Effect of Mepolizumab on Severe Eosinophilic Asthma

Hypotheses Hº1. The levels of certain surface molecules on eosinophils or the presence or absence of certain proteins in the proteome of this leukocyte subset prior mepolizumab treatment can be used as predictive/prognostic markers of response to this biological. Hº2. Mepolizumab alters the abundance of several surface or intracellular proteins in eosinophils as an outcome related to changes in their activation status, migratory ability, regulatory/effector function, or subset composition. Hº3. Late-onset severe eosinophilic asthmatics have elevations in the serum concentration of different members of the IGF family (IGF-1, IGF-BP3, IGF-ALS) and mepolizumab treatment reduces these levels and behaves as a response-biomarker along with the number of eosinophils and clinical exacerbations. H°4. Mepolizumab alters the abundance of several proteins in the medium-low abundance serum proteome of patients with SEA. Therefore, these proteins could be used as predictive/prognostic markers of response to this biological and could provide a better understanding of both eosinophilic and non-eosinophilic-related biological functions of IL-5 in SEA. Objectives or research questions OB (Aspirational objective): Discovery of predictive/prognostic biomarkers of response to mepolizumab using flow cytometry, transcriptomic, and proteomic technologies. * OB1.- To identify changes in surface markers of eosinophils and eosinophil subpopulations in response to treatment with mepolizumab using flow cytometry techniques: * DE(Deliverable)1.1: Selection of 15 healthy controls * DE1.2: 15 patients with severe eosinophilic asthma patients are scheduled to receive mepolizumab, and sign the informed consent. * DE1.3: Generation of the initial database with demographic, clinical, haematologic, and biochemical information. * DE1.4: Collection and processing of serum (1 SST tube) and whole blood samples (1-2 tubes) from healthy donors (T0) and mepolizumab-treated patients (T0, T4, T16, T32). * DE1.5: Flow cytometry assays (see below). * DE1.6: Complete the database with clinical, hematological, biochemical and flow cytometry data generated at times T4, T16 and T32. Final uni-/multivariant statistical analysis. * DE1.7: Publication of results. * OB2.- Transcriptomic analysis to identify mRNAs within the eosinophil transcriptome displaying enhanced or reduced levels in response to treatment with mepolizumab: * DE2.1: Set up an eosinophil isolation protocol. * DE2.2: Purification of eosinophils from healthy donors (T0) and patients (T0, T4, T16, T32). * DE2.3: RNA extraction from eosinophil lysates, assay of quality and quantity of RNA, and storage at -80ºC. * DE2.4: Discovery-based/hypothesis-generating approach. Evaluation of the levels of 770 mRNAs linked to the recruitment, activation, and effector functions of myeloid cells busing a pre-made "nCounter® Human Myeloid Innate Immunity Panel (v2)" (www.nanostring.com/products/gene-expression-panels/gene-expression-panels-overview/ncounter-myeloid-innate-immunity-panel ). Eosinophil samples from healthy donors (T = 0) and patients (T0 and T16). * DE2.5: Processing of the obtained data and initial statistical analysis. * DE2.6: Validation of nCounter® data . RTqPCR analyses of those mRNAs in eosinophils with the greatest abundance changes in response to mepolizumab treatment. In addition, some additional mRNAs not included in the "nanoString Myeloid Innate Immunity" panel, such as FOXP3 (regulatory function), CRLF2, ST2, or IL-7R (cytokine receptors; activation), will be analysed. HPRT1 gene will be used as a house-keeping gene. * DE2.7: Uni- and multivariant statistical analyses. * DE2.8: Publication of results. * OB3.