Received: 9 March 2022 DOI: 10.1002/clt2.12182 ORIGINAL ARTICLE - - Revised: 6 June 2022 Accepted: 1 July 2022 How reliably can algorithms identify eosinophilic asthma phenotypes using non‐invasive biomarkers? Diana Betancor1 | José María Olaguibel2,3 | José Manuel Rodrigo‐Muñoz3,4 | Ebymar Arismendi3,5 | Pilar Barranco3,6 | Blanca Barroso1 | Irina Bobolea3,5 | Blanca Cárdaba3,4 | María Jesús Cruz3,7 | Elena Curto3,8 | Victoria Del Pozo3,4 | Francisco‐Javier González‐Barcala9 | Carlos Martínez‐Rivera3,10 | Joaquim Mullol3,11 | Xavier Muñoz3,12 | Cesar Picado5 | Vicente Plaza3,8 | Santiago Quirce3,6 | Manuel Jorge Rial3,13 | Lorena Soto3,8 | Antonio Valero3,5 | Marcela Valverde‐Monge1,3 | Joaquin Sastre1,3 1 Servicio de Alergología, Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain 2 Servicio de Alergología, Hospital Universitario de Navarra, Pamplona, Navarra, Spain 3 CIBER de Enfermedades Respiratorias (CIBERES), Madrid, Spain 4 Servicio de Inmunología, Instituto de Investigación Sanitaria Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain 5 Allergy Unit & Severe Asthma Unit, Pneumonology and Allergy Department, Hospital Clínic, IDIBAPS, Universitat de Barcelona, Barcelona, Spain 6 Servicio de Alergia, Hospital Universitario La Paz, IdiPAZ, Madrid, Spain 7 Departamento de Biología Celular, Fisiología e Inmunología, Universitat Autónoma de Barcelona, Barcelona, Spain 8 Departamento de Medicina Respiratoria, Hospital de la Santa Creu i Sant Pau, Instituto de Investigación Biomédica Sant Pau (IIB Sant Pau), Universidad Autónoma de Barcelona. Departamento de Medicina, Barcelona, Spain 9 Servicio de Neumología, Complejo Hospitalario Universitario de Santiago, Santiago de Compostela, La Coruña, Spain 10 Servicio de Neumología, Hospital Germans Trias i Pujol, Institut d’Investigació Germans Trias i Pujol, Universitat Autònoma de Barcelona, Badalona, Barcelona, Spain 11 Rhinology Unit & Smell Clinic, ENT Department, Clinical and Experimental Respiratory Immunoallergy (IDIBAPS), Universitat de Barcelona, Barcelona, Catalonia, Spain 12 Servicio de Neumología, Hospital Vall d’Hebron, Barcelona, Spain 13 Servicio de Alergología, Complexo Hospitalario Universitario A Coruña, A Coruña, Spain Correspondence Diana Betancor, Allergy Department, Hospital Abstract Universitario Fundación Jiménez Díaz, Av Background and Aims: Asthma is a heterogeneous respiratory disease that encom- Reyes Católicos, 2, 28040 Madrid, Spain. Email: diana13_b@hotmail.com passes different inflammatory and functional endophenotypes. Many non‐invasive biomarkers has been investigated to its pathobiology. Heany et al proposed a clin- Funding information ical algorithm that classifies severe asthmatic patients into likely‐eosinophilic phe- Instituto de Salud Carlos III, Ministry of Science and Innovation, Government of Spain, notypes, based on accessible biomarkers: PBE, current treatment, FeNO, presence Grant/Award Number: CIBERES of nasal polyps (NP) and age of onset. Materials and Methods: We assessed the concordance between the algorithm proposed by Heany et al. with sputum examination, the gold standard, in 145 asthmatic patients of the MEGA cohort with varying grades of severity. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, pro- vided the original work is properly cited. © 2022 The Authors. Clinical and Translational Allergy published by John Wiley & Sons Ltd on behalf of European Academy of Allergy and Clinical Immunology. Clin Transl Allergy. 