Standardisation method for glycosylation arrays

Abstract

The current invention provides methods to improve inter-assay precision in glycosylation assays and which can also be applied to point-of-care situations. The methods can be enabled in the form of substrates onto which are immobilised capture probes and glycan standards. The methods can be applied in cancer diagnosis.

Description

Standardisation method for glycosylation arraysBackgroundMany proteins are glycosylated after translation. This post-translational modification involves chemical attachment of sugars to a protein by glycosidic bonds to yield a glycoprotein. Protein glycosylation may take a number of forms and is defined according to the type of glycosidic bonds that occur.N-linked glycosylation involves attachment of a sugar molecule to the nitrogen (N4) atom in the side chain of an Asparagine residue within the consensus sequence Asn-Xaa-Ser / Thr (where Xaa is not Proline). This modification occurs in secreted and membrane proteins of eukaryotes and archea, and it is absent in bacteria. This process begins co-translationally in the endoplasmic reticulum, where preassembled blocks of 14 sugars (including 2 N- acetylglucosamines, 9 mannoses and 3 glucoses) are added to the nascent polypeptide chain. After cleavage of 3 glucose and 1 mannose residues, the protein is transferred to the Golgi apparatus where the glycans lose a variable number of mannose residues and acquire a more complex structure during a process called ‘terminal glycosylation’. This process yields 3 types of mature N-glycans: high mannose (those that have escaped terminal glycosylation), hybrid and complex (with different combinations of mannose, N- acetylglucosamine, N-acetylgalactosamine, fucose and sialic acid residues).O-linked glycosylation refers to the attachment of glycans to the oxygen atom in the side chain of Serine and Threonine residues. This modification occurs in secreted and membrane proteins of eukaryotes and resulting O-linked glycans play important roles in protein localization and trafficking, protein solubility, antigenicity and cell-cell interactions. The process is a post-translational event that takes place in the cis-Golgi compartment after N-glycosylation and folding of the protein. O-linked glycans are built in a stepwise fashion with sugars being added incrementally. The most common type of O-glycosylation in secreted and membrane-bound mammalian proteins is the addition of reducing terminal N- acetylgalactosamine (GalNAc) to yield “mucin-type” O-linked glycans. This residue can be further extended with galactose (Gal), N-acetylglucosamine (GIcNAc) or GIcNAc and Gal resulting in 8 common core structures, which are often further decorated with the addition of up to three sialic acid residues. In addition to the “mucin-type” glycans, a variety of mammalian proteins are known to have mannose (Man), fucose (Fuc), glucose (Glc), galactose (Gal) or xylose (Xyl) as reducing terminal linkages.Some eukaryotic cytoplasmic and nuclear proteins have simple O-linked glycans in which a single N-acetylglucosamine is linked to a serine or a threonine residue. This type of glycosylation plays an important role in the modulation of the biological activity ofintracellular proteins, and in some proteins the same residue may be subject to competing phosphorylation and O-linked glycosylation.C-linked glycosylation refers to covalent attachment of mannose residue to Tryptophan residues within an extracellular protein. Two recognition signals for C-mannosylation have been proposed: W-X-X-W (in which the first or both tryptophan residues become mannosylated), and W-S / T-X-C.Glycation refers to the non-enzymatic attachment of reducing sugars to Nitrogen atoms of proteins (N-termini and Lysine and Histidine side chains) in a process known as the “Maillard” reaction. Sugars bound to glycated proteins are gradually modified to become Advanced Glycation End products (AGEs), which have been implicated in a variety of diseases including type II diabetes mellitus, cancer, atherosclerosis, Alzheimer’s disease and Parkinson’s disease.Protein glycosylation has been linked with a variety of human diseases. For example, aberrant protein glycosylation is an established event in cancer development and progression [1 , 2], Additionally, defects in protein glycosylation have been implicated in a number of cellular storage disorders including Gaucher's, Niemann-Pick type C, Sandhoff's, and Tay-Sachs diseases, and more recently protein glycosylation has been linked to the onset of cardiovascular disease. Credible hypotheses supporting a role for protein glycosylation in the pathology and epidemiology of Alzheimer’s disease have also been proposed recently.Sugars covalently attached to proteins are identifiable using lectins, proteins that recognise and bind to specific glycans. Lectins were initially discovered in plants and are now known to be ubiquitous in nature. Each lectin has distinct glycan specificity, and generation of a glycosylation profile for any given protein therefore requires testing against a multitude of individual lectins. Existing technologies for generation of protein glycosylation profiles using this approach requires the use of one or more reporter tags or indirect methods to allow detection of and to distinguish binding of individual lectins and / or antibodies to glycoproteins immobilised on or bound to a surface [3, 4], Alternatively, if a small number of reporter tags are available, generation of protein glycosylation profiles is a substantial task that requires that the glycoproteins immobilised on or bound to a surface are serially exposed to a small number of lectins so as to build up the number of data points required to produce that profile [5], However, mass spectrophotometric methods remain the gold standard for comprehensive one step analysis of protein glycosylation [6],It has been reported that protein glycosylation may interfere with or indeed prevent binding of a protein by an antibody that specifically recognises the protein in a non-glycosylatedstate [7], This can lead to underestimates of protein levels in samples and to false results in diagnostic tests.There is a need for improved methods for the detection and measurement of glycosylated proteins, for the generation of glycosylation profiles for such proteins and for the use of protein glycosylation in the screening, diagnosis, prognostication and treatment of human disease.Cancer screening methods from analysis of biological fluid samples are under constant development aiming to provide more accurate cancer diagnoses and to allow more informed clinical decision making for the ultimate benefit of cancer survival. Each type of cancer has individual needs regarding screening including methodologies and indicative biomarkers with varying degrees of efficacy. Furthermore, certain types of cancer such as ovarian and pancreatic may be asymptomatic until an advanced stage where curative resection may no longer be feasible. A clear and urgent need for applicable and accurate screening methodologies for cancer types in such cases exists which may be implemented both locally at a point-of-application level or through a central laboratory as part of a widespread screening programme. An increasing body of evidence indicates the changes in metabolism and gene expression within the cancer cell may be reflected in changes in protein glycosylation. In the search for novel paradigms to be applied to early cancer detection, alterations in biomarker glycosylation status, is emerging as a novel facet of tumour biology which may be exploitable for next-generation diagnostics.Application W02017 / 109518 described methods of determining the glycosylation signature and determining the level of a protein in a sample obtained from a patient. It also described use of a patient protein glycosylation profile to identify the presence or absence of a disease in subjects. There was a need for a method for standardisation and increasing inter-assay precision of such assays and which could also be applied to point-of-care situations, such a method is provided herein.References[1] HAKOMORI SI (1989), Advances in Cancer Research 52:257-331.[2] HAKOMORI SI and Cummings RD (2012), Glycoconjugate Journal 29:565-566.[3] CHEN S et al (2007), Nature Methods 4(5):437-444.[4] MEANY DL et al (2009), Journal of Proteome Research 8(2):613-619.[5] GOODARZI MT and TURNER GA (1997), Glycoconjugate Journal 14(4):493-496.[6] WADA Y et al (2007), Glycobiology 17(4) :411-422.[7] DE GROOT AS and SCOTT DW (2007), Trends Immunol 28: 482-490.[8] PERKINS G et al (2003), American family physician 68 (6): 1075-1082. [9] GOONETILLEKE KS and SIRIWARDENA AK (2007), Eur J Surg Oncol 33 (3): 266-270.

[0010] O’BRIEN et a / (2015), Clin Cancer Res 21 (3) :622-31

[0011] FERRI MJ et al (2016), PLoS One 11 (1):e0147214.