- Proteomic profiling to identify proteins with differential abundance within the eosinophils in response to treatment with mepolizumab: * DE3.1: Lysis of eosinophils, protein quantification (BCA), and cell supernatants storage at -80°C. * DE3.2: Develop a total proteome analysis protocol with a Data Dependent Acquisition (DDA) method using LC-MS/MS (Triple TOF 6600) * DE3.3: Check the biological (biological replications) and technical (technical replicas) variability. * DE3.4: Create a library for SWATH with as many eosinophil proteins as possible. * DE3.5: Perform SWATH-MS analysis in samples from healthy donors and patients (T0, T4, T16, T32) ("data-independent acquisition" method or DIA; "Targeted label-free proteomics"). * DE3.6: Processing of the obtained data and initial statistical analysis. * DE3.7: Validation of a panel of biomarkers with a different technology (e.g., Selected reaction monitoring/SRM, ELISA). * DE3.8: Final uni- and multivariant statistical analysis. Identify proteins with significant differences between groups (P \< 0.05) and a fold change ≥ 1.5. * DE3.9: Publication of results * OB4. Check whether late-onset severe eosinophilic asthmatics display elevated levels of IGF-1, IGF-BP3, IGF-ALS in serum samples, if the response of mepolizumab depends on the levels of these markers, and if treatment with this biological reduces the concentration in serum of these IGF-family members: * DE4.1: Analysis of IGF-1, IGF-BP3 and IGF-ALS by ELISA * DE4.2: Uni- and multivariant statistical analysis of experimental data * DE4.3: Publication of results * OB5.- Identify proteins with differential abundance within the low abundant serum proteome of patients with severe eosinophilic asthma in response to treatment with mepolizumab. * DE5.1: Optimization of protocols for serum pre-fractioning. Protocol 1, depletion of high abundant proteins using affinity chromatography. Protocol 2, depletion of high abundant proteins using DTT precipitation. * DE5.2: Serum pre-fractioning; protocol 1: To deplete the 14 highest abundant proteins in serum from healthy donors (T0) and mepolizumab-treated severe eosinophilic asthma patients (T0, T4, T32) using high abundant protein depletion columns. * DE5.3: Serum pre-fractioning; protocol 2: To deplete high abundant proteins in serum from healthy donors (T0) and mepolizumab-treated severe eosinophilic asthma patients (T0, T4, T32) using DTT precipitation. * DE5.4-5.10. Same procedure after protocol 1 and protocol 2 in parallel: * DE5.4: Development of a low abundant proteome analysis protocol with a DDA method using LC-MS/MS (Triple TOF 6600). * DE5.5: Check the biological (biological replications) and technical (technical replicas) variability. * DE5.6: Create a library for SWATH with as many low abundant serum proteins as possible. * DE5.7: Perform SWATH-MS analysis in samples from healthy donors and patients (T0, T4, T32) (DIA). * DE5.8: Processing of the obtained data and initial statistical analysis. * DE5.9: Validation of a panel of biomarkers obtained using a different technology (e.g., SRM/MRM, ELISA). * DE5.10: Final uni- and multivariant statistical analysis. Identify proteins with significant differences between groups (P \< 0.05) and at least a fold change ≥ 1.5. * DE5.11: Correlation analyses of proteins with changes (Detected in protocol 1 and/or protocol 2) with clinical, haematological, biochemical and flow cytometry data. * DE5.12: Publication of results. Publication of results: We expect to present 2-3 communications to Spanish Respiratory Congress (SEPAR) and the European Respiratory Congress (ERS) resulting from the study of the clinical and experimental data. In addition, we expect to publish 3 publications in Q1 journals. Study population The study population will include healthy controls (i.e., subjects without asthma, allergy, systemic diseases or scheduled for minor surgeries) and severe eosinophilic asthma patients, who will be recruited from different areas of Galicia (Santiago de Compostela, A Coruña, Lugo, Vigo, and Ourense), Spain. Diagnosis of severe eosinophilic asthma patients at screening will be based on several inclusion criteria and exclusion criteria that we describe below. Inclusion criteria: * Diagnosis of severe uncontrolled asthma according to ERS/ATS criteria. * Persistent eosinophilia in blood (\>300 cells/μL) on ≥ two occasions (≥ 4 weeks between each measurement). * Frequent exacerbations (≥ two per year), defined as a period for ≥ 3 days of lack of asthma control requiring treatment with systemic corticosteroids and/or an ED visit and/or hospitalization. * Signature of informed consent and agree to comply with all the visits of the study and