Figures
- FIGURE 1
Study design. MITT: multiple-inhaler triple therapy; LAMA: long-acting muscarinic receptor antagonist; LABA: long-acting β2-agonist; ICS: inhaled corticosteroid; CAT: COPD Assessment Test; FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; R: randomisation; BEC: blood eosinophil count. #: where available, peripheral BECs were collected using the historical value closest to the patient's consenting visit and no later than 36 months prior to visit 1; ¶: patients were asked, if possible, to withhold short-acting β2-agonists or short-acting anticholinergics for ≥4 h and not to take either their single-inhaler triple therapy or MITT until after the clinic visit at week 24 to enable measurement of trough forced expiratory volume in 1 s (FEV1). If this was not possible or had not been done, FEV1 was still measured; +: safety information was collected at all scheduled or usual care visits recorded in the electronic case report form.
- FIGURE 2
Patient disposition. ITT: intention-to-treat; FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; MITT: multiple-inhaler triple therapy; FEV1, forced expiratory volume in 1 s. #: patients may have been excluded for multiple reasons. Patients who were randomised in error (n=17) are also identified as excluded due to eligibility. For those withdrawing from the study or randomised treatment only one primary reason is recorded. A patient completed randomised study treatment if they did not prematurely discontinue randomised study treatment and attended visit 2 (week 24). A patient who continued on all components of the randomised treatment and added additional medication to their maintenance treatment were considered as modifying their randomised treatment (intercurrent event), but were not considered to have prematurely discontinued from randomised treatment; ¶: one patient who withdrew >1 day after randomisation and did not take any study medication was included in the ITT population.
- FIGURE 3
Proportion of COPD Assessment Test (CAT) responders at week 24. Missing CAT scores were imputed using multiple imputation based on the randomised treatment arm characteristics assuming missing at random. Response is defined as a CAT score ≥2 units below baseline. Data labels above bars are n (%). FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; MITT: multiple-inhaler triple therapy.
- FIGURE 4
Proportion of COPD Assessment Test (CAT) responders at week 24 by prior medication strata. MITT: multiple-inhaler triple therapy; FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; ICS: inhaled corticosteroid; LAMA: long-acting muscarinic receptor antagonist; LABA: long-acting β2-agonist.
Tables
- TABLE 1
Patient characteristics at screening (intention-to-treat population)
FF/UMEC/VI Non-ELLIPTA MITT Total Patients 1545 1547 3092 Age years 67.8±8.78 67.8±8.59 67.8±8.68 Male 837 (54) 818 (53) 1655 (54) BMI kg·m−2 27.84±5.93
n=153628.05±6.05
n=153827.95±5.99
n=3074COPD exacerbation history in the prior 12 months Moderate 0 409 (26) 405 (26) 814 (26) 1 639 (41) 645 (42) 1284 (42) ≥2 497 (32) 497 (32) 994 (32) Severe 0 1349 (87) 1361 (88) 2710 (88) 1 155 (10) 139 (9) 294 (10) ≥2 41 (3) 47 (3) 88 (3) Moderate/severe 0 363 (23) 361 (23) 724 (23) 1 615 (40) 610 (39) 1225 (40) ≥2 567 (37) 576 (37) 1143 (37) CAT score 20.8±6.76
n=154320.5±6.62
n=154720.7±6.69
n=3090Peripheral blood eosinophil count# n=605 n=572 n=1177 <150 cells·µL−1 208 (34) 223 (39) 431 (37) ≥150 cells·µL−1 397 (66) 349 (61) 746 (63) Actual prior medication use strata ICS+LAMA+LABA 1226 (79) 1235 (80) 2461 (80) ICS+LABA 126 (8) 126 (8) 252 (8) LABA+LAMA 192 (12) 183 (12) 375 (12) Missing¶ 1 (<1) 3 (<1) 4 (<1) Data are presented as n, mean±sd or n (%). FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; MITT: multiple-inhaler triple therapy; BMI: body mass index; CAT: COPD Assessment Test; ICS: inhaled corticosteroid; LAMA: long-acting muscarinic receptor antagonist; LABA: long-acting β2-agonist. #: historical eosinophils were recorded as the most recent measure taken within the previous 36 months; ¶: stratum is considered missing if the combination of maintenance treatments taken in the 14 days prior to randomisation do not meet any of the three defined strata groups.
