ASSAY FOR THE DIAGNOSIS OF A NEUROLOGICAL DISEASE

Abstract

A diagnostically useful carrier can include a means for capturing Neurogranin, a means for capturing BACE1 and preferably a means for capturing one or more additional biomarkers. The diagnostically useful carrier can be part of a kit. The kit or the carrier can be used in contacting a sample from a subject with a means for capturing Neurogranin, isolating the means for capturing Neurogranin from the sample, contacting a sample from a subject with a means for capturing BACE1, and isolating the means for capturing BACE1 from the sample. The diagnosis aims to distinguish a neurodegenerative disease, preferably mild Alzheimer's disease, and depression with or without cognitive impairment.

Claims

1: A diagnostically useful carrier comprising a means for capturing Neurogranin, a means for capturing BACE1 and optionally a means for capturing one or more additional biomarkers.

2: A kit comprising the diagnostically useful carrier according to claim 1.

3: The carrier according to claim 1, wherein the means for specifically capturing Neurogranin and the means for capturing BACE1 are spatially separated.

4: The carrier according to claim 3, wherein the diagnostically useful carrier is selected from the group comprising a bead, a test strip, a microtiter plate, a blot, a glass surface, a slide, a biochip, a membrane, a microarray, an electrophoresis gel and a submicrometer sized particle.

5: A method, comprising: (a1) contacting a sample from a subject with a means for capturing Neurogranin, (b1) isolating the means for capturing Neurogranin from the sample and (a2) contacting a sample from a subject with a means for capturing BACE1 and (b2) isolating the means for capturing BACE1 from the sample.

6: The method according to claim 5, further comprising: (c1) contacting the means for capturing Neurogranin with a detectable ligand binding to Neurogranin and (c2) contacting the means for capturing BACE1 with a detectable ligand binding to BACE1.

7: The method according to claim 5, further comprising: (d1) detecting any Neurogranin captured and (d2) detecting any BACE1 captured.

8: The method according claim 5, further comprising: (a3) contacting a sample from a subject with a means for capturing one or more additional biomarkers, (b3) isolating the means for capturing the one or more additional biomarkers from the sample, optionally (c3) contacting the means for capturing the one or more additional biomarkers with a detectable ligand binding to the one or more additional biomarkers, and optionally (d3) detecting any of the one or more additional biomarkers captured

9: The carrier according to claim 1, wherein the carrier comprises the means for capturing one or more additional biomarkers, and wherein the one or more additional biomarkers are selected from the group consisting of Abeta42, Abeta40, p-tau, t-tau and neurofilament protein.

10: The carrier according to claim 1, wherein the carrier is adapted for detecting Neurogranin, BACE1 and optionally one or more additional biomarkers with a method selected from the group consisting of immunodiffusion techniques; immunoelectrophoretic techniques; light scattering immunoassays; agglutination techniques; labeled immunoassays; mass spectrometry; and Surface Plasmon Resonance.

11: The carrier according to claim 10, wherein the carrier is adapted for detecting Neurogranin, BACE1 and optionally one or more additional biomarkers based on a sandwich ELISA.

12: A method of diagnosing a neurological or neurodegenerative disease, the method comprising: diagnosing the disease with the carrier according to claim 1.

13: The method according to claim 12, wherein the diagnosing comprises predicting a rate of future cognitive decline.

14: The method according to claim 12, wherein the diagnosing comprises distinguishing between a neurodegenerative disease and depression with or without cognitive impairment.

15: A method of diagnosing a depression, the method comprising: diagnosing a depression with the carrier according to claim 1.

16: The diagnostically useful carrier according to claim 1, wherein the means for capturing Neurogranin is an antibody, an aptamer, or calmodulin; and wherein the means for capturing BASE1 is an antibody, an aptamer, or calmodulin.

17: The diagnostically useful carrier according to claim 1, wherein the means for capturing Neurogranin is a monoclonal or polyclonal antibody; and wherein the means for capturing BASE1 is a monoclonal or polyclonal antibody.

18: The diagnostically useful carrier according to claim 1, comprising the means for capturing one or more additional biomarkers, wherein the means for capturing one or more additional biomarkers is an antibody, an aptamer, or calmodulin.

19: The method according to claim 5, wherein the means for capturing Neurogranin is an antibody, an aptamer, or calmodulin; and wherein the means for capturing BASE1 is an antibody, an aptamer, or calmodulin.

20: The method according to claim 5, wherein the means for capturing Neurogranin is a monoclonal or polyclonal antibody; and wherein the means for capturing BASE1 is a monoclonal or polyclonal antibody.

