breathless when dressing or undressing. eReversibility to albuterol/salbutamol, defined as FEV ⱖ 12% and ⱖ 200 mL following albuterol/salbutamol inhalation, recalculated for analysis 1 purposes using screening lung function data.
Efficacy Measures Primary End Point: The adjusted mean change from baseline in trough FEV1 on day 29 (ITT population, using LOCF) was significantly increased for all doses (3, 6.25, 12.5, 25, and 50 mg) of VI compared with placebo (P , .001 for all doses) (Fig 2A, Table 2). There were no statistically significant interactions between concurrent use of ICS at baseline (P . .10) or recalculated reversibility stratum (P . .10) and the effect of treatment on trough FEV1; however, the study was not powered for these comparisons. Posterior probabilities for a true . 100 mL increase were . 90% with both the 25-mg (92%) and 50-mg (99%) VI doses but were lower for the 3-mg (37%), 6.25-mg (47%), and 12.5-mg (64%) doses; probabilities for a true . 130-mL increase were 90%, 61%, 122 Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
22%, 11%, and 7% with the 50-, 25-, 12.5-, 6.25-, and 3-mg doses, respectively (Fig 2B). Secondary and Other End Points: Adjusted mean change from baseline in 0- to 24-h weighted mean FEV1 values on days 1 and 28 demonstrated statistically significant (P ⱕ .003) differences for all doses of VI compared with placebo (Table 3). Weighted mean differences vs placebo of ⱖ 130 mL were observed with the 25-mg and 50-mg doses on days 1 and 28, and with the 12.5-mg dose on day 28. Time to achieve a ⱖ 100-mL increase from baseline in FEV1 over the first 4-h postdose on day 1 was significantly shorter for each VI dose compared with placebo (P , .001 for each dose) and shortest in the 25- and 50-mg dose groups (median 6 min for both groups) (Table 4). Similarly, time to achieve a ⱖ 12% Original Research
(Fig 3). Serial FVC results are provided in e-Appendix 5. Over days 1 to 28, greater increases from baseline in the mean percentage of rescue-free 24-h periods were observed with VI (12%-16%) compared with placebo (3%). Statistically significant decreases in adjusted mean change from baseline daily albuterol/salbutamol use (occasions) were observed for all doses (P ⱕ .035) except for 6.25 mg VI (P 5 .84). Safety
Figure 2. A, Adjusted mean change from placebo in trough FEV1 at day 29 (ITT population). B, Posterior probability distribution of the treatment difference in change from baseline (predose, day 1) in trough FEV1 on day 29 (ITT population). Analysis performed using analysis of covariance with covariates of baseline, sex, age, smoking status (at screening), reversibility stratum, and treatment. Error bars are 95% CIs. See Figure 1 legend for expansion of abbreviations.
increase from baseline in FEV1 over the first 4-h postdose on day 1 was significantly shorter for each VI dose group compared with placebo (P , .001 for each dose) and was shortest with 25-mg VI and 50-mg VI (median 18 min and 16 min, respectively) (Table 4). Adjusted mean changes from baseline from repeated measures analysis of serial FEV1 over time on days 1 and 28 showed a distinct dose-response effect and indicated a 24-h duration of effect for all doses of VI
Once-daily administration of VI 3, 6.25, 12.5, 25, and 50 mg was well tolerated over 28 days of treatment, with no adverse signals with increasing dose. The overall incidence of on-treatment AEs in the VI treatment groups (24%-33%) was lower than in the placebo group (36%), with headache as the most commonly reported AE (Table 5). The incidence of drug-related AEs was similar across the VI groups (5%-7%) and lower than for placebo (10%) (Table 5). Increased blood potassium and blood glucose were the two most common drug-related AEs, although the incidence of these events was the same as or less than placebo. Four patients experienced nonfatal SAEs during treatment (3 mg VI, vasovagal syncope; 6.25 mg VI, aortic aneurysm; 12.5 mg VI, atrial fibrillation; 12.5 mg VI, COPD exacerbation and pneumonia); none was considered drug related. No fatal on-treatment SAEs were reported. Four patients experienced posttreatment SAEs, of which COPD exacerbation (3 mg VI) and hyperkalemia (50 mg VI) were considered drug related. In total, 13 patients were withdrawn due to an AE: three in the placebo group and three, four, two, zero, and one in the VI 3-, 6.25-, 12.5-, 25-, and 50-mg groups, respectively.
Table 2—Change From Baseline (Predose, Day 1) in Trough FEV1 on Day 29 (LOCF) (ANCOVA; ITT Population) VI Trough FEV1, L
Placebo (n 5 101)
3 mg (n 5 99)
101 1.255 (0.4672) 0.51, 2.48
99 1.299 (0.4591) 0.56, 2.93
101 1.242 (0.4307) 0.42, 2.56
100 1.222 (0.4265) 0.53, 2.35
100 1.182 (0.4832) 0.48, 2.87
99 1.330 (0.4873) 0.61, 2.60
101 1.284 0.029 (0.0188) … … …
99 1.375 0.120 (0.0190) 0.092 0.039-0.144 , .001
100 1.382 0.127 (0.0188) 0.098 0.046-0.150 , .001
99 1.393 0.138 (0.0190) 0.110 0.057-0.162 , .001
99 1.421 0.166 (0.0190) 0.137 0.085-0.190 , .001
99 1.449 0.194 (0.0190) 0.165 0.112-0.217 , .001
Baseline No. Mean (SD) Min, max Day 29 No. LS mean LS mean change (SE)a Difference vs placebo 95% CI P value
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
Analysis performed using ANCOVA with covariates of baseline, sex, age, smoking status (at screening), reversibility stratum, and treatment. ANCOVA 5 analysis of covariance; LOCF 5 last observation carried forward. LS 5 least squares; max 5 maximum; min 5 minimum. See Table 1 legend for expansion of abbreviations. aSE for both LS mean and LS mean change from baseline. journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
CHEST / 142 / 1 / JULY 2012
123
Table 3—Change From Baseline 0- to 24-h Weighted Mean FEV1 on Days 1 and 28 (ITT Population) VI 0–24 h Weighted Mean FEV1, L Day 1 No.a No.b LS mean LS mean change (SE)c Difference vs placebo 95% CI P value Day 28 No.a No.b LS mean LS mean change (SE)c Difference vs placebo 95% CI P value
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
101 100 1.283 0.028 (0.0135) … … …
99 97 1.340 0.085 (0.0137) 0.057 0.019-0.095 .003
100 100 1.387 0.132 (0.0135) 0.104 0.066-0.141 , .001
99 99 1.404 0.149 (0.0136) 0.120 0.083-0.158 , .001
99 99 1.434 0.178 (0.0136) 0.150 0.112-0.188 , .001
99 97 1.458 0.202 (0.0137) 0.174 0.136-0.212 , .001
101 84 1.265 0.010 (0.0189) … … …
99 88 1.369 0.114 (0.0187) 0.105 0.052-0.157 , .001
100 91 1.390 0.135 (0.0185) 0.125 0.073-0.177 , .001
99 92 1.407 0.152 (0.0185) 0.142 0.090-0.194 , .001
99 92 1.423 0.168 (0.0185) 0.158 0.106-0.210 , .001
99 91 1.441 0.186 (0.0186) 0.177 0.125-0.229 , .001
Analysis performed using mixed model repeated measures with covariates of baseline, sex, age, smoking status (at screening), reversibility stratum, day (nominal), treatment, and day by treatment and day by baseline interactions. Baseline is FEV1 predose on day 1. See Table 1 and 2 legends for expansion of abbreviations. aNo. of patients with analyzable data on 1 or more days. bNo. of patients with analyzable data on the given day. cSE for both LS mean and LS mean change from baseline.