[0012] LIU L ef al (2015), Int J Cancer 136(9):2216-27.

[0013] BALLEHANINNA UK and CHAMBERLAIN RS (2013), Tumour Biology 34(6): 3279-92

[0014] BALLESTA AM et a / (1995), Tumour biology 16 (1): 32-41.

[0015] DUFFY MJ et al (2003), European journal of cancer 39 (6): 718-727.

[0016] CHAO YJ et al (2014), BMC Surg 25 4 72.

[0017] HUANG C et al (2015), Clin Proteomics 27; 12(1): 17.

[0018] MOULE SK et a / (1987), Clinica Chimica Acta. 166; 177-185.

[0019] TURNER GA et al (1985), Journal of Clinical Pathology. 38 588- 592.

[0020] HASHIMOTO S et al (2004), Cancer. 101 ; 2825-2836.

[0021] BALM AN A M et al (2016), J Proteomics 132:144-154.

[0022] RADON TP et a / (2015), Clin Cancer Res 21 (15):3512-3521.

[0023] YUE T et al (2009), Mol Cell Proteomics 8(7): 1697-1707.

[0024] PENG M et al (2019), Cancer Management and Research 11 :2405-2414.

[0025] XU M et al (2015), Int J Clin Exp Med 8(8):12476-12486.

[0026] MERLOTTI A ef a / (2019), Oncoimmunology 8(9):e1629257.

