WO2015020523A1

WO2015020523A1 - Biomarkers for early diagnosis of alzheimer's disease

Info

Publication number: WO2015020523A1
Application number: PCT/NL2014/050550
Authority: WO
Inventors: Davide CHIASSERINI; Charlotte Elisabeth Teunissen; Cornelia Ramona Jimenez; Thorsten MÜLLER; Katrin Marcus; Helmut Erich MEYER; Jens Wiltfang
Original assignee: Ruhr Universitaet Bochum; Universitaet Duisburg Essen; Stichting VU VUMC
Current assignee: Ruhr Universitaet Bochum; Universitaet Duisburg Essen; Vrije Universiteit Amsterdam
Priority date: 2013-08-07
Filing date: 2014-08-05
Publication date: 2015-02-12
Anticipated expiration: 2016-02-07

Abstract

The present invention relates to methods and kits for classifying an individual as being at risk of developing Alzheimer's disease, in particular Alzheimer's disease type-dementia.

Description

Title: Biomarkers for early diagnosis of Alzheimer's disease

FIELD OF THE INVENTION

BACKGROUND OF THE INVENTION

Alzheimer's disease (AD) is the most common cause of dementia in the elderly, with a prevalence rate doubling every five years after 65 years of age (Ziegler- Graham et al., 2008). It is now becoming clear that the neurodegenerative process in sporadic AD begins 10 to 20 years before the clinical onset of the disease (Jack et al., 2013). During this preclinical phase there is a gradual loss of synapses and neurons, leading, later on, to the onset of cognitive symptoms. This condition, which does not yet fulfil chnical criteria for dementia, has been defined as mild cognitive impairment (MCI) (Petersen, 2004).

Mild cognitive impairment does not always progress to AD. It can remain stable over time or even remit. However, in a number of cases, MCI leads to dementia. This dementia may be AD-type dementia, but it may also be vascular dementia, frontotemporal dementia, semantic dementia and dementia with Lewy bodies.

There exists a need in the art to determine which individuals suffering from MCI will develop AD and particular which will develop AD-type dementia.

SUMMARY OF THE INVENTION

One aspect of the disclosure provides methods of determining the risk of developing Alzheimer's disease or speed of onset of Alzheimer's disease in an individual with mild cognitive impairment, said method comprising

determining the concentration of at least three biomarkers in the

cerebrospinal fluid (CSF) of said individual, wherein said biomarkers are each different biomarkers selected from table 2, preferably wherein a first biomarker is NELL2, a second biomarker is CNTNAP5, and a third biomarker is CLSTN1; and determining said risk or speed of onset based on the concentration of at least two of said biomarkers. Preferably, at least one of the at least three biomarkers are selected from NELL2, CLSTN1, MSN,

CNTNAP5, and FCN3. Preferably, wherein at least two of the at least three biomarkers are selected from NELL2, CLSTN1, MSN, CNTNAP5, and FCN3.

Another aspect of the disclosure provides a method of determining the risk of developing Alzheimer's disease or speed of onset of Alzheimer's disease in an individual with mild cognitive impairment, said method comprising

determining the concentration of at least one biomarker selected from table 2 in the cerebrospinal fluid (CSF) of said individual, and determining said risk or speed of onset based on the concentration of said biomarker.

In preferred embodiments of the disclosed methods, the Alzheimer's disease is characterized by the presence of Alzheimer's disease type dementia. In preferred embodiments of the disclosed methods, The method of any of the proceeding claims, wherein the Alzheimer's disease is sporadic Alzheimer's disease.

In preferred embodiments of the disclosed methods, said method further comprises determining the concentration of at least one of beta amyloid peptide 1-42, total tau, and phosphorylated tau. In preferred embodiments of the disclosed methods, an increase in total tau, an increase in phosphorylated tau, or a decrease in beta amyloid peptide 1-42 as compared to a reference value, in combination with an alteration in the concentration of a biomarker, preferably at least two of said biomarkers, in said individual as compared to a reference value indicates that the individual is at risk of developing

Alzheimer's disease. In preferred embodiments of the disclosed methods, the level of said

biomarkers is determined by an immunoassay. In preferred embodiments of the disclosed methods, the level of said biomarkers is determined by mass spectrometry.

In another aspect of the disclosure, a kit for determining the concentration of at least three biomarkers in cerebrospinal fluid is provided, said kit comprising three binding agents, preferably wherein the binding agents are antibodies, each binding agent directed to a different biomarker selected from table 2, preferably wherein a first biomarker is NELL2, a second biomarker is

CNTNAP5, and a third biomarker is CLSTN1. Preferably, said binding agents are immobilized on a substrate surface. Said kits are especially useful for carrying out the disclosed methods.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1

The experimental scheme used in this work is depicted in figure 1. Two independent cohorts of well-characterized patients diagnosed with mild cognitive impairment (MCI) were enrolled in two European centres,

Amsterdam (AMS) and Bochum (BOCH). The AMS cohort was used as first discovery cohort, results were compared to samples from MCI patients analysed in BOCH. Sample processing, including depletion of high abundant proteins, SDS-PAGE and high sensitivity mass spectrometry was performed separately in the two centres. Data processing, including expression profiles clustering and network analysis was performed to select candidates to validate on an independent cohort. Overlap of the candidates, functional annotations and network analysis were used to compare AMS and BOCH datasets in order to prioritize candidates for immune-based validation in an independent cohort. Figure 2

a) Expression profiles for proteins showing up-regulation in MCI-AD in AMS cohort (UP-clusters), on the right the most significant biological processes found using DAVID functional clustering tools are reported, b) Expression profiles for proteins showing down-regulation in MCI-AD patients with respect to MCI-S group in AMS cohort (DOWN-clusters, c) expression profiles of proteins showing no change for the comparison MCI-S vs MCI-AD.

Figure 3

a) Spearman correlation between spectral counts for each single protein across the AMS and BOCH datasets; b) Venn diagram of identified proteins between AMS and BOCH considering all the patients from the two cohorts; c) comparison of total spectral count per sample in the two cohorts, d), e) functional annotation analysis of proteins showing up or down-regulation in BOCH cohort.

Figure 4

a) Network of proteins showing co-up-regulation in AMS and BOCH datasets and their first interactors. The total network was clustered using the fold change for the comparison MCI-S vs MCI-AD in AMS and BOCH and then further clustered using the interaction strength. Eleven sub-networks with at least three nodes were identified. In the lower part of the panel the most significant biological processes for each subnetwork are reported; b) Network of proteins showing parallel down-regulation in AMS and BOCH datasets and their first interactors. Four sub-networks with at least three nodes were identified. In the lower part of the panel the most significant biological processes for each subnetwork are reported. Figure 5

a) and b) Box plot of NELL2 (a) and CLSTN1 (b) validation results using western blot in the validation cohort. For the sake of comparison the proteomic profiles of the cluster in which the two proteins were included is reported above each box plot. * p<0.05, ** p< 0.01, c) Bar graph of Spearman correlation coefficient for NELL2 against the other biomarkers and clinical parameters in each group, d) Bar graph of Spearman correlation coefficient for CLSNT1 against the other biomarkers and clinical parameters in each group. A642 = amyloid beta 42 peptide, t-tau = total tau, p-tau = phosphorylated tau, MMSE- B = mini mental state examination at baseline, MMSE-F = mini mental state examination at follow-up. * = p<0.05, ** = pO.01.

Figure 6

a) Pooled samples of CSF (10 μΐ,, CSF pool 1 and 2) were loaded on a 4-12% polyacrylamide pre-cast gels (NuPAGE). Proteins were transferred and blotted using NELL2 6847 and 6948 antibodies (1:500). Antibodies in some

experiments were pre-incubated with the peptide mix used for immunization (2 times the concentration of the antibody). Both antibodies showed staining of two bands around the predicted molecular weight. After incubation with immunogen peptide mix the upper band disappeared, evidencing the specificity of the signal for this band, b) Linearity of the antibodies. Whole length protein purchased from Abeam (ab 116828, 818 aa, 118 kDa with a N-terminal tag) was loaded on 4-12% polyacrylamide pre-cast gels (NuPAGE) in different amounts (0.5, 0.2, 0.1 μg). Antibody 6487 showed a better linearity with respect to antibody 6488 (r² = 0.90 and r² = 0.53 respectively). For this reason the former was chosen for western blot experiments on the validation cohort. Figure 7

Gene ontology (GO) analysis of all identified proteins in Amsterdam (AMS) and Bochum (BOCH) cohorts. BP = biological process, MF = molecular function, CC = cellular component.

Figure 8

Boxplot of five candidates overlapping between AMS and BOCH cohorts. Data are reported using medians and 25-75% percentiles. DETAILED DESCRIPTION OF THE DISCLOSED EMBODIMENTS

The hope for a disease-modifying treatment for AD has increased attention to biomarker discovery in early phases of AD, i.e., in the so-called "pre-dementia phase of AD" or "MCI-AD" (Dubois et al., 2010; Albert et al., 2011).

Cerebrospinal fluid (CSF) represents a suitable matrix for biomarker discovery in neurodegenerative diseases, reflecting pathological changes occurring in the central nervous system (CNS) structures in real time (Roche et al., 2008).

To date, three CSF biomarkers are reported to predict progression to AD. Several studies have shown that CSF levels of beta amyloid peptide 1-42 (A6l- 42), total tau (t-tau) and phosphorylated tau (p-tau) are able to detect AD not only in the overt phase of the disease (Hulstaert et al., 1999), but also in the pre-dementia phase (Mattsson et al., 2009). However the performance of these three biomarkers for early diagnosis is still not optimal and current

immunoassay methodologies lack reproducibility among different labs

(Mattsson et al., 2011).

The CSF proteome has already been investigated in preclinical and full-blown AD with different proteomic approaches, mainly involving low-throughput, 2D gel-based techniques (Maarouf et al., 2009; Roher et al., 2009; Craig-Schapiro et al., 2011; Perrin et al., 2011). More recently, nanoLC-MS/MS-based proteomics has emerged as the method of choice for in-depth proteomic discovery on biological fluids. First reports used pooled analysis and Isotope Coded Affinity Tags (ICAT) technology on unfractionated CSF to investigate proteome changes in AD patients, yielding to the identification of around 400 proteins and some putative candidates (Zhang et al., 2005; Abdi et al., 2006).