No suggestions of treatment- or dose-related effects on other laboratory analyses were observed, except for a slight increase in eosinophils in the active treatment groups. Five patients across the treatment groups (two placebo, one 3-mg VI, and two 12.5-mg VI) were identified with elevated total bilirubin, alanine transaminase, or aspartate transaminase values, which returned to below levels of clinical concern (1.5 and 3 times upper limit of normal for total bilirubin and
alanine transaminase/aspartate transaminase, respectively) during or by the end of the study. No suggestions of treatment or dose-related effects on blood pressure or pulse rate were observed and the incidence of prolonged (. 450 milliseconds) QTcF interval was low and similar across treatment groups (ⱕ 3%). The overall incidence of COPD exacerbations was low (0%-5% in the VI groups and 3% in the placebo group).
Table 4—Log-Rank Analysis of Time to Increases of ⱖ 100 mL or ⱖ 12% From Baseline FEV1 (0-4 h Postdose) VI Day 1 (0–4 h Postdose) ⱖ 100 mL increase from baseline FEV1 No. No. events No. censored Median time, min P value ⱖ 12% increase from baseline FEV1 No. No. events No. censored Median time, min P value
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
101 42 (42) 59 (58) NDa …
99 73 (74) 26 (26) 32 , .001
101 80 (79) 21 (21) 16 , .001
100 83 (83) 17 (17) 16 , .001
100 89 (89) 11 (11) 6 , .001
99 91 (92) 8 (8) 6 , .001
101 27 (27) 74 (73) NDa …
99 63 (64) 36 (36) 120 , .001
101 73 (72) 28 (28) 30 , .001
100 68 (68) 32 (32) 30 , .001
100 81 (81) 19 (19) 18 , .001
99 80 (81) 19 (19) 16 , .001
ITT population stratified by reversibility (reversible, nonreversible). ND 5 not defined. See Table 1 legend for expansion of other abbreviations. aIf . 50% of patients are censored, median time is ND. 124 Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
Original Research
Figure 3. A, Repeated measures adjusted treatment differences from baseline in serial FEV1 on day 1 (ITT population). B, Repeated measures adjusted treatment differences from baseline in serial FEV1 on day 28 (ITT population). See Figure 1 legend for expansion of abbreviation.
Discussion Our study demonstrates that the novel long-acting b2 agonist VI is effective and well tolerated in patients with COPD. Treatment with VI once daily at doses ranging from 3 mg to 50 mg for 28 days produced statistically significant, dose-dependent improvements in trough FEV1 (primary outcome measure) compared with placebo. An improvement from baseline FEV1 of ⱖ 100 mL represents a clinically relevant treatment difference.11 Adjusted mean treatment differences of ⱖ 100 mL vs placebo in both trough and 0- to 24-h weighted mean FEV1 was consistently observed with VI 12.5-, 25-, and 50-mg doses, whereas a sustained 24-h duration of action was demonstrated at all doses of VI throughout the 28-day treatment period. The Bayesian analysis of trough FEV1 demonstrated that the 25- and 50-mg groups had a higher probability (92% and 99%, respectively) that the true treatment difference vs placebo would be . 100 mL compared with the 3-, 6.25-, and 12.5-mg groups (37%, 47%, and 64%, respectively). The absence of clinically relevant safety signals with increasing dose, including very low incidence of treatment-related AEs/SAEs with no suggestion of an effect on vital signs, QTcF interval, or blood glucose and potassium levels, indicates the favorable safety profile of VI. journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
Although five patients who received VI had increased potassium (considered to be drug related), the incidence of these AEs was the same as or less than placebo and would not be expected to be related to LABA therapy, which typically results in hypokalemia at high doses. The 24-h duration of action of VI is supported by repeated measures analysis of serial FEV1, which showed increasing adjusted mean treatment differences from placebo with increasing VI dose. Although we did observe an apparent improvement in FEV1 with placebo at 24 h, this is most likely to be due to diurnal variability as a similar effect was observed in each VI treatment group. The 24-h duration of effect of VI is further supported by the significant increases from baseline in rescue-free occasions, with all VI doses except 6.25 mg. Although the efficacy and duration of bronchodilator effect of a therapy are important, so is the speed with which a clinically meaningful bronchodilator effect is achieved. VI was shown in this study to have a rapid onset of action (median 6 min for the 25- and 50-mg doses). Collectively, the findings demonstrate that VI is an effective, well-tolerated, and fast-acting LABA with a true 24-h bronchodilator effect. This clinical profile of VI is particularly relevant considering the poor compliance with existing inhaled bronchodilators, CHEST / 142 / 1 / JULY 2012
125
Table 5—Most Common On-Treatment AEs (ⱖ 3% Incidence in Any Treatment Group) and Drug-Related AEs (Occurring in . 1 Patient On Treatment or Posttreatment) VI AEs
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
Any AE Headache Nausea Nasopharyngitis Blood potassium increaseda Blood glucose increased Diarrhea Ventricular extrasystoles Nasal congestion Oropharyngeal pain Any drug-related AE Blood potassium increasedb Blood glucose increased Headache Palpitations Ventricular extrasystoles Nausea Pruritus Tremor
36 (36) 10 (10) 4 (4) 3 (3) 3 (3) 3 (3) 1 (, 1) 2 (2) 3 (3) 0 10 (10) 2 (2) 3 (3) 3 (3) 1 (, 1) 0 1 (, 1) 0 0
24 (24) 6 (6) 1 (1) 2 (2) 0 0 2 (2) 0 0 0 5 (5) 0 0 0 0 0 1 (1) 0 1 (1)
32 (32) 5 (5) 3 (3) 5 (5) 1 (, 1) 1 (, 1) 1 (, 1) 1 (, 1) 2 (2) 3 (3) 5 (5) 1 (, 1) 1 (, 1) 0 0 1 (, 1) 0 0 1 (, 1)
24 (24) 3 (3) 2 (2) 0 2 (2) 3 (3) 1 (, 1) 0 0 0 5 (5) 1 (, 1) 1 (, 1) 0 0 0 0 1 (, 1) 0
33 (33) 3 (3) 2 (2) 1 (, 1) 2 (2) 1 (, 1) 3 (3) 0 0 0 5 (5) 1 (, 1) 1 (, 1) 0 1 (, 1) 0 0 1 (, 1) 0
28 (28) 7 (7) 1 (1) 0 2 (2) 0 0 3 (3) 0 1 (1) 7 (7) 2 (2) 0 0 0 1 (1) 0 0 0
ITT population. Data are presented as No. (%). AE 5 adverse event. See Table 1 legend for expansion of other abbreviation. aAdditionally, the AE term of hyperkalemia was reported for one patient in the 3-mg group and one patient in the 12.5-mg group. bAdditionally, the AE term of hyperkalemia was reported for one patient in the 3-mg group and one patient in the 50-mg group.
