Table 2.
Strengths and limitations of selected RCTs of ICS in patients with COPD.
| Pharmacotherapy | Key trials/reference (year) | Key findings | Main critique | Reference to GOLD COPD documents |
|---|---|---|---|---|
| ICS alone | Yang IA, et al. Cochrane Database Syst Rev (2012)43 |
RCTs of ICS vs. placebo >6 months: • Modest decrease in exacerbations • No decrease in FEV1 decline or mortality • Increased pneumonia risk |
Exacerbation benefit overestimated due to ICS withdrawal effect47 |
GOLD Report 2011 “ICS monotherapy not recommended in COPD as it is less effective than LABA/ICS (Evidence A)”45 |
| LABA/ICS |
TORCH (2007)44 • 3-year survival study • Moderate–severe COPD (FEV1 < 60%), SAL/FP500 vs. its mono-components and placebo |
All-cause mortality (complete survival follow-up): • HR LABA/ICS vs. placebo 0.825 (95% CI 0.681, 1.002), p = 0.052) • Primary endpoint not achieved • Incomplete (discontinuation) follow-up for all non-primary (non-survival) endpoints Pneumonia events: • LABA/ICS and ICS groups (18–20%) • LABA and placebo groups (12–13%) |
• Placebo is not “usual care” • 2-week run-in with ICS and LABA withdrawal (28% dropout) • ICS withdrawal (45–50%) • LABA withdrawal (35%) • Factorial (2*2 analysis) not published by study authors but otherwise calculated to show survival benefit attributable to LABA, not ICS, component47 |
GOLD REPORT 2023—Efficacy of ICS alone4 “In the TORCH trial, a trend toward higher mortality was observed for patients treated with fluticasone propionate alone compared to those receiving placebo or salmeterol plus fluticasone propionate combination” |
| Adding LAMA |
Canadian OPTIMAL Study (2007)49 • 52-week, 3-arm exacerbation RCT • Moderate–severe COPD (FEV1 < 65%); ≥1 exacerbation/year; past asthma excluded - LAMA (Tio, n = 156) - LAMA/LABA (Tio/SAL; n = 148) - LAMA + LABA/ICS (Tio + SAL/FP250; n = 145) |
• Proportion of patients who experienced moderate–severe exacerbations: - LAMA (62.8%) - LAMA + LABA (64.8%) - LAMA + LABA/ICS (60%) - Rate ratios not significantly different • Hospitalisations lower in LAMA + LABA/ICS vs LAMA group: - HR 0.67 (95% CI 0.45, 0.99) • 40% premature discontinuation in non-ICS arms |
• First trial of triple therapy in COPD, non-pharma sponsored • Complete (intent-to-treat) follow-up • High withdrawal rate in non-ICS arms related to ICS withdrawal on randomisation (75% pre-study ICS use) • No exacerbation benefit in ICS-naïve subjects (per-study non-users)47 |
Study not cited |
| LAMA/LABA vs. LABA/ICS |
FLAME (2016)50 • 52-week, 2-arm exacerbation RCT - Moderate–severe COPD (FEV1 25–60%); >1 exacerbation/year; past asthma excluded. - LAMA/LABA (GLY/IND; n = 1680) - LABA/ICS (SAL/FP250; n = 1682) |
• 4-week LAMA (Tio) run-in associated with 32% dropout rate • Annual rate of all COPD exacerbations: - 11% lower in LAMA/LABA group than LABA/ICS group (RR 0.89; 95% CI 0.83, 0.96; p = 0.003) • Incidence of pneumonia: - 3.2% in LAMA/LABA group and 4.8% in LABA/ICS group |
• Study design tending to exclude ICS responders • Run-in bias (4-week Tio run-in; all ICS and LABA discontinued) • Past asthma excluded • Subjects with blood eosinophil count >600 cells/µl excluded • The reported HR for time to first exacerbation may represent a magnification of the real effect51 |
No referral to this study regarding ICS use in COPD |
CI confidence interval, COPD chronic obstructive pulmonary disease, FEV1 forced expiratory volume in 1 s, FP fluticasone propionate, GOLD Global Initiative for Chronic Obstructive Lung Disease, HR hazard ratio, ICS inhaled corticosteroids, LABA long-acting β2-agonist, LAMA long-acting muscarinic antagonist, RCT randomised controlled trial, RR relative risk, SAL salmeterol, Tio tiotropium.