[0027] ZWEIG MH and CAMPBELL G (1993), Clinical Chemistry 39: 561-577.Summary of inventionA first aspect of the current invention is a method for standardising a glycosylation array, said method comprising a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes specific to two or more different target analytes and at separate physical locations glycan standards comprising glycan residues formulatedat two or more different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the capture probes and the glycans on the glycan standards; c) measuring the level of glycan- binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.The measurement of bound glycan-binding agent on the glycan standards allows for a calibration curve to be developed which aids in the quantification of the glycosylation on the target analyte.A second aspect of the invention is a method for aiding in the diagnosis of pancreatic cancer, said method comprising a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes specific to two or more of CA19-9, CEA, CA15-3, CA-125, CA-242, clusterin, A1AG and LYVE-1 and at separate physical locations glycan standards comprising glycan residues formulated at two or more different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the capture probes and the glycans on the glycan standards; c) measuring the level of glycan-binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.As third aspect of the invention is a substrate comprising a solid-state device onto which is immobilised capture probes to two or more target analytes and in separate physical locations glycan standards formulated at two or more different known concentrations.A fourth aspect of the invention is the use of internal glycan standards to calibrate a multiplex glycosylation array.The internal calibration on a single solid-state device provides standardisation for glycosylation assays, increases inter-assay precision and reduces the need for external standards. The internal calibration aspect also allows the device to have utility in a point-of- care screening context in addition to a large scale central laboratory by allowing multiple biomarker measures to be analysed cumulatively from a single multiplexed device.Description of the drawingsFigure 1 - Concept of the internally calibrated multiplexed glycosylated biomarker assay format and application A) Biochip array with analyte-specific captures spotted alongside neoglycoproteins bearing relevant glycans formulated at varying concentrations toform an internal calibration. Glycosylation-specific detection is mediated by labelled lectin- based detectors. B) An example of a biochip array layout consisting of Reference spots, capture antibodies for CEA, CA19-9 and A1AG alongside a nine-point internal calibration. The standards consisted of BSA-Fucose spotted at the following concentrations 0, 0.16, 0.27, 0.47, 0.82, 1.43, 2.48, 4.31 and 7.5 pg / ml, corresponding to spots L1-L9 (L1-L9 refers to concentration level 1 to level 9 of the glycan standard). C) an example of images generated using buffer only controls or three levels of commercially sourced QC sera followed by lectin detection. D) Signal outputs in Relative Light Units from images represented in Panel C. E) Representative calibration curve.Figure 2 - Example of an array with on-chip calibration, discrimination of normal and pancreatic Cancer sera: Fucose-based detection using AAL Lectin. A) RLU output of sera from pancreatic cancer patients or normal subjects (n=15 for each) following analysis using GlycoPanel array and fucose-based detection. B) Glycosylated biomarker readouts expressed in signal relative to BSA-3-fucosyllactose spotted standards (P1-P15 = patient samples, N1-N15 = controls samples). C) ROC analysis of glycosylation associated with CEA, CA19-9, A1AG, CA15-3 and clusterin showing comparison of pancreatic cancer and normal subjects for the individual markers and the 5-marker combination.Figure 3 - Example of an array with on-chip calibration, discrimination of normal and pancreatic Cancer sera. A) Detection of n-acetylglucosamine based on WGA lectin. B) Box-and-whisker plots of glycosylation associated with CA19-9 and LYVE1 showing comparison of pancreatic cancer and normal subjects. C) ROC analysis and combined ROC of CA19-9 and LYVE-1.Detailed description of the inventionProvided herein is a method for the standardisation of glycosylation arrays, said method comprising a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes to two or more different target analytes and in separate physical locations glycan standards comprising glycan residues formulated at two or more different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the capture probes and the glycans on the glycan standards; c) measuring the level of glycan-binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.Such a method provides a previously unavailable solution for multiplexing of glycosylation arrays. It is enabled by an internal calibration format and the use of a single glycan-binding agent for detecting glycan residues on target analytes and glycan standards. This method has the potential to allow the user to develop reference ranges for each target analyte (e.g. protein) for a range of different glycan residues, this would allow sample to be analysed without the need to source (and run) multiple calibrators each time an assay is performed. For example, for an assay to determine glycosylation profiles of ten different proteins typically the skilled person would need to source controls and develop reference ranges for each of the ten proteins for the glycan being determined. This may be time consuming, expensive and for certain proteins which are not obviously glycosylated, may even be impossible to source from commercial suppliers. This enables the method described herein to be useful not only as a research tool but also potentially in point-of-care settings by allowing a clinician to run a patient sample immediately on a solid-state device of the current invention without the need to run external protein standards, calibrators or ladders.The term “internal calibration” means that the reactions used to generate the calibration curve (in this case the determination of glycosylation on the glycan standards) are carried out on the same substrate as the experimental reactions (in this case determining glycosylation on the target analytes). To clarify, this means that there are no physical barriers which prevent the same sample coming into contact with the immobilised capture probes and immobilised glycan standards. In one preferred embodiment glycosylation units may be assigned based on the calibration curve of glycan standards for each glycan to be determined. These can be applied to the glycosylation on target analytes and would allow for simple comparison of glycosylation between targets / samples. For example, when fucosylation is to be determined the concentration of the glycan standard (for example, BSA- fucose neoglycoprotein) spotted at each location on the solid-state device can be determined. The units of the glycan standards spotted (e.g. pg / ml) can then be applied to give a comparable unit of glycosylation for all target analytes for a certain glycan residue (in this example, fucose). As well as allowing comparison of the glycosylation of different analytes in a single multiplex assay the current invention provides a means of developing a universal glycosylation scale for each type of glycosylation which would allow comparison of glycosylation between samples or assays.The term “glycan standard” as used herein refers to the glycans of known concentration which are spotted onto the solid-state device. They may also be referred to as standards or calibrators. Reference to standardisation or calibration herein refers to the use of this range of known glycan concentrations to generate a curve from which the glycosylation on target analyte can be inferred. Glycan residues may be spotted directly onto the solid-state devicebut preferably the glycan standards comprise glycan residues conjugated to a protein (forming a neoglycoprotein) or synthetic carrier (for example, a dendrimer).As used herein, the term “sample” includes any biological samples which may be obtained from a patient or subject, samples may comprise blood, plasma, serum, urine, saliva, seminal plasma, breast milk, cerebrospinal fluid or sputum. The methods of the invention described herein are carried out ex vivo. For the avoidance of doubt, the term “ex vivo” has its usual meaning in the art, referring to methods that are carried out in or on a sample obtained from a subject in an artificial environment outside the body of the subject from whom the sample has been obtained.The term “probe” refers to a molecule that is capable of specifically binding to a target molecule such that the target molecule can be detected as a consequence of said specific binding. Probes may be immobilised on a surface to capture the target molecule they specifically bind. Probes that can be used in the present invention include, for example, antibodies, aptamers and oligonucleotides. Preferred probes include antibodies or ScFvs.The term “capture probe” refers to a probe that is immobilised on the surface of a substrate, it recognises specifically a target in a sample and it binds specifically to and immobilises that target to the surface of the substrate. The phrase “capture probes to two or more different target analytes” means that there are at least two capture probes present, each one specific to a different target analyte. Preferably capture probes specific to one target are present in spatially separate locations on the substrate to capture probes specific to a different target. The capture probe can be selected to recognise specifically an epitope of a target e.g. protein which has no consensus glycosylation sites or an epitope which is sufficiently distant from a consensus glycosylation site to ensure that glycosylation on the target does not affect specific binding by the probe. This ensures efficient binding by the capture probe of its target and can also prevent possible interference during binding of glycan to glycan-binding agent. Preferred capture probes of the current invention are antibodies.The term “antibody” refers to an immunoglobulin which specifically recognises an epitope on a target as determined by the binding characteristics of the immunoglobulin variable domains of the heavy and light chains (VHS and VLS), more specifically the complementarity-determining regions (CDRs). Many potential antibody forms are known in the art, which may include, but are not limited to, a plurality of intact monoclonal antibodies or polyclonal mixtures comprising intact monoclonal antibodies, antibody fragments (for example Fab, Fab’, and Fv fragments, linear antibodies single chain antibodies and multispecific antibodies comprising antibody fragments), single-chain variable fragments (scFvs), multi-specific antibodies, chimeric antibodies, humanised antibodies and fusion proteinscomprising the domains necessary for the recognition of a given epitope on a target. Preferably, references to antibodies in the context of the present invention refer to monoclonal antibodies or ScFvs.Immunoglobulin molecules are known to bear specific glycosylation modifications in the Fc portion which vary across subtype and host species. When used as an immunoassay capture molecule, such glycosylation may interfere with detection by sugar-binding reagents such as lectins by contributing to non-specific background signal. For example, fucosylation, which is a common and prominent form of immunoglobulin glycosylation, may interfere with assays using AAL lectin aimed at detecting fucose-modified biomarkers. Accordingly, strategies have been developed to obviate interference arising from capture antibody glycosylation. Firstly, use of antibody fragments such as Fab fragments, recombinant single chain variable fragments (scFvs) or single chain antibodies not containing glycosylation sites, or with significantly reduced glycosylation may be employed as replacements for glycosylated whole capture molecules. Alternatively, strategies may be employed in the pretreatment of capture molecules aimed at removing glycans including deglycosylase enzymes such as PNGase F, neuraminidase, mannosidase, o-glycosidase or chemical methods such as sodium periodate or phenyl boronate treatment reactive to diols which are a specific feature of many sugar structures but do not occur in proteins. Using such deglycosylation methods, interference from capture antibody glycans can be eliminated. Capture antibodies which are not intrinsically glycosylated are preferred for use in the current methods, but in certain cases when this is unavoidable deglycosylation can be performed.The term “epitope” refers to the portion of a target which is specifically recognised by a given antibody. In instances where the antigen is a protein, the epitope may be formed from either a contiguous or non-contiguous number of amino acids (‘linear’ or ‘conformation’ epitopes respectively), whereby in the case of the latter, residues comprising the epitope are brought together in the three-dimensional fold of the polypeptide. An epitope typically comprises, but is not limited to, 3-10 amino acids in specific positions and orientations with respect to one another. Techniques known in the art for determining the epitope recognised by an antibody (specifically whether or not an epitope comprises a given residue) include but are not limited to, site-directed mutagenesis or the use of suitable homologous proteins to the target biological molecule, e.g. protein, in combination with techniques for determining specific recognition or lack thereof, as exemplified below. By way of example and not limitation, an epitope may be determined as comprising a given residue by comparative analysis with a control comprising specific recognition of the native (non-substituted) target protein by said antibody; wherein diminished binding and / or lack of specific recognition by said antibody when compared with said control identifies a given residue as forming part of an epitope.Furthermore, structural analyses of antibody-target protein complexes via x-ray crystallography and / or nuclear magnetic resonance (NMR) spectroscopy, or suitable derivatives thereof, may also be used to determine the residues which constitute an epitope.The term “binds specifically”, in the context of antibody-epitope interactions, refers to an interaction wherein the antibody and epitope associate more frequently or rapidly, or with greater duration or affinity, or with any combination of the above, than when either antibody or epitope is substituted for an alternative substance, for example an unrelated protein. Generally, but not necessarily, reference to binding means specific recognition. Furthermore, it is appreciated that an antibody may recognise more than one antigen specifically, for example, an antibody that binds specifically to a protein in its unglycosylated form may also bind specifically to it in a glycosylated form. Additionally, an antibody that binds specifically to a protein in its unglycosylated form may be unable to bind specifically to it in a glycosylated form. Techniques known in the art for determining the specific binding of a target by a monoclonal antibody or lack thereof include but are not limited to, FACS analysis, immunocytochemical staining, immunohistochemistry, western blotting / dot blotting, ELISA, affinity chromatography. By way of example and not limitation, specific binding, or lack thereof, may be determined by comparative analysis with a control comprising the use of an antibody which is known in the art to specifically recognise said target and / or a control comprising the absence of, or minimal, specific recognition of said target (for example wherein the control comprises the use of a non-specific antibody). Said comparative analysis may be either qualitative or quantitative. It is understood, however, that an antibody or binding moiety which demonstrates exclusive specific recognition of a given target is said to have higher specificity for said target when compared with an antibody which, for example, specifically recognises both the target and a homologous protein.The term “target analyte” refers to any biological molecule subject to glycosylation. This includes proteins, peptides, carbohydrate antigens or other glycan or glycosaminoglycan- containing molecules or microbial analytes.In preferred embodiments the one or more target analytes of the current invention are selected from Mesothelin, Immunoglobulin A (IgA), Immunoglobulin E (IgE), Immunoglobulin G (IgG), Immunoglobulin M (IgM), Lymphatic vessel endothelial hyaluronan receptor 1 (LYVE-1), Carbohydrate antigen 125 (CA125), Carbohydrate antigen 242 (CA242), Carbohydrate antigen 15-3 (CA15-3), Carcinoembryonic antigen (CEA), Carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM-1), Carbohydrate antigen 72-4 (CA72- 4), Carbohydrate antigen 19-9 (CA19-9), Alpha 1-Acid Glycoprotein (A1AG), Clusterin, Alpha-fetoprotein (AFP), Cytokeratin fraction 21-1 (Cyfra21-1), C3adesArg, Neutrophilgelatinase-associated lipocalin (NGAL), Tissue inhibitor matrix metalloproteinase 1 (TIMP-1), C-reactive protein (CRP), Total prostate-specific antigen (tPSA), free prostate-specific antigen (fPSA), proPSA, nicked PSA, Osteopontin, Tumour necrosis factor alpha (TNFa), Intercellular adhesion molecule 1 (ICAM-1), P-selectin (PSEL), Interleukin 6 (IL6), Interleukin 8 (IL8), beta human chorionic gonadotropin (P-HCG), Prolactin, Zinc alpha-2-glycoprotein (ZAG), prostate-specific membrane antigen (PSMA), Chromogranin A, Leucine-rich alpha-2- glycoprotein (LRG1), Follostatin, Insulin-like growth factor-binding protein 2 (IGFBP-2), Platelet-derived growth factor (PDGF), Fibroblast growth factor 2 (FGF2), Skp1 -Cullin-F box protein (SCF), soluble FAS protein (sFas), Hepatocyte growth factor (HGF), Macrophage migration inhibitory factor (MIF), Human epididymis protein 4 (HE4). Preferred target analytes include CA19-9, CEA, CA15-3, Clusterin, A1AG and LYVE-1.The term “separate physical locations” refers to discrete and non-overlapping test regions on the substrate at which either capture probes or neoglycoproteins may be immobilised. These may be referred to as spatially separate locations or discrete test regions (DTRs). Preferably the capture probes or neoglycoproteins are immobilised at these locations covalently using conventional techniques.The term “neoglycoprotein” as used herein refers to any synthetic or modified glycoprotein comprising at least one glycan covalently attached to a protein carrier. Suitable protein carriers include albumins such as human or bovine serum albumins, Keyhole Limpet Hemocyanin (KLH), Tetanus Toxoid (TT) and Cross-reactive material 197 (CRM197).Neoglycoproteins may be generated by the conjugation of glycosides with the s-amino groups of lysine residues of BSA and have been used for research purposes to identify, characterize and verify lectins or lectin-like molecules, to purify lectins or other carbohydrate- binding proteins, biomarker discovery, and to trigger immune response against carbohydrates moieties. Although many candidate proteins are available, albumins have emerged as the core protein of preference in the production of neoglycoproteins as they are not naturally glycosylated, have many available lysine or arginine residues for conjugation and are therefore multivalent. In the current invention, a glycosylated calibration strategy based on neoglycoproteins bearing relevant monosaccharide or glycan structures was devised to normalise lectin signal output to facilitate interassay precision and cut-off assignment in clinical samples. A diagnostic product resulting from the methods disclosed herein may consist of a base biochip containing spotted capture antibodies, and an internal glycan based standard curve within a single array format (Figure 1). In this format, the array can be offered on automated analyser systems which can facilitate its clinical evaluation and adoption into large scale cancer screening programmes. In a representative preliminaryanalysis of fucosylated biomarkers using the device, discrimination of sera from normal subjects and pancreatic cancer patients was enhanced by inclusion of CEA, CA15-3, clusterin and A1AG with CA19-9 by comparison with CA19-9 alone (Figure 2).Preferably the neoglycoproteins of the current invention comprise an albumin conjugated to at least one glycan residue. Even more preferably the albumin is bovine serum albumin (BSA). In preferred embodiments the neoglycoproteins comprise at least one fucose residue or at least one N-acetylglucosamine residue, even more preferably the neoglycoproteins comprise BSA conjugated to 3-fucosyllactose residues or BSA conjugated to N- acetylglucosamine residues but the neoglycoprotein may comprise any glycan of interest.To generate a suitable calibration curve, it is preferred that there are at least two glycan standards present in different known concentrations. More preferably there are five or more glycan standards, i.e. five, six, seven, eight, nine, ten or more, immobilised at five or more separate physical locations on the solid-state device, each of said glycan standards formulated at different known concentrations.The range of known concentrations of the glycan standards may be determined by the skilled person based on the expected range of an assay. For example, in figure 3A the range of neoglycoprotein standards was 0, 3, 4, 7, 11 , 18, 29 and 47 pg / ml for the determination of n-acetylglucosamine using the WGA lectin for detection. In another example the concentration of BSA-Fucose standards was 0, 0.16, 0.27, 0.47, 0.82, 1.43, 2.48, 4.31 and 7.5 pg / ml for the determination of fucose using AAL as the detector (Figure 1 E). The term “concentrations”, as used in reference to glycan standards (for example, neoglycoproteins) herein, refers to the gravimetric concentration of the glycan standard (for example, neoglycoprotein) within the overall spotting solution used to prepare each of the discrete test regions on a solid-state device. The phrase “Formulated at different known concentrations” can refer to the preparation of these spotting solutions at varying known concentrations of glycan standards. The varying concentrations will generally be an incremental range of increasing amounts starting at zero. It is also conceivable that the concentration of glycan in the standards could be determined by working out the glycan concentration relative to the carrier (for example, carrier protein) within the glycan standard preparation. This could also be expressed as the number of glycan residues per individual carrier molecule (for example a carrier protein such as BSA).The term “glycan” refers to sugar moieties that are found attached to proteins as in glycoproteins and proteoglycans. While typically consisting of a large number of monosaccharides linked glycosidically, “glycans” include the carbohydrate portion of anyglycoconjugate, such as a glycoprotein, glycolipid, or a proteoglycan, even if the carbohydrate is only an oligosaccharide.The term “glycan-binding agent” refers to an entity that binds specifically to a glycan. Glycan binding agents can include, but are not limited to, lectins isolated from plant sources, recombinant lectin molecules identified from microbes or antibodies raised against glycan immunogen targets including monoclonal antibodies, derivative recombinant antibody fragments including antigen binding (Fab) fragments, single chain variable fragments (scFvs), single chain antibodies. Purified or recombinant glycan or glycosaminoglycan binding proteins and peptides may also be employed in glycan-specific detection including R-type, C-type, P-type, C-type, L-type, F-type and l-type lectins, galectins from mammalian, microbial or plant sources. Preferably, the glycan-binding agent is a lectin. The lectin chosen is dependent on the glycan which is to be determined in a particular assay. Preferred lectins include fucose-binding lectins such as Aleuria aurantia lectin (AAL), Ulex europaeus agglutinin (UEA), Lens culinaris lectin (LCA), Lotus tetragonolobus lectin (LTL), Pisum sativum lectin (PSL) and Aspergillus oryzae lectin (AOL) or N-acetylglucosamine lectins such as Wheat germ agglutinin (WGA), Solanum tuberosum lectin (STL), Lycopersicon esculentum lectin (LEL) and Datura stramonium lectin (DSL).The glycan-binding agents as used herein are conjugated to distinct reporter molecules. The term “reporter molecule” refers to a label that is covalently attached to the glycan-binding agent to enable its detection, and such reporters include but are not limited to biotin, radionuclides, fluorophores, dyes or enzymes including, for example horse-radish peroxidase and alkaline phosphatase.The methods of the current invention may be enabled by a substrate comprising a solid-state device onto which is immobilised capture probes to one or more target analytes and in separate physical locations two or more glycan standards formulated at different known concentrations. Even more preferably the substrate comprises five or more glycan standards immobilised at five of more separate physical locations and formulated at five or more different known concentrations. Preferably five or more refers to five, six, seven, eight, nine, ten or more.The solid-state device of the current invention may be any planar surface onto which the capture probes and glycan standards can be immobilised. Such capture probes and glycan standards may be immobilised at discrete areas of an activated surface of the support substrate. The solid-state device may perform multi-analyte assays such that the level of glycosylation on multiple targets in the sample isolated from the patient is determined in parallel. In this context, the support substrate may be one that is used conventionally inmulti-analyte microarray technologies. It may, for example, be a biochip, glass slide or other conventional planar support material, or bead or microparticle. There may be a plurality of Discrete Test Regions (DTRs) present on the solid-state device. These define discrete locations on a substrate and support capture probes or glycan standards. Each DTR is spatially separated from other DTRs, and each may be used for the same or different reactions, depending on how the reactions are to be performed. The DTRs are usually present within a “biochip”, and multiple biochips may be present on the device, each biochip being physically separated from other biochips. In this embodiment, the solid-state device has a multiplicity of DTRs each bearing a desired capture probe covalently bound to the substrate, and in which the surface of the substrate between the DTRs is inert with respect to the target antigen under study. The solid-state, multi-analyte device may therefore exhibit little or no non-specific binding. Different capture probes or neoglycoproteins at varying concentrations may be located in spatially (physically) separate locations i.e. within DTRs on the biochip. In a particular example, the biochip is approximately 1 cm2and there may be 4x4 DTRs present within each biochip, preferably 5x5 DTRs, 6x6 DTRs, 7x7 DTRs, 8x8 DTRs, 9x9 DTRs, 10x10 DTRs, 12x12 DTRs, 15x15 DTRs, 20x20 DTRs, 30x30 DTRs or greater present within each biochip.In one embodiment, each DTR within a solid-state device may contain immobilized thereto a binding capture probe specific for a distinct target analyte or a different concentration of glycan standard. The antigen specificity of each binding molecule immobilized to each DTR upon the solid-state device may be known. The reactions for determining the glycosylation status of different target antigens within the sample isolated from the patient may therefore be carried out at the same time as the determining the signal from the known glycan standards on the same solid-state device. In this embodiment, a glycan-binding agent that recognizes a glycan shared amongst all, or some of the target antigens and the glycan standards may be used. The detection probes (glycan-binding agents) for the target antigens and glycan standards may be labelled with the same detectable label. Alternatively, each detection probes for the target antigens and glycan standards may be labelled with a unique detectable label.A solid-state device of the present invention may be prepared by activating the surface of a suitable substrate and applying an array of capture probes and neoglycoproteins onto discrete sites on the surface. If desired, the other active areas may be blocked. The capture probes, glycan standards and blocking agents may be bound to the substrate via a linker. In particular, it is preferred that the activated surface is activated using an organosilane or polymer coating before reaction with the binding agent. The solid-state device used in the methods of the present invention may be manufactured according to the method disclosedin, for example, GB-A-2324866. Preferably, the solid-state device used in the methods of the present invention is the Biochip Array Technology system (BAT) (available from Randox Laboratories Limited). More preferably, the Evidence, the Evidence Evolution, Evidence Investigator and Multistat apparatus (available from Randox Laboratories) may be used to determine the levels of glycosylation of target analytes in the sample.In one embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to CEA and in separate locations two or more glycan standards formulated at different known concentrations.In one embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to A1AG and in separate locations two or more glycan standards formulated at different known concentrations.In one embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to CA15-3 and in separate locations two or more glycan standards formulated at different known concentrations.In one embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to Clusterin and in separate locations two or more glycan standards formulated at different known concentrations.In one embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to CA19-9 and in separate locations two or more glycan standards formulated at different known concentrations.In one embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to LYVE-1 and in separate locations two or more glycan standards formulated at different known concentrations.In a further embodiment the substrate may comprise a solid-state device onto which is immobilised capture probes to two or more of CEA, A1AG, CA15-3, Clusterin, CA19-9, CEACAM-1 , CA-242, CA-125 and LYVE-1 and in separate locations two or more glycan standards formulated at different known concentrations. One preferred embodiment is a substrate comprising capture probes to CEA, A1AG, CA15-3, Clusterin and CA19-9 and in separate locations two or more glycan standards formulated at different known concentrations. Preferably the substrate comprises five or more glycan standards, immobilised at five or more separate physical locations on the solid-state device, formulated at five or more different known concentrations. Even more preferably the capture probes are antibodies and the glycan standards are neoglycoproteinsA further embodiment of the current invention is a method for aiding in the diagnosis of pancreatic cancer, said method comprising: a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes to CA19-9, CEA, CA15-3, clusterin and A1AG and at separate physical locations two or more glycan standards formulated at different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the protein-specific capture antibodies and the glycans on the glycan standards; c) measuring the level of glycan-binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.Preferably the glycan residues to be detected in this method to aid pancreatic cancer diagnosis are fucose residues. Results from this method (i.e. the extent of fucosylation on each of the target analytes) for the samples analysed may be compared to control samples consisting of non-pancreatic cancer samples and any significant differences therein may aid a clinician’s diagnosis of pancreatic cancer. The extent of fucosylation can mean that more fucose units / residues are present or it can indicate a greater binding affinity or availability of fucose units / residues. The internal calibration method used herein means that control patient samples would not need to be run simultaneously to the samples and reference ranges could be established which would allow a clinician to screen blinded samples as they come into a clinic.In current clinical practice, measurement of the tumour marker carbohydrate antigen 19-9 (CA19-9) also called cancer antigen 19-9 or sialylated Lewis (a) antigen) is used primarily in the management of pancreatic cancer. In people with pancreatic masses, CA 19-9 can be useful in distinguishing between pancreatic cancer and other diseases of the gland [8,9], However, CA 19-9 elevations have been reported in other conditions such as such as colorectal, oesophageal and hepatocellular carcinomas in addition to pancreatitis, cirrhosis, and diseases of the bile ducts [8,11], However, in patients who lack the Lewis antigen (a blood type antigen on erythrocytes), approximately 10% of the Caucasian population, CA 19- 9 is not expressed [9], due to a deficiency of a fucosyltransferase enzyme that is needed to produce CA 19-9. By using a typical cut-off value of 37 U / mL, approximately 25%-35% of patients do not show increases, rendering the test inconclusive for the diagnosis or monitoring of cancer in many patients.On the other hand, further analysis of a subset of the UKCTOCS longitudinal trial focussed on subjects who developed pancreatic cancer during the course of the sampling period