More recently, a high throughput approach was applied to biomarker discovery in the familial form of AD (FAD). Ringman and colleagues reported 56 differentially present proteins in CSF between patients carrying mutations in PSEN1 and APP and age-matched non-carriers. Interestingly, several low abundance proteins were identified, including synaptic and brain-derived proteins (Ringman et al., 2012).

To date, investigations of CSF proteome in sporadic AD and MCI patients using in-depth proteomic approaches are still missing. Importantly, the study in the present disclosure provides biomarkers which can determine the risk that an MCI patient develops full-blown AD, as well as the speed at which the progression to full-blown AD occurs. These biomarkers offer the potential to reduce the inter-lab and inter-assay variabilities observed with the use of beta amyloid peptide 1-42 (A61-42), total tau (t-tau) and phosphorylated tau (p-tau). These biomarkers also offer the potential to determine the risk that an MCI patient develops full-blown AD with greater confidence levels that with the use of beta amyloid peptide 1-42 (A61-42), total tau (t-tau) and phosphorylated tau (p-tau).

Accordingly, one aspect of the disclosure provides for a method of determining the risk of developing Alzheimer's disease in an individual with mild cognitive impairment and a method for determining the speed at which an individual with mild cognitive impairment will develop Alzheimer's disease, said methods comprising determining the concentration of at least one biomarker selected from table 2 in the cerebrospinal fluid (CSF) of said individual. The risk of developing Alzheimer's disease and the speed at which AD develops are based on the concentration of biomarker. It is clear to a skilled person that while all humans may develop AD, the method is directed at determining the risk of developing AD, i.e., whether there is an increased probability of an individual developing AD, in particular whether there is an increased probability of developing AD sooner, as compared to a population of individuals, e.g., a population of individuals of similar age. In particular, the method is directed at determining whether there is an increased probability of an individual afflicted with MCI developing AD as compared to the average of a population of individuals afflicted with MCI.

Preferably, the method determines whether said individual is at risk of developing Alzheimer's disease within 2 years, however, a skilled person will realize that in some individuals it may take as long as 3, 4, 5, or more years for a person classified as at risk in the present methods to develop clinical symptoms of AD or AD type dementia. Preferably, said methods determine the (increased) risk that an individual with mild cognitive impairment will develop Alzheimer's disease within 10 years, more preferably within 5 years.

The methods provide information regarding the speed at which AD will develop. Or rather, the methods determine whether an individual with mild cognitive impairment is at risk of developing Alzheimer's disease sooner than expected. AD occurs in approximately 4% of people under age 65 (although these patients likely suffer from the familial form of AD and not the sporadic form), approximately 6% in ages 65-74, approximately 44% in ages 75-84, and approximately 46% in people above age 85 (Hebert LE, et al.Alzheimer disease in the U.S. population: Prevalence estimates

using the 2000 Census. Archives of Neurology 2003;60(8): 1119-22). As used herein, the terms individual, subject, or patient are used

interchangeably and generally refer to a human. Preferably the human is at least 40 years old, more preferably at least 50 years old.

As used herein, an individual with mild cognitive impairment refers to an individual afflicted with MCI as defined in Petersen (2004). Briefly, the symptoms for inclusion include memory complaint, objective memory impairment, preservation of general cognitive function, and lack of dementia. Patients with a known cause of impairment (alcohol abuse, brain tumor, CNS infection, etc.) were excluded. Preferably, an individual afflicted with MCI exhibits subjective memory complaint, objective memory impairment

(preferably 1.5 SD below normal), normal general cognitive function, normal activities of daily living, and no dementia.

The use of the term Alzheimer's disease refers to the clinical or symptomatic manifestation of the disease, such as diagnoses according to DSM-IV criteria. As used herein, it does not refer to asymptomatic individuals in which the underlying neurodegenerative processes have already begun. Preferably, the methods determine the risk of developing AD-type dementia. Dementia can be diagnosed using, eg., the DSM-IV criteria. Preferably, said biomarkers are useful for predicting the risk of AD type dementia versus other forms of dementia, such as Lewy body dementia (LBD) and frontotemporal dementia (FTD).

Preferably, the methods described herein allow one to determine if an individual is at risk of developing the sporadic form of AD. The familial forms of AD are caused by specific gene mutations. Changes in biomarker

concentration from pre-clinical familial form AD are likely to be quite different from biomarker concentrations in patients that are at risk of developing the sporadic form of AD.

In the methods disclosed herein, the concentration of at least one biomarker selected from table 2 is determined. Table 2 depicts proteins which were found to have differential expression between MCI patients that remained stable after 2 years (MCI-S) and MCI patients that developed AD-type dementia after 2 years (MCI- AD) (see examples herein). An alteration in the concentration of said biomarker (i.e., differential expression) indicates that the patient is at risk of developing AD, in particular that said individual will develop AD sooner that predicted (based on age).

In preferred embodiments, more that one biomarker from table 2 is

determined. The use of multiple biomarkers increases the confidence of a determination of increased risk or speed of developing AD. Therefore, when multiple markers are used, not all of the biomarkers need to exhibit a significant differential expression in order for a risk to be determined. An alteration in the concentration of at least two of the biomarkers (i.e.,

differential expression) indicates that the patient is at risk of developing AD. Preferably an alteration in the concentration of all 3 biomarkers indicates that the patient is at risk of developing AD (i.e., differential expression) indicates that the patient is at risk of developing AD, in particular that said individual will develop AD sooner that predicted (based on age). Preferably, the concentration of at least four biomarkers selected from table 2 is determined and, preferably, the alteration in the concentration of at least two, more preferably three, of the biomarkers indicates that the patient is at risk of developing AD. Preferably an alteration in the concentration of all 4 biomarkers indicates that the patient is at risk of developing AD. Preferably, the concentration of at least 5 biomarkers selected from table 2 is determined and, preferably, the alteration in the concentration of at least 3, more preferably 4, of the biomarkers indicates that the patient is at risk of developing AD. Preferably, the concentration of at least 6 biomarkers selected from table 2 is determined and, preferably, the alteration in the concentration of at least 4, more preferably 5 of the biomarkers indicates that the patient is at risk of developing AD.

The methods disclosed herein comprise determining the concentration of biomarkers. Differential expression of said biomarkers reflects the potential risk of developing AD. For some biomarkers, an increase in expression represents a risk (e.g., CLSTN1), while for others a decrease in expression represents a risk (e.g., FCN3). Preferably, the concentration of biomarkers is compared to a reference value. The difference in concentration which reflects a significant change and the direction of change (increased or decreased expression) can be determined by comparing, e.g., the levels of biomarker in MCI-AD patients versus MCI-S patients (see, e.g., Figure 8).

Preferably, the reference value may be the average concentration of said biomarker in a single, or a collection of MCI-S patients. Preferably, the alteration of biomarker concentration (i.e., differential expression) is at least 20, 30, 40, 50, 60, 70, 80, 90, 100% as compared to a reference value.

In preferred embodiments of the disclosed methods, one or more biomarker from table 2 or the preferred combinations of biomarkers as disclosed herein are used together with one or more biomarkers selected from total tau, phosphorylated tau, and beta amyloid peptide 1-42. Preferably, an increase in total tau, an increase in phosphorylated tau, and/or an decrease in beta amyloid peptide 1-42 as compared to a reference value in combination with an alteration in the concentration of a biomarker from table 2, preferably an alteration in the concentration of at least two biomarkers from table 2, indicates that a patient is at risk of developing Alzheimer's disease. Suitable assays for determining total tau, phosphorylated tau, and beta amyloid peptide 1-42 levels and levels which indicate risk of developing AD are known in the art and are described, e.g., in U.S. Patent Nos. 5,492,812; 6, 114, 133 and 7,700,309. Commercially available tests are also available, see e.g. products from Innogenetics N.V., INNOTEST® β- AMYLOID(l-42) which is an ELISA microplate assay for the quantitative determination of human β- amyloidl-42(ABl-42) in human cerebrospinal fluid (CSF); INNOTEST® hTau Ag is an ELISA microplate assay for the quantitative determination of human tau in cerebrospinal fluid (CSF); INNOTEST® PHOSPHO-TAU(18 IP) is an ELISA microplate assay for the quantitative determination of human tau, phosphorylated at threonine 181, in human cerebrospinal fluid (CSF), INNO- BIA AlzBio3, an immunoassay which allows the simultaneous quantification of ABl-42, total tau, and P-taul81P in CSF using xMAP® technology.

Any number of suitable assays known to one of skill in the art may be used to determine the level of the biomarkers disclosed in table 2. In preferred embodiments, a binding agent is used to bind and detect a biomarker disclosed herein.

Binding agents include antibodies as well as non-immunoglobulin binding agents, such as phage display-derived peptide binders, and antibody mimics, e.g., affibodies, tetranectins (CTLDs), adnectins (monobodies), anticalins, DARPins (ankyrins), avimers, iMabs, microbodies, peptide aptamers, Kunitz domains, aptamers and affilins.

As used herein, the term "antibodies" include, e.g., monoclonal antibodies; polyclonal antibodies, chimeric, human, humanized antibodies; antigen- binding fragments including, but not limited to, Fab, F(ab'), F(ab')2,

complementarity determining region (CDR) fragments, single-chain antibodies (scFv), bivalent single-chain antibodies, diabodies, triabodies, tetrabodies, artificial antibodies, phage display-derived antibodies, and other antigen recognizing immunoglobulin fragments. Preferably, said binding agent is an antibody.

Techniques for producing binding agents are well known in the art. For example, monoclonal antibodies can be made by the conventional method of immunization of a mammal, followed by isolation of plasma B cells producing the monoclonal antibodies of interest and fusion with a myeloma cell.

Antibodies produced by other techniques such as recombinant antibodies are also encompassed in the disclosure ("Recombinant Proteins, monoclonal antibodies and therapeutic genes", 1999, eds Mountain, A., Ney, U. and

Schomburg, D., Wiley- VCH) as are polyclonal antibodies. Specific procedures for immunizing, additional immunogenic substances for boosting the immune system of the animals to be immunized and time scales for immunization are known in the art. Animals to be used for immunization include rabbits, goats, rats and chicks. The blood samples taken after sacrifice of the animals are processed according to standard procedures. Antibodies that specifically recognise the biomarkers can then be purified from the serum by known procedures, such as stepwise affinity purification. In addition, many

commercially available antibodies exist, e.g., anti- NELL2 is available from, e.g., Abeam (ab80885); anti-CLSTNl is available from, e.g., Abeam (ab723141); anti-MSN is available from, e.g., Abeam (ab3196); anti-CNTP5 is available from, e.g., Abeam (abl65317); anti-FCN3 is available from, e.g., Abeam

(ab 112973).

