# paper.pdf ## Metadata - PDFFormatVersion=1.7 - Language=en-US - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260607070513+05'30' - Creator=Adobe InDesign 19.5 (Windows) - ModDate=D:20260607070521+05'30' - Producer=Adobe PDF Library 17.0 - Title=Bronchodilator response is linked with uncontrolled moderate‐to‐severe childhood asthma and elevated IL‐4 and IL‐13 - Trapped=False - Custom.WPS-ARTICLEDOI=10.1111/pai.70392 - Custom.WPS-JOURNALDOI=10.1111/(ISSN)1399-3038 - Custom.WPS-PROCLEVEL=3 - jav:journal_article_version=VoR - prism:doi=10.1111/pai.70392 - pdfx:wps-articledoi=10.1111/pai.70392 - pdfx:wps-journaldoi=10.1111/(ISSN)1399-3038 - pdfx:wps-proclevel=3 - dc:title=Bronchodilator response is linked with uncontrolled moderate‐to‐severe childhood asthma and elevated IL‐4 and IL‐13 ## Contents ### Page 1 Pediatr Allergy Immunol. 2026;37:e70392. | 1 of 15https://doi.org/10.1111/pai.70392wileyonlinelibrary.com/journal/paiReceived: 2 March 2026 | Revised: 19 May 2026 | Accepted: 22 May 2026DOI: 10.1111/pai.70392O R I G I N A L A R T I C L E Bronchodilator response is linked with uncontrolled moderate-to-severe childhood asthma and elevated IL-4 and IL-13 Nariman K. A. Metwally1,2,3,4,5 | Simone Hashimoto1,2,3,4 | Susanne J. H. Vijverberg1,2,3 |Anne H. Neerincx1 | Barbara S. Dierdorp2,6 | Tamara Dekker 2,6 | Eric G. Haarman 4 |Jan Willem Duitman1,2,6 | Mario Gorenjak7 | Antoaneta A. Toncheva 8 |Susanne Harner 8 | Susanne Brandstetter 9 | Christine Wolff 9 |Paula Corcuera-Elosegui10 | Leyre López-Fernández10 | Olaia Sardón-Prado10,11 |Maria Pino-Yanes12,13,14 | Uroš Potočnik7,15,16 | Michael Kabesch 8,9 |Aletta D. Kraneveld17 | René Lutter1,2,6 | Suzanne W. J. Terheggen-Lagro 4 |Mahmoud I. Abdel-Aziz1,2,3,18,19 | Anke H. Maitland-van der Zee1,2,3,4,18 |on behalf of the SysPharmPediA Consortium CorrespondenceMahmoud I. Abdel-Aziz, Department ofPulmonary Medicine, Amsterdam UMC,University of Amsterdam, Amsterdam, theNetherlands. Email: m.i.ibrahim@amsterdamumc.nlFunding informationActive and Assisted Living Programme;ZonMw; The German Ministry ofEducation and Research (BMBF); Ministryof Higher Education, Science andInnovation of the Republic of Slovenia(MVZI)Editor: Ayobami AkenroyeAbstract Background: Bronchodilator response (BDR) is a key clinical feature in childhoodasthma, but its relation to asthma pathophysiology is not fully understood. This studyexamines the associations between BDR, disease control, and serum cytokines/ chemokines in children with moderate-to-severe asthma (MSA).Methods: BDR was assessed in 140 children (aged 6–17 years, 41% females) from theSysPharmPediA cohort using the ERS/ATS 2022 guidelines (ΔFEV1 >10% predicted)and z-score (ΔFEV1-z-score >0.78) post-bronchodilator definitions. Risk of uncon- trolled asthma in relation to BDR was estimated by a logistic regression model, adjust- ing for baseline lung function, age, sex, BMI z-score, ethnicity, country, season, GINAstep, and smoking exposure. Thirty-nine serum proteins were measured using LuminexMultiplex Assay; levels below the detection limit (LOD) were imputed as LOD/√2, andproteins with >40% missing values across samples were excluded. Children with highand low BDR were compared for serum proteins by linear regression model, adjustingfor covariates and applying false discovery rate correction.Results: Children with high BDR (21%) had significantly higher odds of uncontrolledasthma (adjusted OR ≈ 3.22, 95% CIs: 1.07–11.3) and more frequent severe exacer- bations in the past year compared with those with low BDR (70% vs. 46%, p < .05).This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,provided the original work is properly cited. © 2026 The Author(s). Pediatric Allergy and Immunology published by European Academy of Allergy and Clinical Immunology and John Wiley & Sons Ltd. Mahmoud I. Abdel-Aziz and Anke H. Maitland-van der Zee contributed equally to this study.For affiliations refer to page 13. ### Page 2 2 of 15 |METWALLY et al.High-BDR children showed elevated serum levels of IL-13, IL- 4, TNF-α, IL-6, IL-7, IL-8,IL- 1β, and MMP-1 (q < 0.05).Conclusion: In children with MSA, a high BDR is independently associated with poorerasthma control and a distinct systemic inflammatory profile involving T2 and non-T2mediators. BDR may serve as a marker for asthma phenotyping.K E Y W O R D Sasthma control, bronchodilator response, moderate-to-severe pediatric asthma, non-invasivephenotyping, type 2 inflammationG R A P H I C A L A B S T R A C TThe contents of this page will be used as part of the graphical abstract of html only. It will not be published as part of main article. High BDR (≥10% FEV1) predicts 3-fold higher odds of poor asthma control in children with moderate-to-severe asthma. High BDR linksto elevated T2 cytokines (IL-4, IL-13) and non-T2 mediators (IL-1β, IL-6, IL-7, IL-8, TNF-α), as well as MMP-1. BDR offers non-invasivephenotyping potential to advance personalized asthma management. ### Page 3 | 3 of 15METWALLY et al.1 | I NTRO D U C TI O NAsthma is the most common chronic respiratory condition in chil- dren, with moderate-to-severe asthma (MSA) representing a smalleryet clinically significant subgroup that poses major challenges.1 Thissubgroup incurs at least twice the healthcare costs, along with higherrates of hospitalization and emergency department visits, and a di- minished quality of life compared with children who have less severedisease.1 Despite advances in therapeutic management, many chil - dren remain poorly controlled, experiencing frequent exacerbationsand persistent symptoms,1 possibly elevating their long-term risk ofchronic obstructive pulmonary disease (COPD) in adulthood. This complexity of MSA is further amplified by its multi-layered heterogeneity, driven by dynamic gene–environmentinteractions and variable treatment responses. 2 Simple “one-size-fits-all” categorizations fail to adequately reflect the pathophysi- ological and clinical diversity of the disease, especially in children.Therefore, accurate phenotyping of pediatric asthma is essential.Conventional phenotyping utilizes the classification of Type 2(T2)-high and T2-low asthma phenotypes, distinguished by levelsof T2-driven airway inflammation. However, not only can childrenfluctuate between these phenotypes, but many present with over- lapping or ambiguous features.3 While induced sputum remainsthe gold standard for direct assessment of airway inflammation,its invasiveness and procedural requirements limit its widespreaduse in routine pediatric care. 4–6 As an alternative, blood eosino - phil and neutrophil counts have been proposed for their accessi- bility,6 but they still require invasive sampling via venipuncture,which can be particularly challenging in children. 7 Moreover, thesemarkers may not reliably reflect airway- specific inflammation andare susceptible to extrinsic confounders, including infections andatopic comorbidities, diminishing their diagnostic specificity. 