[0010] ,Analysis of serum indicated that samples with CA19-9 above a cut-off of 37 U / mL identified cancer with a sensitivity of 68% up to 1 year, and 53% up to 2 years before diagnosis indicating the utility of measuring CA19-9 as a primary screening measure. In addition, prediagnostic levels of CA19-9 and CA125 were together associated with poor overall survival in pancreatic cancer

[0010] , Therefore, although not ideal, CA19-9 represents an important marker for pancreatic cancer and a good basis on which to build molecular indicators for cancer. The diagnostic performance of CA19-9 assays may be improved by its inclusion with other pancreatic cancer biomarker analyses into a multiplexed assay format allowing generation of a disease prediction model using methods such as logistic regression, Random Forest, or Classification Tree analyses. In this regard, multiple published studies have included addition markers to improve diagnostic performance of CA19-9 alone including IGF-1 and albumin

[0011] , CEA and CA125

[0012] , microRNA signatures and analysis of various sample matrices (reviewed in

[0013] ).CEA describes a set of highly related glycoproteins involved in cell adhesion. CEA, also known as CEACAM5 or CD66e, is normally produced in gastrointestinal tissue during foetal development and stops before birth. Therefore, CEA is usually present only at very low levels in the blood of healthy adults but is raised in some cancer types, hence utility as a tumour marker in clinical tests. Serum from individuals with colorectal carcinoma often has higher levels of CEA than healthy individuals (above a cut-off at approximately 2.5 pg / L)

[0014] , CEA measurement is generally used as a tumour marker to monitor colorectal carcinoma treatment, recurrence after resection, for staging or to localize cancer spread in various biological fluids. CEA levels may also be associated with gastric, pancreatic, lung, breast, and medullary thyroid carcinomas, as well as some non-neoplastic conditions like ulcerative colitis, pancreatitis, cirrhosis, COPD, Crohn's disease, hypothyroidism. The CEA blood test alone is not reliable for diagnosing cancer or as a screening test for early detection of cancer