Non-antibody molecules can be isolated or screened from compound libraries by conventional means. An automated system for generating and screening a compound library is described in U.S. Patents Nos. 5,901,069 and 5,463,564. Suitable assays which utilize binding agents, generally referred to as binding or affinity assays, include, e.g., western blots, radio-immunoassay, ELISA (enzyme -linked immunosorbant assay), "sandwich" immunoassay,

immunoradiometric assay, gel diffusion precipitation reaction,

immunodiffusion assay, precipitation reaction, agglutination assay (e.g., gel agglutination assay, hemagglutination assay, etc.), complement fixation assay, immunofluorescence assay, protein A assay, and immunoelectrophoresis assay.

A typical ELISA assay performed in the art is described as follows. Briefly, the antigen is adsorbed to the wells of a microtiter plate. The wells are typically washed with a blocking buffer to block non-specific antibody binding and to minimize false positive results. Commonly used blocking agents are either protein solutions, such as BSA (typically used at concentrations between 1% and 5% (w/v) in PBS, pH=7.0), non-fat dry milk, or casein (the main protein component of non-fat dry milk).

After the blocking step, the wells of the microtiter plate are typically washed. The adsorbed antigen then undergoes the primary antibody incubation, after which it is typically washed again. Antibody /antigen complexes may then detected using a secondary antibody labeled with chromogenic (e.g.,

horseradish peroxidase and TMB), fluorescent or chemiluminescent (e.g., alkaline phosphatase) means. The amount of color or fluorescence may be measured using a luminometer, a spectrophotometer, or other similar instruments. There are many common variations on the standard ELISA protocol, including e.g., competitive ELISA and sandwich ELISA which are all known to a skilled person.

Binding agents can also be immobilized on a solid support such as a chip or microarray. A biological sample is passed over the solid support. Bound proteins are then detected using any suitable method, such as surface plasmon resonance (SPR) (See e.g., WO 90/05305, herein incorporated by reference). High throughput protein arrays are known in the art and are also described in U.S. Publication No. 20080146459. As used herein, "solid support" means a generally or substantially planar substrate onto which an array of antigens is disposed. A solid support can be composed of any material suitable for carrying the array. Materials used to construct these solid supports need to meet several requirements, such as (1) the presence of surface groups that can be easily derivatized, (2) inertness to reagents used in the assay, (3) stability over time, and (4) compatibility with biological samples. For example, suitable materials include glass, silicon, silicon dioxide (i.e., silica), plastics, polymers, hydrophilic inorganic supports, and ceramic materials. Illustrative plastics and polymers include

poly(tetrafluoroethylene), poly(vinylidenedifluoride), polystyrene,

polycarbonate, polymethacrylate, and combinations thereof.

The disclosed biomarkers are polypeptide based, meaning that they are characterized by mass-to-charge ratio as determined by mass spectrometry, by the shape of their spectral peak in time-of-flight mass spectrometry and by their binding characteristics to adsorbent surfaces.. Preferably, a form of mass spectrometry is used in the disclosed methods. Examples of mass

spectrometers are time-of-flight, magnetic sector, quadrupole filter, ion trap, ion cyclotron resonance, electrostatic sector analyzer and hybrids of these. In a further preferred method, the mass spectrometer is a laser

desorption/ionization mass spectrometer. Mass spectrometry methods are known to one in the art and are described in the examples herein, U.S. Patent Publication 20130122516, U.S. Pat. No. 5,719,060 (describing Surface

Enhanced Laser Desorption and Ionization or SELDI) as well as Regnier et al. Chn Chem 2009 56: 165-171, which discusses approval of such diagnostics with the U.S. Food and Drug Administration. In one embodiment, the biomarkers can be first captured on a chromatographic resin having chromatographic properties that bind the biomarkers. For example, one could capture the biomarkers on a cation exchange resin, wash the resin, elute the biomarkers and detect by MALDI (Matrix-assisted laser desorption/ionization. In another method, one could capture the biomarkers on a probe surface that comprises binding agents that bind the biomarkers, wash the surface to remove unbound material, elute the biomarkers from the surface and detect the eluted biomarkers by MALDI. For some probes, elution from the surface is not necessary and the biomarkers can be detected using mass spectrometry directly from the probe. In some embodiments, the sample is contacted with an affinity capture probe such as a ProteinChip array from Ciphergen Biosystems, Inc. The probe is washed with a buffer that will retain the biomarker while washing away unbound

molecules. The biomarkers are detected by laser desorption/ionization mass spectrometry. These are just a few non-hmiting examples of mass

spectrometry technology known to a skilled person. Additional methods for detecting biomarkers using mass spectrometry are disclosed in U.S.

Publication No. 20110129920.

The biomarkers may be detected in a gas phase ion spectrometer such as a time-of-flight mass spectrometer. The biomarkers are ionized by an ionization source such as a laser, the generated ions are collected by an ion optic assembly, and then a mass analyzer disperses and analyzes the passing ions. The detector then translates information of the detected ions into mass-to- charge ratios. Detection of a biomarker typically will involve detection of signal intensity. Thus, both the quantity and mass of the biomarker can be

determined. Data generated by desorption and detection of biomarkers can be analyzed with the use of a programmable digital computer. The computer program analyzes the data to indicate the number of biomarkers detected, and optionally the strength of the signal and the determined molecular mass for each biomarker detected.

The disclosure also provides kits for determining the concentration of the biomarkers discloses herein. Preferably, said kits comprise at least three binding agents, wherein each agent binds to a different biomarker from table 2. In preferred embodiments, the different biomarkers from table 2 are the preferred biomarkers disclosed herein. Preferably said kits comprise a solid support, such as a chip, a microtiter plate or a bead or resin comprising said binding agents. In some embodiments, the kits comprise mass spectrometry probes, such as ProteinChip™.

The kits may also provide washing solutions and/or detection reagents specific for either unbound binding agent or for said biomarkers (sandwich type assay).

In a preferred embodiment, one of the biomarkers disclosed herein is NELL2. Preferably, an increase in the concentration of NELL2 indicates a risk of developing AD. NELL2 is a secreted neuronal glycoprotein containing six epidermal growth factor (EGF)-like domains. The increase in MCI -AD and the sharp decrease of NELL2 CSF levels in AD patients might be related to its putative functions in neurogenesis and /or synaptic plasticity, as coping mechanism in response to neuronal damage, a mechanism which may be lost during full blown AD due to the extensive neurodegeneration. In preferred embodiments, the biomarkers for the methods and kits disclosed herein comprise at least at least one, preferably at least two, more preferably at least three biomarkers selected from NELL2, CLSTNl, MSN, CNTNAP5, and FCN3. Preferably, an increase in the concentration of CLSTNl correlates with a risk of developing AD. Preferably, an increase in the concentration of

CNTNAP5 correlates with a risk of developing AD. Preferably, a decrease in the concentration of FCN3 correlates with a risk of developing AD. Preferably, the methods comprise determining at least one biomarker from table 2 or a combination thereof. Preferably, the biomarker is selected from NELL2, CLSTN1, MSN, CNTNAP5, and FCN3.

In preferred embodiments, a first biomarker is NELL2, a second biomarker is CLSTN1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CNTNAP5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is MSN, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

NELL2, a second biomarker is FCN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ADAM23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is GALNT6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SEMA3B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is GALNT2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ClQB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SMOCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PGK1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker SPOCKl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SEZ6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PTPRS, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is IGF2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is C6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CSTA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is GC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is LBP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is C3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is DSGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CHIT1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is DKK3 , and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is YWHAB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ITM2B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ODZ1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is TNC, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is PIP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CCL14, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PRDX2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is GM2A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PSAP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SCG2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is MRCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CNTN4, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is CD81, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is HP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is MANlCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is IDH1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is KRT31, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CA2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PTN, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SERPINE2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is HS6ST3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SEMA6A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is MDH2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CECR1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is OLFML3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SORCS3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is SORL1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is COL6A2 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is QSOX2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is ARSA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is C3orf21, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CLSTN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is COL14A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ABI3BP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CDH23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is TUBA1C, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is XYLT1 , and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is AEBPl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PCSK2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is LINGO 1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is TMEM132A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ATP6AP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is GALNT1, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is NELL2, a second biomarker is DSC2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CHGB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is KIAA1199, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is RTN4R, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is NRXN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is CSPG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PCSK1N, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is HRG, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is WFIKKN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

NELL2, a second biomarker is PTPRD, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is NRXN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is F5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is LRRC4B, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is NELL2, a second biomarker is LRP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is PLTP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is APP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is DKK3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is COL6A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is IGFBP7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is FCGBP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is VSIG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ANGPTL7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

NELL2, a second biomarker is CHIT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is ANXA5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is IGHGl, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is NELL2, a second biomarker is EFNB1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is NELL2, a second biomarker is RNASE6, and a third biomarker is selected from table 2.