which leads to inadequate treatment of symptoms and may be due to the dosing regimen (eg, bid) and onset of action. Although the currently marketed LABAs (salmeterol and formoterol) are effective and both have a relatively quick onset of action in the management of the symptoms of COPD,1,12 they have to be administered twice daily. In this regard, the convenience of once-daily treatment with VI may lead to increased patient acceptance, which could improve adherence to therapy, a potential benefit that needs further exploration in future studies. Treatment compliance during this study was ⱖ 99%, with only a small proportion of patients (5%) underusing or overusing their assigned medication. This study has several strengths beyond the fact that it investigated a novel LABA that can be administered once daily. The study included a large number of patients, examined the efficacy of the drug at different doses, and included a comprehensive safety evaluation. A potential limitation of the study is the relatively short treatment period (28 days), and thus long-term studies to examine the safety of VI are needed. In summary, this study demonstrates that oncedaily administration of VI in patients with moderate to severe COPD provides clinically relevant 24-h improvement in lung function with a rapid onset of effect and a safety and tolerability profile comparable with placebo. The 25-mg and 50-mg doses of VI 126 DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
appear to offer the greatest clinical benefit without any safety concerns. Further studies are warranted to evaluate the safety and efficacy of VI as part of a combination ICS/LABA therapy in COPD. Acknowledgments Author contributions: All authors vouch for the accuracy and completeness of the data and the data analysis. Dr Hanania: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Feldman: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Zachgo: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Shim: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Crim: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Ms Sanford: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Dr Lettis: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Dr Barnhart: contributed to developing the design and concept of the study, was the clinical study investigation lead, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Dr Haumann: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Hanania has received fees from GlaxoSmithKline for participation in the speakers bureau and on the National Advisory Board of GlaxoSmithKline; Original Research
his institution has received remuneration for participation in clinical trials sponsored by GlaxoSmithKline. Dr Feldman has received funds as the principle investigator in a clinical trial sponsored by GlaxoSmithKline, which was administered by his employer, S. Carolina Pharmaceutical Research. Dr Zachgo has received lecture fees from GlaxoSmithKline; he has participated in clinical research studies sponsored by GlaxoSmithKline, Novartis AG, Boehringer Ingelheim, Nycomed, and AstraZeneca. Dr Shim has received funds for his participation in a clinical trial sponsored by GlaxoSmithKline, which was administered by his employer, Guro Hospital, Seoul, South Korea. Drs Crim, Lettis, Barnhart, and Haumann and Ms Sanford are employees of and hold stocks in GlaxoSmithKline. Role of sponsors: The charge for color figures was paid for by GlaxoSmithKline. Editorial support in the form of development of draft outline, development of manuscript first draft, editorial suggestions to draft versions of this paper, assembling tables and figures, collating author comments, copyediting, fact checking, referencing and graphic services was provided by David Cutler at Gardiner-Caldwell Communications and was funded by GlaxoSmithKline. Other contributions: We thank M. Patel, MSc; S. Kilbride, MSc; and J. Brooks, MSc (GlaxoSmithKline Research and Development, Uxbridge, England) for their assistance with programming the statistical analyses. Additional information: The e-Appendix and e-Tables can be found in the “Supplemental Materials” area of the online article.
References 1. Global Initiative for Chronic Obstructive Lung Disease guideline: global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease (updated 2009). Global Initiative for Chronic Obstructive Lung Disease website. http://www.goldcopd.com/. Accessed April 13, 2011. 2. Cazzola M, Matera MG. Novel long-acting bronchodilators for COPD and asthma. Br J Pharmacol. 2008;155(3):291-299. 3. Morrison V, Sturton G, Barrett V, Ford A, Knowles R. Pharmacological characterisation of GW642444, a long-acting b2-agonist (LABA) with rapid onset and long duration, on
journal.publications.chestnet.org
DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
4.
5.
6.
7.
8.
9. 10. 11. 12.
isolated large and small human airways [abstract]. Am J Respir Crit Care Med. 2010;181:A4453. Ford A, Hughes S, Morrison V, Barrett V, Knowles R. In vitro and in vivo pharmacological characterisation of GW642444, a novel long-acting b2 agonist (LABA), with fast onset and long duration in the guinea-pig [abstract]. Am J Respir Crit Care Med. 2010;181:A5677. Barrett V, Emmons A, Ford A, Knowles R. In vitro pharmacological characterisation of GW642444, a novel long acting b2-agonist (LABA) using human recombinant b1/2/3 adrenoceptor cAMP assays [abstract]. Am J Respir Crit Care Med. 2010;181:A4451. Kempsford RD, Norris V, Siederer SK. The pharmacodynamics, pharmacokinetics and tolerability of repeat doses of the novel inhaled long-acting b 2 adrenoceptor agonist (LABA) GW642444 (25, 50 and 100mcg) in healthy subjects [abstract]. Am J Respir Crit Care Med. 2010;181:A4461. Kempsford RD, Norris V, Siederer SK. GW642444, a novel inhaled long-acting beta2 adrenoceptor agonist (LABA), at single doses of 25, 50 and 100mcg, is well tolerated and demonstrates prolonged bronchodilation in COPD subjects [abstract]. Am J Respir Crit Care Med. 2010;181:A4447. Lötvall J, Bateman ED, Bleecker ER, et al. Dose-related efficacy of vilanterol trifenatate (VI), a long-acting beta2 agonist (LABA) with inherent 24-hour activity, in subjects with persistent asthma [abstract]. Eur Respir J. 2010; 36(suppl 54):1013s. Hanania NA, Feldman G, Zachgo W, et al. Dose-related efficacy of vilanterol trifenatate (VI) in COPD [abstract]. Eur Respir J. 2010;36(suppl 54):217s. Hanania NA, Feldman G, Zachgo W, et al. Safety of vilanterol trifenatate (VI) in a COPD dose-ranging study [abstract]. Eur Respir J. 2010;36(suppl 54):208s. Donohue JF. Minimal clinically important differences in COPD lung function. COPD. 2005;2(1):111-124. Steiropoulos P, Tzouvelekis A, Bouros D. Formoterol in the management of chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis. 2008;3(2):205-215.
