. 2023 Jul 24;33:27. doi: 10.1038/s41533-023-00347-6

Table 3.

Strengths and limitations of RCTs comparing LAMA/LABA with triple therapy.

Study	TRIBUTE 2018⁵²	IMPACT 2018⁵	ETHOS 2020⁶	KRONOS⁵³
Population	≥40-year-old patients with COPD, CAT ≥ 10, non-current (past) asthma—allowed^a
• FEV₁ < 50% predicted and ≥1 exacerbation/year • 65% pre-study ICS users • 0% pre-study triple therapy (excluded)	• FEV₁ 50–80% predicted and ≥2 (or ≥1 severe) exacerbation/year, OR—FEV₁ < 50% predicted and ≥1 exacerbation/year - 71% pre-study ICS users - 38% pre-study triple therapy (allowed)	• FEV₁ 50–65% predicted and ≥2 (or ≥1 severe) exacerbation/year, OR—FEV₁ 25–50% predicted and ≥1 exacerbation/year • 80% pre-study ICS users • 39% pre-study triple therapy (allowed) • ~60% of enrolled patients had a blood eosinophil count of ≥150 cells/mm³	• FEV₁ 25–80%; prior exacerbations not required • 71% pre-study ICS users • 27% pre-study LAMA/LABA/ICS • 27% ≥1 or more moderate/severe exacerbations in previous year
Study arms	• LAMA/LABA/ICS (BDP/FORM/GLY; n = 764) • LAMA/LABA (IND/GLY; n = 768)	• LAMA/LABA/ICS (FF/UMEC/VI; n = 4151) • LABA/ICS (FF/VI; n = 4134) • LAMA/LABA (UMEC/VI; n = 2070)	• LAMA/LABA/high-dose ICS (BUD320/GLY/FORM; n = 2157) • LAMA/LABA/low-dose ICS (BUD160/GLY/FORM; n = 2137) • LABA/high-dose ICS (BUD320/FORM; n = 2151) • LAMA/LABA (GLY/FORM; n = 2143)	• LAMA/LABA/ICS (BDP/FORM/GLY; n = 640) • LAMA/LABA (FORM/GLY n = 627) • LABA/ICS (BUD/FORM pMDI n = 316) • LABA/ICS open-label (BUD/FORM DPI n = 319)
Design	52-week double-blind RCT	52-week double-blind RCT	52-week double-blind RCT	24-week double-blind RCT
Run-in: 2 weeks LAMA/LABA (IND/GLY)	No run-in	During variable 1–4-week screening period: • Per-study ICS continued • LAMA/LABA replaced by ipratropium 2 puffs QID and albuterol rescue with 46% drop-out—no details provided	During variable 1–4-week screening period: • Per-study ICS continued • LAMA/LABA replaced by ipratropium 2 puffs QID and albuterol rescue with 38% drop-out—no details provided
Findings: Exacerbations & pneumonia	• Decreased adjusted moderate–severe exacerbations • BDP/FORM/GLY vs. IND/GLY - RR 0.848 (95% CI 0.723, 0.995; p = 0.043) • RR of moderate and severe exacerbations analysed separately not significantly different • Pneumonia events: Similar pneumonia rates (4%) in ICS/non-ICS groups (BDP/FORM/GLY; IND/GLY)	• Decreased adjusted moderate–severe exacerbations • FF/UMEC/VI vs. FF/VI - RR 0.85 (95% CI 0.80, 0.90; p < 0.001) • FF/UMEC/VI vs. UMEC/VI - RR 0.75 (95% CI 0.70, 0.81; p < 0.001) • Pneumonia events: - FF/UMEC/VI; FF/VI (7–8%) vs. UMEC/VI (5%) - HR 1.53 (95% CI 1.22, 1.92; p < 0.001). (FF pneumonia rates higher than BUD160–320; see ETHOS)	• Decreased