all the procedures that this entails. Exclusion criteria: * Smoking history: Current smokers or former smokers with a smoking history of ≥10 pack-years (number of pack years = (number of cigarettes per day/20) x number of years smoked). A former smoker is defined as a participant who quit smoking at least 6 months prior to Visit 1. * Clinically important pulmonary disease other than asthma (e.g. active lung infection, COPD, bronchiectasis, pulmonary fibrosis, cystic fibrosis, hypoventilation syndrome associated with obesity, lung cancer, alpha 1 anti-trypsin deficiency, and primary ciliary dyskinesia) or ever been diagnosed with pulmonary or systemic disease, other than asthma, that are associated with elevated peripheral eosinophil counts (e.g. allergic bronchopulmonary aspergillosis/mycosis, Churg- Strauss syndrome, hypereosinophilic syndrome). * Any disorder, including, but not limited to, cardiovascular, gastrointestinal, hepatic, renal, neurological, musculoskeletal, infectious, endocrine, metabolic, haematological, psychiatric, or major physical impairment that is not stable in the opinion of the Investigator. * Malignancy: A current malignancy or previous history of cancer in remission. * Acute upper or lower respiratory infections requiring antibiotics or antiviral medication within 30 days prior to the Visit 1. * Xolair: Participants who have received omalizumab (Xolair) or another monoclonal antibody previously. * Participants who have received systemic corticosteroids within 30 days before Visit 1. * Pregnancy: Participants who are pregnant or breastfeeding. * Obesity class 2 or higher (BMI≥ 35 kg/m2) (https://www.who.int/dietphysicalactivity/childhood\_what/en/). Sample size * Cohort of healthy controls (n=15) only for analysis at T = 0. * Cohort of n=15 subjects with severe eosinophilic asthma that start with mepolizumab therapy with no modification to their currently prescribed medications. Follow-up study visits at 4 (T4), 16 (T16) and 32 (T32) weeks after the original study visit (T=0). * The rationale for sample size is explained in the statistical section. Anticipated rate of enrolment Since this will be a multicentre study, we expect to reach a rate of enrolment of at least 2 severe eosinophilic asthmatics beginning with mepolizumab therapy per month (4 weeks) in each hospital (Total = 8 per month). This means that the 15 subjects should be scheduled to receive mepolizumab along the first 36 weeks of this study. Considering the recruitment and experimental parts, we expect to complete the study in 120 weeks (30 months; \~2.5 years). We also expect that at least 90% of subjects complete this study. Estimated study start date: June 2021 Estimated study completion date: 2.5 years (30 months) Study design and methods This is an observational, longitudinal, prospective, and multicentre study to evaluate both the early response (4 weeks) and late response (16 and 32 weeks) to mepolizumab therapy in severe eosinophilic asthmatics. The study will be headed by Dr. Francisco Javier González Barcala (Pneumology Service at CHUS), the leader of the Translational Research in Airway Diseases (TRIAD) group. Other members of the TRIAD group are Dr. Francisco Javier Salgado Castro, the Project Manager, and Dr. Juan José Nieto Fontarigo, both experts in Immunology, Biochemistry, Proteomics, and Respiratory Diseases. Dra. Marina Blanco Aparicio is responsible for the Asthma Unit at the University Hospital Complex of A Coruña (CHUAC); Dr. Uxío Calvo Álvarez, at the University Hospital Complex of Ferrol (CHUF); Dra. Coral González Fernández at the University Hospital Complex of Ourense (CHUO); Dra. Mar Mosteiro Añón at the University Hospital Complex of Vigo-Alvaro Cunqueiro (CHUVI); and Dolores Corbacho Abelaira at the POVISA Hospital Centre (Vigo). Proteomics experiments will be carried out by Dra. Susana Belén Bravo López at (FIDIS). nCounter® analysis will be carried out by GENVIP group, at FIDIS. The research project will be minimally invasive (e.g., no bronchoscopic examinations) but the protocol needs to be reviewed and approved by the Ethics Committee of Clinical Research of Galicia, Spain. Only fifteen patients who meet the severe asthma diagnosis