. 2020 Sep 1;202(5):660–671. doi: 10.1164/rccm.201912-2384PP

Table 1.

Summary of Studies Evaluating Eosinophils as a Biomarker to Predict Treatment Response for Patients with Chronic Obstructive Pulmonary Disease

Study	N	Key Inclusion Criteria: FEV₁ and Exacerbation History in Previous Year	Comparison	Results: Treatment Difference for Annual Exacerbation Rate for (a) Overall Population and (b) Blood Eosinophil Analysis
Studies of ICS/LABA vs. LABA
Pascoe et al., 2015 (34)	3,177	FEV₁ ≤ 70% predicted; ≥1 exacerbation in the previous year	Fluticasone furoate (ICS)/vilanterol (LABA) vs. vilanterol	a. 30% reduction^* (44)
				b. BEC ≥ 2%: 29% reduction^; BEC < 2%: 10% reduction^
				BEC ≥2 to <4%: 24% reduction^; BEC 4 to <6%: 32% reduction^; BEC ≥ 6%: 42% reduction^*
Siddiqui et al., 2015 (FORWARD) (35)	1,184	FEV₁ < 50% predicted; ≥1 exacerbation	Beclomethasone dipropionate (ICS)/formoterol (LABA) vs. formoterol	a. 28% reduction
				b. BEC ≥ 279.8 cells/μl: 46% reduction
				BEC < 279.8 cells/μl: 28% reduction
Bafadhel et al., 2018 (36)	4,528	Prebronchodilator FEV₁ ≤ 50% predicted; ≥1 exacerbation in the previous year	Budesonide (ICS)/formoterol (LABA) vs. formoterol	a. AER: 0.74 vs. 0.79^*
				b. BEC < 100 cells/μl: 25% increase to 22% reduction^*
				BEC 100–190 cells/μl: 25% reduction^*
				BEC 200–340 cells/μl: 26–50% reduction^*
				BEC 350–630 cells/μl: 51–60% reduction^*
Studies of ICS/LABA vs. LABA/LAMA
Lipson et al., 2018 (IMPACT) (37)	6,204	FEV₁ < 50% predicted and ≥1 moderate to severe exacerbation OR FEV₁ 50–80% predicted and ≥2 moderate exacerbations or 1 severe exacerbation	Fluticasone furoate (ICS)/vilanterol (LABA) vs. umeclidinium (LAMA)/vilanterol (LABA)	a. AER: 1.07 vs. 1.21^*
				b. BEC ≥ 150 cells/μl: 1.08 vs. 1.39^*
				BEC < 150 cells/μl: 1.06 vs. 0.97^*
Wedzicha et al., 2016 (FLAME) (47)	3,362	Post-bronchodilator FEV₁ ≥25% to <60% predicted; ≥1 exacerbation	Glycopyrronium (LAMA)/indacaterol (LABA) vs. fluticasone (ICS)/salmeterol (LABA)	a. 11% reduction^† (LAMA/LABA vs. ICS/LABA)
Post hoc analysis (Roche et al., 2017) (48)		Patients with BEC > 600 cells/μl were excluded		b. BEC < 2%: 20% reduction^† (LAMA/LABA vs. ICS/LABA)
				BEC ≥ 2%: 15% reduction^† (LAMA/LABA vs. ICS/LABA)
				a. 17% reduction^† (LAMA/LABA vs. ICS/LABA)
				b. BEC < 150 cells/μl: 28% reduction^† (LAMA/LABA vs. ICS/LABA)
				BEC 150 to <300 cells/μl: 11% reduction^† (LAMA/LABA vs. ICS/LABA)
				BEC 300–600 cells/μl: 7% reduction^† (LAMA/LABA vs. ICS/LABA)
Studies of ICS/LABA/LAMA vs. LABA/LAMA or LAMA
Papi et al., 2018 (TRIBUTE) (38)	1,532	FEV₁ < 50% predicted; ≥1 moderate to severe exacerbation in the previous year; receiving inhaled maintenance medication	Beclomethasone dipropionate (ICS)/formoterol fumarate (LABA)/glycopyrronium (LAMA) vs. indacaterol (LABA)/glycopyrronium (LAMA)	a. 15% reduction^*
				b. BEC < 200 cells/μl: 13% reduction^*
				BEC ≥ 200 cells/μl: 20% reduction^*
Lipson et al., 2018 (IMPACT) (37)	10,355	FEV₁ < 50% predicted and ≥1 moderate to severe exacerbation OR FEV₁ 50–80% predicted and ≥2 moderate exacerbations or 1 severe exacerbation in the previous year	Fluticasone furoate (ICS)/vilanterol (LABA)/umeclidinium (LAMA) vs. fluticasone furoate (ICS)/vilanterol (LABA) vs. umeclidinium (LAMA)/vilanterol (LABA)	a. ICS/LABA/LAMA vs. LABA/LAMA: 25% reduction^*
				ICS/LABA/LAMA vs. ICS/LABA: 15% reduction^*
				b. ICS/LABA/LAMA vs. LABA/LAMA, BEC ≥ 150 cells/μl: 32% reduction^*
				ICS/LABA/LAMA vs. LABA/LAMA, BEC < 150 cells/μl: 12% reduction^*
Vestbo et al., 2017 (TRINITY) (39)	2,691	FEV₁ < 50%; ≥1 moderate to severe COPD exacerbation	Beclomethasone dipropionate (ICS)/formoterol fumarate (LABA)/glycopyrronium (LAMA) (fixed triple) vs. tiotropium (LAMA) and beclomethasone dipropionate (ICS)/formoterol fumarate (LABA)/tiotropium (LAMA) (open triple)	a. Fixed 20% reduction, open 21% reduction^*
				b. BEC < 2%: fixed 7% reduction,^* open 9% reduction^*
				BEC ≥ 2%: fixed 30% reduction,^* open 31% reduction^*
				BEC < 200 cells/μl: fixed 8% reduction,^* open 9% reduction^*
				BEC ≥ 200 cells/μl: fixed 36% reduction,^* open 38% reduction^*
ICS withdrawal studies
Watz et al., 2016 (WISDOM) (49)	2,296	FEV₁ < 50% predicted; ≥1 exacerbation	Tiotropium (LAMA)/salmeterol (LABA)/fluticasone (ICS). Two arms: first group continues treatment for 52 wk; second initiates stepwise reduction of ICS every 6 wk down to placebo	a. ICS withdrawal vs. continuation: 10% increase^*
				b. BEC < 2%: 2% increase^; BEC ≥ 2%: 22% increase^; BEC ≥ 4%: 63% increase^; BEC ≥ 5%: 82% increase^
				BEC < 300 cells/μl^: 4% increase^; BEC ≥ 300 cells/μl^: 56% increase^
				BEC < 400 cells/μl: 7% increase^; BEC ≥ 400 cells/μl: 73% increase^
Calverley et al., 2017 (50)	2,420			b. ≥1 exacerbation in prior year AND: BEC ≥ 300 cells/μl: 45% increase^; BEC ≥ 400 cells/μl: 25% increase^≥2 exacerbations in prior year AND: BEC ≥ 300 cells/μl: 75% increase^; BEC ≥ 400 cells/μl: 196% increase^

