1. Patent Search
  2. Details

SIALYLATED FETUIN-A AS A MARKER OF IMMUNOTHERAPY EFFICACY

20170370946 · 2017-12-28

    Inventors

    Cpc classification

    International classification

    Abstract

    The invention concerns the identification of specific polypeptides, fragments variants thereof which can be used as markers for the efficacy of immunotherapy, particular for predicting responsiveness of a patient to immunotherapy.

    Claims

    1. A method for predicting responsiveness of a patient to immunotherapy, which method comprises detecting the level of expression of a Fetuin-A polypeptide or a fragment thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1, in a biological sample from said patient, wherein said biological sample has been taken before the commencement of and/or during immunotherapy, and wherein in said immunotherapy an allergen or auto-antigen is administered to said patient in order to treat allergy or auto-immune disease.

    2. The method according to claim 1, which method comprises the steps of: a) detecting the level of expression of a Fetuin-A polypeptide, or a fragment thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1, in a biological sample from said patient; b) comparing said level of expression with a control; c) identifying said patient as likely to be a responder or non-responder to immunotherapy based on the comparison with the control.

    3. The method according to claim 1, wherein step c) is performed by: (i) determining that the patient is likely to be a responder to immunotherapy if the control is derived from a responder subject or group of responder subjects known to respond to said immunotherapy, and if a level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is equal to or greater than the level of expression in the control; or (ii) determining that the patient is likely to be a responder to immunotherapy if the control is derived from a non-responder subject or group of non-responder subjects, and if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is greater than the level of expression in the control; or (iii) determining that the patient is likely to be a responder to immunotherapy if the control is derived from a randomly selected group of subjects, and if a level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is equal to or greater than the level of expression in the control.

    4. The method according to claim 1, wherein: (i) the control consists of a value of the relative abundance of the peptide SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, of at least 0.0046% of the total peptide abundance, or (ii) the control consists of a value of the relative abundance of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, of at least 3,600 units, and wherein the quantification of the abundance of said Fetuin-A polypeptide or fragment thereof is determined by: (a) depleting a serum sample in albumin, IgG, anti-trypsin, IgA, transferrin and haptoglobulin; (b) degrading with trypsin the depleted serum sample; (c) quantifying the relative abundance in said trypsin-digested depleted serum sample of the peptide SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by LC-MS/MS.

    5. A method for selecting a patient for immunotherapy, which method comprises the steps of: a) detecting the level of expression of a Fetuin-A polypeptide, a fragment or a variant thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1, in a biological sample from said patient; b) comparing said level of expression with a control; c) selecting or rejecting said patient for immunotherapy based on the comparison with the control; wherein said biological sample is taken before the commencement of immunotherapy, and wherein said immunotherapy comprises administration of an allergen or auto-antigen to said patient in order to reduce immune response.

    6. The method according to claim 5, wherein selecting or rejecting said patient for immunotherapy based on the comparison with the control is performed by: (i) if the control is derived from a responder subject or group of responder subjects known to respond to said immunotherapy, selecting the patient for immunotherapy if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is equal to or greater than the level of expression in the control; (ii) if the control is derived from a non-responder subject or group of non-responder subjects, selecting the patient for immunotherapy if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is greater than the level of expression in the control; or (iii) if the control is derived from a randomly selected group of subjects, selecting the patient for immunotherapy if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is equal to or greater than the level of expression in the control.

    7. The method according to claim 5, wherein: (i) the control consists of a value of the relative abundance of the peptide SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, of at least 0.0046% of the total peptide abundance, or (ii) the control consists of a value of the relative abundance of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, of at least 3,600 units, and wherein the quantification of the abundance of said Fetuin-A polypeptide or fragment thereof is determined by: (a) depleting a serum sample in albumin, IgG, anti-trypsin, IgA, transferrin and haptoglobulin; (b) degrading with trypsin the depleted serum sample; (c) quantifying the relative abundance in said trypsin-digested depleted serum sample of the peptide SEQ ID NO:2, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by LC-MS/MS.

    8. The method according to claim 1, wherein said Fetuin-A polypeptide or a fragment thereof consists of the sequence SEQ ID NO: 2.

    9. The method according to claim 1, wherein the patient has allergy.

    10. The method according to claim 1, where the immunotherapy comprises administration of allergen to a mucosal surface.

    11. The method according to claim 1, wherein an allergen extract is administered as part of said immunotherapy.

    12. An isolated antibody which binds specifically to a Fetuin-A polypeptide, or a fragment thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1.

    13. A lectin which binds specifically to a Fetuin-A polypeptide, or a fragment thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1.

    14. An aptamer which binds specifically to a Fetuin-A polypeptide, a fragment thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1.

    15. A peptide of sequence SEQ ID NO: 2 bearing an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain associated with two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1.

    16. A method of quantifying absolute amount of sequence SEQ ID NO: 2, comprising the use of a calibration standard polypeptide of sequence SEQ ID NO: 2, optionally labeled with one or more mass-modifying labeling agent.

    17. A method for treating a patient by immunotherapy which comprises the steps of: 1) selecting a patient for immunotherapy by: a) detecting the level of expression of a Fetuin-A polypeptide, or a fragment thereof, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and/or comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by reference to the amino acid positions as shown in sequence SEQ ID NO: 1, in a biological sample from said patient; b) comparing said level of expression with a control; c) selecting or rejecting said patient for immunotherapy based on the comparison with the control; wherein said biological sample is taken before the commencement of immunotherapy, and wherein said immunotherapy comprises administration of an allergen or auto-antigen to said patient in order to treat allergy or auto-immune disease; and 2) administering said allergen or auto-antigen to said patient if the patient is selected at step c).

    18. The method for treating a patient by immunotherapy of claim 17, wherein selecting or rejecting said patient for immunotherapy based on the comparison with the control is performed by: (i) if the control is derived from a responder subject or group of responder subjects known to respond to said immunotherapy, selecting the patient for immunotherapy if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is equal to or greater than the level of expression in the control; (ii) if the control is derived from a non-responder subject or group of non-responder subjects, selecting the patient for immunotherapy if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is greater than the level of expression in the control; or (iii) if the control is derived from a randomly selected group of subjects, selecting the patient for immunotherapy if the level of expression of said Fetuin-A polypeptide, or fragment thereof, in the patient sample is equal to or greater than the level of expression in the control.

    19. The method for treating a patient by immunotherapy of claim 17, wherein (i) the control consists of a value of the relative abundance of the peptide SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, of at least 0.0046% of the total peptide abundance, or (ii) the control consists of a value of the relative abundance of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, of at least 3,600 units, and wherein the quantification of the abundance of said Fetuin-A polypeptide or fragment thereof is determined by: (a) depleting a serum sample in albumin, IgG, anti-trypsin, IgA, transferrin and haptoglobulin; (b) degrading with trypsin the depleted serum sample; (c) quantifying the relative abundance in said trypsin-digested depleted serum sample of the peptide SEQ ID NO:2, comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270, by LC-MS/MS.