criteria, are scheduled to receive mepolizumab, and sign the informed consent will be enrolled in this study. The same protocol will be followed by the different clinical teams. Demographic, as well as clinical, haematological, and biochemical variables will be included in a database. Skin prick test to common allergens and the presence of allergen-specific IgE (ImmunoCAP, Thermo Fisher) will be used to check for allergic sensitization. Lung function parameters (forced expiratory volume in the 1st second (FEV1), forced vital capacity (FVC), and FEV1/FVC ratio) also will be analysed. Spirometry will be performed before and after use of a bronchodilator. The Asthma Control Test (ACT) and the Asthma Quality of Life Questionnaire (AQLQ) questionnaire will be performed. Asthmatics must be in a stable phase of the disease (i.e. absence of exacerbations for at least 4 weeks before sample collection). Exacerbations will be managed in accordance with standard clinical guidelines. Patients (n=15) will receive 100 mg subcutaneous injection of mepolizumab at 4 weeks intervals, and blood and serum samples (2-3 EDTA tubes; 1 SST tube) will be withdrawn at T=0, 4, 16 and 32 weeks, in order to evaluate both early-response (4 weeks) and late-response (16 and 32 weeks) to treatment. Methods * Tubes: EDTA (complete blood) and SST (serum) * Eosinophils purification * Eosinophils can be isolated from whole blood (heparin tubes) using the Miltenyi Human Eosinophil Isolation Kit (Catalog #130-104-466) or the EasySep™ Human Eosinophil Isolation Kit (Catalog #17956), both negative selection procedures that yield untouched subsets of these leukocytes. We expect at least 1.0-4.0 x 10\^6 cells from ∼30 ml blood, but also high viability and purity (\>95%). * ELISA studies. * Serum sample collection: Measurement of IGF-ALS (GENOIT4078 Immunotag Human IGFALS 96 well), IGF-1 (Human IGF-I/IGF-1 DuoSet ELISA, R\&D Systems, catalog #DY291), and IGF-BP3 (Human IGFBP-3 DuoSet ELISA, R\&D Systems, catalog #DY675) by means of ELISA. * Total RNA purification from eosinophils and nCounter nanoString analysis (Discovery-based/hypothesis-generating approach): * Purified eosinophils from healthy controls (T=0) and patients (T=0, 4, 16 and 32 weeks) will be stored at -80ºC in RNAlater solution (Ambion, Paisley, UK). Total RNA will be isolated by means of a RNeasy Mini kit (Qiagen) and stored at -80ºC after checking RNA quality and concentration (Nanodrop). * The nCounter® platform (nanoString; https://www.nanostring.com/scientific-content/technology-overview/ncounter-technology) is a multiplex methodology that allows the quantification of up to 800 RNA, DNA, or protein targets. Regarding mRNA molecules, this technology is based on the in-solution hybridization of every mRNA to two complementary oligonucleotides: a biotinylated mRNA-specific probe and a mRNA-specific oligonucleotide containing a sequential combination of six fluorochromes (four different colours) that create a fluorescent barcode that identifies the specific mRNA being detected. Once the excess of both probes is removed, the hybridised complexes are captured through a biotin-streptavidin interaction and aligned on cartridge in order to the nCounter instrument can read those "barcodes". To carry out these steps, the nCounter platform consists of two instruments the Prep Station, which performs the purification of the hybridized complexes and their immobilization onto the surface of a cartridge, and the Digital Analyzer (DA), a scanner that identifies and counts the barcodes captured for each sample. This quantitative analysis Therefore, each miRNA can be quantified individually (absolute quantification; counts) from difficult samples (e.g., eosinophils) with no need for other requirements such as mRNA-cDNA conversion (RT) or DNA-amplification (qPCR), leading to less data variability (https://www.nanostring.com/scientific-content/technology-overview/challenges-of-rt). In addition, the amount of input material is low (25 ng-300 ng mRNA) and can be derived from FFPE-derived RNA, total RNA, fragmented RNA, cell lysates, and sorted cells. Afterwards, nCounter data will be normalized, background noise subtracted, and further correction performed