Chapman et al., 2018 (SUNSET) (51)	527	Post-bronchodilator FEV₁ ≥40% to <80% predicted; ≤1 exacerbation	Tiotropium (LAMA)/salmeterol (LABA)/fluticasone (ICS). Two arms: first group continues with triple therapy; second switches to glycopyrronium (LAMA)/indacaterol (LABA)	a. ICS withdrawal vs. continuation 8% increase^*
Chapman et al., 2018 (SUNSET) (51)	527			b. BEC ≥ 300 cells/μl: 86% increase^*
Studies of monoclonal antibodies
Pavord et al., 2017 (METREX) (23)	462	Post-bronchodilator FEV₁ >20% to ≤80% predicted; ≥2 moderate or ≥1 severe exacerbation; BEC ≥ 150 cells/μl at baseline or ≥300 cells/μl in the previous year	Mepolizumab vs. placebo	a. 2% reduction^*
Pavord et al., 2017 (METREX) (23)	462		Mepolizumab vs. placebo	b. BEC ≥ 150 cells/μl at baseline or ≥300 cells/μl in the prior year: 18% reduction^*
Pavord et al., 2017 (METREO) (23)	675	Post-bronchodilator FEV₁ >20% to ≤80% predicted; ≥2 moderate or ≥1 severe exacerbation; BEC ≥ 150 cells/μl at baseline or ≥300 cells/μl in the previous year	Mepolizumab vs. placebo	a. NA (all patients had BEC ≥ 150 cells/μl at screening or ≥300 cells/μl during the previous year)
Pavord et al., 2017 (METREO) (23)	675		Mepolizumab vs. placebo	b. BEC ≥ 150 cells/μl at screening or ≥300 cells/μl during the previous year: 20% reduction^*
Criner et al., 2019 (GALATHEA) (24)	1,656	Post-bronchodilator FEV₁ >20% to ≤65% predicted; ≥2 moderate or ≥1 severe exacerbation; BEC ≥ 220 cells/μl	Benralizumab vs. placebo	a. NA
Criner et al., 2019 (GALATHEA) (24)	1,656		Benralizumab vs. placebo	b. BEC ≥ 220 cells/μl: 30 mg, 4% reduction; 100 mg, 17% reduction^*
Criner et al., 2019 (TERRANOVA) (24)	2,254	Post-bronchodilator FEV₁ >20% to ≤65% predicted; ≥2 moderate or ≥1 severe exacerbation; BEC ≥ 220 cells/μl	Benralizumab vs. placebo	a. NA
Criner et al., 2019 (TERRANOVA) (24)	2,254		Benralizumab vs. placebo	b. BEC ≥ 220 cells/μl: 10 mg, 15% reduction; 30 mg, 4% increase; 100 mg, 7% reduction^*
Criner et al., 2019 (GALATHEA/TERRANOVA prespecified analysis of pooled data) (24)	2,665	Post-bronchodilator FEV₁ >20% to ≤65% predicted; ≥2 moderate or ≥1 severe exacerbation; BEC ≥ 220 cells/μl	Benralizumab (100 mg) vs. placebo	a. NA
				b. BEC ≥ 220 cells/μl: 12% reduction^*
				BEC ≥ 220 cells/μl AND: ≥3 exacerbations in prior year: 31% reduction^*
				FEV₁ < 40% predicted: 24% reduction^*
				Post-bronchodilator response ≥ 15%: 33%
				≥3 exacerbations in the prior year and receiving triple therapy: 30% reduction^*
Studies of PDE4 inhibitors
Martinez et al., 2018 (REACT/RE²SPOND) (53)	4,299	FEV₁ ≤ 50% predicted; ≥2 exacerbations	Roflumilast vs. placebo	a. 12% reduction^*
				b. BEC ≥ 150 cells/μl: 19% reduction^*
				BEC ≥150 to <300 cells/μl: 16% reduction^*
				BEC ≥ 300 cells/μl: 23% reduction^*
				Prior hospitalization for COPD exacerbation AND:
				BEC ≥ 150 cells/μl: 35% reduction^*
				BEC ≥ 300 cells/μl: 43% reduction^*