    20. The method for treating a patient by immunotherapy of claim 17, wherein the immunotherapy comprises administration of allergen to a mucosal surface.

    Description

    BRIEF DESCRIPTION OF THE FIGURES

    [0176] FIG. 1: Study design. Following screening and randomization, patients were treated once daily with either a placebo or a 5-grass pollen tablet for 4 months. Grass pollen challenges were performed before treatment and at day 7, and at months 1, 2 and 4, concomitantly with evaluation of ARTSS. The search for predictive biomarkers was performed on 82 serum samples collected from patients before treatment initiation.

    [0177] FIG. 2: Abundances of Fetuin A protein spots measured by 2D-DiGE as mean±SEM in active (A) and placebo (C) groups, and Spearman correlations between Fetuin A abundances and percentages of ARTSS improvement in patients from active (B) and placebo (D) groups after 4 months of AIT. Fetuin A spots are numbered based on their isoelectric point. 4 patient subgroups are defined based on the median percentage of improvement in ARTSS in the active group: active responders (ARs, n=21), active non responders (ANRs, n=20), placebo responders (PRs, n=7), and placebo non responders (PNRs, n=33)

    [0178] FIG. 3: Measurement of intact mass of the A chain of purified Fetuin A from healthy human serum by mass spectrometry.

    [0179] FIG. 4: Post-translational variants OG1, OG2 and OG3 of Fetuin A analyzed by mass spectrometry diamond stands for sialic acid or N-acetylneuraminic acid (Neu5Ac or NeuAc), circle for galactose (Gal), and square for N-acetylgalactosamine (GalNAc). OG3 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270. OG2 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing one terminal sialic acid residue at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270

    [0180] FIG. 5: O-glycopeptide abundances are shown as means±SEM in active (a) and placebo (c) groups. Difference between groups were tested using a Mann-Whitney test (with * and *** indicating p-values≦0.05 and ≦0.001, respectively). Pearson correlation of O-glycopeptide 3 abundance with percentages of ARTSS improvement in patients from active (b) and placebo (d) groups after 4 months of AIT. 4 patient subgroups are defined based on the median percentage of improvement in ARTSS in the active group: active responders (ARs, n=21), active non responders (ANRs, n=21), placebo responders (PRs, n=7), and placebo non responders (PNRs, n=33).

    [0181] FIG. 6: Measurement of intact mass of purified sialylated Fetuin A from healthy human serum by mass spectrometry.

    [0182] a) LC-UV analysis of the reduced forms of natural FetA (in a 20 mM DTT containing buffer). b-d) Mass spectrometry analysis of B chain. Non-glycosylated (b) and sialylated forms (O-linked glycans) can be observed on the mass spectrum (c, d). Square=N-acetylglucosamine; circle=hexose; diamond=N-acetylneuraminic acid (sialic acid).

    [0183] FIG. 7: Measurement of intact mass of purified desialylated Fetuin A from healthy human serum by mass spectrometry.

    [0184] a) LC-UV analysis of the reduced forms of natural FetA (20 mM DTT).

    [0185] b-d) Mass spectrometry analysis of A chain (b) and B chain (c-d). Square=N-acetylglucosamine; circle=hexose; diamond=N-acetylneuraminic acid (sialic acid). The term AsialoFetA means Fetuin A desialylated.

    [0186] FIG. 8: Spearman correlations between O-glycopeptides abundances and percentages of ARTSS improvement in patients from the active group after 4 months of AIT. OG=O-glycopeptide. Total peptide abundance for a sample of 82 patients (42 treated with active tablets and 40 with placebo)=6379678890; Mean of total peptide abundance per patient=77800962.

    [0187] FIG. 9: ROC curves of OG3 abundance levels of 42 treated patients (AUC: area under the ROC curve), threshold to generate groups: (A) Percentage of ARTSS improvement=50%, (B) Percentage of ARTSS improvement=10%. OG3 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270). In these ROC curves, controls are defined as responder patients (i.e. patients with a percentage of improvement in ARTSS greater than or equal to the percentage of ARTSS improvement thresholds defined above).

    [0188] FIG. 10: Patients exclusion based on OG3 abundance and its impact on clinical efficacy. “50% High OG3 abundance” subgroups represent patients from active and placebo groups for which peptide abundances before treatment are higher than the median value of OG3 abundance calculated for all patients (i.e. from both active and placebo groups) Median OG3 Abundance=36301. OG3 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270). Percentages of ARTSS improvement are shown as mean±SEM in active and placebo groups. Mean variations between active and placebo are tested with a Mann-Whitney statistical test.

    [0189] FIG. 11: Sialylated Fetuin-A but not desialylated Fetuin-A synergizes with LPS to engage the TLR4 pathway.

    [0190] a) Optical density values (OD at 655 nm) after 18 hours of stimulation of hTLR4 HEK-293 reporting cells incubated with increasing doses of either LPS, Sialylated Fetuin-A or a combination of both. * and # indicate a statistically significant difference versus Sialylated Fetuin-A-stimulated or LPS-stimulated hTLR4 HEK-293 cells respectively, using a two-way ANOVA test with Tukey's correction. b) The expression of CD80, CD83 and CD86 co-stimulatory markers was assessed at the surface of MoDCs by flow cytometry. c) Cytokine production by stimulated DCs was measured using a Luminex assay. Data are shown as means±SEM (n=4) in FIGS. 4b and 4c. d) hTLR4 HEK-293 reporting cells were stimulated with either LPS (100 ng/mL), Sialylated Fetuin-A (10 μg/mL), Desialylated Fetuin-A (10 μg/mL) or combinations of those molecules. OD values were measured at 655 nm after 18 hours of incubation. e) Cell surface expression of co-stimulatory molecules was assessed by flow cytometry on stimulated MoDCs and f) cytokine production was measured using a Luminex assay. Data are shown as means±SEM (n=6) in FIGS. 4e and 4f. A p-value≦0.05 is considered significant (Friedman test).

    [0191] FIG. 12: Sialylated Fetuin-A, but not Desialylated Fetuin-A, enhances specific DC2 pro-inflammatory characteristics.

    [0192] a) DC2s cultured for 24 hours in serum-free medium in presence of Sialylated Fetuin-A or Desialylated Fetuin-A were tested for expression of co-stimulatory molecules (CD80, CD83 and CD86) by flow cytometry. b) Cytokine production was measured using a Luminex assay. c) Expression of DC1, DC2 or DCreg-related genes was measured by real-time PCR. Data are shown as means±SEM (n=4) in FIG. 5a to c. d) Cytokine production by DC2s cultured in presence of Sialylated Fetuin-A or Desialylated Fetuin-A (both at 10 μg/mL) and a 5 grass pollen extract (20 μg/mL) was measured using a Luminex assay. Results are presented for n=6 (out of 12 donors tested) as means±SEM. e) Quantification of IL-6 and IL-10 produced by DCs in presence of Sialylated Fetuin-A or Desialylated Fetuin-A (both at 10 μg/mL) and a D. far. (1 μg/mL) extract was performed using a Luminex assay (means±SEM on n=6 donors).