CHEST / 142 / 1 / JULY 2012
127
Original Research COPD
The Efficacy and Safety of the Novel Long-Acting b2 Agonist Vilanterol in Patients With COPD A Randomized Placebo-Controlled Trial Nicola A. Hanania, MD, FCCP; Gregory Feldman, MD; Wolfgang Zachgo, MD; Jae-Jeong Shim, MD, PhD; Courtney Crim, MD, FCCP; Lisa Sanford, MSc; Sally Lettis, PhD; Frank Barnhart, DVM; and Brett Haumann, MD
Background: Vilanterol (GW642444M) (VI) is a novel, inhaled, long-acting b2 agonist with inherent 24-h activity under development as a once-daily combination therapy with an inhaled corticosteroid for COPD and asthma. This study assessed the dose response, efficacy, and safety of VI at doses of 3 to 50 mg in patients with moderate to severe COPD. Methods: Six hundred two patients (intent-to-treat) were randomized (double-blind) to VI 3, 6.25, 12.5, 25, or 50 mg or placebo once daily for 28 days. The primary end point was change from baseline in trough FEV1 at the end of the 28-day treatment period. Secondary end points included 0- to 24-h weighted mean FEV1 on days 1 and 28 and time to increases of ⱖ 100 mL or ⱖ 12% from baseline FEV1 on day 1. Safety assessments included adverse events, vital signs, ECG assessment, and clinical laboratory tests. Results: VI once daily for 28 days significantly improved trough FEV1 in a dose-dependent manner vs placebo. Clinically relevant treatment differences of ⱖ 130 mL in trough and 0- to 24-h weighted mean FEV1 were observed with VI 25- and 50-mg doses vs placebo. All doses of VI were associated with a low incidence of treatment-related adverse events/serious adverse events, with no suggestion of effects on BP, pulse rate, QT intervals corrected for heart rate calculated by Fridericia formula, or blood glucose and potassium levels. Conclusions: VI 25 and 50 mg once daily provided both statistically and clinically relevant 24-h improvements in lung function in patients with COPD compared with placebo. All doses of VI had a safety and tolerability profile similar to placebo. Trial registry: ClinicalTrials.gov; No.: NCT00606684; URL: www.clinicaltrials.gov CHEST 2012; 142(1):119–127 Abbreviations: AE 5 adverse event; ANCOVA 5 analysis of covariance; ICS 5 inhaled corticosteroid; ITT 5 intentto-treat; LABA 5 long-acting b2 agonist; LOCF 5 last observation carried forward; QTcF 5 Fridericia formula; SAE 5 serious adverse event; VI 5 vilanterol (GW642444M)
inhaled bronchodilators, comprising Long-acting long-acting b agonists (LABAs) and long-acting 2
muscarinic antagonists, are recommended for symptomatic management of moderate to very severe COPD because they are more efficacious and convenient than short-acting bronchodilators.1 In patients with severe to very severe COPD, combining a longacting bronchodilator with an inhaled glucocorticosteroid can improve health status by reducing exacerbation frequency.1
journal.publications.chestnet.org
DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
Adherence to and compliance with LABA treatment could be improved by once-daily dosing, for which extended duration of activity is needed.2 Vilanterol (GW642444M) (VI) is a novel LABA with inherent 24-h activity currently in development for use as a once-daily treatment combined with an inhaled corticosteroid (ICS) for asthma and COPD and as a oncedaily treatment in combination with a long-acting muscarinic antagonist for COPD. Preclinical models with VI show that this LABA has a significantly faster CHEST / 142 / 1 / JULY 2012
119
onset of action than salmeterol3 and a longer duration of action than either salmeterol or formoterol.3,4 In addition, VI also has significantly greater selectivity for the b2 adrenoceptor than either salmeterol or formoterol.5 Results presented in congress abstracts from studies in healthy volunteers6 and patients with asthma7 or COPD8 have demonstrated that VI is well tolerated with a favorable safety profile. Furthermore, single doses of VI produce a rapid (5 min) increase in FEV1, an effect that is maintained over 24 h.8 The primary objectives of this study were to evaluate the dose response, efficacy, and safety of five doses of VI (3, 6.25, 12.5, 25, and 50 mg) in patients with moderate to severe COPD to determine the optimal dose(s) for further clinical development. Preliminary results have been presented in abstract form.9,10 Materials and Methods Study Population Men and women aged 40 to 80 years with a history of COPD and ⱖ 10 pack-years of smoking were enrolled. At screening, eligible patients had a measured post-albuterol/salbutamol FEV1/FVC ratio of ⱕ 0.70 and an FEV1 of ⱖ 35% and ⱕ 70% of predicted. Medications not permitted during the study are detailed in e-Table 1. Reasons for exclusion included current diagnosis of asthma, a1-antitrypsin deficiency, and receipt of long-term oxygen therapy. Complete inclusion/exclusion criteria are detailed in e-Appendix 1. All participants gave written informed consent. The study was approved by local ethics review committees (e-Table 2) and conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Study Design and Treatments This was a phase IIb, multicenter, randomized, placebo-controlled, double-blind, parallel-group study (GlaxoSmithKline study number: B2C111045; www.clinicaltrials.gov registration number: NCT00606684) conducted at 89 centers from February 2008 to October 2008. Following screening (visit 1) and a 2-week singleManuscript received September 2, 2011; revision accepted December 9, 2011. Affiliations: From the Section of Pulmonary and Critical Care (Dr Hanania), Baylor College of Medicine, Houston, TX; S. Carolina Pharmaceutical Research (Dr Feldman), Spartanburg, SC; Pneumologisches Forschungsinstitut Hohegeest (Dr Zachgo), Geesthacht, Germany; the Department of Pulmonology, Allergy, and Critical Care Medicine (Dr Shim), Guro Hospital, Seoul, South Korea; the Respiratory and Medicines Development Center (Drs Crim and Barnhart), GlaxoSmithKline, Research Triangle Park, NC; and the Respiratory and Medicines Development Centre (Ms Sanford and Drs Lettis and Haumann), GlaxoSmithKline, Uxbridge, England. Funding/Support: This study was funded by GlaxoSmithKline. Correspondence to: Nicola A. Hanania, MD, FCCP, Section of Pulmonary and Critical Care, Baylor College of Medicine, 1504 Taub Loop, Houston, TX 77030; e-mail: [email protected] © 2012 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.11-2231 120 Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
blind placebo run-in period, eligible patients entered a 28-day treatment period. On day 1/visit 2, patients were randomized to one of six treatments (VI 3, 6.25, 12.5, 25, and 50 mg or placebo) given once daily in the morning via a novel dual-strip dry powder inhaler, with predose and 0- to 12-h postdose serial spirometry performed. The dual-strip inhaler contained VI in one strip and placebo in the second, or both strips contained placebo. Patients returned to the clinic the following morning (day 2/visit 3) to capture trough (predose) FEV1. On day 7/visit 4, patients were monitored for compliance, adverse events (AEs), and other safety measurements. On day 14/visit 5, spirometry was performed predose and 0 to 4 h postdose. Predose and 0- to 12-h postdose spirometric evaluations were performed on treatment day 28/visit 6, with trough FEV1 measured the following morning (day 29/visit 7). AEs and concomitant medication were monitored at each visit. BP and pulse rate were assessed predose at each visit and 10 min, 2 h, and 4 h postdose on days 1, 14, and 28. The randomization schedule was generated by the sponsor using a validated computerized system (RandAll; block size 12). Patients were randomized (ratio 1:1:1:1:1:1) using Registration and Medication Ordering System (RAMOS). Further details of the randomization method used are described in e-Appendix 2. At randomization, patients were stratified by reversibility to albuterol/salbutamol (FEV1 ⱖ 12% and ⱖ 200 mL). Investigators and study participants were blinded to study medication by use of identical inhaler devices. Compliance was assessed by reviewing the dose counter on the inhaler. It was possible, however, for the patient to advance the counter without actually inhaling a dose. This was not checked and the device was not opened to measure any unused drug. Outcome Measurements The primary efficacy end point was change from baseline in clinic visit trough (prebronchodilator and predose) FEV1 at the end of the 28-day treatment period (trough FEV1 on day 29). Trough FEV1 was defined as the mean of the FEV1 values obtained 23 and 24 h postdose. Baseline FEV1 was defined as the mean of two assessments made 30 min and immediately predose on day 1. Secondary efficacy end points were 0- to 24-h weighted mean FEV1 on days 1 and 28, time to an increase of ⱖ 100 mL above baseline in FEV1 on day 1 (0-4 h), and time to an increase of ⱖ 12% above baseline in FEV1 on day 1 (0-4 h). Other end points included serial FEV1 on days 1 and 28 (0-24 h); serial FVC on days 1, 14, and 28 (0-24 h); percentage of rescue-free 24-h periods; and mean rescue use during the entire 28-day treatment period (see e-Appendix 3 for further detail and all other end points). Safety Evaluation Safety was evaluated based on AEs and serious AEs (SAEs) (coded using Medical Dictionary for Regulatory Activities), vital signs (including pulse rate and BP), ECG assessment, blood glucose and potassium levels, clinical laboratory tests, and incidence of COPD exacerbations. QT intervals corrected for heart rate were calculated by Fridericia formula (QTcF) and Basset formula. Statistical Analysis This study was designed to have 90% power to detect a treatment difference of 130 mL in change from baseline in trough FEV1 between a VI dose and placebo. Assuming a standard deviation of 250 mL and significance at the two-sided 5% level, a sample size of at least 80 evaluable patients per arm was required (it was planned that 480 patients would be randomized in total). All primary analyses were performed on the intent-to-treat (ITT) population, which comprised all patients randomized to treatment and who had received at least one dose of study medication. Original Research
The primary analysis of the primary end point was performed using an analysis of covariance (ANCOVA) model with covariates of baseline trough FEV1, reversibility stratum, sex, age, smoking status at screening, and study treatment. Missing data were imputed (where possible) using last observation carried forward (LOCF). LOCF trough FEV1 on day 29 was also assessed using a Bayesian analysis with a noninformative prior distribution to estimate the posterior probability distributions for the adjusted mean treatment differences against placebo. Details of other statistical analyses used are in e-Appendix 4.