8,9Fractional exhaled nitric oxide (FeNO) offers a non-invasive optionfor asthma phenotyping and is occasionally used to monitor med-ication adherence.10 However, its clinical value remains debated,as it can be confounded by medication intake, including cortico- steroids, which are commonly used in MSA, reducing its reliabilityin treated patients.11 Emerging evidence suggests that bronchodilator response (BDR)may serve as a non-invasive marker linked to asthma pathophysi-ology.12 BDR, quantified as the percentage increase in forced ex - piratory volume in 1 s (FEV1) after administration of a short-actingβ2-agonist, is a routine component of spirometric evaluation.12 WhileBDR is widely used to support asthma diagnosis, multiple studiesindicate that a higher BDR is associated with poorer asthma control in both children and adults.13,14 These findings raise the possibilitythat BDR can contribute to asthma phenotyping. However, previ-ous studies have primarily evaluated BDR in broader asthma pop- ulations, while its clinical utility specifically in children with MSA,as well as its relationship with systemic inflammatory profiles, re- mains insufficiently explored. We hypothesize that BDR may serveas a non-invasive phenotypic marker in children with MSA and pro- vide additional insights into inflammatory heterogeneity beyondcurrently available biomarkers. The aims of this study were there- fore to (1) assess whether BDR is associated with asthma control and(2) determine whether BDR is linked to distinct systemic inflamma- tory profiles in children with MSA.2 | M E TH O DS 2.1 | Study designA cross-sectional analysis was performed using data from theSysPharmPediA cohort—an international, multicenter study of 145children aged 6–17 years with MSA, recruited from the Netherlands,Germany, Slovenia, and Spain. The research was performed in linewith the principles of the Declaration of Helsinki and ethics approvalwas obtained from all participating sites. Informed consent (with as- sent from children where appropriate) was secured from all partici- pants, and additional cohort details were described previously.15 Aschematic representation of the study design is shown in Figure 1.2.2 | Participant and data collectionsEligibility criteria for the original SysPharmPediA cohort have previ-ously been described in detail at Clini calTrials.gov (NCT04865575).Briefly, children aged 6–17 years with physician-confirmed MSA,treated according to Global Initiative for Asthma (GINA) step 3 orhigher, were eligible for inclusion. Children were excluded in caseof recent antibiotic use within the previous month. In addition, fail- ure to provide written informed consent from parents/caregiversand/or recruited children, when appropriate, resulted in exclusionfrom study participation. Only children with both baseline and post-bronchodilator spirometry data were included in the present study (n = 140). Study-specific questionnaires recorded demographic data,medical history, environmental exposures (e.g., smoking), medica- tion use, and self-reported adherence, which was measured usingthe validated 5-item Medication Adherence Report Scale (MARS-5).16 FeNO was measured at 50 mL/s according to ERS/ATS guide-lines.17 Atopic sensitization (ever) was defined as a positive skinprick test (wheal ≥3 mm) and/or allergen-specific IgE ≥0.35 kU/L.Peripheral blood was collected for differential leukocyte counts (eo- sinophils, neutrophils) via fluorescence flow cytometry.6 AdditionalKey messageIn children with moderate-to-severe asthma, a high bron- chodilator response serves as a simple, noninvasive markerof poor disease control and a distinct systemic inflamma- tory signature characterized by elevated IL-4 and IL-13,supporting precision phenotyping and targeted treatment strategies. ### Page 4 4 of 15 |METWALLY et al.methodological details are available in the Supplementary Materialand described previously.152.3 | Asthma control assessmentUncontrolled asthma was defined as current medication use with ≥1severe exacerbation requiring oral corticosteroids, hospitalization,or emergency room (ER) visit within the past year, or a childhoodAsthma Control Test (cACT) or Asthma Control Test (ACT) ≤19.182.4 | Lung function and BDRA total of 140 out of 145 participants performed spirometry. The pro- cedure took place prior to and 15 min post administering 400 μg ofinhaled salbutamol, following (ERS/ATS) guidelines.17 High BDR wasdefined as an increase in FEV1 >10% of predicted post-bronchodilator,based on ERS/ATS 2022 criteria.19 Additionally, another definitionbased on z-scores was applied, where a high BDR corresponds toan improvement greater than 0.78 in the post-bronchodilator FEV1z- score.20 Both definitions account for demographic differences inspirometry. A representation of the lung function test measurementusing a spirometer is shown in Figure S1.2.5 | Serum protein (cytokine and chemokine)analysisThe serum levels of 39 inflammatory mediators were measured usingLuminex multiplex immunoassays (R&D Systems Inc., Minneapolis,MN; Invitrogen/ProcartaPlex) with fresh, non-thawed aliquots, aspreviously described.6,21 For IL-5, an assay with high sensitivity fromR&D Systems Inc. (Minneapolis, MN) was employed for analysis. Dataacquisitions were carried out using the Bio-Plex 200 system (Bio-Rad,Hercules, CA, USA). Data points with fewer than 25 beads were con- sidered low quality and excluded from further analysis. For analytesbelow the limit of detection (LOD), imputation by LOD/√2, multipliedF I G U R E 1 Study Workflow and Methodology. This flowchart outlines the study design, from participant selection to data analysis. Alldata collection, including asthma control evaluation, blood sampling, and spirometry testing, was performed during patient visits as partof the core SysPharmPediA cohort protocol prior to this analysis. The original cohort included 145 participants, of whom 140 completedspirometry testing and were included in the current study; 5 participants were excluded due to missing spirometry data. Source: Created inBioRender. ### Page 5 | 5 of 15METWALLY et al.by a random factor (0.75–1.25) was performed to minimize skewnessin variable distribution, as previously described.21 Proteins with >40%missing across all samples were excluded. The full list of the 39 meas- ured inflammatory mediators is provided in Table S1.2.6 | Statistical analysisComparisons between high and low BDR groups (both BDR defini- tions) used Mann–Whitney U tests for continuous variables andchi-square or Fisher's exact tests for categorical data, as appro- priate. Logistic regression assessed associations between BDRand risk of uncontrolled asthma, adjusting for baseline FEV1, age,sex, BMI z-score, ethnicity, country, season, GINA step, and cur- rent smoking exposure (as defined by a Directed Acyclic Graph inFigure S2). Since both BDR definitions already incorporate age,sex, and ethnicity in their calculation, we additionally ran modelsexcluding these variables as covariates to assess the robustnessof our findings. Serum protein levels were compared using Mann–Whitney Utests with Benjamini-Hochberg FDR correction applied for mul- tiple testing. Multiple linear regression models further adjustedfor confounders in the DAG to test for differences in protein lev-els; FDR correction was applied to p-values as well. To explorepotential functional relationships between the key cytokines andinflammatory mediators significantly associated with high BDR,we utilized the STRING database to analyze protein–protein inter-actions (PPI). Pathway enrichment analysis was conducted usingthe Reactome (REAC) database via g:Profiler on the proteins sig- nificantly associated with high BDR group, with FDR <0.05 usedto show significantly enriched pathways. All tests were two-tailedwith α < 0.05 considered significant. Analyses were completedusing R version 4.4.3 and RStudio IDE 2024.12.1, employing thestats, plyr, dplyr, tidyr, reshape2, stringr, ggplot2, ggpubr, andrspiro packages.3 | R E S U LT S 3.1 | Demographic and clinicalcharacteristics of the study participantsThis clinical and demographic information, stratified by BDRstatus according to the >10% predicted FEV1 definition, is sum - marized in Tables 1–3. The median age of the participants was11.97 years (IQR: 9.65–14.00), and 39% were female. Based on theERS/ATS 2022 definition, 21% (n = 30) were classified as havinghigh BDR (>10% increase in predicted FEV1). Children with highBDR had significantly poorer asthma control than those with lowBDR: 80% had uncontrolled asthma compared to 58% of low BDRchildren (p = .028). Severe asthma exacerbations in