Lung function effects and safety of fluticasone furoate (FF)/vilanterol (VI) in patients with COPD: low-mid dose assessment Kerwin EM1, Scott-Wilson C2, Sanford L3, Rennard SI4, Agusti A5, Barnes N6, Crim C2 1. Clinical Research Institute of Southern Oregon, Medford, OR, USA; 2. GlaxoSmithKline, Research Triangle Park, NC, USA; 3. GlaxoSmithKline, Stockley Park, London, UK; 4. University of Omaha, Nebraska, OH, USA; 5. Thorax Institute, Hospital Clinic, IDIBAPS, University of Barcelona and FISIB, CIBERES, Mallorca, Spain; 6. Respiratory Medicine, Barts and The London NHS Trust, London, UK

CONFLICT OF INTEREST DISCLOSURE Edward Kerwin, M.D. has served on Advisory Boards for Astra Zeneca, Forest, Ironwood, Mylan, Pearl, Sanofi, Sunovion and Targacept; Speakers Panels for AstraZeneca, Pfizer, Sanofi, and Sunovion; and has received travel reimbursement from Merck, Forest and Novartis. He has conducted clinical trials for multiple pharmaceutical companies, including GlaxoSmithKline

Introduction An inhaled corticosteroid (ICS) combined with a long-acting beta2 agonist (LABA) is an established treatment for COPD patients experiencing impaired airway function or acute exacerbations Fluticasone furoate (FF) and vilanterol (VI) are respectively, a novel ICS and LABA, in development as a once-daily combination therapy for COPD and asthma

Objectives Post-Dose & Trough Lung Function Effects 24-week efficacy of VI Addition of FF to VI Dose Range FF when added to VI Symptoms Safety

FF 100mcg (N=206) VI 25mcg (N=205) FF/VI 50/25mcg (N=206) FF/VI 100/25mcg (N=206) Placebo (N=207) R F S Clinic Visit (Day) 1 (-14 to -17) 2 (1) 3 (2) 4 (7) 5 (14) 6 (28) 7 (56) 8 (84) 9 (112) 10 (140) 11 (168) 12 (169) Phone (176±2) Study design

Endpoints Co-primary Weighted mean FEV1 (0-4hr) post-dose, Day 168 Change from baseline Trough FEV1, Day 169 Secondary CRQ-SAS dyspnoea domain Peak FEV1 (0-4hr) post-dose, Day 1 Time to 100mL (0-4hr) post-dose, Day 1 Safety

VI 25 vs. Placebo FF/VI 100/25 vs. Placebo FF/VI 100/25 vs. FF 100 wmFEV1(0-4hr) Day 168 Trough FEV1 Day 169 VI 25 vs. Placebo FF/VI 100/25 vs. Placebo FF/VI 100/25 vs. VI 25 Co-Primary , FF/VI 100/25 Secondary, FF/VI 100/25 FF/VI 100/25 vs. Placebo for each endpoint FF/VI 50/25 vs. Placebo FF/VI 50/25 vs. Placebo FF/VI 50/25 vs. VI 25 wmFEV1(0-4hr) Day 168 Trough FEV1 Day 169 Co-Primary , FF/VI 50/25 LEVEL 1 LEVEL 2 LEVEL 3 Move to Level 2 if p<0.05 for all Secondary , FF/VI 50/25 FF/VI 50/25 vs. Placebo for each endpoint Move to Level 3 if p<0.05 for all Study power: 80mL for FF/VI vs. VI Trough FEV1 Day 169 100mL for FF/VI vs. Placebo Trough FEV1 Day 169 100mL for FF/VI vs. FF wmFEV1 (0-4hr) Day 168

Characteristics Total (N=1030) Mean Age, yr (SD) 62.7 (9.09) Male Sex, n (%) 685 (67) Mean Post-BD FEV1, L (SD) 1.406 (0.48) Mean Post-BD %pred FEV1, % (SD) 48.3 (12.45) GOLD stage, n (%) II III IV 487 (47) 445 (43) 94 (9) Mean mMRC dyspnoea, score (SD) 2.4 (0.5) Moderate* exacerbations, n (%) 0 1 2 >2 787 (76) 202 (20) 31 (3) 10 (<1) *requiring systemic corticosteroids and/or antibiotics

wmFEV1 (0-4hr): VI vs. PBO Day 168 Test mL (95% CI) p VI – PBO 103 (52, 153) <0.001 0 1 Day wm FEV1 (L), LS mean (95% CI) 1.2 1.3 1.4 1.5 1 14 56 84 168 Placebo VI 25

Trough FEV1: VI vs. PBO Day 169 Test mL (95% CI) p VI – PBO 67 (12, 121) 0.017 -0.05 Day Trough FEV1 (L), LS mean change from baseline (95% CI) 0.00 0.05 0.10 0.15 0.20 0.25 2 7 14 28 56 84 112 140 168 169 Placebo VI 25

wmFEV1 (0-4hr): FF/VI vs. PBO Day 168 Test mL (95% CI) p FF/VI 100/25 – PBO 173 (123, 224) <0.001 FF/VI 50/25 – PBO* 192 (141, 243) - 0 1 Day wm FEV1 (L), LS mean (95% CI) 1.2 1.3 1.4 1.5 1 14 56 84 168 Placebo FF/VI 50/25 FF/VI 100/25 *p-value not shown due to statistical hierarchy; difference is descriptive only