to account for the efficiency of the extraction (calculated based on the expression of spike-in miRNAs that will be added to the sample in a defined amount before the miRNA extraction). Normalizations will be done using the R NanoStringNorm package. After normalization, a log2 transformation of the data will be made and subsequently analysed by means of the LIMMA Bioconductor package to identify those mRNAs displaying a differential abundance upon mepolizumab treatment. This analysis will take no longer than 24 hours. * RTqPCR studies (Hypothesis-driven approach): * To analyse the levels of mRNAs encoding proteins related with alarmin-mediated activation of eosinophils (CRLF2, ST2, IL-7Rα/CD127) and with the regulatory function of eosinophils (FOXP3) from patients treated with mepolizumab, total RNA will be transcribed into cDNA (QuantiTect Rev. Transcription Kit; Qiagen) and stored at -80ºC. qPCR (QuantiTect SYBR Green PCR Kit; Qiagen) will be performed in a LightCycler® 96 Instrument (Roche Life Science) and used to analyse the expression of FOXP3, CRLF2, ST2, IL-7R and the HPRT1 gene (endogenous control). * Flow cytometry studies (Hypothesis-driven approach): * EDTA-treated peripheral blood samples from healthy controls (n=15; T0) and mepolizumab-treated patients (n=15; T0, T4, T16, T32). * Label 100 μL/tube of whole peripheral blood (EDTA) with both specific and isotype-matched control antibodies (BD). Red cells lysis with FACSlyse (BD). Analysis with a FACSCalibur flow cytometer (BD). Use FSC/SSC to select granulocytes; then SSC vs CCR3 (FITC) to separate eosinophils from neutrophils. Gate eosinophils: * Multimarker panel 1 (Regulatory proteins in eosinophils): Measurement of CD16 and galectins-1/10. * Multimarker panel 2 (Activation receptors in eosinophils): Measurement of CD48 (reduced in total eosinophils with moderated-severe asthma compared to healthy controls (HC) \[our studies, 54\]), CD44, and CD11b. * Multimarker panel 3 (Eosinophils subsets): Analysis of subsets based on the expression of Siglec-8, CD62L(L-selectin), and IL-5Rα. * Analysis of eosinophil proteome (Discovery-based/hypothesis-generating approach): * As much as 50 x 103 cells will be necessary to perform proteomic assays. We expect around 50-400 x 103 cells from ∼1 mL of blood. * Isolated eosinophils (50 x 103 cells) will be collected by centrifugation, washed, and resuspended in lysis buffer with proteinase inhibitors. After that, insoluble material will be removed by centrifugation and cell supernatants stored at -80°C. * For eosinophils total proteome characterization will be made after trypsin digestion using a DDA method in a LC-MSMS system. For this approach we will use samples from 15 healthy donors and 15 patients (T0, T4, T16, T32). The proteins selected will be only those that reported a 1% Global false discovery rate (FDR) or better. * Protein "pools" from the 5 groups of study (healthy donors and patients at time T0, T4, T16, and T32 after treatment) will be used, dividing them (1-DE) in 5-6 bands, extracting the proteins from each band, generating the corresponding peptides and analysing them by MS / MS to produce a library for SWATH with a high number of proteins, on which then the quantification will be carried out. Once the library was made and maintaining the standardized conditions of LC-MS / MS (TripleTOF), we will perform a SWATH-MS analysis ("information-dependent acquisition" method or IDA; "Targeted label-free proteomics") in samples from 15 healthy donors and 15 patients (T0, T4, T16, T32). This assay will let us identify proteins with significant differences between the groups of study. The proteins selected will be only those with a P\<0.05 and a fold change ≥1.5. * Analysis of medium-low abundant serum proteome (Discovery-based/hypothesis-generating approach): * Serum from healthy controls (T=0) and mepolizumab-treated patients (T=0, 4 and 32 weeks) will be isolated from peripheral venous blood. Once the blood has been collected in a BD Vacutainer™ SST™ Serum Separation Tube, blood coagulation will be allowed to proceed for a minimum of 45 min in a vertical position at room temperature (RT). Afterward, the tube will be centrifugated at 1100-1300 xg for 10 min (RT) in swinging bucket rotor units. After centrifugation, the serum will be located above the polymer barrier. The serum will be harvested and aliquoted into Eppendorf tubes. * After that two different protocols for high abundant protein removal will be performed. * Pre-fractioning protocol 1: High abundant serum protein removal using affinity chromatography. 