the past yearwere also more frequent in the high BDR group (70% vs. 46%,p = .022). Asthma Control Test (ACT) z-scores were lower in thehigh BDR group, and this difference was statistically significant(median z-score: 0.55 vs. 1.03, p = .05). Ethnic differences were noted, with mixed/other ethnicitiesoverrepresented among high BDR children (20% vs. 7%), whileCaucasian children were more prevalent in the low BDR group (63%vs. 81%, p < .05). Sensitization to aeroallergens was also more com- mon in high BDR children: sensitization to dog dander (61% vs. 25%,p = .001) and, to a lesser extent and not statistically significant, catdander (52% vs. 33%, p = .078). Baseline lung function was markedly reduced in high BDR chil- dren, with a median pre-bronchodilator FEV1 % predicted of 85.1compared to 96.5 in low BDR children (p < .001). Post-bronchodilatorFEV1 did not differ significantly between groups. No differenceswere observed for blood eosinophil (at 0.3 × 109/L or 0.5 × 109/Lcut-offs) or neutrophil counts or for FeNO. All children were treatedwith inhaled corticosteroids. Use of biologics (omalizumab or mepo- lizumab) tended to be higher among high BDR children (20% vs. 8%,p = .091), as was the use of leukotriene receptor antagonists (LTRA)(30% vs. 15%, p = .05). Medication adherence was high across allgroups (≥93%), with no significant differences between high and lowBDR children.When applying the alternative z-score definition (ΔFEV1 z-score >0.78), 27% of children (n = 38) were classified as having highBDR (Figure S3). Findings were consistent with those observed forthe 10% FEV1 definition, with additional significant differences iden- tified in the high BDR group, specifically higher rates of cat sensitiza- tion and use of biologics (Table S2).3.2 | Association of BDR with asthma controlMultivariable logistic regression showed that children with high BDR (>10%) had significantly higher odds of uncontrolled asthmacompared to those with low BDR, after adjustment for covariates(OR: 3.22; 95% CI: 1.07–11.3; p = .048) (Table 4). GINA treatmentstep was also strongly associated with asthma control: children atstep 5 were associated with increased odds of uncontrolled asthmacompared to those at step 3 (OR: 16.2; 95% CI: 2.22–343; p = .018),and step 4 also showed a significant association (OR: 2.69; 95% CI:1.03–7.33; p = .046). Other covariates, including baseline FEV1, age,sex, BMI z-score, ethnicity, country, inclusion season, and smokingexposure, were not predictive of asthma control. Analyses using thez-score definition showed a similar pattern as the 10% FEV1 defini-tion (Table S3). Comparable results were also obtained using twodefinitions when models were adjusted for a reduced set of covari- ates excluding age, sex, and ethnicity (Tables S4 and S5).3.3 | Association of BDR with serum inflammatorymarkersChildren with high BDR (>10% FEV1) demonstrated significantlyhigher serum levels of interleukin (IL)-4, IL-13, tumor necrosis ### Page 6 6 of 15 |METWALLY et al.factor-α (TNF-α), matrix metalloprotease (MMP-1), IL-6, IL-7, mono- cyte chemoattractant protein-4 (MCP-4, CCL13) thymus and acti- vation regulated chemokine (TARC, CCL17), and IL-1β compared tochildren with low BDR. After correction for multiple comparisonsusing FDR, IL-4, IL-13, TNF-α, and MMP-1 remained significantly el-evated (q < 0.05). These findings are illustrated in Figure 2, which presents box- plots of cytokine and chemokine levels of inflammatory mediators,and in Figure 3, where heatmaps demonstrate distinct separationbetween high and low BDR groups based on the abundance of theseinflammatory mediators. For the z-score definition and followingFDR correction, additional proteins to the previous proteins, IL-6,IL-7, and IL-10 had significantly higher serum levels in children withhigh BDR (all q < 0.05) (Figures S4 and S5).After adjustment for potential confounders and FDR correc- tion, high BDR (>10% FEV1) remained significantly associated withTA B L E 1 Demographic and clinical characteristics of children with high and low BDR according to the 10% FEV1 definition.Characteristics Total (N = 140) Low BDR (ΔFEV1≤10%) (n = 110) High BDR (ΔFEV1>10%) (n = 30) p valueAge in years, median (IQR) 11.97 (9.65, 14.00) 11.52 (9.56, 13.80) 13.35 (10.21, 14.10) 0.164Sex (Female), n (%) 55/140 (39%) 39/102 (38%) 16/38 (42%) 0.677Ethnicity, n (%) 0.013Caucasian 107/139 (77%) 88/109 (81%) 19/30 (63%) Latino 10/139 (7%) 9/109 (8%) 1/30 (3%) African 6/139 (4%) 4/109 (4%) 2/30 (7%) Asian 2/139 (1%) 0/109 (0%) 2/30 (7%)Mixed/Others 14/139 (10%) 8/109 (7%) 6/30 (20%)BMI z-score, median (IQR) 0.57 (−0.32, 1.38) 0.63 (−0.30, 1.36) 0.47 (−0.48, 1.67) 0.953 Birth and early life factors, n (%) Mode of delivery (Caesarean section) 26/136 (19%) 21/107 (20%) 5/29 (17%) 0.772Breastfeeding >4 months 58/134 (43%) 4.00 (0.00, 9.00)(n = 105) 4.00 (1.00, 11.00)(n = 29)0.801Smoking Exposure, n (%) Smoking exposure during pregnancy 34/127 (27%) 27/102 (26%) 7/25 (28%) 0.877Current smoke exposure 40/135 (30%) 32/107 (30%) 8/28 (29%) 0.89Country of Inclusion, n (%) 0.098Spain 50/140 (36%) 43/110 (39%) 7/30 (23%)Germany 37/140 (26%) 30/110 (27%) 7/30 (23%)The Netherlands 32/140 (23%) 20/110 (18%) 12/30 (40%) Slovenia 21/140 (15%) 17/110 (15%) 4/30 (13%) Inclusion season, n (%) 0.06Winter 27/140 (19%) 23/110 (21%) 4/30 (13%) Spring 38/140 (27%) 33/110 (30%) 5/30 (17%) Summer 46/140 (33%) 30/110 (27%) 16/30 (53%) Autumn 29/140 (21%) 24/110 (22%) 5/30 (17%) Clinical Status, n (%) Uncontrolled Asthmaa 88/140 (63%) 64/110 (58%) 24/30 (80%) 0.028(Childhood) Asthma Control Test ((c) ACT) z- score,median (IQR) 0.95 (0.30, 1.55)(n = 138) 1.03 (0.30, 1.55)(n = 109) 0.55 (0.12, 1.03)(n = 29)0.05Severe asthma exacerbations in the past year 72/140 (51%) 51/110 (46%) 21/30 (70%) 0.022Note: Demographic, and clinical characteristics are presented for the total cohort and stratified by BDR status according to the >10% increase inpredicted FEV1 definition (high BDR). The low BDR group comprised children who did not meet the respective BDR criterion. Continuous variablesare presented as medians with interquartile ranges, while categorical variables are shown as counts and percentages. Comparisons between highBDR and low BDR groups were conducted using the Mann–Whitney U test for continuous data and either the chi-squared test or Fisher's exact testfor categorical data, as appropriate. A p-value below 0.05 showed evidence of a significant difference between groups and is highlighted in boldface. Abbreviations: BMI, body mass inded; IQR, interquartile range.aUncontrolled asthma is characterized by a (childhood) Asthma Control Test (cACT) score of 19 or below and/or the occurrence of severeexacerbations within the previous year that necessitated hospitalization, emergency room visits, or the use of oral corticosteroids (OCS). ### Page 7 | 7 of 15METWALLY et al.elevated serum levels of multiple inflammatory mediators in themultiple linear regression models. The strongest associations wereobserved for MMP-1, IL-13, and IL-4, with additional significant find- ings for IL-7, IL-8, IL-6, IL-1β, and TNF-α (Figure 4). Results basedon the z-score definition were largely consistent, although IL-10additionally emerged as significant, whereas MMP-1 was no longerassociated (Figure S6).STRING analysis of the significantly elevated proteins in chil- dren with high BDR revealed an interconnected network, and theseproteins likely contribute to overlapping signaling pathways relatedto immune activation and cytokine-mediated airway inflammation,suggesting functional associations among these inflammatorymediators (Figure S7).3.4 | Pathway enrichment analysisPathway enrichment analysis for the high BDR enriched proteins(MMP-1, IL-4, IL-13, IL-6, IL-7, IL-8, TNF, IL-10, and IL-1β) revealedstrong enrichment