Trough FEV1: FF/VI vs. PBO Day 169 Test mL (95% CI) p FF/VI 100/25 – PBO 115 (60, 169) <0.001 FF/VI 50/25 – PBO* 129 (74, 184) - -0.05 Day Trough FEV1 (L), LS mean change from baseline (95% CI) 0.00 0.05 0.10 0.15 0.20 0.25 2 7 14 28 56 84 112 140 168 169 Placebo FF/VI 50/25 FF/VI 100/25 *p-value not shown due to statistical hierarchy; difference is descriptive only

wmFEV1 (0-4hr): FF/VI vs. FF Day 168 Test mL (95% CI) p FF/VI 100/25 – FF 120 (70, 170) <0.001 0 1 Day wm FEV1 (L), LS mean (95% CI) 1.2 1.3 1.4 1.5 1 14 56 84 168 FF 100 FF/VI 50/25 FF/VI 100/25

Trough FEV1: FF/VI vs. VI Day 169 Test mL (95% CI) p FF/VI 100/25 – VI 48 (-6, 102) 0.082 FF/VI 50/25 – VI* 62 (8, 117) - -0.05 Day Trough FEV1 (L), LS mean change from baseline (95% CI) 0.00 0.05 0.10 0.15 0.20 0.25 2 7 14 28 56 84 112 140 168 169 VI 25 FF/VI 50/25 FF/VI 100/25 *p-value not shown due to statistical hierarchy; difference is descriptive only

CRQ-SAS dyspnoea domain: Day 168 Test Units(95% CI) VI – Placebo 0.14 (-0.10, 0.38) FF 100 – Placebo 0.06 (-0.18, 0.30) FF/VI 50/25 – Placebo 0.19 (-0.05, 0.43) FF/VI 100/25 – Placebo 0.30 (0.06, 0.54) FF/VI 50/25 – VI 25 0.05 (-0.19, 0.29) FF/VI 100/25 – VI 25 0.16 (-0.08, 0.40) FF/VI 100/25 – FF 100 0.24 (0.01, 0.48) Differences are descriptive only due to statistical hierarchy Range: 1 – maximum impairment, 7 – no impairment; MCID > 0.5 Units

Peak FEV1 (0-4hr): Day 1 Test mL(95% CI) VI – Placebo 142 (114, 169) FF 100 – Placebo 12 (-15, 39) FF/VI 50/25 – Placebo 148 (120, 175) FF/VI 100/25 – Placebo 139 (112, 166) FF/VI 50/25 – VI 25 6 (-22, 33) FF/VI 100/25 – VI 25 -3 (-30, 25) FF/VI 100/25 – FF 100 127 (100, 154) Differences are descriptive only due to statistical hierarchy

Time to 100mL (0-4h): Day 1 MEDIAN TIME 100mL Treatment Min Placebo n/a FF 100 n/a VI 25 16 FF/VI 100/25 17 FF/VI 50/25 17 Serial FEV1: Day 1 (0-4h) n/a = not applicable as >50% of patients did not achieve a 100mL improvement by 4hr Placebo FF 100 VI 25 FF/VI 50/25 FF/VI 100/25 Time - Day 1 Change from baseline FEV1 (L) mean (SD) 0.0 0.2 0.4 5min 15min 30min 1hr 2hr 4hr 0.1 0.3

Safety Placebo N=207 FF 100 N=206 VI 25 N=205 FF/VI 50/25 N=206 FF/VI 100/25 N=206 ANY EVENT, n (%) On-Treatment AEs 100 (48) 123 (60) 111 (54) 114 (55) 111 (54) On-Treatment SAEs 11 (5) 16 (8) 15 (7) 6 (3) 11 (5) EVENTS OF SPECIAL INTEREST, n (%) Cardiovascular Effects 16 (8) 18 (9) 15 (7) 15 (7) 11 (5) Local Steroid Effects 7 (3) 13 (6) 6 (3) 24 (12) 16 (8) LRTI excluding pneumonia 8 (4) 8 (4) 6 (3) 3 (1) 5 (2) Pneumonia 3 (1) 4 (2) 5 (2) 3 (1) 5 (2) Hypersensitivity 2 (<1) 5 (2) 3 (1) 1 (<1) 6 (3) Effects on Glucose 1 (<1) 4 (2) 1 (<1) 3 (1) 5 (2) Bone Disorders 4 (2) 2 (<1) 2 (<1) 1 (<1) 1 (<1) Ocular Effects 1 (<1) 2 (<1) 1 (<1) 1 (<1) 1 (<1) Systemic Steroid Effects 2 (<1) 0 0 2 (<1) 1 (<1) Effects on Potassium 1 (<1) 1 (<1) 0 0 0 Tremor 0 1 (<1) 0 0 0

VI 25mcg provides sustained bronchodilation over 24 weeks in moderate-to-severe COPD Addition of FF to VI provides further bronchodilation Additional lung function effect not statistically significantly greater than VI alone VI has a negligible effect on dyspnoea Addition of FF at 100mcg provides a numerical improvement VI provides rapid substantial post-dose bronchodilation from Day 1 onward Addition of FF provides no more rapid or increased bronchodilation FF/VI exhibits a safety profile similar to that of its components and placebo Summary

FF/VI represents a once-daily treatment option for moderate-to-severe COPD which is efficacious and well tolerated Conclusion