100 μL of serum will be applied to High Select™ Top14 Abundant Protein Depletion Midi Spin Columns from ThermoFisher Scientific according to manufacturer´s instructions. * Pre-fractioning protocol 2: High abundant serum protein removal using DTT precipitation. Use 30 μL of serum, according to the protocol we have published. * Low abundant serum protein concentration from protocol 1 and 2 will be quantified by RCDC Protein Assay Kit (BIORAD) as per the manufacturer's protocol. After depletion of the highest abundant proteins (99% of total protein; 60-80 mg/mL) we expect to have around 60-80 μg of low abundant proteins. * Proteome characterization will be made following the same protocol as the one described for eosinophil homogenates. Study endpoints: Demographic data for all individuals enrolled in the study will be obtained at basal. In addition, several data will be collected, including asthma history, lung function parameters, skin prick test, allergen-specific IgE, AQLQ score, ACT score, the number of exacerbations, and consumption of prednisone. During the following visits to the Pneumology service at T0, 4, 16, and 32, patients treated with mepolizumab will be followed up. This includes measurements of lung function (FEV1, FEV1/FVC), biochemical and haematological parameters. Peripheral blood and serum samples will be collected, and eosinophils will be magnetically purified, at T0, T4, T16, and T32. Flow cytometry, RTqPCR, and proteomic analyses, as well as immunoassays, will be performed. All the experimental variables (e.g., the abundance of eosinophil proteins in proteomic assays, the abundance of serum proteins, eosinophil activation markers, …) will be correlated with clinical parameters (e.g., lung function, asthma control, number of exacerbations) in order to assess the association of these variables with the response to treatment. We will consider a favourable response to mepolizumab if one of the following criteria is met: * To obtain adequate asthma control ACT ≥20 \[60\], or/ a change of ≥3 points representing a minimally important difference. * To achieve a reduction in the annual rate of exacerbations of 48%. Exacerbation is defined as the increase in symptoms requiring treatment with systemic corticosteroids for ≥3, or an unscheduled medical consultation, similar to that reflected in clinical trials with mepolizumab \[20, 61\]. * Get a 50% reduction in the annual rate of hospital admissions due to asthma exacerbation, similar to that reflected in clinical trials \[62\]. * To achieve a reduction in the median annual dose of systemic corticosteroids of 50% \[63\]. * Study primary endpoints: * IGF-1, IGF-BP3 and IGF-ALS levels in serum * Transcriptomic (nanoString)/mRNA expression data: FOXP3, CRLF2, ST2, IL-7R * Proteomic data * Flow cytometry data: Expression of CD16, galectins-1/10, CD48, CD44, CD11b, Siglec-8, CD62L, and IL-5Rα * Study secondary endpoints: * Lung function parameters (FEV1, FEV1/FVC) * Haematological parameters (e.g., eosinophils number). * Other clinical and biochemical variables (e.g., IgE or other immunoglobulins). * Number of exacerbations, prednisone consumption, ACT score, AQLQ score. Statistical plan or data analysis: Graph Pad Prism will be used to create graphics. IBM SPSS, Statistics 22.0, or R. will be used for the statistical study. During the analyses, we will be assisted by the USC Statistics and Operational Research area (Dr. Rosa María Crujeiras Casais). Sample size The calculation of sample size (N) has been carried out by using G\*Power 3.1.9.4 \[64\]. During these analyses we calculate N necessary get statistical significance in a F test (ANOVA: Repeated measures, within factors), given α (0.05), power (1-β, 0.95), the number of measurements (T0, T4, T16 and T32), and the effect size (f = 0.4; large effect size, which gives as a more