In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CNTNAP5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is NELL2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is MSN, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is FCN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is ADAM23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is GALNT6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is SEMA3B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is GALNT2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CLSTNl, a second biomarker is ClQB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is SMOCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PGK1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker SPOCKl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is SEZ6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PTPRS, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is IGF2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is C6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CSTA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is GC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is LBP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is C3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is DSGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CHIT1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is DKK3 , and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is YWHAB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is ITM2B, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is ODZ1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is TNC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PIP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CCL14, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is PRDX2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is GM2A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PSAP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is SCG2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is MRCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CNTN4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CD81, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is HP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is MANlCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CLSTNl, a second biomarker is IDH1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is KRT31, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CA2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PTN, and a third

biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is SERPINE2, and a third

biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is HS6ST3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CLSTNl, a second biomarker is SEMA6A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is MDH2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CECR1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is OLFML3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is SORCS3, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is SORL1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is COL6A2 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is QSOX2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is ARSA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is C3orf21, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CLSTNl, a second biomarker is CLSTN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is COL14A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is ABI3BP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is CDH23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is TUBA1C, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is XYLT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is AEBP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PCSK2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is LINGO 1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is TMEM132A, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is ATP6AP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CLSTNl, a second biomarker is GALNTl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is DSC2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CHGB, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is KIAA1199, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is RTN4R, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is NRXN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CSPG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PCSK1N, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is HRG, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is WFIKKN2, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is PTPRD, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is NRXN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is F5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is LRRC4B, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is LRP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is PLTP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is APP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is DKK3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is COL6A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CLSTNl, a second biomarker is IGFBP7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is FCGBP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is VSIG4, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is ANGPTL7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is CHIT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CLSTNl, a second biomarker is ANXA5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is IGHGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CLSTNl, a second biomarker is EFNB1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CLSTNl, a second biomarker is RNASE6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CLSTNl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is NELL2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is MSN, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is FCN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is ADAM23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is GALNT6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is SEMA3B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is GALNT2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is ClQB, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is SMOCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is PGK1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker SPOCK1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is SEZ6, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is PTPRS, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is IGF2, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is C6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is CSTA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is GC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is LBP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is C3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is DSGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is CHIT1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is DKK3 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is YWHAB, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is ITM2B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is ODZ1, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is TNC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is PIP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CCL14, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is PRDX2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is GM2A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is PSAP, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is SCG2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is MRCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CNTN4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CD81, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is HP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is MANlCl, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is IDHl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is KRT31, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CA2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is PTN, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is

SERPINE2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is HS6ST3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is SEMA6A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is MDH2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is CECR1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is OLFML3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is SORCS3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is SORL1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is COL6A2 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is QSOX2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is ARSA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is C3orf21, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CLSTN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is

COL14A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is ABI3BP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CDH23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is TUBA1C, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is XYLT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is AEBP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is PCSK2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is LINGO 1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is

TMEM132A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is

ATP6AP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is GALNTl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is DSC2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CHGB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is KIAA1199, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is RTN4R, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is NRXN3, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is CSPG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is PCSK1N, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is HRG, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is WFIKKN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is PTPRD, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is NRXN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is F5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is LRRC4B, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is LRPl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is PLTP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is APP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is DKK3, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is CNTNAP5, a second biomarker is COL6A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is IGFBP7, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is FCGBP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is

CNTNAP5, a second biomarker is VSIG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is ANGPTL7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is CHIT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is ANXA5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is IGHGl, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is CNTNAP5, a second biomarker is EFNB1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is CNTNAP5, a second biomarker is RNASE6, and a third biomarker is selected from table 2.

In preferred embodiments, a first biomarker is MSN, a second biomarker is CNTNAP5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is NELL2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CLSTN1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is FCN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ADAM23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is GALNT6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SEMA3B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is GALNT2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ClQB, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is SMOCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PGK1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker SPOCK1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SEZ6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PTPRS, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is IGF2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is C6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CSTA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is GC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is LBP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is C3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is DSGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CHIT1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is DKK3 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is YWHAB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ITM2B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ODZ1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is TNC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PIP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CCL14, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PRDX2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is GM2A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PSAP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SCG2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is MRCl, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is CNTN4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CD81, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is HP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is MA 1C1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is IDH1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is KRT31, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CA2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PTN, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SERPINE2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is HS6ST3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SEMA6A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is MDH2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is CECRl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is OLFML3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SORCS3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is SORL1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is COL6A2 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is QSOX2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ARSA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is C3orf21, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is CLSTN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is COL14A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ABI3BP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CDH23, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is TUBAlC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is XYLT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is AEBP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PCSK2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is LINGO 1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is TMEM132A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is ATP6AP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is GALNTl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is DSC2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CHGB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is KIAA1199, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is RTN4R, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is NRXN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CSPG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PCSK1N, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is HRG, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is WFIKKN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PTPRD, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is NRXN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is F5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is LRRC4B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is LRP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is PLTP, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is APP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is DKK3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is COL6A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is IGFBP7, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is FCGBP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is VSIG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is A GPTL7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is CHIT1 , and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is MSN, a second biomarker is ANXA5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is IGHGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is EFNBl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is MSN, a second biomarker is RNASE6, and a third biomarker is selected from table 2.

In preferred embodiments, a first biomarker is FCN3, a second biomarker is CNTNAP5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is NELL2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CLSTN1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is MSN, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ADAM23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is GALNT6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is SEMA3B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is GALNT2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ClQB, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is SMOCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is PGK1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker SPOCK1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is SEZ6, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is PTPRS, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is IGF2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is C6, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CSTA, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is GC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is LBP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is C3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is DSGl, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is CHITl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is DKK3 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is YWHAB, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ITM2B, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is ODZ1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is TNC, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is PIP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CCL14, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is PRDX2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is GM2A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is PSAP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is SCG2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is MRCl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CNTN4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CD81, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is HP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is MANlCl, and a third biomarker is selected from table 2. In preferred embodiments, a first

biomarker is FCN3, a second biomarker is IDH1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is KRT31, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CA2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is PTN, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is SERPINE2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is HS6ST3, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is SEMA6A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is MDH2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CECR1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is OLFML3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is SORCS3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is SORL1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is COL6A2 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is QSOX2, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is ARSA, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is C3orf21, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CLSTN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is COL14A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ABI3BP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CDH23, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is TUBA1C, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is XYLT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is AEBP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is PCSK2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is LINGO 1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is TMEM132A, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ATP6AP1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is GALNTl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is DSC2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CHGB, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is KIAAl 199, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is RTN4R, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is NRXN3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CSPG4, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is PCSK1N, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is HRG, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is WFIKKN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is PTPRD, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is NRXN2, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is F5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is LRRC4B, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is LRP1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is PLTP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is APP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is DKK3, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is COL6A1, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is IGFBP7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is FCGBP, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is VSIG4, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ANGPTL7, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is CHIT1 , and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is ANXA5, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is IGHGl, and a third biomarker is selected from table 2. In preferred embodiments, a first biomarker is FCN3, a second biomarker is EFNB1, and a third biomarker is selected from table 2. In preferred

embodiments, a first biomarker is FCN3, a second biomarker is RNASE6, and a third biomarker is selected from table 2. A further aspect the disclosure provides a method of treating an individual with mild cognitive impairment comprising a) classifying said individual as being at risk of developing AD or AD type dementia by determining the concentration of at least three biomarkers in the cerebrospinal fluid (CSF) of said individual using the methods described herein and b) treating an individual classified as being at risk of developing AD or AD type dementia with an AD therapeutic.

Although there is not yet a cure for AD, there are therapeutics which can reduce or slow the symptoms of AD. Therapeutics include cholinesterase inhibitors (ChEIs), such as tacrine, donepezil, rivastigmine and galantamine.

The administration of such compounds is described in U.S. Publication No.

20060160079, which is hereby incorporated by reference. Therapeutics may also include inflammatory mediaters as described in U.S. Publication No.

20110142795 or memantine.

References

Abdi F, Quinn JF, Jankovic J, Mcintosh M, Leverenz JB, Peskind E, Nixon R, Nutt J, Chung K, Zabetian C, Samii A, Lin M, Hattan S, Pan C, Wang Y, Jin J, Zhu D, Li GJ, Liu Y, Waichunas D, Montine TJ, Zhang J (2006) Detection of biomarkers with a multiplex quantitative proteomic platform in cerebrospinal fluid of patients with neurodegenerative disorders. J Alzheimers Dis 9:293- 348.

Aihara K, Kuroda S, Kanayama N, Matsuyama S, Tanizawa K, Horie M (2003) A neuron-specific EGF family protein, NELL2, promotes survival of neurons through mitogen-activated protein kinases. Brain Res Mol Brain Res 116:86- 93.

Albert MS, Dekosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, Snyder PJ, Carrillo MC, Thies B, Phelps CH (2011) The diagnosis of mild cognitive impairment due to Alzheimer's disease: Recommendations from the National Institute on Aging and Alzheimer's Association workgroup. Alzheimers Dement 7:270-279.

Albrethsen J, Knol JC, Piersma SR, Pham TV, de Wit M, Mongera S, Carvalho B, Verheul HM, Fijneman RJ, Meijer GA, Jimenez CR (2010) Subnuclear proteomics in colorectal cancer: identification of proteins enriched in the nuclear matrix fraction and regulation in adenoma to carcinoma progression. Mol Cell Proteomics 9:988-1005.

Blennow K, Hampel H, Weiner M, Zetterberg H (2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol 6:131-144.

Bossers K, Wirz KT, Meerhoff GF, Essing AH, van Dongen JW, Houba P,

Kruse CG, Verhaagen J, Swaab DF (2010) Concerted changes in transcripts in the prefrontal cortex precede neuropathology in Alzheimer's disease. Brain 133:3699-3723.

Craig-Schapiro R, Perrin RJ, Roe CM, Xiong C, Carter D, Cairns NJ, Mintun MA, Peskind ER, Li G, Galasko DR, Clark CM, Quinn JF, DAngelo G, Malone JP, Townsend RR, Morris JC, Fagan AM, Holtzman DM (2011) YKL-40: a novel prognostic fluid biomarker for preclinical Alzheimer's disease. Biol Psychiatry 68:903-912.

Dubois B, Feldman HH, Jacova C, Cummings JL, Dekosky ST, Barberger- Gateau P, Delacourte A, Frisoni G, Fox NC, Galasko D, Gauthier S, Hampel H, Jicha GA, Meguro K, O'Brien J, Pasquier F, Robert P, Rossor M, Salloway S, Sarazin M, de Souza LC, Stern Y, Visser PJ, Scheltens P (2010) Revising the definition of Alzheimer's disease: a new lexicon. Lancet Neurol 9:1118-1127. Fratantoni SA, Piersma SR, Jimenez CR (2011) Comparison of the

performance of two affinity depletion spin filters for quantitative proteomics of CSF: Evaluation of sensitivity and reproducibility of CSF analysis using GeLC-MS/MS and spectral counting. Proteomics Clin Appl 4:613-617.

Hintsch G, Zurlinden A, Meskenaite V, Steuble M, Fink-Widmer K, Kinter J, Sonderegger P (2002) The calsyntenins-a family of postsynaptic membrane proteins with distinct neuronal expression patterns. Mol Cell Neurosci 21:393- 409.

Huang da W, Sherman BT, Lempicki RA (2009a) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37: 1-13.

Huang da W, Sherman BT, Lempicki RA (2009b) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44-57.