Description

EXAMPLES

Materials and Methods

1.1 Study Population

[0083] CSF samples were obtained by lumbar puncture (LP) at the L3/L4 or L4/L5 interspace, collected in polypropylene tubes, immediately frozen in liquid nitrogen and stored at 80 C. until analysis. Within 3 months prior to or after LP, neuropsychological examinations, at least consisting of Mini-Mental State Examination (MMSE) scores, were performed for each patient.

[0084] Patients with MCI due to AD (referred to as MCI hereafter) (n=38) as well as age-matched patients with dementia due to AD (AD) (n=50) were included and diagnosed according to the NIA-AA criteria (Albert et al. (2011) The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 7(3), 270-279; McKhann et al. (2011) The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 7(3), 263-269).

[0085] To discriminate AD from cognitively normal, CSF biomarkers A.sub.1-42, T-tau and P-tau181P were already analyzed prior to this study at the BIODEM lab using commercial kits (Innotest -AMYLOID(1-42), Innotest hTAU-Ag and Innotest PHOSPHO-TAU(181P); Fujirebio Europe, Ghent, Belgium), whereby the lab's cutoff values were applied [Van der Mussele et al. (2014) Depression in Mild Cognitive Impairment is associated with Progression to Alzheimer's Disease: A Longitudinal Study. J Alzheimers Dis 42(4), 1239-1250]. All MCI patients, of which 2 patients with a pathogenic AD gene mutation, had high probability of AD etiology according to the NIA-AA criteria (see also Supplementary Table 1). Regarding the participants of the age- and gender-matched cognitively healthy control group (CTRL) (n=20), cognitive deterioration was ruled out by means of a neuropsychological screening. CTRL subjects had no neurological or psychiatric antecedents, central nervous system disorders or inflammatory syndromes.

[0086] If available, MMSE total scores from follow-up neuropsychological examinations (at least one year after LP (=baseline) were included in the study. This allowed calculating the rate of cognitive decline based on a linear mixed model adjusted for age and gender.

1.2 Research Neurogranin ELISA for CSF

[0087] Ninety-six-well microtiter plates were coated with mAb ADNGCI2 in Phosphate-buffered Saline (PBS), for 1 h at 21 C. After washing the wells with PBS containing Tween-20 (PBST), the plates were blocked with blocking reagent for 2 h at 21 C. The calibrator, a synthetic peptide containing the C-terminus of neurogranin truncated at Pro75 (custom-made by Proteogenix, France), was prepared in sample diluent (SD) (PBS based buffer with blocking reagent and detergents) at an initial concentration of 900 pg/mL and subsequent three-fold dilutions were made. Samples as well as calibrator dilutions were simultaneously incubated during 3 h at 21 C. with the biotinylated mAb ADNGCT1 in SD. After washing the plates with PBST, horseradish peroxidase-conjugated streptavidin was added. After incubation of 30 min at 21 C., the plates were washed and a 3,3,5,5-tetramethylbenzidine solution was added to engage development. After 30 min at 21 C., the coloring reaction was stopped and plates were read in a BioTek microplate reader, at 450 nm. Levels of neurogranin trunc P75 were calculated via intrapolation (5PL curve fit; log(X)) based on the calibrator curve.

1.3 BACE1 ELISA for CSF

[0088] The BACE1 ELISA used is based on the ELISA method by Baro et al. ((2013) BACE1 levels correlate with phospho-tau levels in human cerebrospinal fluid. Curr Alzheimer Res 10(7), 671-678). This ELISA was further developed by implementing modifications allowing a simultaneous incubation (during 3 h) of 15 L undiluted CSF sample and the detector antibody. Reagents from EUROIMMUN Medizinische Labordiagnostika AG, Lbeck, Germany, were used (BACE-1 ELISA, product number EQ 6541-9601-L). Also, the inventors performed additional rounds of subcloning of the hybridoma cell line of one of the two involved antibodies to ensure the monoclonal character and production of the antibodies. This resulted in the mAb AD402 (clone 10B8F1), which is the detector antibody in the ELISA, and which is biotinylated instead of conjugated with peroxidase as in the previous format (described by Baro et al., 2013). The hybridoma cell line of the capture antibody, mAb AD401 (clone 5G7) underwent no additional rounds of subcloning. The inventors analyzed the BACE1 levels according to the new protocol, where concentrations were calculated via interpolation (5PL curve fit; log (X)) based on a calibrator curve.