for several immune-related pathways, with themost significant being IL-4 and IL-13 signaling (padj = 2.20 × 10 −13 ),and IL-10 signaling (padj = 3.27 × 10 −8) (Figure S8).TA B L E 2 Lung function, FeNO, white blood cell counts, and atopic sensitization in children with high and low BDR according to the 10%FEV1 definition.Characteristics Total (N = 140) Low BDR (ΔFEV1≤10%) (n = 110) High BDR (ΔFEV1>10%) (n = 30) p valueLung functionFEV1 pre-salbutamol % predicted, median (IQR) 94.88 (82.68, 103.29) 96.47 (86.12, 104.44) 85.14 (79.83, 94.84) 0.002FEV1 post-salbutamol % predicted, median (IQR) 99.55 (89.97, 108.76) 99.66 (89.44, 108.16) 99.08 (95.81, 109.95) 0.631 FEV1 pre-salbutamol z-score, median (IQR) −0.42 (−1.45, 0.28) −0.30 (−1.18, 0.37) −1.25 (−1.68, −0.45) 0.002FEV1 post-salbutamol z-score, median (IQR) −0.04 (−0.84, 0.73) −0.03 (−0.92, 0.71) −0.08 (−0.36, 0.84) 0.633FeNO (ppb), median (IQR) 16.00 (8.83, 36.90)(n = 122) 14.00 (8.15, 33.92)(n = 98) 18.80 (10.38, 46.83)(n = 24)0.244White blood cell count (×10 9/L), median (IQR)Eosinophil 0.37 (0.21, 0.64)(n = 123) 0.37 (0.20, 0.62)(n = 97) 0.36 (0.24, 0.66)(n = 26) 0.546Neutrophil 3.24 (2.42, 4.20)(n = 123) 3.20 (2.43, 4.12)(n = 97) 3.29 (2.33, 4.42)(n = 26)0.66Lymphocyte 2.58 (2.23, 3.10)(n = 123) 2.57 (2.23, 3.10)(n = 97) 2.62 (2.11, 3.10)(n = 26)0.68Eosinophilia, n (%)>0.3 × 109/L 78/123 (63%) 61/97 (63%) 17/26 (65%) 0.814>0.5 × 109/L 44/123 (36%) 35/97 (36%) 9/26 (35%) 0.89Atopy and allergic sensitization, n (%)aAtopy 119/134 (89%) 93/105 (89%) 26/29 (90%) 1Aeroallergen combined 119/136 (88%) 94/107 (88%) 25/29 (86%) 0.759 HDM 95/133 (71%) 75/104 (72%) 20/29 (69%) 0.74 Grass pollen 75/133 (56%) 57/104 (55%) 18/29 (62%) 0.486 Mold 13/110 (12%) 8/86 (9%) 5/24 (21%) 0.152 Cat 46/125 (37%) 33/100 (33%) 13/25 (52%) 0.078 Dog 38/119 (32%) 24/96 (25%) 14/23 (61%) 0.001IgE, abnormal 98/113 (87%) 74/86 (86%) 24/27 (89%) 1Allergic rhinitis (ever), n (%) 101/133 (76%) 82/104 (79%) 19/29 (66%) 0.138Atopic dermatitis (Eczema) 52/128 (41%) 40/100 (40%) 12/28 (43%) 0.786Note: Lung function, FeNO, white blood cell counts, and atopy characteristics are presented for the total cohort and stratified by BDR statusaccording to the >10% increase in predicted FEV1 definition (high BDR). The low BDR group comprised children who did not meet the respectiveBDR criterion. Continuous variables are presented as medians with interquartile ranges, while categorical variables are shown as counts andpercentages. Comparisons between high BDR and low BDR groups were conducted using the Mann–Whitney U test for continuous data and eitherthe chi-squared test or Fisher's exact test for categorical data, as appropriate. A p-value below 0.05 showed evidence of a significant differencebetween groups. Boldface values indicate statistically significant differences of p-values p < (0.05). Abbreviations: FEV1, forced expiratory volume in 1 second; FeNO, fractional exhaled nitric oxide; IQR, interquartile range.aAtopic sensitization is defined based on a physician-documented history of sensitization to airborne allergens, demonstrated by a positive skin pricktest (wheal diameter ≥3 mm) and/or elevated allergen-specific IgE levels (≥0.35 kU/L). ### Page 8 8 of 15 |METWALLY et al.4 | D I S CUS S I O NThis study provides novel evidence that BDR, a widely used, non-invasive measure of airflow reversibility, can serve as a meaningfulphenotypic marker of disease burden and systemic inflammationin children with MSA. Our findings demonstrate that children withhigh BDR, despite inhaled corticosteroid treatment, are morelikely to have uncontrolled asthma and a systemic inflammatoryprofile characterized by elevations in both T2 and non-T2 inflam- matory mediators. Taken together, these results support the useof BDR as more than a physiologic diagnostic tool, revealing itspotential role in pediatric asthma phenotyping and personalizedtreatment strategies. Our analyses demonstrated that high BDR is an independentpredictor of uncontrolled asthma, among children with MSA exhibit- ing approximately a threefold increased risk of poor asthma controlafter adjusting for multiple confounders. The association persistedacross additional analyses that excluded demographic covariatesalready accounted for in BDR reference equations, confirming therobustness of the observed relationship. Importantly, high BDR may reflect improvement arising from anunderlying deficit in airway function rather than a superior therapeu- tic response. This indicates significant airway hyper-responsivenesscharacterized by exaggerated airway narrowing and instability.Accordingly, the ‘room for improvement’ signaled by a high BDRserves as a marker of airway hyper-responsiveness, conferring in- creased risk of uncontrolled asthma and exacerbations despite bron-chodilator responsiveness.Although children with high BDR exhibited lower baseline FEV1,their post-bronchodilator values were normalized, consistent with areversible airflow obstruction pattern. However, FEV1 itself did notindependently predict uncontrolled asthma in the adjusted model. This contrast highlights that BDR provides additional prognosticvalue as a dynamic measure of airway responsiveness rather than astatic marker of airflow limitation. Medication patterns reinforced the clinical relevance of highBDR. While all children received ICS and exhibited similar use ofshort- and long-acting β2-agonists, those with high BDR were morelikely to be treated with LTRA and biologic therapies. This patternreflects more severe disease and the need for controller therapy inchildren exhibiting high BDR. Notably MARS-5 adherence was highacross groups as only compliant children were included in this studyby design,22 suggesting treatment differences are unlikely due topoor adherence. Our results align with previous pediatric data. Galant et al.demonstrated that high BDR phenotype (≥10%–12% FEV1), evenin untreated children with otherwise normal baseline spirom- etry, was strongly associated with poor asthma control, atopy,nocturnal symptoms, and increased β2-agonist use.23 Similarly,Coverstone et al. (2019) linked BDR with severe asthma, reducedlung function, and higher healthcare utilization.13 In contrast toour findings, Coverstone et al. (2019) observed elevated FeNOlevels in children with high BDR 13; however, FeNO levels did notsignificantly differ between BDR groups in our cohort. This dif- ference might reflect variability in ICS use or adherence, as ICStherapy is known to reduce airway inflammation and FeNO levels,TA B L E 3 Asthma medication use and treatment steps in children with high and low BDR according to the 10% FEV1 definition.Characteristics Total (N = 140) Low BDR (ΔFEV1 ≤10%) (n = 110) High BDR (ΔFEV1 >10%) (n = 30) p valueAsthma medication n (%) ICS 140 (100%) 110 (100%) 30 (100%) N/A SABA 130/140 (93%) 102/110 (93%) 28/30 (93%) 1LABA 132/140 (94%) 105/110 (95%) 27/30 (90%) 0.368OCS 4/140 (3%) 2/110 (2%) 2/30 (7%) 0.201LTRA 25/140 (18%) 16/110 (15%) 9/30 (30%) 0.05Anticholinergics 17/140 (12%) 15/110 (14%) 2/30 (7%) 0.527 Biologics (Omalizumab/Mepolizumab) 15/140 (11%) 9/110 (8%) 6/30 (20%) 0.091GINA Steps, n (%) 0.15 Step 3 (GINA) 63/140 (45%) 53/110 (48%) 10/30 (33%) Step 4 (GINA) 61/140 (44%) 47/110 (43%) 14/30 (47%) Step 5 (GINA) 16/140 (11%) 10/110 (9%) 6/30 (20%) MARS-5 (≥21), n (%) 117/126 (93%) 95/101 (94%) 22/25 (88%) 0.38Note: Treatment characteristics are presented for the total cohort and stratified by BDR status according to the >10% increase in predicted FEV1definition (high BDR). The low BDR group comprised children who did not meet the respective BDR criterion. Continuous variables are presented asmedians with interquartile ranges, while categorical variables are shown as counts and percentages. Comparisons between high BDR and low BDRgroups were conducted using the Mann–Whitney U test for continuous data and either the chi-squared test or Fisher's exact test for categoricaldata, as appropriate. A p-value below 0.05 showed evidence of a significant difference between groups and is highlighted in boldface. Abberviations: GINA, global initiative for asthma; ICS, inhaled corticosteroid; IQR, interquartile range; LABA, long-acting β-agonist; LTRA, leukotrieneantagonist; MARS-5, Medication Adherence Report Scale-5; OCS, oral corticosteroid; SABA, short-acting β- agonist. ### Page 9 | 9 of 15METWALLY et al.which is consistent with the high treatment adherence reported in our study.13,24 Notably, similar patterns have been observed in adult asthmapopulations. Godoy Fernandes et al., Ferrer Galván et al., andHeffler et al. showed that persistent high BDR, even with treat- ment, predicted poor asthma control and future risk.14,25,26 Morerecently, Kaminsky et al. examined BDR across diverse study pop- ulations and found no association with overall asthma control orsymptom burden. 