- TABLE 2
Change from baseline in forced expiratory volume in 1 s (FEV1) and trough FEV1 at week 24
FF/UMEC/VI population Non-ELLIPTA MITT population FF/UMEC/VI versus non-ELLIPTA MITT Patients 910 904 FEV1# Patients 691 675 FEV1 mL 1446 (1425–1467) 1396 (1375–1418) 50 (26–73); p<0.001 FEV1 mL change from baseline 77 (57–98) 28 (6–49) Trough FEV1 Patients 301 292 FEV1 mL 1498 (1462–1534) 1445 (1404–1486) 53 (9–96); p=0.017 FEV1 mL change from baseline¶ 100 (64–135) 47 (6–88) Data are presented as n or least-squares mean (95% CI), unless otherwise stated. FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; MITT: multiple-inhaler triple therapy. #: data include both trough and non-trough values; ¶: patients with imputed FEV1, n=82 (FF/UMEC/VI), n=115 (non-ELLIPTA MITT). Trough FEV1 is defined as the FEV1 value recorded while patients have withheld COPD maintenance, short-acting β2-agonist and short-acting muscarinic receptor antagonist treatment. For COPD maintenance treatments taken once daily, FEV1 was considered as trough if the patient withheld the LABA and LAMA components of the maintenance treatment for ≥16 h. For COPD maintenance treatments taken twice daily, FEV1 was considered as trough if the patient withheld the LABA and LAMA components of the maintenance treatment for ≥8 h.
- TABLE 3
Incidence of on-randomised-treatment adverse events#
FF/UMEC/VI Non-ELLIPTA MITT Patients n (%) Events rate¶ (n) Patients n (%) Events rate¶ (n) Patients n 1545 1547 Total duration at risk patient-years 636.7 685.8 Any adverse event 250 (16) 590.6 (376) 151 (10) 322.2 (221) Any treatment-related adverse event 145 (9) 329.8 (210) 44 (3) 77.3 (53) Any adverse event leading to study withdrawal 115 (7) 279.6 (178) 32 (2) 70.0 (48) Any SAE 114 (7) 257.6 (164) 114 (7) 255.2 (175) Any treatment-related SAE 13 (<1) 20.4 (13) 6 (<1) 10.2 (7) Any fatal SAE 8 (<1) 20.4 (13) 8 (<1) 23.3 (16) Any treatment-related fatal SAE 0 0 0 0 Serious AESIs Cardiovascular effects 29 (2) 55.0 (35) 23 (1) 39.4 (27) Decreased BMD and associated fractures 6 (<1) 9.4 (6) 4 (<1) 7.3 (5) Infective pneumonia 27 (2) 44.0 (28) 32 (2) 46.7 (32) LRTI excluding infective pneumonia 7 (<1) 11.0 (7) 10 (<1) 14.6 (10) FF: fluticasone furoate; UMEC: umeclidinium; VI: vilanterol; MITT: multiple-inhaler triple therapy; SAE: serious adverse event; AESI: adverse event of special interest; BMD: bone mineral density; LRTI: lower respiratory tract infection. #: the recording of adverse events was limited to treatment-related adverse events, SAEs and adverse events leading to study treatment discontinuation or study withdrawal. Refer to supplementary tables S6 and S7 for further details; ¶: event rate per 1000 patient-years, calculated as number of events × 1000, divided by the total duration at risk.
Supplementary Material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material 00950-2020.SUPPLEMENT