Hulstaert F, Blennow K, Ivanoiu A, Schoonderwaldt HC, Riemenschneider M, De Deyn PP, Bancher C, Cras P, Wiltfang J, Mehta PD, Iqbal K, Pottel H, Vanmechelen E, Vanderstichele H (1999) Improved discrimination of AD patients using beta-amyloid(l-42) and tau levels in CSF. Neurology 52: 1555- 1562.

Ikeuchi T, Hirayama S, Miida T, Fukamachi I, Tokutake T, Ebinuma H, Takubo K, Kaneko H, Kasuga K, Kakita A, Takahashi H, Bujo H, Saito Y,

Nishizawa M (2010) Increased levels of soluble LRU in cerebrospinal fluid of patients with Alzheimer disease. Dement Geriatr Cogn Disord 30:28-32.

Jack CR, Jr., Knopman DS, Jagust WJ, Petersen RC, Weiner MW, Aisen PS,

Shaw LM, Vemuri P, Wiste HJ, Weigand SD, Lesnick TG, Pankratz VS, Donohue MC, Trojanowski JQ (2013) Tracking pathophysiological processes in

Alzheimer's disease: an updated hypothetical model of dynamic biomarkers.

Lancet Neurol 12:207-216.

Jeong JK, Kim HR, Hwang SM, Park JW, Lee BJ (2008) Region- and neuronal phenotype-specific expression of NELL2 in the adult rat brain. Mol Cells 26: 186-192.

Kuroda S, Oyasu M, Kawakami M, Kanayama N, Tanizawa K, Saito N, Abe T, Matsuhashi S, Ting K (1999) Biochemical characterization and expression analysis of neural thrombospondin-l-like proteins NELL1 and NELL2.

Biochem Biophys Res Commun 265:79-86. Ma QL, Galasko DR, Ringman JM, Vinters HV, Edland SD, Pomakian J, Ubeda OJ, Rosario ER, Teter B, Frautschy SA, Cole GM (2009) Reduction of SorLA/LRl l, a sorting protein limiting beta-amyloid production, in Alzheimer disease cerebrospinal fluid. Arch Neurol 66:448-457.

Maarouf CL, Andacht TM, Kokjohn TA, Castano EM, Sue LI, Beach TG, Roher AE (2009) Proteomic analysis of Alzheimer's disease cerebrospinal fluid from neuropath ologically diagnosed subjects. Curr Alzheimer Res 6:399-406.

Maere S, Heymans K, Kuiper M (2005) BiNGO: a Cytoscape plugin to assess overrepresentation of gene ontology categories in biological networks.

Bioinformatics 21:3448-3449.

Matsuyama S, Doe N, Kurihara N, Tanizawa K, Kuroda S, Iso H, Horie M (2005) Spatial learning of mice lacking a neuron-specific epidermal growth factor family protein, NELL2. J Pharmacol Sci 98:239-243.

Mattsson N, Zetterberg H, Hansson O, Andreasen N, Parnetti L, Jonsson M, Herukka SK, van der Flier WM, Blankenstein MA, Ewers M, Rich K, Kaiser E, Verbeek M, Tsolaki M, Mulugeta E, Rosen E, Aarsland D, Visser PJ, Schroder J, Marcusson J, de Leon M, Hampel H, Scheltens P, Pirttila T, Wallin A, Jonhagen ME, Minthon L, Winblad B, Blennow K (2009) CSF biomarkers and incipient Alzheimer disease in patients with mild cognitive impairment. Jama 302:385-393.

Mattsson N, Andreasson U, Persson S, Arai H, Batish SD, Bernardini S, Bocchio-Chiavetto L, Blankenstein MA, Carrillo MC, Chalbot S, Coart E, Chiasserini D, Cutler N, Dahlfors G, Duller S, Fagan AM, Forlenza O, Frisoni GB, Galasko D, Galimberti D, Hampel H, Handberg A, Heneka MT,

Herskovits AZ, Herukka SK, Holtzman DM, Humpel C, Hyman BT, Iqbal K, Jucker M, Kaeser SA, Kaiser E, Kapaki E, Kidd D, Klivenyi P, Knudsen CS, Kummer MP, Lui J, Llado A, Lewczuk P, Li QX, Martins R, Masters C,

McAuliffe J, Mercken M, Moghekar A, Molinuevo JL, Montine TJ, Nowatzke W, O'Brien R, Otto M, Paraskevas GP, Parnetti L, Petersen RC, Prvulovic D, de Reus HP, Rissman RA, Scarpini E, Stefani A, Soininen H, Schroder J, Shaw LM, Skinningsrud A, Skrogstad B, Spreer A, Talib L, Teunissen C,

Trojanowski JQ, Tumani H, Umek RM, Van Broeck B, Vanderstichele H, Vecsei L, Verbeek MM, Windisch M, Zhang J, Zetterberg H, Blennow K (2011) The Alzheimer's Association external quality control program for cerebrospinal fluid biomarkers. Alzheimers Dement 7:386-395 e386.

McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM

(1984) Clinical diagnosis of Alzheimer's disease: report of the NINCDS- ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology 34:939-944.

Morris JH, Apeltsin L, Newman AM, Baumbach J, Wittkop T, Su G, Bader GD, Ferrin TE (2011) clusterMaker: a multi- algorithm clustering plugin for Cytoscape. BMC Bioinformatics 12:436.

Nelson BR, Claes K, Todd V, Chaverra M, Lefcort F (2004) NELL2 promotes motor and sensory neuron differentiation and stimulates mitogenesis in DRG in vivo. Dev Biol 270:322-335.

Newman AM, Cooper JB (2010) AutoSOME: a clustering method for

identifying gene expression modules without prior knowledge of cluster number. BMC Bioinformatics 11:117.

Perrin RJ, Craig-Schapiro R, Malone JP, Shah AR, Gilmore P, Davis AE, Roe CM, Peskind ER, Li G, Galasko DR, Clark CM, Quinn JF, Kaye JA, Morris JC, Holtzman DM, Townsend RR, Fagan AM (2011) Identification and validation of novel cerebrospinal fluid biomarkers for staging early Alzheimer's disease. PLoS One 6:el6032.

Petersen RC (2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256: 183-194.

Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E (1999) Mild cognitive impairment: clinical characterization and outcome. Arch Neurol 56:303-308. Pham TV, Piersma SR, Warmoes M, Jimenez CR (2010) On the beta-binomial model for analysis of spectral count data in label -free tandem mass

spectrometry-based proteomics. Bioinformatics 26:363-369.

Picotti P, Aebersold R (2012) Selected reaction monitoring-based proteomics: workflows, potential, pitfalls and future directions. Nat Methods 9:555-566. Porcellini E, Ianni M, Carbone I, Franceschi M, Licastro F (2013) Monocyte chemoattractant protein- 1 promoter polymorphism and plasma levels in alzheimer's disease. Immun Ageing 10:6.

Rifai N, Gillette MA, Carr SA (2006) Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat Biotechnol 24:971-983.

Ringman JM, Schulman H, Becker C, Jones T, Bai Y, Immermann F, Cole G, Sokolow S, Gylys K, Geschwind DH, Cummings JL, Wan HI (2012) Proteomic changes in cerebrospinal fluid of presymptomatic and affected persons carrying familial Alzheimer disease mutations. Arch Neurol 69:96-104.

Roche S, Gabelle A, Lehmann S (2008) Clinical proteomics of the cerebrospinal fluid: Towards the discovery of new biomarkers. Proteomics Clin Appl 2:428- 436.

Roher AE, Maarouf CL, Sue LI, Hu Y, Wilson J, Beach TG (2009) Proteomics- derived cerebrospinal fluid markers of autopsy-confirmed Alzheimer's disease. Biomarkers 14:493-501.

Steuble M, Diep TM, Schatzle P, Ludwig A, Tagaya M, Kunz B, Sonderegger P (2012) Calsyntenin-1 shelters APP from proteolytic processing during anterograde axonal transport. Biol Open 1:761-774.

Teunissen CE, Petzold A, Bennett JL, Berven FS, Brundin L, Comabella M,

Franciotta D, Frederiksen JL, Fleming JO, Furlan R, Hintzen RQ, Hughes SG, Johnson MH, Krasulova E, Kuhle J, Magnone MC, Raj da C, Rejdak K,

Schmidt HK, van Pesch V, Waubant E, Wolf C, Giovannoni G, Hemmer B, Tumani H, Deisenhammer F (2009) A consensus protocol for the standardization of cerebrospinal fluid collection and biobanking. Neurology 73: 1914-1922.

Thomas PD, Kejariwal A, Campbell MJ, Mi H, Diemer K, Guo N, Ladunga I, Ulitsky-Lazareva B, Muruganujan A, Rabkin S, Van der griff JA, Doremieux O (2003) PANTHER: a browsable database of gene products organized by biological function, using curated protein family and subfamily classification. Nucleic Acids Res 31:334-341.

Vagnoni A, Perkinton MS, Gray EH, Francis PT, Noble W, MiUer CC (2012) Calsyntenin-1 mediates axonal transport of the amyloid precursor protein and regulates Abeta production. Hum Mol Genet 21:2845-2854.

von Mering C, Huynen M, Jaeggi D, Schmidt S, Bork P, Snel B (2003)

STRING: a database of predicted functional associations between proteins. Nucleic Acids Res 31:258-261.

Watanabe TK, Katagiri T, Suzuki M, Shimizu F, Fujiwara T, Kanemoto N, Nakamura Y, Hirai Y, Maekawa H, Takahashi E (1996) Cloning and characterization of two novel human cDNAs (NELL1 and NELL2) encoding proteins with six EGF-like repeats. Genomics 38:273-276.

Yin GN, Lee HW, Cho JY, Suk K (2009) Neuronal pentraxin receptor in cerebrospinal fluid as a potential biomarker for neurodegenerative diseases. Brain Res 1265: 158-170.

Zhang J, Goodlett DR, Quinn JF, Peskind E, Kaye JA, Zhou Y, Pan C, Yi E, Eng J, Wang Q, Aebersold RH, Montine TJ (2005) Quantitative proteomics of cerebrospinal fluid from patients with Alzheimer disease. J Alzheimers Dis 7: 125-133; discussion 173-180.

Zhou L, Barao S, Laga M, Bockstael K, Borgers M, Gijsen H, Annaert W, Moechars D, Mercken M, Gevaer K, De Strooper B (2012) The neural cell adhesion molecules LI and CHL1 are cleaved by BACE1 protease in vivo. J Biol Chem 287:25927-25940. Ziegler-Graham K, Brookmeyer R, Johnson E, Arrighi HM (2008) Worldwide variation in the doubling time of Alzheimer's disease incidence rates.