    [0193] FIG. 13: ROC curves of OG3 abundance levels of 42 treated patients (AUC: area under the ROC curve) in which the controls are defined as non-responder patients (i.e. patients with a percentage of improvement in ARTSS lower than a threshold). Threshold used to generate groups: (A) Percentage of ARTSS improvement=50%, (B) Percentage of ARTSS improvement=43.9%, (C) Percentage of ARTSS improvement=10%. OG3 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270).

    EXAMPLES

    Example 1

    Materials and Methods

    [0194] Clinical Samples from VO56.07A Pollen Chamber Study

    [0195] Eighty-nine allergic patients were randomized 1:1 to receive either a grass pollen or placebo tablet through the sublingual route. Challenges were performed before treatment and after 1 week and 1, 2, and 4 months of treatment. Because patients were challenged before treatment, individual clinical responses were evaluated by calculating percentages of improvement in Average Rhinoconjunctivitis Total Symptom Scores (ARTSSs) between baseline and after 4 months of treatment. The median percentage ARTSS improvement in the active group (corresponding to at least a 43.9% decrease of ARTSS after treatment) was considered a threshold to identify clinical responders. Subjects with a percentage of ARTSS improvement greater than or equal to this threshold were considered responders, and those with improvement lower than the threshold were considered non-responders. Whole blood was collected in 82 patients before and after treatment for immunological measurements, cellular assays or comparative proteomics experiments: 4 patient subgroups, including active responders (ARs; n=21), active non-responders (ANRs; n=21), placebo responders (PRs; n=7), and placebo non-responders (PNRs; n=33).

    [0196] Affinity Depletion of High Abundance Proteins for Proteomics Analyses

    [0197] Serum samples (340 μL each) were processed using a human Multiple Affinity Removal System (MARS) Human 6 (Hu-6HC column, 10×100 mm, Agilent Technologies, Les-Ulis, France), which selectively and specifically removed albumin, IgG, antitrypsin, IgA, transferrin and haptoglobin. An Ultimate 3000 HPLC (Thermo scientific, Villebon-sur-Yvette-Courtaboeuf, France) consisting of a binary pump, a thermostatted autosampler with extended injection volume option, a thermostatted column compartment, a diode array detector, and a thermostatted analytical scale fraction collector was used for the affinity chromatography. Flow-through proteins were collected and concentrated according to the manufacturer's instructions (Hu-6HC column, Agilent Technologies). Samples were stored at −80° C. until analysis and run-to-run reproducibility of depletion was confirmed with the chromatographic data and SDS-PAGE analysis under reducing conditions (4-12% NuPAGE gel, Life Technologies, Saint-Aubin, France).

    [0198] Two-Dimensional DiGE Experimental Design

    [0199] Serum samples were precipitated with 2D clean-up kit according to GE Healthcare's protocol (Velizy-Villacoublay, France), solubilized in a buffer containing 7 M urea, 2 M thiourea, 4% CHAPS and 30 mM Tris pH 8.8 (all obtained from Sigma-Aldrich, Saint-Quentin-Fallavier, France) and stored at −80° C. Protein concentrations were determined by means of the Bradford assay (Bio-Rad, Marnes-la-Coquette, France) and depleted serum proteins (50 μg) were minimally labeled with 400 pmol of Cy2 (internal standard), Cy3, or Cy5 DiGE fluors (GE Healthcare) as described in the instruction manual. Samples consisted of depleted serum from ARs (n=21), ANRs (n=21), PRs (n=7) and PNRs (n=33). A dye-swapping scheme was used to avoid any specific dye-labeling artifacts. The Cy2 internal standard was obtained by pooling equal amounts of proteins (25 μg) from all 82 patients sera included in analysis. Two SDS-PAGE tests were performed, with each gel containing two different samples (Cy3- and Cy5-labeled) and an internal standard sample (Cy2-labeled). Protein samples labeled with Cy2, Cy3, and Cy5 dyes were then mixed and diluted with rehydration buffer, containing 7 M urea, 2 M thiourea, 4% CHAPS, 100 mM DTT, 0.5% IPG pH 4-7 buffer, to a final volume of 450 μl. 150 μg (combination of the three labeled protein samples) of protein was applied to 24-cm-long immobilized pH 4-7 gradient strips (GE Healthcare) via the passive rehydration technique for 15 h. First-dimension isoelectric focusing (IEF) was performed using an IPGPhor 3 electrophoresis unit (GE Healthcare) cooled to 18° C. for a total of 74 kVh. After IEF, strips were equilibrated in urea-containing buffer for full protein reduction, alkylation and placed on top of pre-cast SDS-containing 12.5% polyacrylamide gels (GE Healthcare). SDS-PAGE was carried out at 0.5 W per gel for 1 h followed by 1 W per gel for 16 h (Ettan DALT Twelve Electrophoresis System, GE Healthcare). DiGE gels were scanned using an Ettan DiGE Imager (GE Healthcare) according to the manufacturer's instructions. Based on quality control gel, gel number 79 was excluded from analysis (i.e. ANR sample). Differentially expressed spots were determined by image analysis with SameSpots program (Nonlinear Dynamics, Newcastle upon Tyne, UK) and selected for automatic spot picking (EXquest™ Spot Cutter, Bio-Rad). Preparative gels post-stained with Sypro Ruby (Life Technologies) were used for spot picking and protein identification was performed by tandem mass spectrometry (MS/MS).

    [0200] Identification of Protein Spots by Tandem Mass Spectrometry

    [0201] Gel plugs were digested with trypsin and peptides were separated by reversed-phase chromatography using an Ultimate 3000 RS-nanoLC system (Thermo scientific). Peptides were injected and trapped on an Acclaim® PepMap100 (100 μm×2 cm; C.sub.18, 5 μm particles and pore size at 100 Å, Thermo scientific) and separation was then performed using a C.sub.18 column (Acclaim® PepMap RSLC 75 μm ID, 15 cm, 2 μm particles, and pore size at 100 Å, Thermo Scientific). The nanoLC system was coupled to a high-resolution maXis 4G ESI-Qq-TOF mass spectrometer (Bruker Daltonics, Wissembourg, France). NanoLC-MS/MS data were analyzed using an in-house Mascot server (Matrix Science Ltd, London, UK) or PEAKS program (Bioinformatics Solutions Inc., Waterloo, Canada) to search public databases such as the Uniprot/Swiss-Prot database and the nonredundant National Center for Biotechnology Information (nrNCBI) database, assuming tryptic digestion. Precursor mass and fragment mass were searched with initial mass tolerance of 8 ppm and 0.05 Da, respectively. The search included fixed modification of carbamidomethyl (CAM) cysteine. Minimal peptide length was set to 6 amino acids and a maximum of one miscleavage was allowed. Peptide identifications were accepted if they could be established at a greater than 95% probability as specified by Mascot or PEAKS software.