1.4 ELISA for the CSF Biomarkers Total-Tau, AB1-42, AB1-40 and AB1-38 CSF levels of total-tau, A.sub.1-42 and A.sub.1-40 were re-analyzed during this study, using the total-tau ELISA, Beta-Amyloid (1-42) ELISA and Beta-Amyloid (1-40) ELISA kits commercially available from EUROIMMUN Medizinische Labordiagnostika AG, Lbeck, Germany (Beta-Amyloid 1-38, product number EQ 6501-9601 for plasma, EQ 6501-9601-L for CSF; Beta-Amyloid 1-40 ELISAEQ 6501-9601 for plasma, EQ 6501-9601-L for CSF, Beta-Amyloid 1-42 ELISA, product number EQ 6501-9601 for plasma, EQ 6501-9601-L for CSF).

[0089] Also, CSF A.sub.1-38 was measured using the assay supplied by EUROIMMUN Medizinische Labordiagnostika AG, Lbeck, Germany.

1.4 Statistical Analysis

[0090] GraphPad Prism 6.02 and R version 3.1.2 were used for statistical analyses and figures. The strength of the correlation between analytes and/or parameters was expressed by Spearman's correlation coefficient. To compare quantitative variables data between groups, a Kruskal-Wallis test was applied. Linear mixed models were fitted with MMSE as dependent variable, fixed effects included time (continuous variable), CSF marker value and their interaction. The random effects included a random intercept for individual and a random slope for time. The significance of the interaction between the CSF marker value and time entailed whether the level of CSF marker had a significant effect on the MMSE change over time. The significance of this term was tested using an F-test with a Kenward-Roger correction for the number of degrees of freedom. Results were considered significant for P-values <0.05.

Results

1.5 Generation of Neurogranin Monoclonal Antibodies

[0091] The anti-neurogranin mAbs ADNGCI2 and ADNGCT1 were generated by immunization of mice with synthetic peptides encompassing the C-terminus of neurogranin. The mAb ADNGCI2 (isotype IgG2a) was produced at PharmAbs (K U Leuven, Belgium) in Balb/c mice after injections with a KLH-coupled peptide containing an internal sequence of C-terminal neurogranin (R51-D78) and additional boosts with full-length synthetic neurogranin. Both the peptide and full-size neurogranin were synthesized at Proteogenix (France). The mAb ADNGCT1 (isotype IgG1) was isolated (at BIOTEM, France) from OF1 mice following an immunization with a KLH-conjugated peptide (synthesized at BIOTEM) corresponding to the C-terminus truncated at P75 (G60-P75). These two mAbs ADNGCI2 and ADNGCT1 each recognize a different epitope on neurogranin as depicted in FIG. 1a. A scan by ELISA with streptavidin coating and biotinylated peptides covering the C-terminal end of neurogranin revealed that the epitope of ADNGCI2 is located within the sequence R51-A66. The mAb ADNGCT1 targets the amino acid range from G62, ending at P75. This specificity of both mAbs was also confirmed by gel electrophoresis on synthetic neurogranin, full size as well as truncated at P75. Three abundant types of human collagen were included as well since ADNGCI2 and ADNGCT1 target the collagen like domain of neurogranin (FIG. 1b).

1.6 Design of an ELISA Targeting Neurogranin Truncated at P75

[0092] For measuring neurogranin, the inventors combined ADNGCI2 as capture antibody and ADNGCT1 as biotinylated detector antibody, into the research sandwich ELISA specific for neurogranin truncated at P75 (FIG. 1c). This form of neurogranin is more abundant in human CSF than the intact C-terminus [Kvartsberg et al. (2015) Cerebrospinal fluid levels of the synaptic protein neurogranin correlates with cognitive decline in prodromal Alzheimer's disease. Alzheimers Dement 11(10), 1180-1190]. The sensitivity of the new mAbs based assay, was assessed with three CSF samples with low endogenous levels of neurogranin trunc P75, run in 4 replicates and the lowest quantifiable concentration with a coefficient of variation (% CV) of 20 was 26 pg/mL (based on the standard curve in duplicate). Three-fold serial dilutions of a synthetic peptide encompassing the C-terminal sequence of neurogranin, ending at P75, were used to generate the standard curve, ranging from 900 to 4 pg/mL.