clinically relevant results). The output N was 15, with a critical F= 2.82705. For clinical, flow cytometry and transcriptomic data. Cross-sectional comparisons between HC and patients in T0 (before treatment) following a normal distribution and having homogeneity of variances will be made by using t-test. For non-normal distributed variables, we will use Mann-Whitney U test. Changes in the different study variables in response to treatment with mepolizumab (longitudinal study; T0, T4, T16, and T32) will be tested using RM-ANOVA. Multivariate analysis (e.g., PCA, unsupervised clustering) as well as functional enrichment analysis will be performed with flow cytometry, and above all, transcriptomic data. For total proteome characterization and quantitative SWATH analysis We will use ProteinPilotTM 5.0.1 software from ABSciex which have the algorithm ParagonTM for database search and ProgroupTM for data grouping. Data will be searched using a Human specific Uniprot database. False discovery rate will be performed using a non-lineal fitting method displaying only those results that reported a 1% Global false discovery rate or better \[65\]. Functional analysis will be performed by different open-access software. FunRich (Functional Enrichment analysis tool) for functional enrichment and interaction network analysis (http://funrich.org/index.html). For statistics, FunRich uses hypergeometric test, BH and Bonferroni \[66, 67\]. We will use DAVID (https://david.ncifcrf.gov/tools.jsp) or GO (http://geneontology.org/page/go-enrichment-analysis) for gene ontology enrichment and for protein-protein interaction, network construction and clustering, we will use String (https://string-db.org/) or Cytoscape 3.7 (https://cytoscape.org/ ) \[68\]. For SWATH data, MarkerView software will give us a multivariate statistical analysis using principal component analysis (PCA) to compare the data across the samples. The average MS peak area of each protein will be derived from the replicates of the SWATH-MS of each sample followed by Student's t-test analysis using the MarkerView software for comparison among the samples based on the averaged area sums of all the transitions derived for each protein. The t-test will indicate how well each variable distinguishes the two groups, reported as a P-value. For the library, its set of differentially abundant proteins (p-value \<0.05) with a 1.5 up-regulated or down-regulated proteins will be selected. Limitations * As previously commented, 15 subjects will be scheduled to receive mepolizumab during the first half of the study (36 weeks). We expect that at least 90% subjects complete this study. However, patient dropouts and non-adherence (or non-compliance) are common events in clinical studies. In such a case, sample size will be proportionally inflated. * The present project has been proposed as a study of the discovery of molecular biomarkers in response to mepolizumab. This kind of studies can be boarded through Targeted/hypothesis-driven or broader/untargeted ("-omics" technologies) approaches. We are aware that it might be challenging to find predictive markers in this small and prospective/proof of concept study. We propose a double approach to minimize this risk. On the one hand, modern and untargeted methodologies to work and highly sensitive to detect low-abundant proteins (SWATH MS) or simplified protocols to work with difficult samples and reduce technical variance (e.g., nCounter nanoString) in order to shorten sample sizes. On the other hand, hypothesis-driven approaches (e.g., flow cytometry, RT-qPCR, ELISA), with the advantages of greater credence, less risk of type I and II errors, and easy to future replication of results. These targeted-methodologies will be also used to confirm only clinically relevant (high effect-size) and significant (p-value \< 0.05) differences obtained with untargeted transcriptomic/proteomic approaches. Finally, we purpose the use of two independent but complementary protocols for serum pre-fractioning. This will increase the probability of reaching a higher number of proteins with changes in serum and this will, therefore, increase the probability of discovering a higher number of biomarkers.