    [0202] Quantitation by Label-Free Mass Spectrometry

    [0203] 50 μg of serum proteins were mixed with a urea-containing buffer, reduced with 20 mM DTT, alkylated with 50 mM iodoacetamide and digested with trypsin (37° C., overnight, enzyme/substrate ratio of 1/25). After digestion, peptides were acidified with 2.5% FA and analyzed by LC-MS or LC-MS/MS using the Ultimate 3000 RSLC system coupled to a maXis 4G ESI-Qq-TOF mass spectrometer. 5 μg of tryptic peptides were injected on an Acquity C.sub.18 BEH130 column (10 mm ID, 10 cm, 1.7 μm particles, Waters) equilibrated at 40° C. with 95% solvent A (0.15% FA); 5% solvent B (100% ACN 0.15% formic acid) at a flow rate of 0.4 mL/min. Separation was performed at a flow rate of 0.4 ml/min with a linear gradient from 5 to 40% solvent B over 30 min. Ion intensities recorded in LC-MS data were analyzed using Progenesis LC-MS v4.1 software (Nonlinear Dynamics, Newcastle upon Tyne, UK) to provide reliable measurements of peptide abundance across samples. Lock mass calibration was performed and peptide detection was performed with peptide intensity>1000, peptide abundance>2000 and 2+≦peptide charge≦12+. Data were then normalized by the “normalize to all features” method and comparison between the four groups (obtained from ARs, ANRs, PRs and PNRs) was performed to choose which peptides were statistically differentially represented. LC-MS/MS data were analyzed using an in-house Mascot server against the Swiss-Prot database, taxonomy Homo sapiens, assuming tryptic or semi-tryptic digestion as described above, and were subsequently imported into software.

    [0204] Enzymatic Desialylation and Characterization of Human Fetuin-A

    [0205] 31.15 μg of native purified Fetuin-A were mixed with 10 μL of 5× Reaction Buffer (EDEGLY kit, Sigma). 1 μL of α-(2.fwdarw.3,6,8,9)-Neuraminidase was added (EDEGLY kit, Sigma), to reach a final enzymatic activity of 0.109 U/mL, and the mixture was incubated overnight at 37° C. Desialylation efficiency was assessed by SDS-PAGE and MS characterization.

    [0206] Mass Spectrometry Analyses of Purified Fetuin A

    [0207] Measurements of average mass (Mav) of intact proteins were performed on a maXis ESI-Qq-TOF mass spectrometer coupled to an Ultimate 3000 RSLC system (Thermo Scientific). Proteins were denatured and reduced in a buffer containing 8M urea, 75 mM Tris pH 8.5, 20 mM DTT for 20 min and acidified by 2.5% formic acid (FA). Protein samples were then desalted and concentrated for 4 min onto a Acquity C.sub.4 BEH300 column (10 mm ID, 10 cm, 1.7 μm particles, Waters, Saint-Quentin-en-Yvelines, France) equilibrated at 70° C. with 95% solvent A (0.15% FA); 5% solvent B (100% ACN 0.15% formic acid) at a flow rate of 0.4 mL/min. Proteins were eluted directly into the mass spectrometer at a flow rate of 0.4 mL/min with a linear gradient from 5 to 60% solvent B over 30 min. Mass spectra were deconvoluted using MaxEnt, and the precise Mav of proteins was determined with Data Analysis (Bruker Daltonic).

    [0208] The presence of post-translational modifications (PTMs) on purified proteins was determined by nanoLC-MS/MS as described above. Purified Fetuin A or desialylated Fetuin A proteins were solubilized in a buffer containing 8 M urea, 75 mM Tris pH 8.5 and 5 mM TCEP. Proteins were then alkylated with iodoacetamide (10 mM) for 20 min and subjected to a trypsin digestion in-solution (protein/protease mass ratio 50/1) for 3 h at 37° C. in presence of 0.018% of ProteaseMax surfactant (Promega, Charbonnieres, France). 2.5% FA was added to the mixture to quench enzymatic activity and peptides were stored at −80° C. until the day of analysis. Peptide samples were then spun at 18000 g and the peptide mixture was separated using a C.sub.18 column (Acclaim® PepMap RSLC 75 μm ID, 25 cm, 2 μm particles, and pore size at 100 Å, Thermo Scientific). N- and O-glycopeptides were manually identified by the presence of glycan-specific oxonium ion fragments.

    [0209] Immunomodulatory Properties of Sialylated Fetuin-A and Desialylated Fetuin-A

    [0210] Monocytes isolated from PBMCs of healthy volunteers with CD14 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) were cultured in presence of GM-CSF and IL-4 (Miltenyi Biotec) for 6 days. DC differentiation was confirmed by flow cytometry based on the loss of CD14 expression and the upregulation of CD1a and CD11c surface expression. Immature monocyte-derived DCs (MoDCs) or MoDCs polarized towards a DC2 pattern were cultured in serum free medium (CelIGro DC medium, Cellgenix, Freiburg, Germany) for 24 hours, in presence of either medium alone, or medium supplemented with either ultra-pure LPS-EB from Escherichia coli (100 ng/mL, Invivogen, Toulouse, France), native purified sialylated Fetuin-A (10 μg/mL), desialylated Fetuin-A (10 μg/mL), a 5-grass pollen (from Lolium perenne, Poa pratensis, Phleum pratense, Dactylis glomerata and Anthoxanthum odoratum) or Dermatophagoïdes farinae allergen extracts (20 or 1 μg/mL, respectively, Stallergenes SA, Antony, France), or combinations of these reagents. In some experiments, the TLR4 pathway was blocked by adding LPS from Rhodobacter sphaeroides (LPS-RS, 10 μg/mL, Invivogen) to MoDCs for 30 minutes. Supernatants were harvested to measure cytokine (IL-1β, IL-6, IL-8, IL-10, IL-12p70 and TNF-α) production using a Luminex assay (Milliplex, Millipore). Surface expression of maturation markers (CD80, CD83 and CD86,) was assessed on DCs by flow cytometry (BD Biosciences, San Jose, US). Total RNA (RNeasy Mini Kit, Qiagen, Venlo, Netherlands) was isolated to evaluate the expression of C1QA, MX1, and PADI2 genes by real-time PCR using β-actin as an endogenous reference gene (the following primers, all from Thermo scientific, were used: Hs00381122_m1, Hs00895608_m1 and Hs00247108_m1, respectively).

    [0211] Statistical Analysis

    [0212] Data are expressed as means±SEMs. Statistical differences between groups were assessed by using 2-tailed nonparametric Mann-Whitney test or by using 2-tailed parametric t-test and the Friedman test for multiple comparisons. Correlation analyses were performed by using the nonparametric Spearman test, or the parametric Pearson test. p-values of less than 0.05 were considered significant. Statistical and graphic analyses were performed with GraphPrism 5 software (GraphPad Software, Inc, La Jolla, Calif.). Significant differences in protein expression changes in DiGE analysis, and in peptide abundance in label-free MS experiments were assessed by using an anova p-value threshold of 0.01. A fold-change filter (≧1.5) was also used. Statistics on proteomic data were performed with two programs from Nonlinear Dynamics (Newcastle upon Tyne, United Kingdom) called Samespots or Progenesis LC-MS.