adjusted moderate–severe exacerbations • BUD160 and 320/GLY/FORM vs. BUD320/FORM - BUD160/GLY/FORM: RR 0.86 (95% CI 0.79, 0.95; p = 0.002) - BUD320/GLY/FORM: RR 0.87 (95% CI 0.79, 0.95; p = 0.003) • BUD160 and 320/GLY/FORM vs. GLY/FORM - BUD160/GLY/FORM: RR 0.75 (95% CI 0.69, 0.83; p < 0.001) - BUD320/GLY/FORM: RR 0.76 (95% CI 0.69, 0.83; p < 0.001) • Pneumonia events: - BUD/GLY/FORM (3.5–4.2%); BUD/FORM (4.5%) vs. GLY/FORM (2.3%) - BUD160/GLY/FORM vs. BUD320/GLY/FORM achieved similar results with less pneumonia risk (3.5% vs. 4.2%)	Decreased moderate or severe exacerbations (secondary endpoint) • BGF MDI vs. GFF MDI: rate ratio 0.48 (95% CI 0.37, 0.64; p < 0.0001) • BGF MDI vs. BFF MDI: rate ratio 0.82 (0.58, 1.17; p = 0.2792) • BGF MDI vs. BUD/FORM DPI: rate ratio 0.83 (0.59, 1.18; p = 0.3120)
Findings: Mortality	Study not powered for mortality	Mortality¹²¹: • FF/UMEC/VI (2.36%) vs. UMEC/VI (3.19%) - HR for death 0.72 (95% CI 0.53, 0.99; p = 0.042) • FF/UMEC/VI (2.36%) vs. FF/VI (2.64%) - HR for death 0.89 (95% CI 0.67, 1.16; p = 0.387)	Mortality¹²² • BUD320/GLY/FORM vs. BUD320/FORM - HR 0.72 (95% CI 0.44, 1.16; p = 0.1721) • BUD320/GLY/FORM vs. GLY/FORM - HR 0.51 (95% CI 0.33, 0.80; p = 0.0035)	Not powered for mortality
Critique	• Customary study population (exacerbators with FEV₁ < 50%); allowing non-current asthma • 2-week LAMA/LABA run-in • Prior triple therapy: 0% • Prior ICS therapy: 65% • ICS withdrawal (mixed intervention)	• Unusual study population: - allowing non-current asthma, - inclusion of frequent exacerbators without severe airflow limitation (asthma like, FEV₁ > 50% GOLD 2D sub-cohort) • Withdrawal of prior ICS (71%) and triple therapy (38%); mixed intervention • Both exacerbation⁵⁶ and mortality⁶⁰ benefit essentially confined to the first 90 days of the study—representing ICS withdrawal in an “ICS-sensitive” sub-cohort • Mortality indication rejected by regulatory agencies^116,117	• Unusual study population: - allowing non-current asthma, - inclusion of frequent exacerbators without severe airflow limitation (asthma like, FEV₁ > 50% GOLD 2D sub-cohort) • Excluding patients with very severe airflow limitation (FEV₁ < 25%) • Unusual 1–4-week pre-randomisation screening with long-acting bronchodilator withdrawal and 46% dropout, suggestive of significant run-in bias⁵¹ • Withdrawal of prior ICS (80%) and triple therapy (39%)—mixed intervention • Mortality benefit⁶⁰ essentially confined to the first 90 days of the study—representing ICS withdrawal in an “ICS-sensitive” sub-cohort	• Allowed non-current asthma • 71% of patients on ICS therapy prior to study • Run-in withdrawal of long-acting bronchodilators associated with 38% pre-randomisation drop-out