2022;e12182. wileyonlinelibrary.com/journal/clt2 1 of 6 https://doi.org/10.1002/clt2.12182 2 of 6 - BETANCOR ET AL. Results: No correlation was found between both classifications 0.025 (CI = 0.013– 0.037). Moreover, no relationship was found between sputum eosinophilia and pe- ripheral blood eosinophilia count in the total studied population. Discussion and Conclusion: In conclusion, our results suggest that grouping the biomarkers proposed by Heany et al. are insufficient to diagnose eosinophilic phe- notypes in asthmatic patients. Sputum analysis remains the gold standard to assess airway inflammation. KEYWORDS asthma, biomarkers, eosinophils, exhaled nitric oxide, non‐eosinophilic, phenotypes, sputum 1 | INTRODUCTION Quantitative variables were described as mean and standard deviation, and qualitative variables as absolute and relative fre- Asthma is a respiratory syndrome characterised by airway inflam- quencies. Inter‐group comparisons were performed using chi‐ mation and reversible airway obstruction.1 Due to its heterogeneity, square test or Fisher exact test for qualitative and ANOVA or considerable efforts have been made to subclassify the disease into Kruskal–Wallis for quantitative variables. Agreement was assessed different phenotypes and identify non‐invasive biomarkers that with the kappa coefficient. Correlations were estimated by Spear- reflect its pathobiology. Sputum examination is the gold standard for man's R. Statistical analysis was carried out using the GraphPad determining airway inflammation; other non‐invasive biomarkers Instat 6 (GraphPad Software). p values <0.05 were considered studied to date, such as peripheral blood eosinophil (PBE) count, significant. serum periostin, fraction of exhaled nitric oxide (FeNO), and serum IgE levels, have low specificity and sensitivity.2 Heany et al3 proposed a clinical algorithm that classifies severe asthmatic patients based on 3 | RESULTS the likelihood of an eosinophilic phenotype using easily accessible biomarkers such as PBE, current treatment, FeNO, presence of nasal Data from 145 asthmatic subjects aged 18–75 years were cat- polyps (NP), and age of onset. This algorithm reflects he criteria of egorised according to the phenotypes proposed by Heany et al.3 the Global Initiative for Asthma (GINA). Grade 3 (likely eosinophilic) was the most prevalent (69.6%), followed We analysed the consistency between the Heany et al. algo- by grade 2 (likely eosinophilic; 20.8%), grade 1 (less likely; 16.8%); rithm3 and sputum examination in a retrospective analysis of asth- and grade 0 (non‐eosinophilic; 5.9%). The average patient age was matic patients with varying degrees of severity from eight Spanish 48 years, and a majority were female. There was no significant dif- hospitals, previously described as the MEGA cohort.4 As secondary ference in demographic characteristics, asthma severity, exacerba- outcomes, we evaluated the clinical characteristics, asthma severity, tions, or lung function between grades. As FeNO, PBE, and NP were and lung function in these phenotypes.4,5 classification criteria, higher levels were shown in grade 3 (p < 0.05). Data are summarised in Table 1. The agreement between the eosinophilia grades proposed by 2 | MATERIAL AND METHODS Heany et al3 (grade 2–3) and eosinophilic sputum for the MEGA cohort was 0.025 (95% CI = 0.013–0.037). Patients from the MEGA cohort with a valid sputum analysis and an Though the sputum eosinophilia rate was higher in grade 3, it 4,5 accurate asthma diagnosis were selected. A total of 145 patients was not statistically significant compared to the other grades (see were included. A retrospective observational study was conducted by Table 1 