Results Study Population A total of 605 patients were randomized; the ITT population comprised 602 patients, and 539 (90%)
completed the study (Fig 1). The primary reasons for withdrawing early from the study were protocol deviation (n 5 20), AE (n 5 12), investigator discretion (n 5 12), or reaching protocol-defined withdrawal criteria (n 5 10). Baseline demographic and clinical characteristics were similar across treatment groups (Table 1). The majority of patients in each treatment group were male, white, and graded 2 or 3 on the Modified Medical Research Council dyspnea scale (for scale definition, see Table 1 footnote). Mean treatment compliance during the study was ⱖ 99% in each treatment group, with at least 50% of patients in each treatment group being 100% compliant. Eight patients (1%) were , 80% complaint and 20 patients (3%) were ⱖ 110% compliant.
Figure 1. Patient disposition. *One patient not randomized but took study drug (considered as a withdrawal, not a run-in failure). †Three patients excluded from the ITT population due to not taking any doses of study medication. ITT 5 intent-to-treat; VI 5 vilanterol. journal.publications.chestnet.org
DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
CHEST / 142 / 1 / JULY 2012
121
Table 1—Patient Demographics and Baseline Clinical Characteristics (ITT Population) VI Demographic/Clinical Parameter Male sex Age, y BMI,a kg/m2 Race White Asian African American American Indian or Alaska Native Mixed Tobacco history Current smoker Former smoker Smoking pack-yc mMRC Scale graded 1 2 3 4 5 FEV1 % predicted FEV1/FVC, % Reversibility in FEV1, % Reversibility stratum,e Reversible ICS use at baseline
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
Total (N 5 602)
57 (56) 61.6 ⫾ 8.53 28.51 ⫾ 6.028
68 (69) 61.1 ⫾ 8.57 27.62 ⫾ 7.119
64 (63) 62.0 ⫾ 7.94 27.23 b⫾ 6.693
57 (56) 62.6 ⫾ 8.03 27.44 ⫾ 6.833
59 (58) 62.6 ⫾ 8.88 28.01 ⫾ 6.872
65 (66) 61.4 ⫾ 8.12 27.27 ⫾ 6.719
370 (61) 61.9 ⫾ 8.34 27.68 ⫾ 6.705
90 (89) 7 (7) 3 (3) 0
84 (85) 11 (11) 1 (1) 2 (2)
84 (83) 13 (13) 3 (3) 0
86 (85) 9 (9) 2 (2) 4 (4)
84 (83) 10 (10) 4 (4) 2 (2)
87 (88) 6 (6) 3 (3) 3 (3)
515 (86) 56 (9) 16 (3) 11 (2)
1 (, 1)
1 (1)
1 (, 1)
0
46 (46) 55 (54) 47.7 ⫾ 28.25
50 (51) 49 (49) 52.4 ⫾ 32.43
48 (48) 53 (52) 49.6 ⫾ 29.66
52 (51) 49 (49) 49.2 ⫾ 28.66
45 (45) 56 (55) 49.5 ⫾ 29.12
44 (44) 55 (56) 43.1 ⫾ 20.58
285 (47) 317 (53) 48.6 ⫾ 28.38
9 (9) 37 (37) 34 (34) 17 (17) 4 (4) 51.5 ⫾ 10.67 52.5 ⫾ 8.66 16.5 ⫾ 15.08
7 (7) 43 (43) 30 (30) 18 (18) 1 (1) 50.3 ⫾ 10.14 51.4 ⫾ 10.17 15.1 ⫾ 13.01
7 (7) 48 (48) 25 (25) 14 (14) 7 (7) 49.9 ⫾ 8.57 49.9 ⫾ 8.33 12.8 ⫾ 10.86
8 (8) 37 (37) 39 (39) 13 (13) 4 (4) 50.2 ⫾ 9.20 51.0 ⫾ 9.12 15.8 ⫾ 14.81
8 (8) 32 (32) 42 (42) 19 (19) 0 49.9 ⫾ 10.44 50.1 ⫾ 10.34 18.1 ⫾ 16.45
12 (12) 39 (39) 30 (30) 18 (18) 0 51.0 ⫾ 9.64 52.6 ⫾ 9.58 13.3 ⫾ 14.78
51 (8) 236 (39) 200 (33) 99 (16) 16 (3) 50.5 ⫾ 9.78 51.2 ⫾ 9.41 15.3 ⫾ 14.34
39 (39) 33 (33)
36 (36) 33 (33)
34 (34) 30 (30)
37 (37) 23 (23)
34 (34) 30 (30)
34 (34) 36 (36)
1 (, 1)
0
4 (, 1)
214 (36) 185 (31)
Data are presented as No. (%) or mean ⫾ SD. ICS 5 inhaled corticosteroid; ITT 5 intent-to-treat; mMRC 5 Modified Medical Research Council Dyspnea Scale; VI 5 vilanterol. aBMI categories: underweight (, 18.5 kg/m2), normal (18.5 kg/m2 to , 25 kg/m2), overweight (25 kg/m2 to , 30 kg/m2), and obese (ⱖ 30 kg/m2). bn 5 100. cSmoking pack-years 5 (number of cigarettes smoked per day/20) 3 number of years smoked. dmMRC scale grade: 1 5 not troubled with breathlessness except with strenuous exercise; 2 5 troubled by shortness of breath when hurrying on level ground/walking up slight hill; 3 5 walks slower than others of same age on level ground because of breathlessness; 4 5 stops to breathe after walking about 100 yards/meters, or after few minutes on level ground; 5 5 too breathless to leave house or breathless when dressing or undressing. eReversibility to albuterol/salbutamol, defined as FEV ⱖ 12% and ⱖ 200 mL following albuterol/salbutamol inhalation, recalculated for analysis 1 purposes using screening lung function data.