Study

TRIBUTE 2018⁵²

IMPACT 2018⁵

ETHOS 2020⁶

KRONOS⁵³

Population

≥40-year-old patients with COPD, CAT ≥ 10, non-current (past) asthma—allowed^a

• FEV₁ < 50% predicted and ≥1 exacerbation/year

• 65% pre-study ICS users

• 0% pre-study triple therapy (excluded)

• FEV₁ 50–80% predicted and ≥2 (or ≥1 severe) exacerbation/year, OR—FEV₁ < 50% predicted and ≥1 exacerbation/year

- 71% pre-study ICS users

- 38% pre-study triple therapy (allowed)

• FEV₁ 50–65% predicted and ≥2 (or ≥1 severe) exacerbation/year, OR—FEV₁ 25–50% predicted and ≥1 exacerbation/year

• 80% pre-study ICS users

• 39% pre-study triple therapy (allowed)

• ~60% of enrolled patients had a blood eosinophil count of ≥150 cells/mm³

• FEV₁ 25–80%; prior exacerbations not required

• 71% pre-study ICS users

• 27% pre-study LAMA/LABA/ICS

• 27% ≥1 or more moderate/severe exacerbations in previous year

Study arms

• LAMA/LABA/ICS (BDP/FORM/GLY; n = 764)

• LAMA/LABA (IND/GLY; n = 768)

• LAMA/LABA/ICS (FF/UMEC/VI; n = 4151)

• LABA/ICS (FF/VI; n = 4134)

• LAMA/LABA (UMEC/VI; n = 2070)

• LAMA/LABA/high-dose ICS (BUD320/GLY/FORM; n = 2157)

• LAMA/LABA/low-dose ICS (BUD160/GLY/FORM; n = 2137)

• LABA/high-dose ICS (BUD320/FORM; n = 2151)

• LAMA/LABA (GLY/FORM; n = 2143)

• LAMA/LABA/ICS (BDP/FORM/GLY; n = 640)

• LAMA/LABA (FORM/GLY n = 627)

• LABA/ICS (BUD/FORM pMDI n = 316)

• LABA/ICS open-label (BUD/FORM DPI n = 319)

Design

52-week double-blind RCT

24-week double-blind RCT

Run-in: 2 weeks LAMA/LABA (IND/GLY)

No run-in

During variable 1–4-week screening period:

• Per-study ICS continued

• LAMA/LABA replaced by ipratropium 2 puffs QID and albuterol rescue with 46% drop-out—no details provided

During variable 1–4-week screening period:

• Per-study ICS continued

• LAMA/LABA replaced by ipratropium 2 puffs QID and albuterol rescue with 38% drop-out—no details provided

Findings: Exacerbations &

pneumonia

• Decreased adjusted moderate–severe exacerbations

• BDP/FORM/GLY vs. IND/GLY

- RR 0.848 (95% CI 0.723, 0.995; p = 0.043)

• RR of moderate and severe exacerbations analysed separately not significantly different

• Pneumonia events:

Similar pneumonia rates (4%) in ICS/non-ICS groups (BDP/FORM/GLY; IND/GLY)

• Decreased adjusted moderate–severe exacerbations

• FF/UMEC/VI vs. FF/VI

- RR 0.85 (95% CI 0.80, 0.90; p < 0.001)

• FF/UMEC/VI vs. UMEC/VI

- RR 0.75 (95% CI 0.70, 0.81; p < 0.001)

• Pneumonia events:

- FF/UMEC/VI; FF/VI (7–8%) vs. UMEC/VI (5%)

- HR 1.53 (95% CI 1.22, 1.92; p < 0.001).

(FF pneumonia rates higher than BUD160–320; see ETHOS)

• Decreased adjusted moderate–severe exacerbations

• BUD160 and 320/GLY/FORM vs. BUD320/FORM

- BUD160/GLY/FORM: RR 0.86 (95% CI 0.79, 0.95; p = 0.002)

- BUD320/GLY/FORM: RR 0.87 (95% CI 0.79, 0.95; p = 0.003)

• BUD160 and 320/GLY/FORM vs. GLY/FORM

- BUD160/GLY/FORM: RR 0.75 (95% CI 0.69, 0.83; p < 0.001)

- BUD320/GLY/FORM: RR 0.76 (95% CI 0.69, 0.83; p < 0.001)

• Pneumonia events:

- BUD/GLY/FORM (3.5–4.2%); BUD/FORM (4.5%) vs. GLY/FORM (2.3%)

- BUD160/GLY/FORM vs. BUD320/GLY/FORM achieved similar results with less pneumonia risk (3.5% vs. 4.2%)

Decreased moderate or severe exacerbations (secondary endpoint)

• BGF MDI vs. GFF MDI: rate ratio 0.48 (95% CI 0.37, 0.64; p < 0.0001)

• BGF MDI vs. BFF MDI: rate ratio 0.82 (0.58, 1.17; p = 0.2792)

• BGF MDI vs. BUD/FORM DPI: rate ratio 0.83 (0.59, 1.18; p = 0.3120)

Findings: Mortality

Study not powered for mortality

Mortality¹²¹:

• FF/UMEC/VI (2.36%) vs. UMEC/VI (3.19%)

- HR for death 0.72 (95% CI 0.53, 0.99; p = 0.042)

• FF/UMEC/VI (2.36%) vs. FF/VI (2.64%)

- HR for death 0.89 (95% CI 0.67, 1.16; p = 0.387)

Mortality¹²²

• BUD320/GLY/FORM vs. BUD320/FORM

- HR 0.72 (95% CI 0.44, 1.16; p = 0.1721)

• BUD320/GLY/FORM vs. GLY/FORM

- HR 0.51 (95% CI 0.33, 0.80; p = 0.0035)

Not powered for mortality

Critique

• Customary study population (exacerbators with FEV₁ < 50%); allowing non-current asthma

• 2-week LAMA/LABA run-in

• Prior triple therapy: 0%

• Prior ICS therapy: 65%

• ICS withdrawal (mixed intervention)

• Unusual study population:

- allowing non-current asthma,

- inclusion of frequent exacerbators without severe airflow limitation (asthma like, FEV₁ > 50% GOLD 2D sub-cohort)

• Withdrawal of prior ICS (71%) and triple therapy (38%); mixed intervention

• Both exacerbation⁵⁶ and mortality⁶⁰ benefit essentially confined to the first 90 days of the study—representing ICS withdrawal in an “ICS-sensitive” sub-cohort

• Mortality indication rejected by regulatory agencies^116,117

• Unusual study population:

- allowing non-current asthma,

- inclusion of frequent exacerbators without severe airflow limitation (asthma like, FEV₁ > 50% GOLD 2D sub-cohort)

• Excluding patients with very severe airflow limitation (FEV₁ < 25%)

• Unusual 1–4-week pre-randomisation screening with long-acting bronchodilator withdrawal and 46% dropout, suggestive of significant run-in bias⁵¹

• Withdrawal of prior ICS (80%) and triple therapy (39%)—mixed intervention

• Mortality benefit⁶⁰ essentially confined to the first 90 days of the study—representing ICS withdrawal in an “ICS-sensitive” sub-cohort

• Allowed non-current asthma

• 71% of patients on ICS therapy prior to study

• Run-in withdrawal of long-acting bronchodilators associated with 38% pre-randomisation drop-out

^aPopulation criteria same across all four studies.

BDP beclometasone dipropionate, BGF budesonide/glycopyrrolate/formoterol fumarate, BUD budesonide, CAT COPD Assessment Test, CI confidence interval, COPD chronic obstructive pulmonary disease, DPI dry powder inhaler, FEV₁ forced expiratory volume in 1 s, FF fluticasone furoate, FORM formoterol fumarate, GFF glycopyrrolate/formoterol fumarate, GLY glycopyrronium, GOLD Global Initiative for Chronic Obstructive Lung Disease, HR hazard ratio, ICS inhaled corticosteroids, IND indacaterol, LABA long-acting β₂-agonist, LAMA long-acting muscarinic antagonist, MDI metered dose inhaler, OR odds ratio, pMDI pressurised metered dose inhaler, QID four times/day, RCT randomised controlled trial, RR relative risk, UMEC umeclidinium, VI vilanterol.