[0015] however may strengthen cancer prediction when analysed in combination with other biomarkers in a multiplexed context. As a predictor of resection outcome, post-operative CEA levels of <6 pg / Lt is predictive of longer overall survival in patients with locally advanced pancreatic cancer

[0016] , CEA has 28 known glycosylation sites yielding up to 61 different glycoforms

[0017] and as such may be assessed as a glycosylated biomarker target.Orosomucoid (ORM) or alpha-1 -acid glycoprotein (AGP or A1AG) is an abundant acute phase plasma alpha-globulin glycoprotein synthesized primarily in hepatocytes with a normal plasma concentration between 0.6-1.2 mg / mL (1-3% plasma protein). A1AG levels increase in association with obstructive jaundices. In the case of pancreatic cancer, jaundice and pale stools indicative of bile duct obstruction may be indicative of pancreatic tumour growth.Reported heterogeneity in A1AG glycosylation status include an increase in the sialic acid content of cancer A1AG glycans compared to normal controls [18,19], A more recent study on A1AG branching and fucosylation in cancer concluded that glycosylation status can indicate carcinoma progression and prognosis

[0020] , Similar findings can be associated with surgery and other traumas in general. Alpha 1-3 fucosylated glycoforms of A1AG have recently been identified as being increased in pancreatic cancer by lectin assay, mass spectrometry analysis of glycans and capillary zone electrophoresis

[0021] , In our preliminary studies, fucosylated A1AG detected by AAL lectin following anti-A1AG scFv-mediated capture revealed a superior specificity and sensitivity for pancreatic cancer detection by comparison with A1AG total antigen (ROC AUC 0.687 versus 0.507).Extracellular link domain containing 1 , also known as XLKD1 or lymphatic vessel endothelial hyaluronan receptor (LYVE)-1 , is a human gene. This gene encodes a type I integral membrane glycoprotein. The encoded protein acts as a receptor and binds to both soluble and immobilized hyaluronan. This protein may function in lymphatic hyaluronan transport and have a role in tumour metastasis. LYVE-1 is a cell surface receptor on lymphatic endothelial cells that can be used as a lymphatic endothelial cell marker and sort these cells for experimental purposes. So far, the physiological role for this receptor is still the subject of debate, but evolutionary conservation suggests an important role. Recently, LYVE-1 was detected as part of a three-protein biomarker panel that is able to detect patients with early- stage pancreatic cancer in urine specimens

[0022] , LYVE-1 binds hyaluronan (HA), an extracellular matrix mucopolysaccharide, and transports it across the lymphatic vessel wall, particularly in the lymph nodes, a site of HA degradation. Most studies utilise LYVE-1 as a marker of lymphatic endothelium in the context of lymph angiogenesis, however, markedly different findings regarding active lymph angiogenesis across different tumour types exist. EP 2911681 A1 and US2015 / 0246118 A1 disclose the use of LYVE-1 antagonists in the treatment of conditions associated with lymph angiogenesis.Mucin 1 (MUC1) bearing carcinoma antigen 15-3 (CA15-3) was first identified as a breast cancer-associated antigen in breast carcinomas and in human milk in the 1980s and has found clinical utility in the monitoring of metastatic breast cancer (including the response to treatment). CA 15-3 is a tumour marker for many types of cancer, most prominently breast cancer but may also be elevated normal in cancer of the lung, pancreas, ovary and prostate. MUC1 is a polymorphic mucin-like protein that contains a large extracellular domain consisting of a glycosylated polypeptide made up of 30-100 tandem repeats of a 20-amino acid sequence, a transmembrane domain, and a cytoplasmic tail. The CA15-3 antigen is derived from o-linked threonine glycosylation related modifications of a particular repeating PDTRPAP amino acid sequence motif within the MUC1 protein. Distinct, disease-specificglycan modifications can be detected on MUC1 relating to pancreatic cancer independently of core protein elevations

[0023] ,Clusterin, also known as apolipoprotein J is a heterodimeric protein involved in the clearance of cellular debris, lipids in particular, through its chaperone function. In a cancer context, clusterin can confer resistance to TNF-alpha-induced apoptosis for example in breast cancer cells

[0024] as well as promoting resistance to gemcitabine therapy in the MIA-PaCa2 pancreatic cancer cell line

[0025] , Fucosylation of clusterin as with many acute phase proteins is common. Fucosylated clusterin glycoforms are produced by luminal breast cancer cells and play a role in tumour progression by promoting the release of pro-angiogenic factors by intratumoral macrophages

[0026] in addition to prosurvival actions.Detection of cancer biomarker glycoforms may be performed using a modified immunoassay format in which proteins may be captured by immobilised target-specific antibodies and glycosylation status detected using glycan binders such as lectins from sources such as plants or microbes or glycan-specific antibodies. Multiplexed immunoassays for glycosylated markers based on glycan detection reagents are novel in the clinical diagnostics arena and offer a number of advantages over current methodologies including (a) the ability to acquire multiple lines of evidence supporting a diagnosis from a plurality of biomarker readouts for a sample; (b) the use of a single glycan detection reagent for detection of a particular glycan modification that may be common to multiple circulating biomarkers and (c) the potential for a glycosylation-based assay readout to provide information that is additive to the conventional total antigen measurement in disease state detection reflecting changes in biomarker glycosylation status. An assay format may be devised comprising capture of a multiplicity of glycoproteins on a solid phase followed by glycan detection using lectins with defined specificity for separate biomarker panels which may be analysed cumulatively. In many previously described enzyme-linked lectin assay (ELLA) formats, standardization has been achieved through comparison of signal output from the assay, e.g. Absorbance or fluorescence intensity from control samples, or used an external standard for assay calibration i.e. a glycosylated protein standard prepared from cell culture.The following definitions apply to terms used throughout this description and in relation to any of the aspects of the invention described herein.The terms “patient” and “subject” are used interchangeably herein and refer to any animal (e.g. mammal), including, but not limited to, humans, non-human primates, canines, felines, rodents and the like, which is to be the recipient of the diagnosis. Preferably, the subject or patient is a human.The “level” of a glycan refers to the amount or concentration of the glycan within the sample. This can also be referred to as the level of glycosylation in reference to a specific target analyte. The glycan level may be expressed as a ratio or percentage of the level of glycosylation of one or more other analytes or of the same analyte in a different sample or a control sample. The ratio of total protein to glycosylated protein may be measured for a target analyte. The level of glycosylation may differ between controls and a clinical condition of interest. A preferred clinical condition of interest is cancer.The term “cancer” refers to or describes the physiological condition in mammals in which a population of cells are characterised by unregulated cell growth.The terms “cancer cell” and “tumour cell” are grammatical equivalents referring to the total population of cells derived from a tumour or a pre-cancerous lesion. The terms “tumour” and “neoplasm” are used interchangeably herein and refer to any mass of tissue that results from excessive cell growth, proliferation and / or survival, either benign (noncancerous) or malignant (cancerous), including pre-cancerous lesions.The term “metastasis” and related terms “metastases” and “metastatic” refers to a cancer or tumour that has spread from the original site in which it developed and has invaded and / or is growing in other tissues, forming new tumour growths and obstructions.The terms “immunoassay”, “immuno-detection” and immunological assay” are used interchangeably herein and refer to antibody-based techniques for identifying the presence of or levels of a target molecule (e.g. protein) in a sample. Examples of such assays and methods are well known to those of skill in the art.The term “glycosylation profile” or “glycosylation signature” of a target or a defined group of targets refers to the type of glycans attached to one or more targets; the profile can be expressed qualitatively in terms of which glycan-binding agents it binds or quantitatively in terms of the amount of a particular glycan-binding agent it binds, for example as a concentration or a ratio. For example, the glycosylated protein level may be expressed as a ratio of total protein level to glycosylated protein level. The terms “profile” and “signature” are used interchangeably.Glycosylation present in a sample isolated from a patient having a disease may have levels which are different to that of a control. A suitable control group to use can be determined by the skilled person. For example, suitable controls used herein may include healthy control groups (i.e. persons not knowingly suffering from an illness or condition which could affect the results, this may also be referred to as a normal control), non-cancer control groups (persons not suffering from any form of cancer) or could be persons suffering from a benigncondition or unrelated cancer to the cancer to be determined. However, the levels of glycosylation on some proteins that are different compared to a control may not show a strong enough correlation with disease such that they may be used to diagnose disease with an acceptable accuracy. Accuracy of a diagnostic method is often described by its receiveroperating characteristics (ROC)