    Example 2

    Results

    [0213] Sialylation of Serum Fetuin A Discriminates Clinical Responders

    [0214] The clinical study described in Example 1 was used to compare serum proteome profiles and predict clinical responders from non-responders to treatment with AIT. Serum samples, from 4 patient subgroups, including active responders (ARs; n=21), active non-responders (ANRs; n=20), placebo responders (PRs; n=7), and placebo non-responders (PNRs; n=33) were retrospectively compared by using 2D-DiGE. After immune-affinity depletion of the top 6 high abundant proteins to enhance the detection of lower abundance proteins, protein spots whose volume was significantly different between subgroups were cut out from 2D-gels, trypsin digested and analyzed by mass spectrometry (MS). Differences in levels of expression of Fetuin A protein spots (also termed Alpha-2-HS-glycoprotein, AHSG) were observed between AR and ANR patients (FIG. 2). The abundance of four proteoforms of A chain (sp419, 428, 439 and 448) was increased in AR individuals (p<0.01 or 0.001, FIG. 2A) whereas the one of sp469 was decreased (p<0.05). When plotted against percentages of symptom score improvement for each individual patient, the abundance of the more acidic spots, i.e. sum of sp419, 428 and 439, was significantly correlated with clinical benefit in patients from the active group (with Spearman correlation of r=0.5, p=0.0009, FIG. 2B), whereas no difference in levels of expression and no such correlation were observed in placebo-treated patients (FIG. 2C-D). Collectively, those observations indicate that Fetuin A-spot train could shift toward acidic pH values, suggesting post-translational modification (PTM) changes in AR patients.

    [0215] The inventors subsequently purified Fetuin A from healthy human serum by affinity chromatography to gain further insight into the type of PTMs associated. The A chain resolved on SDS-PAGE as a single band at ˜40 kDa and measurements of intact mass by using MS revealed the contribution of a large number of sialic acid (Neu5Ac) branched on glycan moieties (FIG. 3 and FIG. 6A). Strikingly, Neu5Ac is anionic and therefore likely contributes to Fetuin A's charge and the migration toward acidic pH during 2D-DiGE, as observed in AR donors. The MS analysis also revealed that a portion of B chain is sialylated with one or two Neu5Ac (FIG. 6B-D). Lastly, the treatment of purified Fetuin A with sialidase, attested by MS analysis of asialofetuin A (FIG. 7), markedly changed the 2D-gel profile and allowed the complete disappearance of the spot charge-train.

    [0216] Taken together, the shift of Fetuin A-spot train toward acidic pH is dependent on sialylation levels, suggesting that the more the patient is responder to AIT, the greater the sialylation of Fetuin A before AIT is.

    [0217] Sialylation Levels of Fetuin A Depend of O-Linked Oligosaccharide Chains

    [0218] As a glycoprotein, Fetuin A carries N-linked and O-linked oligosaccharide chains that terminate with sialic acid residues (Swiss-Prot AC P02765) and the inventors observed that the positive AIT response is associated with the increase of sialylated glycoforms before treatment. The inventors thus comprehensively examined for its glycan structures by using MS in order to provide a reliable quantitative comparability of those glycoforms in clinical samples. Glycopeptides, derived from purified Fetuin A by trypsin digestion, were assigned based on a combination of the MS/MS data and the accurate precursor ion mass measurement. Particularly unique to glycopeptides containing sialylated glycans are B-type ions corresponding to neutral mass 291 Da (Neu5Ac) and neutral mass 273 Da (Neu5Ac—H.sub.2O). Mass peaks also include those observed as 203 Da, 365 Da, 656 Da, and 947 Da which correspond to the neutral masses of HexNAc, (Hex+HexNAc), (Hex+HexNAc+Neu5Ac) and (Hex+HexNAc+Neu5Ac2). In this work, two sites of N-glycan attachment were evidenced for residues N.sub.156 and N.sub.176 and the most abundant N-glycan found was disialylated (Hex5HexNAc4Neu5Ac2, Table 2). In addition to N-glycans, a number of O-glycopeptides were present, mostly attached with sialylated glycans. The O-glycopeptide from A chain contained two mucin-type glycans for residues T.sub.256 and T.sub.270 and all of which were sialylated, with one or two Neu5Ac (termed OG1, OG2, and OG3 depending on the number of Neu5Ac, FIG. 4 and Table 2). The B chain was observed in both non-, mono- and disialylated form (residue S.sub.346, FIG. 6B-D).

    [0219] Furthermore, considering the site-specific glycosylation of Fetuin A, it is now possible to conduct analysis of clinical samples to understand the diagnostic potential of sialylated Fetuin A. A label-free MS analysis was developed for the direct comparison of MS signals corresponding to Fetuin A-peptides and -glycopeptides (Table 2). With this method in place, the inventors conducted analysis of clinical samples and observed that O-glycopeptides containing core 1 type with two Neu5Acs were the most significantly increased in AR individuals (OG2, p<0.01 and OG3, p<0.001, FIG. 5A and disialylated B chain p<0.01, Table 2). Strikingly, as previously observed for the abundance of the more acidic spots of Fetuin A, the tetrasialylated glycopeptide OG3 significantly correlated with improvement of clinical symptoms in patients from the active group (with Spearman correlation of r=0.5, p=0.0002, FIG. 5B and FIG. 8), whereas no such correlation were observed in placebo-treated patients (FIG. 5C-D).

    [0220] In summary, sialylation levels of Fetuin A dismiminate clinical responders from non-responders to treatment with grass pollen allergy vaccine and are dependent, at least in part, of the number of Neu5Acs attached to O-linked oligosaccharide chains. The sialylation of Fetuin A may thus play a significant role in its functional heterogeneity.

    [0221] The inventors showed that the peptide variants herein named OG1, OG2 and OG3 were present at a significantly higher level in the serum of patients that responded well to sublingual immunotherapy. More importantly, although all three peptides showed a significant level of correlation, peptide variant OG3 presented a better statistical correlation with the response to treatment. Similarly, OG3 presented a more significant correlation than global Fetuin-A (OG1+OG2+OG3).

    [0222] Peptide OG3 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270.

    [0223] Peptide OG2 consists of SEQ ID NO:2 comprising an O-linked oligosaccharide chain bearing one terminal sialic acid residue at position 256 and an O-linked oligosaccharide chain bearing two terminal sialic acid residues at position 270.