and Figure 1). We examined the relationship of sputum eo- reviewing the MEGA cohort electronic database. Asthma was diag- sinophils with PBE count, finding no significant correlation in the total nosed in patients with an FEV1 increase of greater than 200 ml and 12% population (r = 0.11, p = 0.22). No correlation was found considering on spirometry and/or methacholine PC20 < 16 mg/ml. Sputum eosin- different grades (r = −0.17 (p = 0.93), r = 0.08 (p = 0.78), r = 0.35 ophilia was defined as >3% eosinophils. The ethics committees of each (p = 0.12), and r = 0.16 (p = 0.16)) for grades 0, 1, 2, and 3, participating hospital approved this study. All subjects provided signed respectively. informed consent. Study data included demographic and clinical characteristics, asthma severity (following GINA guidelines1) and control (assessed 4 | DISCUSSION using the Asthma Control Test), treatment, number of exacerbations, and exacerbation severity. Lung function tests (spirometry and Several biomarkers have been considered to phenotype asthmatic plethysmography), sputum eosinophil count, PBE, and FeNO were patients. Heany et al. proposed a system of eosinophilic probability also collected at baseline. depending on easily accessible biomarkers.3 The low agreement BETANCOR ET AL. - 3 of 6 TABLE 1 Demographic, clinical, and diagnostic test results of studied patients Grade 0 (non‐ Grade 1 (less likely Grade 2 (likely Grade 3 (most likely eosinophilic) eosinophilic) eosinophilic) eosinophilic) No. subjects (%) 6 (5.9) 17 (16.8) 21 (20.8) 101 (69.6) Demographic characteristics Individual characteristics Female sex, N (%) 3 (50) 9 (52.9) 12 (57.1) 66 (65.3) NS Mean age, years, mean (SD) 47 (13) 52.1 (11.4) 47 (13) 48 (13) NS Body mass index (BMI), mean (SD) 27.2 (5.3) 26.9 (5.3) 26.9 (5.2) 26.8 (5.2) NS Obesity,1 N (%) 2 (33.3) 3 (17.6) 2 (9.5) 16 (15.8) NS Residency, urban area, N (%) 5 (83.3) 8 (47.1) 16 (76.2) 78 (77.2) NS (0.06) Comorbidities, N (%) Atopy2 5 (83.3) 13 (76.5) 19 (90.5) 75 (74.3) NS Allergic rhinitis 3 (50) 11 (64.7) 13 (61.9) 58 (57.4) NS Bronchiectasis 1 (16.6) 1 (5.9) 2 (9.5) 11 (10.9) NS CRSwNP 0 (0) 4 (23.5) 1 (4.8) 52 (51.5) 0.02 CRSsNP 3 (50) 1 (5.9) 4 (19.1) 13 (12.9) 0.05 Obstructive sleep apnoea syndrome (OSAS) 1 (16.6) 0 (0) 1 (4.8) 4 (3.9) NS Smoking habit, N (%) Never smoker 4 (66.6) 9 (52.9) 12 (57.1) 55 (54.5) NS Current smoker 0 (0) 2 (17.6) 1 (4.8) 11 (10.9) NS Ex‐smoker 2 (33.3) 6 (35.3) 8 (38.1) 31 (30.7) NS Education level, N (%) Higher education 3 (50) 12 (70.6) 11 (52.4) 42 (48.3) NS Primary education 1 (16.6) 3 (17.6) 10 (47.6) 44 (50.6) NS a a a No studies 2 (33.3) 2 (11.8) 0 (0) 1 (1.1) <0.0001 Clinical characteristics Treatment, N (%) ICS/LABA 6 (100) 16 (94.11) 19 (90.5) 89 (88.1) NS Long‐term OCS 0 (0) 0 (0) 2 (9.5) 14 (13.9) NS 3 Asthma severity, N (%) Intermittent 0 (0) 0 (0) 1 (4.8) 5 (4.9) NS Mild persistent 2 (33.3) 2 (11.8) 4 (19.1) 15 (14.8) NS Moderate persistent 2 (33.3) 6 (35.3) 4 (19.1) 28 (27.7) NS Severe persistent 2 (33.3) 7 (41.2) 12 (57.1) 53 (52.5) NS Exacerbations, N (%) Patients with asthma exacerbation during 4 (66.6) 6 (35.3) 16 (76.2) 53 (52.5) NS (0.07) previous year Severe asthma exacerbation 0 (0) 1 (5.8) 7 (33.3) 16 (15.8) NS (0.07) Exacerbations over the previous year, mean 2.7 (1.2) 1.5 (0.8) 1.9 (1.6) 3.5 (3.7) NS (SD) Emergency department (ED) visits 1 (1.3) 0.4 (0.8) 0.6 (0.95) 0.6 (1.4) NS ≥5 ED visits 0 (0) 0 (0) 0 (0) 7 (6.9) NS (Continues) 4 of 6 - BETANCOR ET AL. T A B L E 1 (Continued) Grade 0 (non‐ Grade 1 (less likely Grade 2 (likely Grade 3 (most likely