[0027] , The ROC graph is a plot of all the sensitivity / specificity pairs resulting from continuously varying the decision threshold over the entire range of data observed.A ROC plot depicts the overlap between the two distributions by plotting the sensitivity versus 1 - specificity for the complete range of decision thresholds. On the y-axis is sensitivity, or the true-positive fraction defined as [(number of true-positive test results) / (number of true-positive + number of false-negative test results)]. This has also been referred to as positivity in the presence of a disease or condition. It is calculated solely from the affected subgroup. On the x-axis is the false-positive fraction, or 1 - specificity [defined as (number of false-positive results) / (number of true-negative + number of false-positive results)]. It is an index of specificity and is calculated entirely from the unaffected subgroup. Because the true- and false-positive fractions are calculated entirely separately, by using the test results from two different subgroups, the ROC plot is independent of the prevalence of disease in the sample. Each point on the ROC plot represents a sensitivity / specificity pair corresponding to a particular decision threshold. A test with perfect discrimination (no overlap in the two distributions of results) has a ROC plot that passes through the upper left corner, where the true-positive fraction is 1.0, or 100% (perfect sensitivity), and the falsepositive fraction is 0 (perfect specificity). The theoretical plot for a test with no discrimination (identical distributions of results for the two groups) is a 45° diagonal line from the lower left corner to the upper right corner. Most plots fall in between these two extremes. Qualitatively, the closer the plot is to the upper left corner, the higher the overall accuracy of the test.One convenient goal to quantify the diagnostic accuracy of a laboratory test is to express its performance by a single number. The most common global measure is the area under the curve (AUC) of the ROC plot. The area under the ROC curve is a measure of the probability that the perceived measurement will allow correct identification of a condition. By convention, this area is always > 0.5. Values range between 1.0 (perfect separation of the test values of the two groups) and 0.5 (no apparent distributional difference between the two groups of test values). The area does not depend only on a particular portion of the plot such as the point closest to the diagonal or the sensitivity at 90% specificity, but on the entire plot. This is a quantitative, descriptive expression of how close the ROC plot is to the perfect one (area = 1.0). In the context of the present invention, the two different conditions are whether a patient has or does not have cancer or alternatively whether a patient who has cancer hasonly a primary tumour or has metastatic disease. The ROC plot data and the clinical requirements of the test may be considered together when calculating a threshold or “cut-off’ value to be used in future application of the diagnostic test. When the analyte value is measured above (or below) this cut-off value, the test is considered “positive” and further action may be taken appropriate to the clinical condition. An important feature in setting the cut-off value is the required specificity of the test (i.e. the true positive rate). By convention, the required specificity for many diagnostic tests is stated in advance to be 90%, 95%, or as close to 100% as practical. For cancer biomarker tests, it is likely that an effective test will need to approach 100% due to the relatively low prevalence of cancer in the general population (or even targeted populations) compared to those with inflammatory diseases and because of the severe consequences of false negative results for patients. The analyte cutoff value required to achieve these specificities may then be read from the ROC plot. This point on the plot will also denote a value for test sensitivity (true negative rate). Alternatively, the optimum cut-off value may be obtained by selecting the point on the ROC curve closest to the top-left corner of the graph. In one embodiment of the present invention the AUC measurements could be improved by calculating ratios based on total protein levels of a biomarker protein and glycosylated protein levels of the same biomarker protein.The determination of the level of glycosylation of a target analyte in the sample may be determined by immunological methods such as an ELLA or ELISA-based assay. Preferably, the methods of the present invention use a solid-state device for determining the level of a glycosylation in the sample isolated from the patient. The solid-state device comprises a substrate on which is immobilised two or more neoglycoproteins and a probe that binds specifically to a target analyte. Such probes may be immobilised at discrete areas of an activated surface of the substrate. The solid-state device may perform multi-analyte assays such that the level of a glycosylation of one target analyte in a sample isolated from the patient may be determined in parallel with the level of glycosylation of a further biomarker of interest in the sample. The solid-state device may have a plurality of discrete test regions present. Each discrete test region is spatially separated from other discrete test regions, and each may be used for the same or different reactions, depending on how the reactions are to be performed. The DTRs are usually present within a “biochip”, and multiple biochips may be present on the device, each biochip being physically separated from other biochips. In this embodiment, the solid-state device has a multiplicity of discrete test regions each bearing a desired antibody covalently bound to the substrate, and in which the surface of the substrate between the reaction sites is inert with respect to the target biomarker. The solid- state, multi-analyte device may therefore exhibit little or no non-specific binding. The biochip is preferably an essentially planar substrate onto which a biological molecule, e.g. antibodyor neoglycoprotein may be immobilised. Different biological molecules may be located in spatially separate locations i.e. within DTRs on the biochip. In a particular example the biochip is approximately 1 cm2and there may be 4x4 DTRs present within each biochip, preferably 7x7 DTRs, 8x8 DTRs, 9x9 DTRs, 10x10 DTRs, 12x12 DTRs, 15x15 DTRs or 20x20 DTRs present within each biochip.A device that may be used in the invention may be prepared by activating the surface of a suitable substrate and applying an array of probes on to discrete sites on the surface. If desired, the other active areas may be blocked. The ligands may be bound to the substrate via a linker. In particular, it is preferred that the activated surface is reacted successively with an organosilane, a bifunctional linker and the antibody. The solid-state device used in the methods of the present invention may be manufactured according to the method disclosed in, for example, GB-A-2324866 the contents of which is incorporated herein in its entirety. Preferably, the solid-state device used in the methods of the present invention is the Biochip Array Technology system (BAT) (available from Randox Laboratories). More preferably, the Evidence Evolution, Evidence Investigator and Multistat apparatus (available from Randox Laboratories) may be used to determine the levels of biomarkers in the sample.MethodsExample 1 - Preparation and use of a multiplexed array of capture reagents specific for glycosylated biomarkers alongside glycan standards.For glycan standards, a stock solution was prepared at 4x the indicated final working concentration of neoglycoprotein in phosphate buffered saline. Working strength spotting solutions were prepared by 1 :4 dilution into a high pH spotting buffer and arrayed according to internal procedures. Multiplexed analyses of glycosylated antigens within samples were performed using the Randox Biochip Array platform featuring the Evidence Investigator™ analyser. Briefly, samples were diluted 1 : 10 in assay diluent and 100 pl volume of each were placed in a biochip well with 200 pl assay diluent. After incubation at 25°C for 30 mins with rotation at 370 RPM, the biochips were washed using a TBST-based wash buffer. A detection solution dilution containing HRP-conjugated lectins was incubated in each well at 25°C for 30 mins with rotation at 370 RPM followed by washing. In the case of biotinylated lectins, a subsequent incubation of Streptavidin-HRP was employed. Glycosylated biomarkers were detected using chemiluminescence and quantified using the Evidenceinvestigator analyser. Biomarker-associated glycosylation signals are quantified through use of arbitrary units which in the current embodiment derive from the final spotting concentration of the neoglycoprotein within the spotting solution. Relative light units were reported by the analyser were plotted against input neoglycoprotein concentrations and a four-parameter logistic curve fit model used to calculate concentrations of samples or controls. Standardisation of product batches for the glycan standards may be achieved using master lots of glycan standards (the process of lyophilisation may be employed) and / or master biochip lots spotted with glycan standards whereas standardisation of the capture probes may be achieved using master lots of control material containing glycosylated or nonglycosylated variants of the target analytes. Control material may be formulated in a buffer or serum base and may contain recombinant and / or native protein preparations. The material or combinations of the material described above can be used to ensure commutability and uniformity across product batches.The 5-marker combination of A1AG, CEA, CA19-9, CA15-3 and clusterin with fucose detection using AAL showed good discriminatory ability for pancreatic cancer vs normal controls (AUC 0.969, Figure 2C, this combination had enhanced sensitivity over CA19-9 on its own, 93.75% compared to 75%, and the specificity remained 100%) and also for pancreatic cancer vs other cancers (AUC 0.886, other cancers referring to a group consisting of colorectal, lung, ovarian and liver cancer patients).The combination of CA19-9 and LYVE-1 with N-acetylglucosamine detection using WGA showed good discriminatory ability for pancreatic cancer vs normal controls (AUC 0.899, Figure 3).