Alzheimers Dement 4:316-323. Definitions

As used herein, "to comprise" and its conjugations is used in its non-limiting sense to mean that items following the word are included, but items not specifically mentioned are not excluded. In addition the verb "to consist" may be replaced by "to consist essentially of meaning that a compound or adjunct compound as defined herein may comprise additional component(s) than the ones specifically identified, said additional component(s) not altering the unique characteristic of the invention.

The articles "a" and "an" are used herein to refer to one or to more than one (i.e., to at least one) of the grammatical object of the article. By way of example, "an element" means one element or more than one element.

The word "approximately" or "about" when used in association with a

numerical value (approximately 10, about 10) preferably means that the value may be the given value of 10 more or less 1% of the value.

The term "treating" includes prophylactic and/or therapeutic treatments. The term "prophylactic or therapeutic" treatment is art-recognized and includes administration to the host of one or more of the subject compositions. If it is administered prior to clinical manifestation of the unwanted condition (e.g., disease or other unwanted state of the host animal) then the treatment is prophylactic (i.e., it protects the host against developing the unwanted condition), whereas if it is administered after manifestation of the unwanted condition, the treatment is therapeutic, (i.e., it is intended to diminish, ameliorate, or stabilize the existing unwanted condition or side effects thereof). All patent and literature references cited in the present specification are hereby incorporated by reference in their entirety.

The invention is further explained in the following examples. These examples do not limit the scope of the invention, but merely serve to clarify the

invention.

EXAMPLES

Materials and methods

Patients inclusion and CSF collection

The proteomic analysis has been carried out on a total of 30 subjects which were enrolled in two European centres belonging to the cNEUPRO consortium. We selected CSF from 20 patients from the NUBIN biobank containing biomaterial of the Amsterdam Dementia Cohort at the VU University Medical Center (VUMC). The other 10 patients were selected from the bio-bank of the memory clinic of the University of Eastern Finland, Kuopio, Finland and used a first validation cohort of the proteomic results. Patients and samples were retrospectively chosen accordingly to the guidelines decided internally to the consortium, which followed our guidelines published for CSF bio-banking and CSF biomarkers assays (Teunissen et al., 2009). All the patients underwent a thorough clinical examination by experienced neurologists including

neuropsychological assessment (Mini Mental State Examination, MMSE) and laboratory analysis. Patients were diagnosed with AD accordingly to NINCDS- ADRDA criteria (McKhann et al., 1984). The MCI group was composed of patients diagnosed according to Petersen criteria (Petersen et al., 1999), who were followed up for at least two years. The Amsterdam cohort (from now on AMS) was composed of the following groups, each including 5 patients: a control group composed of subjects with subjective memory complaints (SMC), a group of patients diagnosed with MCI with a stable disease over a period of at least two years (MCI-S), a group of patients diagnosed with MCI progressing to AD-type dementia within two years (MCI-AD) and a group of patients diagnosed with AD (AD). The patients retrospectively selected in Finland and analyzed in Bochum (from now on BOCH) were composed of one group of MCI-S patients (n=5) and one group of MCI-AD patients (n=5). The validation cohort was selected from the Amsterdam Dementia Cohort / NUBIN biobank. It was composed of a total of 80 patients, with 20 subjects for each of the groups reported above for Amsterdam cohort. A detailed description of patients characteristics including neuropsychological scores and CSF biomarkers (A61-42, t-tau and p-tau) is reported in table l.CSF was obtained by lumbar puncture between the L3/L4 or L4/L5 intervertebral space, using a 25-gauge needle, and collected in 12-ml polypropylene tubes. Within 2 h, CSF samples were centrifuged at 2000 x g for 10 min at 4°C. A small amount of CSF was used for routine analysis, including total cells (leukocytes and erythrocytes), total protein and glucose concentration analysis. CSF was aliquoted in polypropylene tubes of 0.5 mL and stored at -80°C pending analysis.All the procedures involving human subjects were performed following Helsinki Declaration. All patients or their closest relatives gave informed consent for the study that was approved by the local Ethics

Committees.

CSF sample preparation and gel electrophoresis

CSF samples were initially coded and processed in a blinded fashion. To minimize inter-run bias, each gel contained 2 patients from each clinical group. All samples were processed within the same batch of spin columns to avoid batch-to-batch variability. The depletion of top- 14 high abundant proteins was achieved as previously reported (Fratantoni et al., 2011). Briefly, 1 mL ahquots of CSF from each patient were applied directly to the spin filters (Agilent Human 14 or Genway), following instructions from the manufacturer. Depleted CSF was further concentrated using 3kDa filters prior (Millipore) to loading the whole depleted CSF fraction on gradient gels from Invitrogen (NuPAGE 4-12% Bis-Tris gel, 1.5mmxl0 wells). The gels were then stained with Coomassie brilliant blue G-250 (Pierce, Rockford, IL).

In- Gel Digestion

Before MS analysis, separated proteins were in-gel digested as previously described (Piersma et al., 2012). Briefly, gels were washed and dehydrated once in 50 mM ammonium bicarbonate (ABC) and twice in 50 mM ABC/50% acetonitrile (ACN). Cysteine bonds were reduced by incubation with 10 mM DTT/50 mM ABC at 56°C for 1 h and alkylated with 50 mM iodoacetamide/50 mM ABC at room temperature (RT) in the dark for 45 minutes. After washing sequentially with ABC and ABC/50% ACN, the whole gel was sliced in 10 bands for each lane. Gel parts were sliced up into approximately 1-mm cubes and collected in tubes, washed in ABC/ACN and dried in a vacuum centrifuge. Gel cubes were incubated overnight at 23°C with 6.25 ng/mL trypsin and covered with ABC to allow digestion. Peptides were extracted once in 1% formic acid and twice in 5% formic acid/50% ACN. The volume was reduced to 50 μΐ_^ in a vacuum centrifuge prior to LC-MS analysis.

NanoLC-MS/MS Analysis.

Peptides were separated by an Ultimate 3000 nanoLC system (Dionex LC- Packings, Amsterdam, The Netherlands) equipped with a 20 cm x 75 pm ID fused silica column custom packed with 3 pm 120 A ReproSil Pur C18 aqua (Dr Maisch GMBH, Ammerbuch-Entringen, Germany). After injection, peptides were trapped at 30 pL/min on a 5 mm x 300 pm ID Pepmap C18 cartridge (Dionex LCPackings, Amsterdam, The Netherlands) at 2% buffer B (buffer A: 0.05% formic acid in MQ; buffer B: 80% ACN + 0.05% formic acid in MQ) and separated at 300 nL/min in a 10-40% buffer B gradient in 60 min. Eluting peptides were ionized at 1.7 kV in a Nanomate Triversa Chip-based nanospray source using a Triversa LC coupler (Advion, Ithaca, NJ). Intact peptide mass spectra and fragmentation spectra were acquired on a LTQ-FT hybrid mass spectrometer (Thermo Fisher, Bremen, Germany). Intact masses were measured at resolution 50.000 in the ICR cell using a target value of 1 x 106 charges. In parallel, following an FT prescan, the top 5 peptide signals (charge- states 2+ and higher) were submitted to MS/MS in the linear ion trap (3 amu isolation width, 30 ms activation, 35% normalized activation energy, Q value of 0.25 and a threshold of 5000 counts). Dynamic exclusion was applied with a repeat count of 1 and an exclusion time of 30s.

The BOCH cohort of samples was analysed using nanoLC -MS/MS on an

UltiMate 3000 RSLCnano LC system (Dionex, Idstein, Germany). Samples were loaded on a trap column (Dionex, 75 μιη x 2 cm, particle size 3 μιη, pore size 100 A) with 0.1 % TFA (flow rate 10 μΐ/min). After washing, the trap column was connected with an analytical C18 column (Dionex, 75 μιη x 25 cm, particle size 2 μιη, pore size 100 A). Peptides were separated with a flow rate of 400 nl/min using the following solvent system: (A) 0.1% FA; (B) 84% ACN, 0.1% FA. In a first step, a gradient from 5% B to 40% B (95 min) was used, followed by a second gradient from 40% B to 95% B within 5 min and finally a gradient from 95% to 5 % B within 25 min. ESI-MS/MS was performed on a LTQ Orbitrap (Thermo Fisher Scientific), which was directly coupled to the LC system. MS spectra were scanned between 300 and 2000 m/z with a resolution of 30,000 and a maximal acquisition time of 500 ms. The m/z values initiating MS/MS were set on a dynamic exclusion list for 35 sec. Lock mass

polydimethylcyclosiloxane (m/z 445.120) was used for internal recalibration. The 10 most intensive ions (charge>l) were selected for MS/MS-fragmentation in the ion trap. Fragments were generated by low-energy collision induced dissociation (CID) on isolated ions with collision energy of 35% and maximal acquisition time of 50 ms.

Database Searching and Statistics

Data analysis was performed in Amsterdam, to minimize possible errors related to database search differences and normalization of datasets. MS/MS spectra were searched against the human IPI database 3.31(67511 entries) using Sequest (version 27, rev 12), which is part of the Bio Works 3.3 data analysis package (Thermo Fisher, San Jose, CA). MS/MS spectra were searched with a maximum allowed deviation of 10 ppm for the precursor mass and 1 amu for fragment masses. Methionine oxidation and cysteine

carbamidomethylation were allowed as variable modifications, two missed cleavages were allowed and the minimum number of tryptic termini was 1. After database searching the DTA and OUT files were imported into Scaffold 2.01.01 (Proteome software, Portland, OR). Scaffold was used to organize the gel-band data and to validate peptide identifications using the Peptide Prophet algorithm, only identifications with a probability > 95% were retained.

Subsequently, the Protein Prophet algorithm was applied and protein identifications with a probability of >99% with 2 peptides or more in at least one of the samples were retained. Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped. For each identified protein, the number of spectra was exported to a spreadsheet and normalized according to previously published procedures (Albrethsen et al., 2010; Pham et al., 2010). The beta-binomial test was performed to identify differentially expressed proteins among the different groups (Pham et al., 2010). ELISA and immunoblot data of the validation experiments were analysed with Mann-Whitney test and Spearman correlations (p<0.05).