eosinophilic) eosinophilic) eosinophilic) eosinophilic) ICU Admission 0 (0) 0.1 (0.3) 0 (0) 0.1 (0.3) NS Asthma control in ACT, N (%) Completely controlled 2 (33.3) 5 (29.4) 7 (33.2)b 18 (17.8)b 0.01 Well‐controlled 3 (50) 6 (35.3) 9 (42.8) 49 (48.5) NS Poorly controlled 1 (16.6) 6 (35.3) 5 (23.8) 34 (33.6) NS Respiratory function tests and other biomarkers Total IgE, IU/mL, mean (SD) 574.4 (855.2) 417.0 (532.1) 615.0 (171.0) 414.5 (634.9) NS Peripheral eosinophilia cells/μL mean (SD) 92.2 (37.9)c 103.9 (34.8)c 357.1 (347.1)c 357.1 (337.3)c <0.001 Spirometry, litres, mean (SD) FEV1 2.2 (1.1) 2.3 (0.7) 2.7 (0.8) 2.6 (0.8) NS FVC 3.1 (1.3) 3.5 (1.1) 3.7 (0.8) 3.6 (0.9) NS FEV1/FVC 72.4 (10.1) 67.9 (7.6) 71.9 (11.4) 71.2 (9.3) NS 4 Positive spirometry bronchodilator test, N 4 (66.7) 4 (23.5) 4 (19.1) 21 (20.8) (%) Plethysmography, litres, mean (SD) TLC 5.2 (1.5) 6.2 (1.3) 6.8 (0.5) 5.9 (1.1) NS RV 2.1 (0.4) 2.3 (0.8) 2.1 (0.7) 2.1 (0.8) NS 5 Functional spirometry phenotype, N (%) Normal 2 (33.3) 8 (47.1) 10 (47.6) 63 (63.6) NS Obstructive 0 (0) 3 (17.6) 5 (23.9) 15 (15.2) NS Air trapping 4 (66.7) 6 (35.3) 6 (28.6) 21 (21.2) NS (0.06) d d d FeNO, ppb, mean (SD) 18.8 (4.2) 28.8 (22.5) 44.2 (39.7) 55.3 (46.2) 0.03 Methacholine challenge, PC20 mean (SD) 0.5 (0.6) 1.8 (2.3) 0.2 (0) 2.0 (2.9) NS Sputum analysis Sputum eosinophilia, mean (SD) 1.7 (1.8) 8.9 (16.4) 7.6 (15.1) 11.4 (20.2) NS Patients with sputum eosinophils >3%, N (%) 2 (33.3) 8 (47.1) 11 (52.4) 48 (47.5) NS Cellular profile, N (%) Eosinophilic 1 (16.7) 6 (35.3) 7 (33.3) 45 (44.5) NS c c Mixed 1 (16.6) 2 (11.8) 4 (19.0) 3 (2.9) 0.03 Neutrophilic 1 (16.6) 5 (29.4) 4 (19.0) 10 (9.9) NS Paucigranulocytic 3 (50) 4 (23.5) 6 (28.6) 43 (42.6) NS Abbreviations: ACT, Asthma Control Test; CRSsNP, chronic rhinosinusitis without nasal polyposis; CRSwNP, chronic rhinosinusitis with nasal polyposis; FEV1, forced expiratory volume during the first second; FVC, forced vital capacity; ICS/LABA, inhaled corticosteroids/long‐acting beta 2‐agonists; ICU, intensive care unit; NS, not statistically significant; OCS, oral corticosteroids; RV, residual volume; TLC, total lung capacity. a Significant differences were determined between grades 0–2 and 0–3. b Significant difference between both. c Significant differences were determined between all inter‐group analysis except 0–1. d Significant differences were determined between grade 0–3 and 1–3. 1 Obesity was defined as BMI over 30 kg/m2. 2 Atopy was defined as the presence of at least 1 skin prick test or specific serum IgE positive to common allergens. 3 Asthma severity was assessed following the Global Initiative for Asthma (GINA) guidelines. 4 Positive bronchodilator test was defined as an increase in FEV1 of greater than 200 ml and more than 12% of the baseline value 15 min after the administration of 2 puffs of salbutamol. 5 Functional phenotype was defined as normal when FEV1 >80% and FEV1/FVC >70%, obstructive when FEV1/FVC <70%, and air trapping when FVC <80% or a 12% FVC increase after bronchodilation. BETANCOR ET AL. - 5 of 6 one limitation of our study. Further limitations include a possible bias toward recruiting predominantly eosinophilic patients in tertiary centres and allergy clinics, which could be confounding given the limitations in finding significant differences in unequally sized groups. In conclusion, our results suggest that the biomarker groupings proposed by Heany et al. are insufficient to diagnose eosinophilic phenotypes in asthmatic patients. Sputum analysis remains the gold standard to assess airway inflammation. A U T H O R C O N T R I B U T I