Efficacy Measures Primary End Point: The adjusted mean change from baseline in trough FEV1 on day 29 (ITT population, using LOCF) was significantly increased for all doses (3, 6.25, 12.5, 25, and 50 mg) of VI compared with placebo (P , .001 for all doses) (Fig 2A, Table 2). There were no statistically significant interactions between concurrent use of ICS at baseline (P . .10) or recalculated reversibility stratum (P . .10) and the effect of treatment on trough FEV1; however, the study was not powered for these comparisons. Posterior probabilities for a true . 100 mL increase were . 90% with both the 25-mg (92%) and 50-mg (99%) VI doses but were lower for the 3-mg (37%), 6.25-mg (47%), and 12.5-mg (64%) doses; probabilities for a true . 130-mL increase were 90%, 61%, 122 Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
22%, 11%, and 7% with the 50-, 25-, 12.5-, 6.25-, and 3-mg doses, respectively (Fig 2B). Secondary and Other End Points: Adjusted mean change from baseline in 0- to 24-h weighted mean FEV1 values on days 1 and 28 demonstrated statistically significant (P ⱕ .003) differences for all doses of VI compared with placebo (Table 3). Weighted mean differences vs placebo of ⱖ 130 mL were observed with the 25-mg and 50-mg doses on days 1 and 28, and with the 12.5-mg dose on day 28. Time to achieve a ⱖ 100-mL increase from baseline in FEV1 over the first 4-h postdose on day 1 was significantly shorter for each VI dose compared with placebo (P , .001 for each dose) and shortest in the 25- and 50-mg dose groups (median 6 min for both groups) (Table 4). Similarly, time to achieve a ⱖ 12% Original Research
(Fig 3). Serial FVC results are provided in e-Appendix 5. Over days 1 to 28, greater increases from baseline in the mean percentage of rescue-free 24-h periods were observed with VI (12%-16%) compared with placebo (3%). Statistically significant decreases in adjusted mean change from baseline daily albuterol/salbutamol use (occasions) were observed for all doses (P ⱕ .035) except for 6.25 mg VI (P 5 .84). Safety
Figure 2. A, Adjusted mean change from placebo in trough FEV1 at day 29 (ITT population). B, Posterior probability distribution of the treatment difference in change from baseline (predose, day 1) in trough FEV1 on day 29 (ITT population). Analysis performed using analysis of covariance with covariates of baseline, sex, age, smoking status (at screening), reversibility stratum, and treatment. Error bars are 95% CIs. See Figure 1 legend for expansion of abbreviations.
increase from baseline in FEV1 over the first 4-h postdose on day 1 was significantly shorter for each VI dose group compared with placebo (P , .001 for each dose) and was shortest with 25-mg VI and 50-mg VI (median 18 min and 16 min, respectively) (Table 4). Adjusted mean changes from baseline from repeated measures analysis of serial FEV1 over time on days 1 and 28 showed a distinct dose-response effect and indicated a 24-h duration of effect for all doses of VI
Once-daily administration of VI 3, 6.25, 12.5, 25, and 50 mg was well tolerated over 28 days of treatment, with no adverse signals with increasing dose. The overall incidence of on-treatment AEs in the VI treatment groups (24%-33%) was lower than in the placebo group (36%), with headache as the most commonly reported AE (Table 5). The incidence of drug-related AEs was similar across the VI groups (5%-7%) and lower than for placebo (10%) (Table 5). Increased blood potassium and blood glucose were the two most common drug-related AEs, although the incidence of these events was the same as or less than placebo. Four patients experienced nonfatal SAEs during treatment (3 mg VI, vasovagal syncope; 6.25 mg VI, aortic aneurysm; 12.5 mg VI, atrial fibrillation; 12.5 mg VI, COPD exacerbation and pneumonia); none was considered drug related. No fatal on-treatment SAEs were reported. Four patients experienced posttreatment SAEs, of which COPD exacerbation (3 mg VI) and hyperkalemia (50 mg VI) were considered drug related. In total, 13 patients were withdrawn due to an AE: three in the placebo group and three, four, two, zero, and one in the VI 3-, 6.25-, 12.5-, 25-, and 50-mg groups, respectively.
Table 2—Change From Baseline (Predose, Day 1) in Trough FEV1 on Day 29 (LOCF) (ANCOVA; ITT Population) VI Trough FEV1, L
Placebo (n 5 101)
3 mg (n 5 99)
101 1.255 (0.4672) 0.51, 2.48
99 1.299 (0.4591) 0.56, 2.93
101 1.242 (0.4307) 0.42, 2.56
100 1.222 (0.4265) 0.53, 2.35
100 1.182 (0.4832) 0.48, 2.87
99 1.330 (0.4873) 0.61, 2.60
101 1.284 0.029 (0.0188) … … …
99 1.375 0.120 (0.0190) 0.092 0.039-0.144 , .001
100 1.382 0.127 (0.0188) 0.098 0.046-0.150 , .001
99 1.393 0.138 (0.0190) 0.110 0.057-0.162 , .001
99 1.421 0.166 (0.0190) 0.137 0.085-0.190 , .001
99 1.449 0.194 (0.0190) 0.165 0.112-0.217 , .001
Baseline No. Mean (SD) Min, max Day 29 No. LS mean LS mean change (SE)a Difference vs placebo 95% CI P value
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
Analysis performed using ANCOVA with covariates of baseline, sex, age, smoking status (at screening), reversibility stratum, and treatment. ANCOVA 5 analysis of covariance; LOCF 5 last observation carried forward. LS 5 least squares; max 5 maximum; min 5 minimum. See Table 1 legend for expansion of abbreviations. aSE for both LS mean and LS mean change from baseline. journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
CHEST / 142 / 1 / JULY 2012
123
Table 3—Change From Baseline 0- to 24-h Weighted Mean FEV1 on Days 1 and 28 (ITT Population) VI 0–24 h Weighted Mean FEV1, L Day 1 No.a No.b LS mean LS mean change (SE)c Difference vs placebo 95% CI P value Day 28 No.a No.b LS mean LS mean change (SE)c Difference vs placebo 95% CI P value
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
101 100 1.283 0.028 (0.0135) … … …
99 97 1.340 0.085 (0.0137) 0.057 0.019-0.095 .003
100 100 1.387 0.132 (0.0135) 0.104 0.066-0.141 , .001
99 99 1.404 0.149 (0.0136) 0.120 0.083-0.158 , .001
99 99 1.434 0.178 (0.0136) 0.150 0.112-0.188 , .001
99 97 1.458 0.202 (0.0137) 0.174 0.136-0.212 , .001
101 84 1.265 0.010 (0.0189) … … …
99 88 1.369 0.114 (0.0187) 0.105 0.052-0.157 , .001
100 91 1.390 0.135 (0.0185) 0.125 0.073-0.177 , .001
99 92 1.407 0.152 (0.0185) 0.142 0.090-0.194 , .001
99 92 1.423 0.168 (0.0185) 0.158 0.106-0.210 , .001
99 91 1.441 0.186 (0.0186) 0.177 0.125-0.229 , .001
Analysis performed using mixed model repeated measures with covariates of baseline, sex, age, smoking status (at screening), reversibility stratum, day (nominal), treatment, and day by treatment and day by baseline interactions. Baseline is FEV1 predose on day 1. See Table 1 and 2 legends for expansion of abbreviations. aNo. of patients with analyzable data on 1 or more days. bNo. of patients with analyzable data on the given day. cSE for both LS mean and LS mean change from baseline.