27 However, their pediatric subgroup analysisrevealed that higher BDR correlated with increased symptom bur- den, underscoring the importance of pediatric-specific researchsince findings from mixed-age cohorts may not fully apply to chil-dren with asthma.19,27 Interestingly, in our study, children with high BDR demon- strated disproportionately high sensitization to dog and cat al- lergens. This is in line with previous pediatric studies linkingsensitization to dog and cat allergens with greater asthma sever- ity, impaired pulmonary function, and increased symptom bur-den.28,29 Persistent exposure to pet allergens may contribute tochronic airway inflammation and airway hyperresponsiveness,potentially reinforcing the association between allergen sensitiza-tion and elevated BDR observed in our cohort.Ethnic disparities were also evident, with mixed and non-Caucasian children overrepresented in the high BDR group. Thisobservation is consistent with prior evidence demonstrating greaterasthma severity among non-Caucasian populations, driven by acomplex interplay of social determinants, including environmentalexposures, housing conditions, allergen burden, and healthcare ac- cess, as well as genetic susceptibilities.19,27,30 Importantly, geneticfactors linked to ethnicity may independently influence asthma out-comes or interact with socioeconomic variables to modulate disease severity. 30One of the most novel and impactful findings of our analysis is thecharacterization of systemic inflammatory profiles in relation to BDR.High BDR was linked to significantly elevated levels of classical type 2cytokines, including IL-4 and IL-13. This pattern reflects type 2 airwayinflammation, which is associated with frequent exacerbations andpoor asthma control. 31 IL- 4 and IL-13 are key mediators of T2 airwayinflammation and are associated with airway hyperresponsiveness,mucus overproduction, and structural remodeling in asthma.31TA B L E 4 Multivariate logistic regression model showing the association between high BDR according to the FEV1 >10% definition andasthma control after adjusting for confounders.Characteristic (n = 134) OR SE 95% CI p-valueBDR (FEV1 >10% of predicted) 3.22 0.591 1.07, 11.3 0.048Baseline FEV1 (L) 1.09 0.510 0.40, 2.99 0.870Age 0.94 0.135 0.72, 1.22 0.630Sex (male) 0.80 0.452 0.33, 1.94 0.620 BMI (Z- score) 0.99 0.178 0.70, 1.41 0.971 Ethnicity (other/mixed) 2.41 0.669 0.69, 10.0 0.188Country of inclusiona Netherlands 1.53 0.743 0.36, 6.76 0.566 Slovenia 1.56 0.706 0.39, 6.34 0.527 Spain 2.38 0.583 0.76, 7.66 0.137 Inclusion season bSpring 0.83 0.582 0.26, 2.59 0.751 Summer 0.53 0.657 0.14, 1.88 0.337Winter 1.08 0.672 0.29, 4.14 0.906GINA treatments stepcStep4 2.69 0.496 1.03, 7.33 0.046Step5 16.2 1.18 2.22, 343 0.018Current smoking exposure 0.81 0.471 0.32, 2.06 0.657Note: Multivariate logistic regression model showing the association between high bronchodilator response (BDR) according to the FEV1 >10%definition and asthma control after adjusting for different confounders. Odds ratios (ORs) and their 95% confidence intervals (CIs) for uncontrolledasthma are shown for each variable, with controlled asthma as the reference group. Variables include bronchodilator response, defined as >10%increase in predicted FEV1 (high1 BDR), baseline FEV1, age, sex, ethnicity, country of inclusion, BMI z-score, season of inclusion, GINA treatment step,and current smoking exposure. A p-value < .05 was considered statistically significant and is highlighted in boldface. Abbreviations: BDR, bronchodilator response; BMI, body mass index; CI, confidence interval; FEV1, forced expiratory volume in one second; GINA,Global Initiative for Asthma; OR, odds ratio; SE, standard error.aGermany was selected randomly as a reference group for comparison of country of inclusion.bAutumn was selected randomly as a reference group for comparison of inclusion season.cStep 3 was selected as a reference group for comparison of GINA treatment. ### Page 10 10 of 15 |METWALLY et al.Importantly, our data also revealed significant elevations in keypleiotropic pro-inflammatory cytokines including IL-6, IL-1β, andTNF-α. These cytokines, characteristic of non-T2 inflammation 32,33are known to drive neutrophilic recruitment, tissue injury, airwayremodeling, and steroid resistance. 34,35 Notably, IL-1β, TNF-α, andIL-6 may contribute to both non-T2 and T2 inflammatory responses,further highlighting the complexity of inflammatory pathways asso-ciated with high BDR. 36–38 Elevated serum IL-6 has been associatedwith a 24% increased odds of corticosteroid-requiring exacerbationsand more asthma severity symptoms. 39 Moreover, elevated IL-6 lev- els have been implicated in asthma exacerbations characterized byoverlapping eosinophilic and neutrophilic inflammation, which areoften associated with poorer pulmonary function relative to otherinflammatory phenotypes. 33,40 Collectively, these complex cyto- kine interactions contribute to asthma heterogeneity and treatmentresponsiveness.33,35 In addition, IL-8 and IL-7 were significantly elevated in chil- dren with high BDR. Elevated IL-8 (CXCL8) levels have been linkedto neutrophilic airway inflammation, airway remodeling, diseaseseverity, and corticosteroid resistance in asthma.33 On the otherhand, IL-7 has been implicated in eosinophilic inflammatory re- sponses in asthma through promotion of eosinophil activation andsurvival. 41Beyond these, MMP1 was significantly elevated only under thefirst BDR definition (10% FEV1). MMP-1, an interstitial collagenaseinduced partly by IL-4,42 contributes to bronchial hyperresponsive-ness and asthma exacerbations through its role in airway remodel-ing. 43 IL-10 was significantly elevated only using the second BDRdefinition (FEV1 z-score). Although traditionally regarded as anti-inflammatory, IL-10 has also been linked to T2 cytokine production,airway hyperreactivity, and virus-induced exacerbations.44–46 Theelevated IL-10 levels in this study may reflect both ongoing diseaseactivity and the known effect of corticosteroid therapy increasingendogenous IL-10.44 These evidence might suggest that these in- flammatory mediators might be involved in T2 and non-T2 pathways.The absence of statistical significance under the 10% FEV1 defini- tion may relate to differences in participant classification betweenBDR definitions, as well as the relatively larger sample size of thehigh-BDR group identified under the z-score definition, which mayhave slightly improved statistical power to detect differences in thiscytokine. Our study advances prior work from our group (AlizadehBahmani et al.), which found blood eosinophilia, defined by >300 cells/μL or >500 cells/μL, associated with higher IL-5 levels.6 Inour study, high BDR was not linked