Protein expression clustering and biological functions

Functional annotation analysis has been carried out using different tools. As an initial step, all the identified proteins in AMS and BOCH datasets were searched using PANTHER classification system (Thomas et al., 2003) to retrieve a general gene ontology (GO) description of the proteomes in the two datasets. Then, putative biomarkers were analysed using DAVID

knowledgebase (Huang da et al., 2009b, a) to find specific enrichment of GO annotations either with respect to whole genome and the whole dataset. In particular, to reduce the redundancy of GO categories, the functional annotation clustering tool embedded in DAVID website was used. Results were reported using the most significant GO categories for each of the cluster found with the software.

Expression data from AMS dataset were clustered using an unsupervised approach, a K-means method implemented in-house with a MATLAB script. Protein abundances were normalized to zero, mean and unit variance per protein. The number of clusters was arbitrarily set to 16. Proteins in the expression profiles were then analysed with DAVID bioinformatics tools in order to: i) gather as much information as possible on the biological functions of the proteins ii) see if the different expression profiles could match specific sets of biological functions, defining the possible involvement of those proteins in AD pathological process. Network analysis

Functional and physical interactions were retrieved from STRING database (von Mering et al., 2003). Data were imported for visualization and network analysis in Cytoscape (v 2.8.3). To identify modules of proteins showing co- expression in MCI patients between AMS and BOCH datasets we used the AUTOSOME procedure (Newman and Cooper, 2010) embedded in

clusterMaker v 1.1 (Morris et al., 2011). As clustering variables we used the fold change for the comparison MCI-S vs MCI-AD and the strength of the interaction (combined score from STRING database). Sub-networks were then searched for functional annotations using BiNGO plug-in (Maere et al., 2005). Immunoblotting

NELL2 antibody was produced by Biogenes GmbH (Germany). CLSTN1 polyclonal antibody (HP AO 12412) was purchased from Sigma Aldrich.

For Western blot analysis, CSF samples (10 uL) were denatured by boiling for 5 min and loaded onto 4-12%. NuPAGE 4-12% Bis-Tris gel, (1 mm x 15 wells). After electrophoresis, proteins were transferred to PVDF membranes for 1 h. Membranes were blocked for 1 h RT with blocking buffer (Li-Cor Biosciences) then incubated for overnight with NELL2 (1:500) and CLSTN1 (1:500) antibodies, diluted in 50% blocking buffer and PBS-T (Tween20 0.1%). After washing with PBS-T six times, the membranes were incubated for 1 h RT with swine anti-rabbit immunoglobulins/biotinylated secondary antibody (Dako Denmark A/S). Blots were washed six times with PBS-T and then incubated for 45 minutes RT with IRDye (1: 15000, LI-COR Biosciences). Blots were then washed 4 times with PBS-T followed with 4 times rinsing in PBS. Proteins were visualized by scanning the membrane on an Odyssey Infrared Imaging System (LI-COR Biosciences) with 700-nm channel.

NELL2 antibody validation

For NELL2 validation a polyclonal antibody created within the cNEUPRO consortium (Biogenes GmbH, Germany) has been used. Two polyclonal antibodies were initially produced in rabbit and labelled 6847 and 6848.

Immunization was achieved injecting a mix of two peptides into different animals. The sequence of the peptide used for immunization of rabbit 6847 was CTAEQFFQKLRNKHE-/ CLHQNGETLYNSGDT-amide. The sequence of the peptide used for immunization of rabbit 6848 was CTAEQFFQKLRNKHE- / CLHQNGETLYNSGDT-amide.

Results

The examples of the present disclosure describe a two-centre study to identify biomarkers for early AD diagnosis using an established workflow (Fratantoni et al., 2011). This workflow consisted of immuno-depletion of high abundant proteins, mono-dimensional SDS-PAGE, label-free protein quantification and pathway analysis . We analysed the CSF proteome from a well-characterised discovery cohort composed of patients diagnosed with MCI either with a stable disease over a follow-up of two years (MCI-S) or progressing to AD type dementia (MCI-AD), patients with full blown AD and control subjects. We prioritized the putative candidates using overlap analysis with an independent cohort composed of MCI patients recruited in a second centre, expression clustering and functional annotations. Our analysis showed a significant overlap of identified CSF proteins in MCI patients in the two datasets, together with a differential CSF protein profile in MCI-S and MCI-AD patients.

Biomarker selection process

Figure 1 depicts the workflow for AD biomarker discovery analysis, selection and validation. Our main focus for biomarker prioritization was to detect candidates for early AD, analysing the CSF of MCI-AD and MCI-S patients taking advantage from the two-cohort setup. First we analysed differential protein expression in AMS dataset, highlighting the protein expression profiles in the four groups of patients and searching for specific functional annotation matching to each profile. We subsequently focused on the comparison MCI-S vs MCI-AD and performed overlap analysis between AMS and BOCH datasets, in order to identify possible correspondences between the two datasets for expression levels and functional annotations. Identification of AD CSF biomarkers and functions in AMS dataset

Depletion of the top- 14 abundant proteins was unsuccessful for two AMS samples (one control and one MCI-S) as judged from SDS-PAGE. These samples were excluded from the final analysis. In total 898 proteins were identified in AMS dataset . We first analysed the AMS dataset to highlight differential proteins in the four groups (p<0.05, table 2). The beta binomial test for the MCI-S vs. MCI-AD comparison yielded 19 differentially regulated proteins in the AMS cohort with 12 proteins up-regulated in MCI-AD patients and 7 proteins down-regulated. We found the highest increase for Moesin (MSN), Disintegrin and metalloproteinase domain-containing protein 23 (ADAM23), Polypeptide N-acetylgalactosaminyltransferase 6 (GALNT6), and Semaphorin-3B (SEMA3B). Vitamin D-binding protein precursor (GC),

Lipopolysaccharide-binding protein (LBP), complement C3 (C3) and

Desmoglein 1 (DSGl) showed the highest decrease in MCI-AD patients.

The SMC vs. AD comparison indicated 11 proteins up-regulated in AD patients with respect to SMC control subjects while 9 proteins were decreased (table 2). The majority of the proteins had a fold change > 2 or > 1.5. To have a global view of the protein expression profiles across the four diagnostic groups we analysed the AMS dataset with cluster analysis. Sixteen distinct expression profiles were generated using a k-means clustering method: a profile

exhibiting increasing expression during disease progression (profile 11), profiles displaying first an increase in expression in the MCI stages followed by a decrease in AD (profiles 1, 7, 6), down-up profiles (4, 10, 13), profiles with decreasing expression (3, 16) and up-down profile (12). To explore whether the different expression profiles were associated with specific functions, the proteins in each profile were searched with the ontology tools embedded in DAVID knowledgebase. The profiles of the sixteen clusters are reported in figure 2 together with their functional annotations as retrieved with DAVID knowledgebase. Looking at the single profiles, we found the profile number 3 and profile number 11 resembled those of classical CSF AD biomarkers

(Blennow et al., 2010). Profile 11 was similar to that of t-tau and p-tau, with an increase in MCI-AD and AD patients with respect to controls and MCI-S patients (figure 2a). Profile 3 was similar to that of A61-42, with a decrease in MCI-AD and AD groups (figure 2b). The proteins included in this profile were involved in cell adhesion and glycosylation processes. According to the biological processes obtained by ontology analysis, cluster 11 was composed by proteins involved in different pathways such as glycolysis, synaptic functions, and nervous system development. Amyloid precursor protein (APP) was present in this cluster, though not significantly regulated in AMS dataset. Several peptides of APP were identified, principally at the N-terminal belonging to the soluble isoforms beta and alpha. Globally, the association of functional annotations for each profile highlighted that the profiles with a specific increase of protein expression in MCI-AD patients (UP-profiles No 1, 4, 6, 7, 10, 11) were more frequently associated with GO biological processes like neurogenesis, axonal guidance, cell adhesion, or synaptic functions (figure 2a). Instead the profiles in which MCI-AD patients had a decreased level of protein expression (DOWN-profiles No 3, 9, 12, 13, 14, 16) were more frequently associated with immune pathways, such as complement activation, but also with cell adhesion and stress response processes (figure 2b).

Verification of candidates and functions in BOCH dataset

We used the BOCH cohort of MCI patients as first verification dataset, performing an overlap analysis and comparing functional annotations with the AMS dataset. A similar proteomic workflow was applied in the BOCH cohort, in which protein depletion was successful in all patients (n=10). Eight hundreds-eleven proteins were identified in BOCH dataset. To ensure that the AMS and BOCH cohorts were comparable in terms of quantitative features, we evaluated the correlation of spectral count quantification between AMS and BOCH datasets. We noticed a quite strong correlation of spectral count data (r= 0.76, p<0.0001, figure 3a). The two datasets were also similar for the total spectral counts for each sample, with an average of count per sample of 8858 ± 818 in AMS dataset and of 8568 ± 669 in BOCH dataset (figure 3c). The overlap of identified proteins between the two datasets, considering all the proteins identified in the four AMS groups and the two groups in BOCH was high (81%, n=763, total = 944) (figure 3b).

General functional analysis of AMS and BOCH datasets using PANTHER classification system, confirmed the high overlap in protein identifications between the two datasets (Figure 7). Most of the identified proteins were extracellular (63% AMS, 59% BOCH), while around 25% of the identified proteins were intracellular in both datasets. Only one protein, contactin-associated protein-like 5 (CNTNAP5), showed differential regulation at p<0.05 in both datasets. At the significance level of p<0.1, five proteins overlapped between the two datasets. Protein kinase C- binding protein NELL2 (NELL2), Calsyntenin-1 (CLSTN1) and CNTNAP5 were increased in MCI-AD patients while Ficolin 3 (FCN3) was decreased in MCI-AD patients with respect to MCI-S group. These four proteins showed similar trends in the two datasets, while for Moesin (MSN) the trend was the opposite in AMS and BOCH datasets, with an increase in AMS dataset and a decrease in BOCH (Figure 8). Using the same validation approach for functional annotations, we compared biological processes of the up and down- profiles in AMS for MCI patients with the annotations of proteins up or down regulated in BOCH MCI patients. Figure 3d shows associated functions of up- regulated proteins in CSF of MCI-AD patients in the BOCH cohort. DAVID functional association clustering showed four main groups of different biological processes, related to cell adhesion, neuronal development and differentiation, endocytosis and behaviour / cognition. For the down-regulated proteins (figure 3e) the main functional annotations were related to immune response, response to stress, cell adhesion and metabolic processes. Several other proteins (n=56) were differentially expressed in MCI-AD group when compared with MCI-S in BOCH cohort. 46 were up-regulated in MCI-AD patients while 10 were decreased with respect to MCI-S patients (table 2).