No suggestions of treatment- or dose-related effects on other laboratory analyses were observed, except for a slight increase in eosinophils in the active treatment groups. Five patients across the treatment groups (two placebo, one 3-mg VI, and two 12.5-mg VI) were identified with elevated total bilirubin, alanine transaminase, or aspartate transaminase values, which returned to below levels of clinical concern (1.5 and 3 times upper limit of normal for total bilirubin and
alanine transaminase/aspartate transaminase, respectively) during or by the end of the study. No suggestions of treatment or dose-related effects on blood pressure or pulse rate were observed and the incidence of prolonged (. 450 milliseconds) QTcF interval was low and similar across treatment groups (ⱕ 3%). The overall incidence of COPD exacerbations was low (0%-5% in the VI groups and 3% in the placebo group).
Table 4—Log-Rank Analysis of Time to Increases of ⱖ 100 mL or ⱖ 12% From Baseline FEV1 (0-4 h Postdose) VI Day 1 (0–4 h Postdose) ⱖ 100 mL increase from baseline FEV1 No. No. events No. censored Median time, min P value ⱖ 12% increase from baseline FEV1 No. No. events No. censored Median time, min P value
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
101 42 (42) 59 (58) NDa …
99 73 (74) 26 (26) 32 , .001
101 80 (79) 21 (21) 16 , .001
100 83 (83) 17 (17) 16 , .001
100 89 (89) 11 (11) 6 , .001
99 91 (92) 8 (8) 6 , .001
101 27 (27) 74 (73) NDa …
99 63 (64) 36 (36) 120 , .001
101 73 (72) 28 (28) 30 , .001
100 68 (68) 32 (32) 30 , .001
100 81 (81) 19 (19) 18 , .001
99 80 (81) 19 (19) 16 , .001
ITT population stratified by reversibility (reversible, nonreversible). ND 5 not defined. See Table 1 legend for expansion of other abbreviations. aIf . 50% of patients are censored, median time is ND. 124 Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
Original Research
Figure 3. A, Repeated measures adjusted treatment differences from baseline in serial FEV1 on day 1 (ITT population). B, Repeated measures adjusted treatment differences from baseline in serial FEV1 on day 28 (ITT population). See Figure 1 legend for expansion of abbreviation.
Discussion Our study demonstrates that the novel long-acting b2 agonist VI is effective and well tolerated in patients with COPD. Treatment with VI once daily at doses ranging from 3 mg to 50 mg for 28 days produced statistically significant, dose-dependent improvements in trough FEV1 (primary outcome measure) compared with placebo. An improvement from baseline FEV1 of ⱖ 100 mL represents a clinically relevant treatment difference.11 Adjusted mean treatment differences of ⱖ 100 mL vs placebo in both trough and 0- to 24-h weighted mean FEV1 was consistently observed with VI 12.5-, 25-, and 50-mg doses, whereas a sustained 24-h duration of action was demonstrated at all doses of VI throughout the 28-day treatment period. The Bayesian analysis of trough FEV1 demonstrated that the 25- and 50-mg groups had a higher probability (92% and 99%, respectively) that the true treatment difference vs placebo would be . 100 mL compared with the 3-, 6.25-, and 12.5-mg groups (37%, 47%, and 64%, respectively). The absence of clinically relevant safety signals with increasing dose, including very low incidence of treatment-related AEs/SAEs with no suggestion of an effect on vital signs, QTcF interval, or blood glucose and potassium levels, indicates the favorable safety profile of VI. journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ on 07/18/2012
Although five patients who received VI had increased potassium (considered to be drug related), the incidence of these AEs was the same as or less than placebo and would not be expected to be related to LABA therapy, which typically results in hypokalemia at high doses. The 24-h duration of action of VI is supported by repeated measures analysis of serial FEV1, which showed increasing adjusted mean treatment differences from placebo with increasing VI dose. Although we did observe an apparent improvement in FEV1 with placebo at 24 h, this is most likely to be due to diurnal variability as a similar effect was observed in each VI treatment group. The 24-h duration of effect of VI is further supported by the significant increases from baseline in rescue-free occasions, with all VI doses except 6.25 mg. Although the efficacy and duration of bronchodilator effect of a therapy are important, so is the speed with which a clinically meaningful bronchodilator effect is achieved. VI was shown in this study to have a rapid onset of action (median 6 min for the 25- and 50-mg doses). Collectively, the findings demonstrate that VI is an effective, well-tolerated, and fast-acting LABA with a true 24-h bronchodilator effect. This clinical profile of VI is particularly relevant considering the poor compliance with existing inhaled bronchodilators, CHEST / 142 / 1 / JULY 2012
125
Table 5—Most Common On-Treatment AEs (ⱖ 3% Incidence in Any Treatment Group) and Drug-Related AEs (Occurring in . 1 Patient On Treatment or Posttreatment) VI AEs
Placebo (n 5 101)
3 mg (n 5 99)
6.25 mg (n 5 101)
12.5 mg (n 5 101)
25 mg (n 5 101)
50 mg (n 5 99)
Any AE Headache Nausea Nasopharyngitis Blood potassium increaseda Blood glucose increased Diarrhea Ventricular extrasystoles Nasal congestion Oropharyngeal pain Any drug-related AE Blood potassium increasedb Blood glucose increased Headache Palpitations Ventricular extrasystoles Nausea Pruritus Tremor
36 (36) 10 (10) 4 (4) 3 (3) 3 (3) 3 (3) 1 (, 1) 2 (2) 3 (3) 0 10 (10) 2 (2) 3 (3) 3 (3) 1 (, 1) 0 1 (, 1) 0 0
24 (24) 6 (6) 1 (1) 2 (2) 0 0 2 (2) 0 0 0 5 (5) 0 0 0 0 0 1 (1) 0 1 (1)
32 (32) 5 (5) 3 (3) 5 (5) 1 (, 1) 1 (, 1) 1 (, 1) 1 (, 1) 2 (2) 3 (3) 5 (5) 1 (, 1) 1 (, 1) 0 0 1 (, 1) 0 0 1 (, 1)
24 (24) 3 (3) 2 (2) 0 2 (2) 3 (3) 1 (, 1) 0 0 0 5 (5) 1 (, 1) 1 (, 1) 0 0 0 0 1 (, 1) 0
33 (33) 3 (3) 2 (2) 1 (, 1) 2 (2) 1 (, 1) 3 (3) 0 0 0 5 (5) 1 (, 1) 1 (, 1) 0 1 (, 1) 0 0 1 (, 1) 0
28 (28) 7 (7) 1 (1) 0 2 (2) 0 0 3 (3) 0 1 (1) 7 (7) 2 (2) 0 0 0 1 (1) 0 0 0
ITT population. Data are presented as No. (%). AE 5 adverse event. See Table 1 legend for expansion of other abbreviation. aAdditionally, the AE term of hyperkalemia was reported for one patient in the 3-mg group and one patient in the 12.5-mg group. bAdditionally, the AE term of hyperkalemia was reported for one patient in the 3-mg group and one patient in the 50-mg group.