to blood eosinophils or IL-5 butrather with elevations in IL- 4 and IL-13, pointing to a distinct T2F I G U R E 2 Boxplots of serum protein levels of inflammatory markers in children with high BDR and low BDR, shown separately for BDRdefinition: >10% increase in predicted FEV1 (high BDR). Boxplots display the distribution of significantly different cytokines and proteinsbetween groups. Blue boxes represent the low BDR group, while red boxes represent the high BDR group. BDR, bronchodilator response;FEV1, forced expiratory volume in 1 s; CRP, C-reactive protein; IL, interleukin; MCP, monocyte chemoattractant protein; MMP, matrixmetalloproteinase; TARC, thymus and activation regulated chemokine; TNF-α, tumor necrosis factor alpha. A statistically significant q-value< .05 is highlighted in boldface. ### Page 11 | 11 of 15METWALLY et al.signature. IL-4 and IL-13 signal through a shared receptor complex,making this pathway crucial in the pathophysiology of asthma andtargeted for treatment.31 This emphasizes that BDR may provideadditional insights into pediatric asthma endotypes. While bloodeosinophils and IL-5 remain critical markers, especially for iden- tifying candidates for anti-IL-5 biologics, they are only part ofthe story. Blood eosinophil counts fluctuate widely, sometimesvarying by over 40%, reducing the reliability of single time-pointmeasures. 47 They also do not always mirror airway- specific in - flammation. 8,9 FeNO, though non-invasive, is greatly attenuatedby ICS therapy, limiting its discriminative use in treated cohortssuch as ours. 24 Although sputum analysis remains the clinical stan- dard for assessing airway inflammation, its routine use in childrenis limited by procedural complexity. 4–6 In contrast, BDR testing isnon-invasive, accessible, and easily repeatable, offering a valuabletool in clinical settings where frequent monitoring is needed. Therelationship between high BDR, disease burden, and systemic in- flammation highlights its utility in precision asthma care. Notably,elevated IL- 4 and IL-13 across both BDR definitions could supportthe use of BDR to identify children likely to benefit from biologicstargeting the IL- 4/IL-13 axis (e.g., dupilumab), which has demon- strated efficacy in improving lung function and reducing exacer- bations in pediatric MSA.48Although asthma is primarily an airway disease, systemic cyto- kine elevations may reflect ongoing airway inflammation and im- mune activation, particularly in children with persistent or severedisease. Elevated serum IL-4 and IL-13 levels in children with highBDR may therefore represent indirect systemic signatures of un- derlying T2 airway inflammation and airway hyperresponsiveness.Nevertheless, serum cytokine concentrations should be interpretedcautiously, as systemic inflammatory markers do not fully reflectlocal airway inflammatory processes.F I G U R E 3 Heatmaps of serum cytokine and chemokine profiles in children with high and low BDR shown for BDR >10% increasein predicted FEV1 (high BDR). Each row represents a cytokine or protein, and each column represents an individual participant. Colorintensity reflects relative expression levels, illustrating the separation between high BDR and low BDR groups; CRP = C-reactive protein; IL,interleukin; MCP, monocyte chemoattractant protein; MMP, matrix metalloproteinase; TARC, thymus and activation regulated chemokine;TNF-α, tumor necrosis factor alpha. ### Page 12 12 of 15 |METWALLY et al.Our study has several strengths, including a comprehensive,multi-dimensional phenotyping approach that integrates clinical,physiological, and molecular data. We evaluated BDR using two rec- ommended thresholds and observed consistent results, indicatingthat the choice of cut-off did not affect the associations. This sup- ports BDR as a robust and reproducible marker of pediatric asthmaphenotypes. Additionally, we applied rigorous adjustment for poten- tial confounders to strengthen the validity of our findings. Anotherkey strength is the focus on serum cytokine profiling, through whichwe are the first to link BDR with the immunologic heterogeneity ofpediatric MSA, providing valuable insights that complement tradi- tional clinical and physiological markers. However, some limitations should be noted. The cross-sectionaldesign precludes causal inference. The relatively small number ofchildren with high-BDR may have limited statistical power for sub- group analyses and reduced the precision of some effect estimates.In addition, serum cytokines may not fully reflect airway inflam- mation due to compartmental differences. Finally, the inclusion ofchildren with uniformly high medication adherence may introduceselection bias and limit generalizability to broader real-world pedi- atric asthma populations, where adherence is often more variable. Future studies should validate these findings in larger pedi- atric cohorts and assess the longitudinal stability of BDR-relatedphenotypes. Evaluating whether BDR-related inflammatory signa- tures predict long-term outcomes and treatment responses will becritical. One promising direction is the use of oscillometry, whichnon-invasively measures both large (proximal) and small (distal)airway function by assessing airway resistance and reactance andrequires minimal patient cooperation.49 Its adoption in clinical andresearch settings could greatly enhance the assessment of lungfunction and treatment response in pediatric asthma. Moreover,smart inhaler technologies, which track medication use in real-time, offer a valuable opportunity to obtain objective adherencedata.50 Their use could overcome the limitations of self-reportedadherence measures and provide a more accurate understandingof how medication use relates to BDR, asthma control, and inflam-matory patterns.In conclusion, our results show that high BDR is associated withpoorer asthma control. In addition, high BDR significantly associ- ates with a distinct systemic inflammatory profile in children withMSA. This suggests that BDR can be a useful non-invasive markerfor asthma phenotyping.F I G U R E 4 Adjusted associations between high bronchodilator response (BDR) and serum protein levels. Associations for high BDRdefined as >10% increase in predicted FEV1. Beta coefficients (β) represent the direction and magnitude of association, with FDR-adjusted p-values indicating statistical significance. β coefficients are displayed as point estimates, with horizontal lines representing thecorresponding 95% confidence intervals. All models were adjusted for age, sex, ethnicity, BMI z-score, GINA treatment step, baselinelung function, study site, season of inclusion, and current smoking exposure. FEV1, forced expiratory volume in 1 s; BDR, bronchodilatorresponse, IL, interleukin, TNF-α, tumor necrosis factor alpha, MMP, matrix metalloproteinase, FDR, false discovery rate. A pseudo-logarithmic x-axis was used only for visualization purposes to allow plotting cytokines in different pg/mL ranges in one plot. ### Page 13 | 13 of 15METWALLY et al.AU T H O R CO N T R I B U T I O N SN. K. A. M. performed the analysis and drafted the initial version ofthe manuscript. M.I.A.