Claims

Claims1) A method for the standardisation of a glycosylation array, said method comprising: a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes to two or more different target analytes and at separate physical locations glycan standards formulated at two or more different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the capture probes and on the glycan standards; c) measuring the level of glycan-binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.2) The method of claim 1 wherein the glycan standards are neoglycoproteins.3) The method of claim 2 wherein the neoglycoproteins comprise an albumin, Keyhole Limpet Hemocyanin (KLH), Tetanus Toxoid (TT) or Cross-reactivity-material 197 (CRM197) conjugated to at least one glycan residue.4) The method of claim 1 wherein the glycan standards comprise at least one fucose residue.5) The method of claim 1 wherein the glycan standards comprise at least one N- acetylglucosamine residue.6) The method of claim 3 wherein the albumin is bovine serum albumin (BSA).7) The method of claim 6 wherein the neoglycoproteins comprise BSA conjugated to 3- fucosyllactose residues.8) The method of claim 6 wherein the neoglycoproteins comprise BSA conjugated to N- acetylglucosamine residues.9) The method of any previous claim wherein there are five or more glycan standards, immobilised in five or more separate physical locations on the solid-state device, each of said glycan standards formulated at a different known concentration.10) The method of claim 1 wherein the glycan-binding agent is a lectin.11) The method of claim 10 wherein the lectin is a fucose-binding lectin.12) The method of claim 11 wherein the fucose-binding lectin is Aleuria Aurantia Lectin (AAL)13) The method of claim 10 wherein the lectin is a N-acetylglucosamine binding lectin.14) The method of claim 13 wherein the N-acetylglucosamine-binding lectin is Wheat germ agglutinin (WGA).15) The method according to claim 1 wherein the glycan-binding agent is conjugated to a distinct reporter molecule.16) The method of any previous claim wherein the immobilised capture probes are antibodies.17) The method of claim 16 wherein the capture antibodies are deglycosylated.18) The method of claim 16 wherein the capture antibodies are recombinant antibodies, preferably recombinant antibody fragments including antigen binding (Fab) fragments, single chain variable fragments (scFvs) or single chain antibodies.19) The method of any previous claim wherein the capture probes lack, or have minimal, glycosylation.20) The method of any previous claim wherein the solid-state device is a biochip.21) The method of claim 1 wherein the two or more target analytes are selected from Mesothelin, Immunoglobulin A (IgA), Immunoglobulin E (IgE), Immunoglobulin G (IgG), Immunoglobulin M (IgM), Lymphatic vessel endothelial hyaluronan receptor 1 (LYVE-1), Carbohydrate antigen 125 (CA125), Carbohydrate antigen 242 (CA242), Carbohydrate antigen 15-3 (CA15-3), Carcinoembryonic antigen (CEA), Carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM-1), Carbohydrate antigen 72-4 (CA72-4), Carbohydrate antigen 19-9 (CA19-9), Alpha 1-Acid Glycoprotein (A1AG), Clusterin, Alpha-fetoprotein (AFP), Cytokeratin fraction 21-1 (Cyfra21-1), C3adesArg, Neutrophil gelatinase-associated lipocalin (NGAL), Tissue inhibitor matrix metalloproteinase 1 (TIMP-1), C-reactive protein (CRP), Total prostate-specific antigen (tPSA), free prostate-specific antigen (fPSA), proPSA, nicked PSA, Osteopontin, Tumour necrosis factor alpha (TNFa), Intercellular adhesion molecule 1 (ICAM-1), P-selectin (PSEL), Interleukin 6 (IL6), Interleukin 8 (IL8), beta human chorionic gonadotropin (P-HCG), Prolactin, Zinc alpha-2-glycoprotein (ZAG), prostatespecific membrane antigen (PSMA), Chromogranin A, Leucine-rich alpha-2-glycoprotein (LRG1), Follostatin, Insulin-like growth factor-binding protein 2 (IGFBP-2), Platelet-derivedgrowth factor (PDGF), Fibroblast growth factor 2 (FGF2), Skp1 -Cullin-F box protein (SCF), soluble FAS protein (sFas), Hepatocyte growth factor (HGF), Macrophage migration inhibitory factor (MIF), Human epididymis protein 4 (HE4).22) A method for aiding in the diagnosis of pancreatic cancer, said method comprising: a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes to two or more of CA19-9, CEA, CA15-3, clusterin, LYVE-1 , CA-242, CA-125 and A1AG and at separate physical locations two or more glycan standards formulated at two or more different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the protein-specific capture antibodies and the glycans on the glycan standards; c) measuring the level of glycan-binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.23) A method for aiding in the diagnosis of pancreatic cancer, said method comprising: a) contacting the sample to be analysed with a solid-state device onto which is immobilised capture probes to CA19-9, CEA, CA15-3, clusterin and A1AG and at separate physical locations glycan standards formulated at two or more different known concentrations; b) contacting the array with a glycan-binding agent having affinity for both the glycan residues on the target analytes bound by the protein-specific capture antibodies and the glycans on the glycan standards; c) measuring the level of glycan-binding agent bound to the target analyte(s) and comparing to the levels of glycan-binding agent bound to the glycan standards.24) The method of claim 22 wherein the glycan residues to be detected are fucose residues.25) A substrate comprising a solid-state device onto which is immobilised capture probes to one or more target analytes and in separate physical locations glycan standards formulated at two or more different known concentrations.26) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to CEA and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.27) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to A1AG and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.28) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to CA19-9 and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.29) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to CA15-3 and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.30) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to Clusterin and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.31) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to LYVE-1 and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.32) The substrate of claim 25 comprising a solid-state device onto which is immobilised capture probes to two or more of CEA, A1AG, CA15-3, Clusterin, LYVE-1 , CA-242, CA-125 and CA19-9 and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.33) The substrate of claim 32 comprising a solid-state device onto which is immobilised capture probes to CA19-9, CEA, CA15-3, clusterin and A1AG and in separate physical locations two or more glycan standards formulated at two or more different known concentrations.34) The substrate of any of claims 25-33 comprising five or more glycan standards immobilised in separate physical locations on the solid-state device, comprising glycan residues formulated at five or more different known concentrations.35) The use of internal glycan standards to calibrate a multiplex glycosylation array.

Images