which leads to inadequate treatment of symptoms and may be due to the dosing regimen (eg, bid) and onset of action. Although the currently marketed LABAs (salmeterol and formoterol) are effective and both have a relatively quick onset of action in the management of the symptoms of COPD,1,12 they have to be administered twice daily. In this regard, the convenience of once-daily treatment with VI may lead to increased patient acceptance, which could improve adherence to therapy, a potential benefit that needs further exploration in future studies. Treatment compliance during this study was ⱖ 99%, with only a small proportion of patients (5%) underusing or overusing their assigned medication. This study has several strengths beyond the fact that it investigated a novel LABA that can be administered once daily. The study included a large number of patients, examined the efficacy of the drug at different doses, and included a comprehensive safety evaluation. A potential limitation of the study is the relatively short treatment period (28 days), and thus long-term studies to examine the safety of VI are needed. In summary, this study demonstrates that oncedaily administration of VI in patients with moderate to severe COPD provides clinically relevant 24-h improvement in lung function with a rapid onset of effect and a safety and tolerability profile comparable with placebo. The 25-mg and 50-mg doses of VI 126 DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
appear to offer the greatest clinical benefit without any safety concerns. Further studies are warranted to evaluate the safety and efficacy of VI as part of a combination ICS/LABA therapy in COPD. Acknowledgments Author contributions: All authors vouch for the accuracy and completeness of the data and the data analysis. Dr Hanania: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Feldman: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Zachgo: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Shim: contributed as a study principal investigator, had full access to and interpreted the data, and wrote the manuscript. Dr Crim: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Ms Sanford: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Dr Lettis: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Dr Barnhart: contributed to developing the design and concept of the study, was the clinical study investigation lead, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Dr Haumann: contributed to developing the design and concept of the study, approved the statistical plan, had full access to and interpreted the data, and wrote the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Hanania has received fees from GlaxoSmithKline for participation in the speakers bureau and on the National Advisory Board of GlaxoSmithKline; Original Research
his institution has received remuneration for participation in clinical trials sponsored by GlaxoSmithKline. Dr Feldman has received funds as the principle investigator in a clinical trial sponsored by GlaxoSmithKline, which was administered by his employer, S. Carolina Pharmaceutical Research. Dr Zachgo has received lecture fees from GlaxoSmithKline; he has participated in clinical research studies sponsored by GlaxoSmithKline, Novartis AG, Boehringer Ingelheim, Nycomed, and AstraZeneca. Dr Shim has received funds for his participation in a clinical trial sponsored by GlaxoSmithKline, which was administered by his employer, Guro Hospital, Seoul, South Korea. Drs Crim, Lettis, Barnhart, and Haumann and Ms Sanford are employees of and hold stocks in GlaxoSmithKline. Role of sponsors: The charge for color figures was paid for by GlaxoSmithKline. Editorial support in the form of development of draft outline, development of manuscript first draft, editorial suggestions to draft versions of this paper, assembling tables and figures, collating author comments, copyediting, fact checking, referencing and graphic services was provided by David Cutler at Gardiner-Caldwell Communications and was funded by GlaxoSmithKline. Other contributions: We thank M. Patel, MSc; S. Kilbride, MSc; and J. Brooks, MSc (GlaxoSmithKline Research and Development, Uxbridge, England) for their assistance with programming the statistical analyses. Additional information: The e-Appendix and e-Tables can be found in the “Supplemental Materials” area of the online article.
References 1. Global Initiative for Chronic Obstructive Lung Disease guideline: global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease (updated 2009). Global Initiative for Chronic Obstructive Lung Disease website. http://www.goldcopd.com/. Accessed April 13, 2011. 2. Cazzola M, Matera MG. Novel long-acting bronchodilators for COPD and asthma. Br J Pharmacol. 2008;155(3):291-299. 3. Morrison V, Sturton G, Barrett V, Ford A, Knowles R. Pharmacological characterisation of GW642444, a long-acting b2-agonist (LABA) with rapid onset and long duration, on
journal.publications.chestnet.org
DownloadedFrom:http://journal.publications.chestnet.org/on07/18/2012
4.
5.
6.
7.
8.
9. 10. 11. 12.
isolated large and small human airways [abstract]. Am J Respir Crit Care Med. 2010;181:A4453. Ford A, Hughes S, Morrison V, Barrett V, Knowles R. In vitro and in vivo pharmacological characterisation of GW642444, a novel long-acting b2 agonist (LABA), with fast onset and long duration in the guinea-pig [abstract]. Am J Respir Crit Care Med. 2010;181:A5677. Barrett V, Emmons A, Ford A, Knowles R. In vitro pharmacological characterisation of GW642444, a novel long acting b2-agonist (LABA) using human recombinant b1/2/3 adrenoceptor cAMP assays [abstract]. Am J Respir Crit Care Med. 2010;181:A4451. Kempsford RD, Norris V, Siederer SK. The pharmacodynamics, pharmacokinetics and tolerability of repeat doses of the novel inhaled long-acting b 2 adrenoceptor agonist (LABA) GW642444 (25, 50 and 100mcg) in healthy subjects [abstract]. Am J Respir Crit Care Med. 2010;181:A4461. Kempsford RD, Norris V, Siederer SK. GW642444, a novel inhaled long-acting beta2 adrenoceptor agonist (LABA), at single doses of 25, 50 and 100mcg, is well tolerated and demonstrates prolonged bronchodilation in COPD subjects [abstract]. Am J Respir Crit Care Med. 2010;181:A4447. Lötvall J, Bateman ED, Bleecker ER, et al. Dose-related efficacy of vilanterol trifenatate (VI), a long-acting beta2 agonist (LABA) with inherent 24-hour activity, in subjects with persistent asthma [abstract]. Eur Respir J. 2010; 36(suppl 54):1013s. Hanania NA, Feldman G, Zachgo W, et al. Dose-related efficacy of vilanterol trifenatate (VI) in COPD [abstract]. Eur Respir J. 2010;36(suppl 54):217s. Hanania NA, Feldman G, Zachgo W, et al. Safety of vilanterol trifenatate (VI) in a COPD dose-ranging study [abstract]. Eur Respir J. 2010;36(suppl 54):208s. Donohue JF. Minimal clinically important differences in COPD lung function. COPD. 2005;2(1):111-124. Steiropoulos P, Tzouvelekis A, Bouros D. Formoterol in the management of chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis. 2008;3(2):205-215.
CHEST / 142 / 1 / JULY 2012
127