-A., and A.H.M.Z. contributed to the design ofthe analysis plan. All authors contributed to the acquisition of data,interpretation of the analysis, revision, drafting, and ensuring the ac- curacy and integrity of the analysis. All authors provided final ap- proval of the version to be published.A FFI L I AT I O N S1Department of Pulmonary Medicine, Amsterdam UMC, Universityof Amsterdam, Amsterdam, the Netherlands2Amsterdam Institute for Infection and Immunity, InflammatoryDiseases, Amsterdam, the Netherlands3Amsterdam Public Health, Personalized Medicine, Amsterdam, TheNetherlands 4Department of Pediatric Pulmonology and Allergy, Emma Children'sHospital, Amsterdam UMC, Amsterdam, The Netherlands 5 Department of Public and Occupational Health, Amsterdam UMC,University of Amsterdam, Amsterdam, the Netherlands6Department of Experimental Immunology, Amsterdam UMC,University of Amsterdam, Amsterdam, the Netherlands7Center for Human Molecular Genetics and Pharmacogenomics,Faculty of Medicine, University of Maribor, Maribor, Slovenia8Department of Pediatric Pneumology and Allergy, UniversityChildren's Hospital Regensburg (KUNO), Regensburg, Germany9University Children's Hospital, University of Regensburg,Regensburg, Germany10Division of Pediatric Respiratory Medicine, Donostia UniversityHospital, San Sebastián, Spain11Department of Pediatrics. Faculty of Medicine and Nursing,University of the Basque Country (UPV/EHU), San Sebastián, Spain12Genomics and Health Group, Department of Biochemistry,Microbiology, Cell Biology, and Genetics, Universidad de La Laguna(ULL), La Laguna, Tenerife, Spain13CIBER de Enfermedades Respiratorias, Instituto de Salud CarlosIII, Madrid, Spain14Instituto de Tecnologías Biomédicas (ITB), Universidad de LaLaguna (ULL), La Laguna, Tenerife, Spain 15 Faculty of Chemistry and Chemical Engineering, University ofMaribor, Maribor, Slovenia16Department for Science and Research, University Medical CentreMaribor, Maribor, Slovenia17Division of Pharmacology, Utrecht Institute for PharmaceuticalSciences, Faculty of Science, Utrecht University, Utrecht, CG, TheNetherlands 18Department of Genetics, UMC Groningen, University ofGroningen, Groningen, Netherlands19Department of Clinical Pharmacy, Faculty of Pharmacy, AssiutUniversity, Assiut, EgyptAC K N OW L E D G M E N T SThe SysPharmPediA is a consortium effort and we would like toacknowledge the help and expertise of the individuals and groupswhose names are mentioned in the attached SysPharmPediA con- sortium list. In addition, we would like to thank all the children andparents/caregivers who participated in the SysPharmPediA study.F U N D I N G I N FO R M AT I O NThe SysPharmPediA consortium has been supported by ZonMW[project number: 9003035001], Ministry of Higher Education,Science and Innovation of the Republic of Slovenia (MVZI)[contract number C330-16-500106]; the German Ministry ofEducation and Research (BMBF) [project number FKZ 031 L0088];Instituto de Salud Carlos III (ISCIII) through Strategic Action forHealth Research (AES) and European Community (EC) withinthe Active and Assisted Living (AAL) Program framework [awardnumbers AC15/00015 and AC15/00058] under the frame of theERACoSysMed JTC-1 Call.CO N FL I C T O F I N T E R E S T S TAT E M E N TA.H. Neerincx (AHN) reported support for the present manuscriptvia a grant paid to the institution from ERANET Systems Medicineand ZonMW. Mario Gorenjak (MG) disclosed reported supportfor the present manuscript including the SysPharmPedia grant co-financed by the Ministry of Education, Science and Sport Slovenia(MIZS), and research core funding from the Slovenian ResearchAgency. Uroš Potočnik (UP) also reported support for the pre- sent manuscript from the SysPharmPedia grant, co-financed bythe Ministry of Higher Education, Science and Innovation Slovenia(MVZI), as well as core funding and a research grant from theSlovenian Research Agency. Maria Pino-Yanes (MPY) reported sup- port for the present manuscript from the Instituto de Salud CarlosIII (ISCIII) and the SysPharmPedia grant from ERACoSysMed; fur- thermore, within the past 36 months, MPY received grants from theSpanish Ministry of Science, Innovation, and Universities (includingthe Ramon y Cajal program), grant support outside the submittedwork from GlaxoSmithKline, Spain and CSL Behring, and fundingfrom the Fundación Canaria Instituto de Investigación Sanitaria deCanarias (FIISC), alongside lecture fees from AstraZeneca Spainoutside the submitted work. Michael Kabesch (MK) reported sup- port from German Ministry of Education and Research and theEuropean Research Council for this manuscript, and within the past36 months reported grants/contracts from the European Union,German Ministry of Education and Research, and Bavarian Ministryof Health, consulting fees from AstraZeneca, and honoraria for lec- tures from EAACI, Novartis, Nutricia, and Pari. Anke H Maitland-vander Zee (AHMZ) disclosed receiving grants (past 36 months) as thePI of the P4O2 consortium sponsored by Health Holland, involv- ing private partners (AbbVie, Boehringer Ingelheim, Breathomix,Fluidda, Ortec Logiqcare, PeXA, Philips, Quantib-U, Smartfish,Clear, SODAQ, Thirona, Roche, TopMD, Novartis, RespiQ), an un- restricted research grant from Boehringer Ingelheim, and a VertexInnovation Award grant; her institution received consulting hono- raria from Boehringer Ingelheim and Astra Zeneca; and she servesas the Chair of a DSMB of a study on BPD in neonates. Jan WillemDuitman (JWD) reported grants or contracts (past 36 months) from ### Page 14 14 of 15 |METWALLY et al.Abbvie, Boehringer Ingelheim, Dutch Lung Foundation, NWO, andHealth~Holland. Susanne Vijverberg (SV) acts as the 3TR ABCProject Manager (past 36 months) for a project funded by theInnovative Medicines Initiative 2 Joint Undertaking (JU), supportedby the European Union and EFPIA, listing pharmaceutical fundersAstraZeneca, Sanofi & GSK, and holds a leadership role as an ERSOfficer in Assembly 7 (Pediatrics). Olaia Sardón Prado (OSP) receivedsupport paid to her institution (past 36 months) from Faes Farmaand Sanofi for attending the SENP 2025 congress. Finally, SusanneHarner (SHar) reported receiving payment or honoraria for lec- tures from Nutricia and Allergopharma, and support for travel fromNutricia (past 36 months). All other co-authors, including Rene Lutter(RL), A.D. Kraneveld (ADK), Antoaneta A. Toncheva (AAT), BarbaraS. Dierdorp (BSD), Christine Wolff (CW), E.G. Haarman (EGH), LeyreLópez-Fernández (LLF), Mahmoud I. Abdel-Aziz (MIAA), NarimanKotb Abbas Metwally (NKAM), Paula Corcuera Elosegui (PCE), Dr.S.W.J. Terheggen-Lagro (SWJTL), Simone Hashimoto (SH), SusanneBrandstetter (SB), and Tamara Dekker (TD) have no conflicts of in-terest to disclose.P E E R R E V I E WThe peer review history for this article is available at https://www.webof scien ce. com/api/gatew ay/ wos/ peer- review/10. 1111/ pai.70392.O R C I DSusanne J. H. Vijverberg https://orcid.org/0000-0002-4579-4081Susanne Harner https://orcid.org/0000-0001-6733-5823Maria Pino-Yanes https://orcid.org/0000-0003-0332-437XMichael Kabesch https://orcid.org/0000-0003-0697-1871Mahmoud I. Abdel-Aziz https://orcid.org/0000-0001-9263-7137R E F E R E N C E S1. Pijnenburg MW, Fleming L. Advances in understanding and re- ducing the burden of severe asthma in children. Lancet Respir Med.2020;8(10):1032-1044. 2. Vijverberg SJH, Farzan N, Slob EMA, Neerincx AH, Maitland-vander Zee AH. Treatment response heterogeneity in asthma: the roleof genetic variation. Expert Rev Respir Med. 2018;12(1):55-65. 3. Ricciardolo FLM, Guida G, Bertolini F, Di Stefano A, Carriero V.Phenotype overlap in the natural history of asthma. Eur Respir Rev. 2023;32(168):220201.4. Dragonieri S, Bikov A, Capuano A, Scarlata S, Carpagnano GE.Methodological aspects of induced sputum. Adv Respir Med.2023;91(5):397- 406. 5. Djukanović R, Sterk PJ, Fahy JV, Hargreave FE. Standardised meth- odology of sputum induction and processing. Eur Respir J Suppl.2002;37:1s-2s. 6. Alizadeh Bahmani AH, Vijverberg SJH, Hashimoto S, et al.Association of blood inflammatory phenotypes and asthma bur- den in children with moderate-to-severe asthma. ERJ Open Res.2024;10(6):222-2024. 7. Corneanu LE, Petriș OR, Lionte C, et al. Peripheral venipuncture inpediatric patients: a mini-review of clinical practice and technolog-ical advances. J Clin Med. 2025;14(18):6397. 8. Hirano T, Matsunaga K. Measurement of blood eosinophils inasthma and chronic obstructive pulmonary disease. Intern Med.2023;62(1):21-25. 