    TABLE-US-00003 TABLE 2 Summary of the main FetuinA tryptic-peptides identified by nanoESI-Qq-TOF MS/MS. Measured (Meas.). Peptide charge state (z). Carbamidomethyl Cys (CAM). Phosphorylation (Phospho). Hexose (Hex). N-acetylhexosamine (HexNAc). Sialic acid (NeuAc). Range, numbering according to the pro-protein sequence. Label- free MS SEQ Glycosyl- Label- (Anova Acce- ID m/z Δ m/z ation Sialic free p- ssion Chain NO: meas. z [ppm] Sequence Modifications type acid Range MS value) FETUA_ A  5 1107.041 4 4.00 APHGPGLIYRQPNCDDPETEEAAL CAM: 14  20-57 HUMAN VAIDYINQNLPWGYKH FETUA_ A  6 1065.0150 4 0.79 PHGPGLIYRQPNCDDPETEEAALV CAM: 12  21-57 HUMAN AIDYINQNLPWGYKH FETUA_ A  7 1121.8587 3 4.17 RQPNCDDPETEEAALVAIDYINQN CAM: 4  29-57 HUMAN LPWGYKH FETUA_ A  8 1136.0470 4 2.09 RQPNCDDPETEEAALVAIDYINQN CAM: 4  29-67 HUMAN LPWGYKHTLNQIDEVKV FETUA_ A  9 1317.6619 2 2.97 DPETEEAALVAIDYINQNLPWGYK  35-57 HUMAN H FETUA_ A 10  971.8927 5 2.02 KHTLNQIDEVKVWPQQPSGELFEI CAM: 32  58-99 HUMAN EIDTLETTCHVLDPTPVARC FETUA_ A 11 1098.3058 4 2.35 NQIDEVKVWPQQPSGELFEIEIDTL CAM: 28  62-99 HUMAN ETTCHVLDPTPVARC FETUA_ A 12  508.2569 4 3.86 RQLKEHAVEGDCDFOLLKL CAM: 11 104-120 X ns HUMAN FETUA_ A 13  554.2603 3 1.70 KEHAVEGDCDFQLLKL CAM: 8 107-120 X ns HUMAN FETUA_ A 14  519.2541 4 1.81 KEHAVEGDCDFQLLKLDGKF CAM: 8 107-124 HUMAN FETUA_ A 15  407.2298 2 1.59 KFSVVYAKC 125-131 X ns HUMAN FETUA_ A 16 1326.2169 3 4.29 KVCQDCPLLAPLNDTRV CAM: 2, 5; N-linked 2 145-159 X ns HUMAN Hex(5)HexNAc(4) glycan NeuAc(2): 14 FETUA_ A 16 1544.9553 3 1.03 KVCQDCPLLAPLNDTRV CAM: 2, 5; N-linked 2 145-159 X ns HUMAN Hex(6)HexNAc(5) glycan NeuAc(3): 14 FETUA_ A 17 1183.0825 2 2.00 KAALAAFNAQNNGSNFOLEEISRA 166-187 ns HUMAN FETUA_ A 17 1523.9822 3 KAALAAFNAQNNGSNFOLEEISRA Hex(5)HexNAc(4) N-linked 2 166-187 X ns HUMAN NeuAc(2): 11 glycan FETUA_ A 17 1143.2366 4 0.30 KAALAAFNAQNNGSNFOLEEISRA Hex(5)HexNAc(4) N-linked 2 166-187 x ns HUMAN NeuAc(2): 11 glycan FETUA_ A 17 1742.7277 3 KAALAAFNAQNNGSNFOLEEISRA Hex(6)HexNAc(5) N-linked 3 166-187 HUMAN NeuAc(3): 11 glycan FETUA_ A 18 1290.1579 2 0.94 RAQLVPLPPSTYVEFTVSGTDCVA CAM: 21 188-211 X ns HUMAN KE FETUA_ A 19 1093.8886 3 1.83 RAQLVPLPPSTYVEFTVSGTDCVA CAM: 21 188-218 HUMAN KEATEAAKC FETUA_ A 20  424.2214 2 0.94 KCNLLAEKQ CAM: 1 219-225 X ns HUMAN FETUA_ A 21  401.6823 2 1.99 KQYGFCKA CAM: 5 226-231 X p < 0.05 HUMAN FETUA_ A 2 1741.847 5 2.91 LGGAEVAVTCTVFQTQPVTSQPQ CAM: 10; O-linked 2 238-311 X ns HUMAN PEGANEAVPTPVVDPDAPPSPPLG Hex(1)HexNAc(1) glycan APGLPPAGSPPDSHVLLAAPPGH NeuAc(1): 19, 33 QLHR FETUA_ A 2 1800.0704 5 4.71 LGGAEVAVTCTVFQTQPVTSQPQ CAM: 10; O-linked 3 238-311 X p < 0.01 HUMAN PEGANEAVPTPVVDPDAPPSPPLG Hex(1)HexNAc(1) glycan APGLPPAGSPPDSHVLLAAPPGH NeuAc(1): 19 or 33; QLHR Hex(1)HexNAc(1) NeuAc(2): 19 or 33 FETUA_ A 2 1858.2795 5 -1.30 LGGAEVAVTCTVFQTQPVTSQPQ CAM: 10; O-linked 4 238-311 X p < 0.01 HUMAN PEGANEAVPTPVVDPDAPPSPPLG Hex(1)HexNAc(1) glycan APGLPPAGSPPDSHVLLAAPPGH NeuAc(2): 19, 33 QLHR FETUA_ A 22  724.3871 5 2.43 DPDAPPSPPLGAPGLPPAGSPPD 275-311 HUMAN SHVLLAAPPGHQLHRA FETUA_ A 23  387.6989 2 0.09 RAHYDLRH 312-317 X ns HUMAN FETUA_ A 24  568.0879 5 5.50 RAHYDLRHTFMGVVSLGSPSGEV 312-337 HUMAN SHPRK FETUA_ A 25  642.3136 3 2.16 RHTFMGVVSLGSPSGEVSHPR 318-336 HUMAN FETUA_ A 26  694.3468 3 1.16 RHTFMGVVSLGSPSGEVSHPRK 318-337 X ns HUMAN FETUA_ A 26  541.0047 4 3.36 RHTFMGVVSLGSPSGEVSHPRK Phospho: 13 318-337 X ns HUMAN FETUA_ A 26  721.0030 3 2.16 RHTFMGVVSLGSPSGEVSHPRK Phospho: 13 318-337 X ns HUMAN FETUA_ A 27  553.0377 4 5.00 RHTFMGVVSLGSPSGEVSHPRKT 318-338 HUMAN FETUA_ B 28 1008.5432 2 4.76 RTVVQPSVGAAAGPVVPPCPGRI CAM: 18 341-361 X ns HUMAN FETUA_ B 28 1336.6567 2 3.40 RTVVQPSVGAAAGPVVPPCPGRI CAM: 18; O-linked 1 341-361 X ns HUMAN Hex(1)HexNAc(1) glycan NeuAc(1): 1 FETUA_ B 28  891.4380 3 0.90 RTVVOPSVGAAAGPVVPPCPGRI CAM: 18; O-linked 1 341-361 X ns HUMAN Hex(1)HexNAc(1) glycan NeuAc(1): 1 FETUA_ B 28  988.4692 3 0.14 RTVVQPSVGAAAGPVVPPCPGRI CAM: 18; O-linked 2 341-361 X p < 0.01 HUMAN Hex(1)HexNAc(1) glycan FETUA_ B 29  958.0155 2 0.94 TVVQPSVGAAAGPVVPPCPGRI CAM: 17 342-361 HUMAN FETUA_ B 30  908.4811 2 0.81 VVQPSVGAAAGPVVPPCPGRI CAM: 16 343-361 HUMAN