O NS Diana Betancor: Data curation (equal); Formal analysis (equal); Methodology (equal); Writing – review & editing (equal). Jose Maria Olaguibel: Data curation (equal); Formal analysis (equal); Investigation F I G U R E 1 Concordance between values of eosinophils in (equal); Methodology (equal); Project administration (equal); Writing – sputum and Heany et al grades original draft (equal); Writing – review & editing (equal). Jose Manuel Rodrigo Munoz: Data curation (equal); Resources (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal). Ebymar found in our study between this algorithm and sputum analysis Arismendi: Formal analysis (equal); Resources (equal); Validation suggests no relationship between the two criteria. (equal); Writing – review & editing (equal). Pilar Barranco: Formal We observed higher sputum eosinophilia levels in grade 3 of analysis (equal); Resources (equal); Validation (equal); Writing – re- Heany's classification, but without reaching statistical significance. view & editing (equal). Blanca Barroso: Data curation (equal); Re- The percentage of patients with sputum eosinophils >3% was similar sources (equal); Writing – original draft (equal); Writing – review & in grades 2–3 (eosinophilic) and 1, indicating an inability of the pro- editing (equal). Irina Bobolea: Formal analysis (equal); Resources posed system to phenotype these patients. (equal); Validation (equal); Writing – review & editing (equal). Blanca As in other studies showing a 70%–80% prevalence of the Cardaba: Formal analysis (equal); Resources (equal); Validation eosinophilic phenotype in tertiary centers,3,6 the 69.6% rate in our (equal); Writing – review & editing (equal). Maria Jesus Cruz: Formal study contrasts with the 50% previously proposed.1 Furthermore, analysis (equal); Resources (equal); Validation (equal); Writing – re- our study found a 5.9% prevalence of non‐T2 asthma, similar to the view & editing (equal). Elena Curto: Formal analysis (equal); Resources 5% reported by Kerkhof et al.7 (equal); Validation (equal); Writing – review & editing (equal). Victoria Worse lung function is associated with eosinophilic pheno- Del Pozo: Formal analysis (equal); Resources (equal); Validation types,6,8,9 as demonstrated in our previous report characterising (equal); Writing – review & editing; (equal). Francisco‐Javier 5 MEGA patients. No differences in lung function were found in the Gonzalez‐Barcala: Formal analysis (equal); Resources (equal); Vali- present study, likely owing to the presence of sputum eosinophilia in dation (equal); Writing – review & editing (equal). Carlos Martinez‐ all grades. Rivera: Formal analysis (equal); Resources (equal); Validation (equal); A clear correlation has been reported between eosinophilic Writing – review & editing (equal). Joaquim Mullol: Formal analysis 6,8,9 asthma and greater severity and exacerbations. Higher exacer- (equal); Resources (equal); Validation (equal); Writing – review & bation rates and severity were found in the “eosinophilic grades” of editing (equal). Xavier Munoz: Formal analysis (equal); Resources our study, without reaching a statistical difference (p = 0.07). (equal); Validation (equal); Writing – review & editing (equal). Cesar Some biomarkers used in the Heany classification are easily Picado: Formal analysis (equal); Resources (equal); Validation (equal); influenced by external factors, such as treatments and other illnesses Writing – review & editing (equal). Vicente Plaza: Formal analysis in the case of PBE, and age, gender, atopy, and tobacco and food/ (equal); Resources (equal); Validation (equal); Writing – review & beverage consumption for FeNO. Wagener et al.10 estimated the editing (equal). Santiago Quirce: Formal analysis (equal); Resources sensitivity of PBE and FeNO in diagnosing eosinophilic phenotypes at (equal); Validation (equal); Writing – review & editing (equal). Manuel around 89% and 78%, respectively. Nevertheless, Lemière et al.9 Rial Prado: Formal analysis (equal); Resources (equal); Validation found no correlation between FeNO and sputum eosinophilia. PBE (equal); Writing – review & editing (equal). Lorena Soto‐Retes: Formal has been suggested as a sputum eosinophilia predictor,10 though we analysis (equal); Resources (equal); Validation (equal); Writing – re- found no correlation between sputum eosinophils and PBE, thus view & editing (equal). Antonio Valero: Formal analysis (equal); Re- resembling other studies.2,8 Due to this external variability, evidence sources (equal); Validation (equal); Writing – review & editing (equal). supporting these biomarkers in phenotyping asthma remains unclear. Marcela Valverde: Formal analysis (equal); Resources (equal); Vali- Kjarsgaard et al demonstrated the presence of free eosinophil dation (equal); Writing – review & editing (equal). Joaquin Sastre: Data granules in the airway in the absence of intact eosinophils.11 curation (equal); Formal analysis (equal); Methodology (equal); Project Detecting intact eosinophils only, as in our study, may give misleading administration (equal); Resources (equal); Validation (equal); Writing – information on the real prevalence of eosinophilic sputum, marking original draft (equal); Writing – review & editing (equal). 6 of 6 - BETANCOR ET AL. A CK N O WL ED GE M EN T S Joaquim Mullol https://orcid.org/0000-0003-3463-5007 This study was funded by CIBERES, Instituto de Salud Carlos III, Lorena Soto https://orcid.org/0000-0001-6046-9204 Ministry of Science and Innovation, Government of Spain. R E F E R E NC E S C O N F LI C T O F IN T ER E S T 1. Masoli M, Fabian D, Holt S, Beasley R. Global Initiative for Asthma Dr. Betancor is supported by a Rio Hortega Research Contract from (GINA) Program. The global burden of asthma: executive summary of the GINA Dissemination Committee report. Allergy. 2004;59(5): Instituto Carlos III, Spanish Ministry of Science. Dr. Valverde has 469‐478. https://doi.org/10.1111/j.1398‐9995.2004.00526.x received lecturing fees from GSK and sits on the advisory board for 2. Aleman F, Lim HF, Nair P. Eosinophilic endotype of asthma. Immunol Organon. Dr. Rial reports receiving personal fees from GSK, Allergy Allergy Clin. 2016;36(3):559‐568. https://doi.org/10.1016/j.iac.2016. Therapeutics, and AstraZeneca outside the submitted work. Dr. 03.006 3. Heaney LG, Perez de Llano L, Al‐Ahmad M, et al. Eosinophilic and González Barcala reports personal fees from ALK, AstraZeneca, Bial, noneosinophilic asthma: an expert consensus framework to char- Boehringer Ingelheim, Chiesi, Gebro Pharma, GlaxoSmithKline, Lab- acterize phenotypes in a global real‐life severe asthma cohort. Chest. oratorios Esteve, Menarini, Mundipharma, Novartis, Rovi, Roxall, 2021;160(3):814‐830. https://doi.org/10.1016/j.chest.2021.04.013 Stallergenes‐Greer, Teva, as well as grants from Mundipharma 4. Muñoz X, Álvarez‐Puebla MJ, Arismendi E, et al. The MEGA project: a study of the mechanisms involved in the genesis and disease outside the submitted work. Dr. Quirce reports personal fees from course of asthma. Asthma cohort creation and long‐term follow‐up. AstraZeneca, Novartis, Sanofi, Boehringer Ingelheim, Teva, ALK, Arch Bronconeumol (Engl Ed). 