9. Ullmann N, Bossley CJ, Fleming L, Silvestri M, Bush A, Saglani S.Blood eosinophil counts rarely reflect airway eosinophilia in chil-dren with severe asthma. Allergy. 2013;68(3):402- 406. 10. Menzies-Gow A, Mansur AH, Brightling CE. Clinical utility of frac-tional exhaled nitric oxide in severe asthma management. Eur Respir J. 2020;55(3):1901633. 11. Dweik RA, Boggs PB, Erzurum SC, et al. An official ATS clini- cal practice guideline: interpretation of exhaled nitric oxide lev- els (FENO) for clinical applications. Am J Respir Crit Care Med.2011;184(5):602-615. 12. Sim YS, Lee JH, Lee WY, et al. Spirometry and bronchodilator test.Tuberc Respir Dis (Seoul). 2017;80(2):105-112. 13. Coverstone AM, Bacharier LB, Wilson BS, et al. Clinical significanceof the bronchodilator response in children with severe asthma.Pediatr Pulmonol. 2019;54(11):1694-1703. 14. Heffler E, Crimi C, Campisi R, et al. Bronchodilator response as amarker of poor asthma control. Respir Med. 2016;112:45-50. 15. Abdel-Aziz MI, Neerincx AH, Vijverberg SJH, et al. A system phar- macology multi-omics approach toward uncontrolled pediatricasthma. J Pers Med. 2021;11(6):484.16. Cohen JL, Mann DM, Wisnivesky JP, et al. Assessing the validityof self-reported medication adherence among inner-city asthmaticadults: the medication adherence report scale for asthma. Ann Allergy Asthma Immunol. 2009;103(4):325-331. 17. ATS/ERS recommendations for standardized procedures for theonline and offline measurement of exhaled lower respiratory ni- tric oxide and nasal nitric oxide, 2005. Am J Respir Crit Care Med.2005;171(8):912-930. 18. Nathan RA, Sorkness CA, Kosinski M, et al. Development of theasthma control test: a survey for assessing asthma control. J Allergy Clin Immunol. 2004;113(1):59-65. 19. Stanojevic S, Kaminsky DA, Miller MR, et al. ERS/ATS technicalstandard on interpretive strategies for routine lung function tests.Eur Respir J. 2022;60(1):2101499.20. Quanjer PH, Ruppel GL, Langhammer A, et al. Bronchodilator re- sponse in FVC is larger and more relevant than in FEV(1) in severeairflow obstruction. Chest. 2017;151(5):1088-1098. 21. Abdel-Aziz MI, Hashimoto S, Neerincx AH, et al. Metabotypes arelinked to uncontrolled childhood asthma, gut microbiota, and sys- temic inflammation. J Allergy Clin Immunol. 2025;156:339-351. 22. Alizadeh Bahmani AH, Slob EMA, Bloemsma LD, et al.Medication use in uncontrolled pediatric asthma: results from theSysPharmPediA study. Eur J Pharm Sci. 2023;181:106360.23. Galant SP, Morphew T, Newcomb RL, Hioe K, Guijon O, Liao O.The relationship of the bronchodilator response phenotype topoor asthma control in children with normal spirometry. J Pediatr.2011;158(6):953-9. 24. Matsunaga K, Hirano T, Akamatsu K, Minakata Y. Predictors foridentifying the efficacy of systemic steroids on sustained exhalednitric oxide elevation in severe asthma. Allergol Int. 2013;62(3):359-365. 25. Ferrer Galván M, Javier Alvarez Gutiérrez F, Romero Falcón A,Romero Romero B, Sáez A, Medina Gallardo JF. Is the broncho- dilator test an useful tool to measure asthma control? Respir Med.2017;126:26-31. 26. Fernandes AL, Amorim MM, Caetano LB, et al. Bronchodilator re- sponse as a hallmark of uncontrolled asthma: a randomised clinicaltrial. J Asthma. 2014;51(4):405- 410. 27. Kaminsky DA, He J, Henderson R, et al. Bronchodilator re-sponse does not associate with asthma control or symptom bur-den among patients with poorly controlled asthma. Respir Med.2023;218:107375. ### Page 15 | 15 of 15METWALLY et al.28. Madulara GM, Andaya AG. Effects of aeroallergen sensitizationon symptom severity, pulmonary function, and bronchodilatorresponse in children with bronchial asthma. Journal of Medicine, University of Santo Tomas. 2022;6(2):959-970. 29. Lombardi C, Savi E, Ridolo E, Passalacqua G, Canonica GW. Is aller- gic sensitization relevant in severe asthma? Which allergens may beculprit? World Allergy Organ J. 2017;10(1):2.30. Forno E, Celedon JC. Asthma and ethnic minorities: socioeco-nomic status and beyond. Curr Opin Allergy Clin Immunol. 2009;9(2):154-160. 31. Pelaia C, Heffler E, Crimi C, et al. Interleukins 4 and 13 in asthma:key pathophysiologic cytokines and Druggable molecular targets.Front Pharmacol. 2022;13:851940. 32. Adrish M, Akuthota P. Approach to non-type 2 asthma. Respir Med. 2023;216:107327.33. Sze E, Bhalla A, Nair P. Mechanisms and therapeutic strategies fornon-T2 asthma. Allergy. 2020;75(2):311-325. 34. Padrón-Morales J, García-Solaesa V, Isidoro-García M, et al.Implications of cytokine genes in allergic asthma. Allergol Immunopathol (Madr). 2014;42(6):603-608. 35. Barnes PJ. The cytokine network in asthma and chronic obstructivepulmonary disease. J Clin Invest. 2008;118(11):3546-3556. 36. Caucheteux SM, Hu-Li J, Guo L, et al. IL-1β enhances inflammatoryTH2 differentiation. J Allergy Clin Immunol. 2016;138(3):898-901. 37. Choi JP, Kim YS, Kim OY, et al. TNF-alpha is a key mediator in thedevelopment of Th2 cell response to inhaled allergens induced by aviral PAMP double-stranded RNA. Allergy. 2012;67(9):1138-1148. 38. Neveu WA, Allard JL, Raymond DM, et al. Elevation of IL-6 in theallergic asthmatic airway is independent of inflammation but associ- ates with loss of central airway function. Respir Res. 2010;11(1):28.39. Jackson DJ, Bacharier LB, Calatroni A, et al. Serum IL-6: a biomarkerin childhood asthma? J Allergy Clin Immunol. 2020;145(6):1701- 4. 40. Chu DK, Al-Garawi A, Llop-Guevara A, et al. Therapeutic poten- tial of anti-IL-6 therapies for granulocytic airway inflammation inasthma. Allergy Asthma Clin Immunol. 2015;11(1):14. 41. Kelly EA, Koziol-White CJ, Clay KJ, et al. Potential contributionof IL-7 to allergen-induced eosinophilic airway inflammation inasthma. J Immunol. 2009;182(3):1404-1410. 42. Sasaguri T, Arima N, Tanimoto A, Shimajiri S, Hamada T, SasaguriY. A role for interleukin 4 in production of matrix metallopro-teinase 1 by human aortic smooth muscle cells. Atherosclerosis.1998;138(2):247-253. 43. Naveed SU, Clements D, Jackson DJ, et al. Matrix Metalloproteinase-1activation contributes to airway smooth muscle growth and asthma severity. Am J Respir Crit Care Med. 2017;195(8):1000-1009. 44. Ogawa Y, Duru EA, Ameredes BT. Role of IL-10 in the resolution ofairway inflammation. Curr Mol Med. 2008;8(5):437- 445. 45. Qian G, Jiang W, Sun D, et al. B-cell-derived IL-10 promotes aller- gic sensitization in asthma regulated by Bcl-3. Cell Mol Immunol.2023;20(11):1313-1327. 46. Grissell TV, Powell H, Shafren DR, et al. Interleukin-10 gene ex- pression in acute virus-induced asthma. Am J Respir Crit Care Med.2005;172(4):433- 439. 47. Mathur SK, Fichtinger PS, Evans MD, Schwantes EA, Jarjour NN.Variability of blood eosinophil count as an asthma biomarker. Ann Allergy Asthma Immunol. 2016;117(5):551-553. 48. Bacharier LB, Maspero JF, Katelaris CH, et al. Dupilumab in chil- dren with uncontrolled moderate-to-severe asthma. N Engl J Med.2021;385(24):2230-2240. 49. Smith EF, Bradshaw TK, Urs RC, et al. Oscillometry and spirome-try are not interchangeable when assessing the bronchodilator re-sponse in children and young adults born preterm. Pediatr Pulmonol.2023;58(11):3122-3132. 50. Reddel HK, Bateman ED, Schatz M, Krishnan JA, Cloutier MM. Apractical guide to implementing SMART in asthma management. J Allergy Clin Immunol Pract. 2022;10(1s):S31.S U P P O R T I N G I N FO R M AT I O NAdditional supporting information can be found online in theSupporting Information section at the end of this article.How to cite this article: Metwally NKA, Hashimoto S,Vijverberg SJH, et al. Bronchodilator response is linked withuncontrolled moderate-to-severe childhood asthma andelevated IL-4 and IL-13. Pediatr Allergy Immunol. 2026;37:e70392. doi:10.1111/pai.70392