    [0224] The pertinence of the polypeptide OG3 was further assessed by a receiver operating characteristic (ROC) analysis. The ROC curve of OG3 levels of 42 active patients divided in 2 subgroups based on a percentage of improvement in ARTSS of 50% is shown in FIG. 9A. The AUC was of 0.7015 (with p-value of 0.03509). The ROC curve of OG3 levels of 42 active patients divided in 2 subgroups based on a percentage of improvement in ARTSS=10%, % is shown in FIG. 9B. The AUC was of 0.7978 (with p-value of 0.0095). In these ROC curves, controls are defined as responder patients (i.e. patients with a percentage of improvement in ARTSS greater than or equal to the percentage of ARTSS improvement thresholds defined above).

    [0225] Moreover ROC curve of OG3 levels of 42 active patients divided in 2 subgroups based on a percentage of improvement in ARTSS of 50% and in which the controls are defined as non-responder patients (i.e. patients with a percentage of improvement in ARTSS lower than 50%) is shown in FIG. 13.A. The AUC was of 0.7015 (with p-value of 0.03509). The ROC curve of OG3 levels of 42 active patients divided in 2 subgroups based on a percentage of improvement in ARTSS of 43.9% and in which the controls are defined as non-responder patients (i.e. patients with a percentage of improvement in ARTSS lower than 43.9%) is shown in FIG. 13.B. The AUC was of 0.8050 (with p-value of 0.0007). The ROC curve of OG3 levels of 42 active patients divided in 2 subgroups based on a percentage of improvement in ARTSS of 10% and in which the controls are defined as non-responder patients (i.e. patients with a percentage of improvement in ARTSS lower than 10%) is shown in FIG. 13.C. The AUC was of 0.7978 (with p-value of 0.0095).

    [0226] These latter ROC curves in which the controls are defined as non-responder patients were associated with the following data on sensitivity and specificity.

    TABLE-US-00004 TABLE 3 Threshold at 10% improvement in ARTSS Cutoff OG3 Sensitivity % Specificity % >8619 100 12.5 >9377 100 25 >10984 100 37.5 >14446 100 50 >20117 97.06 50 >23537 97.06 62.5 >23893 94.12 62.5 >24170 91.18 62.5 >24893 88.24 62.5 >26369 85.29 62.5 >27576 82.35 62.5 >28037 79.41 62.5 >28520 76.47 62.5 >28950 73.53 62.5 >29308 70.59 62.5 >30786 67.65 62.5 >32026 64.71 62.5 >32918 61.76 62.5 >33777 58.82 62.5 >34916 55.88 62.5 >36447 55.88 75 >37862 52.94 75 >39205 50 75 >41113 50 87.5 >43103 47.06 87.5 >43967 44.12 87.5 >45310 41.18 87.5 >46825 38.24 87.5 >48242 35.29 87.5 >49345 35.29 100 >50051 32.35 100 >50788 29.41 100 >51984 26.47 100 >57652 23.53 100 >62528 20.59 100 >62952 17.65 100 >64570 14.71 100 >67985 11.76 100 >71033 8.824 100 >73660 5.882 100 >83772 2.941 100

    TABLE-US-00005 TABLE 4 Threshold at 43.9% improvement in ARTSS Cutoff OG3 Sensitivity % Specificity % >8619 100 4.762 >9377 100 9.524 >10984 100 14.29 >14446 100 19.05 >20117 95.24 19.05 >23537 95.24 23.81 >23893 95.24 28.57 >24170 95.24 33.33 >24893 95.24 38.1 >26369 95.24 42.86 >27576 90.48 42.86 >28037 90.48 47.62 >28520 90.48 52.38 >28950 90.48 57.14 >29308 85.71 57.14 >30786 80.95 57.14 >32026 80.95 61.9 >32918 80.95 66.67 >33777 76.19 66.67 >34916 76.19 71.43 >36447 76.19 76.19 >37862 71.43 76.19 >39205 71.43 80.95 >41113 71.43 85.71 >43103 66.67 85.71 >43967 61.9 85.71 >45310 57.14 85.71 >46825 52.38 85.71 >48242 47.62 85.71 >49345 47.62 90.48 >50051 42.86 90.48 >50788 38.1 90.48 >51984 33.33 90.48 >57652 28.57 90.48 >62528 23.81 90.48 >62952 19.05 90.48 >64570 19.05 95.24 >67985 19.05 100 >71033 14.29 100 >73660 9.524 100 >83772 4.762 100

    TABLE-US-00006 TABLE 5 Threshold at 50% improvement in ARTSS Cutoff Sensitivity % Specificity % >8619 100 3.571 >9377 100 7.143 >10984 100 10.71 >14446 100 14.29 >20117 92.86 14.29 >23537 92.86 17.86 >23893 92.86 21.43 >24170 92.86 25 >24893 92.86 28.57 >26369 92.86 32.14 >27576 92.86 35.71 >28037 92.86 39.29 >28520 92.86 42.86 >28950 92.86 46.43 >29308 85.71 46.43 >30786 78.57 46.43 >32026 78.57 50 >32918 78.57 53.57 >33777 78.57 57.14 >34916 78.57 60.71 >36447 78.57 64.29 >37862 71.43 64.29 >39205 71.43 67.86 >41113 71.43 71.43 >43103 64.29 71.43 >43967 64.29 75 >45310 57.14 75 >46825 50 75 >48242 42.86 75 >49345 42.86 78.57 >50051 35.71 78.57 >50788 28.57 78.57 >51984 21.43 78.57 >57652 21.43 82.14 >62528 21.43 85.71 >62952 14.29 85.71 >64570 14.29 89.29 >67985 14.29 92.86 >71033 7.143 92.86 >73660 7.143 96.43 >83772 7.143 100

    [0227] These results confirm that OG3 is useful to discriminate clinical responders from non-responders before AIT.

    [0228] Finally, patients from each treatment group (i.e. active and placebo) were divided into 2 subgroups depending upon levels of OG3 found in their plasma before treatment (subgroups “50% high OG3 abundance”). 50% High OG3 abundance” subgroups represent patients from active and placebo groups for which peptide abundances before treatment are higher than the median value of OG3 abundance calculated for all patients (i.e. from both active and placebo groups) Median OG3 Abundance=36301. As shown in FIG. 10, a more pronounced improvement in clinical symptoms was observed in active patients who had the 50% highest OG3 levels.