Definition of abbreviations: AER = annualized exacerbation rate; BEC = blood eosinophil count; COPD = chronic obstructive pulmonary disease; FORWARD = Foster 48-Week Trial to Reduce Exacerbations in COPD; ICS = inhaled corticosteroids; IMPACT = Informing the Pathway of COPD Treatment; LABA = long-acting β₂-agonist; LAMA = long-acting muscarinic antagonist; METREO = Mepolizumab vs. Placebo as Add-on Treatment for Frequently Exacerbating COPD Patients Characterized by Eosinophil Level; METREX = Mepolizumab vs. Placebo as Add-on Treatment for Frequently Exacerbating COPD Patients; NA = not available; PDE4 = phosphodiesterase-4; REACT = Roflumilast in the Prevention of COPD Exacerbations While Taking Appropriate Combination Treatment; RE²SPOND = Roflumilast Effect on Exacerbations in Patients on Dual (LABA/ICS) Therapy; SUNSET = Study to Understand the Safety and Efficacy of ICS Withdrawal from Triple Therapy in COPD; WISDOM = Withdrawal of Inhaled Steroids during Optimized Bronchodilator Management.

Exacerbation rate for moderate and severe exacerbations.

^{^†}

Exacerbation rate for mild, moderate, and severe exacerbations.