Network analysis of CSF datasets

After the evaluation of the expression profiles and functional annotations we investigated the overlap between AMS and BOCH datasets at the interaction and pathway level. We used network analysis to find subnetworks of proteins sharing functional or physical interactions which presented also a correlation at the expression level between AMS and BOCH datasets. In this way we could define if sub -networks of interacting proteins had similar expression level between the two datasets. We first built a total network of the complete dataset including 944 proteins from both cohorts, using different types of functional interactions retrieved from STRING database. This first network contained 918 nodes, representing proteins and 1734 edges representing known functional and physical interactions. The total network was then clustered using an AUTOSOME procedure. Protein fold change in the MCI patients for AMS and BOCH datasets and the interaction score from STRING database were used as clustering variables. We obtained 7 sub-networks of which two showed co- regulation of protein expression in AMS and BOCH datasets. We subsequently focused our attention on these two networks for further analyses. The first network contained up-regulated proteins in MCI-AD patients in both datasets and their neighbouring proteins (UP-network, figure 4a). Seventeen proteins belonging to this network were included in our candidate list.

Further clustering of this network using the interaction strength revealed 11 smaller modules of containing candidate proteins which showed co-regulation in MCI-AD patients. The most interesting one (sub-network 2, figure 4a) was composed of several proteins involved in the amyloid pathway, including APP, CLSNT1 and integral membrane protein 2B (ΊΤΜ2Β), the latter showing a significant increase in AD patients when compared to SMC control subjects (fold change 100, p-value = 0.022, only in AMS). It is interesting to note that the enrichment of neurogenesis related-processes we found in the analysis of ontologies associated with AMS expression profiles was conserved also in these sub-networks of proteins showing co-regulation between AMS and BOCH datasets (figure 4a). The network of proteins showing decreased levels in MCI- AD across AMS and BOCH datasets is represented in figure 4b (DOWN- network). Further clustering using interaction strength produced 4 subnetworks including candidate proteins. The first sub-network included complement 6 (C6) which was decreased in MCI-AD patients in AMS cohort (fold change -1.6, p-value = 0.024) while the second network included two proteins of our candidate list, Insulin-like growth factor-2 (IGF2, fold change - 1.5; p-value = 0.029 in AMS MCI-S vs MCI-AD) and insulin-like growth factor binding protein 7 (IGFBP7, fold change -1.3, p-value = 0.038 in BOCH MCI-S vs MCI-AD). Gene ontology enrichment using BINGO plug-in showed overrepresentation of processes related to complement activation and cell-cell adhesion. The presence of co-regulated proteins networks in independent cohort of MCI-AD patients, may indicate that amyloid metabolism,

neurogenesis and complement activation are key pathways in early AD development. Immuno-based validation of candidates in independent CSF cohort

We then performed validation of two candidates in an independent cohort of patients retrospectively selected from the NUBIN biobank in Amsterdam. We chose to validate NELL2 and CLSNT1 because both proteins showed

differential expression in the overlap analysis of AMS and BOCH datasets. Moreover CLSNT1 belonged to the sub-network related to APP processing, supporting its involvement in AD pathogenesis. Clinical details and classical AD biochemical markers of the validation cohort are reported in table 3. Four diagnostic groups were included in the validation cohort, a control group composed of SMC subjects (n=20), a group of MCI patients (n=40), of which half had progressed to AD and a group of AD patients (n=20).

Immunoblot results evidenced a significant increase of NELL2 protein in MCI- AD with respect to the other groups (p<0.05 vs MCI-S and AD, p<0.01 vs SMC), with a very similar profile with respect to proteomic results (figure 5a). Also for CLSTN1 western blot results yielded a relatively similar profile when compared with proteomic results, with an increasing trend in MCI-AD patients (figure 5b), but the differences were not significant. We then analysed the correlations of NELL2 and CLSTN1 with classical CSF biomarkers and cognitive measures in the validation cohort. Spearman coefficients for each correlation are reported in the bar graphs in figure 5c and d. NELL2 had a high significant correlation with age (r = 0.33, p<0.01) and with t-tau and p- tau in the whole sample (r= 0.58, pO.0001 and r = 0.57, pO.0001

respectively). When we analysed the correlations in the separate groups, we noticed an increase of the correlation strength in the pathological groups with respect to MSI-S and SMC groups (figure 5c; r= 0.79, p<0.0001 for t-tau, r= 0.65, pO.0001 for p-tau, r= 0.63, pO.0001 for age) In the AD group, NELL2 levels correlated also with MMSE score at follow-up (r = 0.46, p<0.05).

CLSTNl instead had a weak but significant correlation with t-tau, p-tau and age (r= 0.26, p<0.05 for t-tau, r= 0.28, p<0.05 for p-tau and r= 0.32, pO.01 for age). In the separate groups only the correlation with t-tau was still

significant, but only in the SMC and AD group (figure 5 d).

The possible reason for the lack of significant CLSNT1 regulation in the validation experiment may be related to the variability of the antibody -based assay. However the differential expression of CLSTNl in MCI-AD patients and the parallel increase of CLSTN2 in BOCH dataset point out a possible dysregulation of calsyntenins-related pathways in early AD. Western blot validation methods are not optimal to screen a large number of candidates because of their low-throughput and the scarce availability of antibody to establish quantitative immunoassays (Rifai et al., 2006). Table 1

Data are reported as median and 25-75% percentiles range unless indicated. SMC= subjective memory complaints, MCI-S= MCI with stable disease, MCI- AD= MCI converting to AD, AD= probable AD. a=at least p<0.05 from SMC, b=at least p<0.05 from MCI-S.

Table 2

The list of differentially expressed proteins in the datasets is described in table 2. Protein and gene names, IPI identifier, fold changes and p-values for the comparisons SMC vs AD and MCI-S vs MCI-AD are reported SMc = subjective memory complains, MCI-S stable mild cognitive impairment, MCI-AD mild cognitive impairment progressing to Alzheimer's disease, AD= Alzheimer's disease.

Table 3

Table 1

Patient Age No Αβι-42 (pg/mL) t-tau p-tau Total MMSE MMSE

Groups (mean ± (M/F (pg/mL) (pg/mL) protein (baseline) last ye.

SD) ) s follow-

(fig fiL) up

(mean

SD)

Amsterdam cohort

SMC 60.29 ± 4(2/2) 837.50 ± 133.45 200.25 ± 75.87 47.00 ± 0.34 ± 29.50 ± 1.00 /

4.46 16.02 0.11

MCI-S 62.05 ± 4(1/3) 874.75 ± 201.21 420.75 ± 72.50 ± 0.43 ± 27.40 ± 2.19 28.40 ±

3.21 346.73 47.77 0.24 1.95

MCI-AD 66.17 ± 5(2/3) 499.20 ± 78.47 1071.20 ± 137.60 ± 0.31 ± 27.00 ± 1.41 25.00 ±

6.41 247.63 a 36.51 ^« 0.09 2.45

AD 63.89 ± 5(2/3) 384.00 ± 145.55 525.75 ± 101.60 ± 0.46 ± 21.40 ± 6.31 20.75 ±

6.57 a, b 120.15 39.66 0.17 5.62

Bochum cohort

MCI-S 73.20 ± 5(3/2) 756.80 ± 307.48 235.80 ± 71.86 48.20 ± 0.48 ± 25.20 ± 2.58

5.54 13.48 0.17

MCI-AD 68.00 ± 5(3/2) 345.40 ± 87.46 520.40 ± 77.74 82.00 ± 0.51 ± 25.80 ± 2.77 26.0 ±

7.52 151.15 0.12 1.4

Table 2.

List of differentially expressed proteins in Amsterdam and Bochum datasets

Table 3. Demographic and CSF AD biomarkers in the validation cohort.

Claims

1. A method of determining the risk of developing Alzheimer's disease or speed of onset of Alzheimer's disease in an individual with mild cognitive impairment, said method comprising determining the concentration of at least three biomarkers in the cerebrospinal fluid (CSF) of said individual, wherein said biomarkers are each different biomarkers selected from table 2, preferably wherein a first biomarker is NELL2, a second biomarker is CNTNAP5, and a third biomarker is CLSTNl; and

determining said risk or speed of onset based on the concentration of at least two of said biomarkers.

2. The method of any of the preceding claims, wherein at least one of the at least three biomarkers are selected from NELL2, CLSTNl, MSN, CNTNAP5, and FCN3. 3. The method of any of the preceding claims, wherein at least two of the at least three biomarkers are selected from NELL2, CLSTNl, MSN, CNTNAP5, and FCN3.

4. A method of determining the risk of developing Alzheimer's disease or speed of onset of Alzheimer's disease in an individual with mild cognitive

impairment, said method comprising determining the concentration of at least one biomarker selected from table 2 in the cerebrospinal fluid (CSF) of said individual, and

determining said risk or speed of onset based on the concentration of said biomarker.

5. The method of any of the preceding claims wherein the Alzheimer's disease is characterized by the presence of Alzheimer's disease type dementia.

Claims

6. The method of any of the preceding claims, wherein the Alzheimer's disease is sporadic Alzheimer's disease. 7. The method of any of the preceding claims, wherein said method further comprises determining the concentration of at least one of beta amyloid peptide 1-42, total tau, and phosphorylated tau.

8. The method of claim 7, wherein an increase in total tau, an increase in phosphorylated tau, or a decrease in beta amyloid peptide 1-42 as compared to a reference value, in combination with an alteration in the concentration of at least two of said biomarkers in said individual as compared to a reference value indicates that the individual is at risk of developing Alzheimer's disease. 9. The method of any one of the preceding claims wherein the level of said biomarkers is determined by an immunoassay.

10. The method of any one of claims 1-8, wherein the level of said biomarkers is determined by mass spectrometry.

11. A kit for determining the concentration of at least three biomarkers in cerebrospinal fluid, said kit comprising three binding agents, preferably wherein the binding agents are antibodies, each binding agent directed to a different biomarker selected from table 2, preferably wherein a first biomarker is NELL2, a second biomarker is CNTNAP5, and a third biomarker is

CLSTN1.

12. The kit of claim 11, wherein said binding agents are immobilized on a substrate surface.