    [0229] Sialylated Fetuin-A Synergizes with LPS in a TLR4-Dependent Pathway

    [0230] Functional interactions between sialylated Fetuin-A and TLR4 were examined using HEK-293 cells expressing human TLR4 and an inducible secreted embryonic alkaline phosphatase as a reporter gene. Those cells were stimulated with either LPS, sialylated Fetuin-A at various doses, or a mixture of both. As shown in FIG. 11,a, sialylated Fetuin-A alone had no effect, whereas a combination of sialylated Fetuin-A and LPS acted in synergy to activate TLR4 in a dose-dependent manner.

    [0231] Since dendritic cells (DCs) are critical for T cell priming against allergens, and with the notion that bacterial LPS is an important sensitization cofactor, it was then examined the capacity of sialylated Fetuin-A to synergize with LPS and/or allergens, to impact DC polarization. MoDCs generated in serum-free medium were incubated with either LPS, sialylated Fetuin-A at various doses, or a combination of both. Stimulation of MoDCs with a mixture of LPS and sialylated Fetuin-A enhanced the expression of CD83 and CD86, and to a lower extent of CD80 co-stimulatory molecules (FIG. 11,b), whereas LPS or sialylated Fetuin-A alone had no effect. Similarly, a combination of the two molecules increased the secretion of IL-6, IL-10, IL-12 p-70 and TNF-α cytokines, when compared with LPS alone, whereas sialylated Fetuin-A per se had no activity (FIG. 11,c). No stimulation of MoDCs was observed in presence of α-(2.fwdarw.3,6,8,9)-Neuraminidase, desialylation buffer or free sialic acids (data not shown). Noteworthy, the synergy observed between LPS and sialylated Fetuin-A was totally abrogated by adding the TLR4 antagonist LPS-RS (FIG. 11,b-c), establishing that sialylated Fetuin-A and LPS synergized to activate the TLR4 pathway.

    [0232] Sialylated Fetuin-A and Neuraminidase-treated sialylated Fetuin-A (i.e. desialylated Fetuin-A) were subsequently compared for their capacity to synergize with LPS in those human cellular assays. The synergistic modulation of TLR4 activity by LPS was substantially decreased when hTLR4 HEK-293 cells were co-stimulated with desialylated Fetuin-A as opposed to sialylated Fetuin-A (FIG. 11,d). Consistent with this observation, no synergistic effect was observed between LPS and desialylated Fetuin-A on MoDCs, both in terms of the induction of co-stimulatory molecules (FIG. 11,e) and cytokine secretion (FIG. 11,f). Collectively, our results establish the synergistic activation of the TLR4 pathway by LPS and FetA, with evidence for sialylation of the latter in this functional interaction.

    [0233] Sialylated Fetuin-A, but not Desialylated Fetuin-A, Enhances the Pro-Allergic Features of Type 2 MoDCs (DC2s)

    [0234] Engagement of TLR4 is known to contribute to allergic inflammation, for example during concomitant exposure to allergens with endotoxins, or as a consequence of a functional mimicry of MD2 by the allergen. It was investigated the potential contribution of sialylated Fetuin-A glycoforms to such TLR4-mediated allergic inflammation. Immature MoDCs were polarized towards a DC2 phenotype (i.e. capable to polarize naïve CD4.sup.+ T cells towards IL-5 and IL-13 secreting T.sub.H2 cells), using a mixture of TSLP, IL-25, IL-33 and low doses (i.e. 10 ng/mL) of LPS, in presence of 10 μg/mL sialylated Fetuin-A or desialylated Fetuin-A, respectively. A flow cytometry analysis revealed an up-regulation of CD83 and CD86 co-stimulatory markers in presence of sialylated Fetuin-A but not desialylated Fetuin-A (FIG. 12,a). Similarly, DC2s differentiated in presence of sialylated Fetuin-A secreted higher levels of IL-6, IL-8, IL-10 and TNF-α (FIG. 12,b), when compared to DC2s treated with desialylated Fetuin-A. The latter observation was further confirmed when monitoring the expression of genes specifically associated with a DC2 polarization, such as PADI2, GATA3 and NMES. Such DC2 marker genes were overexpressed in presence of sialylated Fetuin-A but not desialylated Fetuin-A. In contrast, the expression of genes rather associated with DC1 or regulatory DCs, such as MX1 or C1QA, respectively, was highly down-regulated when DC2s were cultured in the presence of sialylated Fetuin-A, but not desialylated Fetuin-A (FIG. 12,c). In addition, following a stimulation of DC2s with a combination of aqueous allergen extracts and sialylated Fetuin-A or desialylated Fetuin-A, sialylated Fetuin-A synergized with a grass pollen extract to induce the secretion of IL-6, IL-8, IL-10 and TNF-α cytokines (FIG. 12,d), whereas no such effect was observed with desialylated Fetuin-A. Similarly, DCs cultured with house dust mite (HDM) allergen extract secreted higher amounts of IL-6, IL-10, and TNF-α in presence of sialylated Fetuin-A, whereas this synergy was not observed with desialylated Fetuin-A (FIG. 12,e). No clear influence of sialylated Fetuin-A or desialylated Fetuin-A was observed on IL-8 secretion, already very high in presence of house dust mites allergens. Collectively, those experiments established that sialylated Fetuin-A can enhance the proallergic profile of DC2s in presence of natural allergens, in relationship with sialylation levels of the Fetuin-A molecule.

    Example 3

    Clinical Samples from House Dust Mite Study

    [0235] Study Design

    [0236] The objectives of this multicenter, double-blind, parallel-group comparative study were to evaluate the efficacy and safety of the house dust mite extract in patients with perennial allergic rhinitis due to house dust mite. Patients were randomized to receive house dust mite extract at dose A, house dust mite extract at dose B or a placebo tablets. Patients were treated over a period of 12 months. Serum samples were collected and symptoms monitored at weeks 0, 16 and 52.

    [0237] Serum Analysis

    [0238] Serum samples (340 μL each) were processed using a human Multiple Affinity Removal System (MARS) Human 14 (Hu-14 column, 10×100 mm, Agilent Technologies, Palo Alto, USA), which selectively removes α1-acid glycoprotein, α1-antitrypsin, α2-macroglobuin, albumin, apolipoprotein A1, apolipoprotein A2, complement C3, fibrinogen, haptoglobin, IgA, IgG, IgM, transferrin, and transthyretin. An Ultimate 3000 HPLC apparatus (Thermo scientific, Waltham, USA) was used for the affinity depletion. Flow-through proteins were collected and concentrated according to the manufacturer's instructions. Following depletion, samples were stored at −80° C. until analysis. The run-to-run reproducibility of depletion was confirmed by chromatography and SDS-PAGE analyses under reducing conditions, using 4-12% NuPAGE gels (Thermo scientific).