2018;54(7):278‐385. https://doi.org/10. Mundipharma, GSK, Chiesi, and Leti outside the submitted work. Dr. 1016/j.arbr.2018.05.011 Soto reports non‐financial support from CIBER de Enfermedades 5. Rial MJ, Álvarez‐Puebla MJ, Arismendi E, et al. Clinical and inflam- matory characteristics of patients with asthma in the Spanish MEGA Respiratorias (CIBERES) during the conduct of the study; outside of project cohort. Clin Transl Allergy. 2021;11(1):e12001. https://doi. the present work, she reports personal fees from Stallergennes‐ org/10.1002/clt2.12001 Greer, Menarini, and Novartis, personal fees from GSK, Hal Allergy, 6. Moore WC, Meyers DA, Wenzel SE, et al. Identification of asthma Allergy Therapeutics, AstraZeneca, and grants from Sociedad Espa- phenotypes using cluster analysis in the Severe Asthma Research ñola de Alergología e Inmunología Clínica SEAIC and Sociedad Program. Am J Respir Crit Care Med. 2010;181(4):315‐894. https:// doi.org/10.1164/rccm.200906‐0896oc Española de Neumología y Cirugía Torácica SEPAR. Dr. Martinez 7. Kerkhof M, Tran T, Zangrilli J, Carter V, Price D. Eosinophilic asthma Rivera reports having received grants and personal fees from phenotypes in the UK population. Eur Respir J. 2020;56(64):2059. AstraZeneca, Teva, GSK, Novartis, and Mundipharma outside the 8. Ten Brinke A, Zwinderman AH, Sterk PJ, Rabe KF, Bel EH. Factors submitted work. Dr. Munoz reports personal fees from AstraZeneca, associated with persistent airflow limitation in severe asthma. Am J Respir Crit Care Med. 2001;164(5):744‐748. https://doi.org/10.1164/ Boehringer Ingelheim, GlaxoSmithKline, Novartis, Teva, Mundifarma, ajrccm.164.5.2011026 Chiesi, and Faes outside the submitted work. Dr. Sastre reports 9. Lemière C, Ernst P, Olivenstein R, et al. Airway inflammation grants and personal fees from Sanofi, GSK, Novartis, AstraZeneca, assessed by invasive and noninvasive means in severe asthma: Mundipharma, and Faes Farma outside the submitted work. Dr. eosinophilic and noneosinophilic phenotypes. J Allergy Clin Immunol. 2006;118(5):1033‐1039. https://doi.org/10.1016/j.jaci.2006.08.003 Olaguibel reports grants from Sanofi and/or personal fees from 10. Wagener AH, de Nijs SB, Lutter R, et al. External validation of blood AstraZeneca and Mundipharma outside the submitted work. Dr. eosinophils, FENO and serum periostin as surrogates for sputum Plaza reports grants and personal fees from AstraZeneca, Boehringer eosinophils in asthma. Thorax. 2015;70(2):115‐120. https://doi.org/ Ingelheim, Merck, Chiesi, Novartis, Menarini, and Sanofi outside the 10.1136/thoraxjnl‐2014‐205634 11. Kjarsgaard M, Adatia A, Bhalla A, et al. Underestimation of airway submitted work. Dr. Mullol reports personal and other fees from luminal eosinophilia by quantitative sputum cytometry. Allergy Asthma Sanofi‐Genzyme & Regeneron, Novartis, Viatris (Mylan pharma), Clin Immunol. 2021;17(1):63. https://doi.org/10.1186/s13223‐021‐ Uriach group, Mitsubishi‐Tanabe, Menarini, UCB, AstraZeneca, GSK, 00567‐w and MSD outside the submitted work. Dr. del Pozo reports personal and other fees from Sanofi, AstraZeneca, and GSK outside the sub- mitted work. All other authors have no conflicts of interest. How to cite this article: Betancor D, Olaguibel JM, Rodrigo‐ Muñoz JM, et al. How reliably can algorithms identify O RC ID eosinophilic asthma phenotypes using non‐invasive Diana Betancor https://orcid.org/0000-0003-2765-3581 biomarkers? Clin Transl Allergy. 2022;e12182. https://doi.org/ Francisco‐Javier González‐Barcala https://orcid.org/0000-0001- 10.1002/clt2.12182 5847-4784