1.7 Age Dependence of CSF Neurogranin, CSF BACE1 and the Classic CSF Ad Biomarkers

[0093] In preparation of the clinical study, the inventors first verified the neurogranin trunc P75 assay as a proof-of-concept in a large set of CSF samples. Specifically, the inventors analyzed the control group for the clinical study (n=20) in addition to a cohort of diagnostically undefined CSF samples (n=161) (Biomnis, France). The inventors selected the latter group of samples within a wide range of age and with a similar number of male and female subjects. In total, 181 samples (104 female/77 male subjects) were analyzed with a median age of 50.3 (years) (range 7.0-92.1). Apart from neurogranin trunc P75, the inventors analyzed the CSF levels of BACE1, tau, A.sub.1-42, A.sub.1-40 and A.sub.1-38. Table 1 summarizes the concentrations of all analytes. As depicted, only in a very minor number of samples neurogranin trunc P75 levels could not be determined, i.e. 2.8%, which was highly comparable to the other five analytes, i.e. 1.7% for total-tau, A.sub.1-42 and A.sub.1-40, 1.1% for A.sub.1-38, and 0% for BACE1. In addition, all CSF analytes showed an association with age, albeit very weak in case of A.sub.1-42: neurogranin trunc P75 (=0.253; P=0.0007), BACE1 (=0.412; P<0.0001), tau (=0.366; P<0.0001), A.sub.1-42 (=0.147; P=0.050), A.sub.1-40 (=0.305; P<0.0001) and A.sub.1-38 (p=0.326; P<0.0001). FIG. 2, which illustrates this relationship between the several CSF analytes and age, also confirms the accurate selection of the age-matched control group for the clinical study. Finally, there was a strong correlation between neurogranin trunc P75, BACE1 and tau on the one hand, and a solid relationship between the three A-species on the other hand (Table 2).

1.8 CSF Levels of Neurogranin Trunc P75, BACE1 and the Classic CSF Ad Biomarkers in MCI and AD Patients

[0094] After establishing the proof-of-concept of the research assay for neurogranin, the inventors investigated the levels of neurogranin trunc P75 and BACE1 in the CSF of MCI (n=38) and AD (n=50) patients versus cognitively normal participants (CTRL) (n=20). In addition, total-tau, A.sub.1-42, A.sub.1-40 and A.sub.1-38 were quantified in the samples, which were tested blinded from clinical diagnosis. Table 3 sums up the levels of every analyte, combined with the demographic and clinical data of the population.

[0095] Compared to the CTRL group, CSF neurogranin trunc P75 was significantly increased in MCI patients (P<0.01), but was not significantly different in AD patients (Table 3, FIG. 3). Likewise, no significant differences were noted between the MCI and AD groups. On the contrary, CSF BACE1 levels were higher in the MCI group compared to the AD patients (P<0.01), whereas no significant differences were seen between CTRL and MCI, nor between CTRL and AD. For both proteins, there was a trend of increasing group levels when progressing from cognitively normal to MCI, followed by a decrease in case of progression to AD. These findings are reflected by Receiver Operating Characteristics (ROC) analyses, comparing the CTRL group versus both groups of patients (Table 4). CSF neurogranin trunc P75, as single analyte, performed well in discriminating MCI from cognitively healthy with an area under the ROC curve (AUC) of 0.741 (P=0.003), but failed to distinguish AD from control (AUC=0.615, ns). Regarding BACE1, no significant discriminating power was noted for CSF BACE1 as single analyte, for MCI (AUC=0.620, ns), nor for AD (AUC=0.554, ns). As expected, the ROC statistics demonstrated strong diagnostic value of the classic CSF biomarkers total-tau and A.sub.142. AUC values for CSF total-tau were 0.842 (P<0.0001) in case of MCI and 0.854 (P<0.0001) for AD, whereas for CSF A.sub.1-42, AUC values were 0.765 (P=0.001) in case of MCI and 0.849 (P<0.0001) for AD. Combining the biomarkers into ratios clearly resulted in higher AUC values as well as increased statistical strength (Table 4). The AUC value for the ratio of CSF neurogranin trunc P75/A.sub.1-42 was 0.874 (P<0.0001) for discriminating MCI from CTRL and 0.903 (P<0.0001) in case of AD versus CTRL, paralleling the largest discriminating performance of the ratio CSF A.sub.1-42/total-tau (AUC(MCI)=0.871, P<0.0001; AUC(AD)=0.923, P<0.0001). The combination of both synaptic proteins neurogranin trunc P75 and BACE1 also resulted in an AUC value that was superior to the values of both analytes as single parameter. The AUC value for CSF neurogranin trunc P75/BACE1 was 0.715 (P=0.008) regarding MCI patients and 0.806 (P<0.0001) for the AD group.

[0096] Also, the inventors investigated the relationship between all CSF analytes in the clinical groups, and similar to the proof-of-concept analyses described above, there was a strong correlation between neurogranin trunc P75 and BACE1, in MCI (0.793, P<0.0001) as well as AD patients (0.711, P<0.0001) (Table 5). Moderately strong correlations between these synaptic analytes and total-tau were observed as well, i.e. neurogranin trunc P75 and tau were both correlated in MCI (0.635, P<0.0001), and AD (0.617, P<0.0001), as were BACE1 and tau in MCI (0.588, P<0.001), and AD (0.497, P<0.001). Even though neurogranin trunc P75, BACE1 and tau appeared closely associated, subtle differences became apparent when studying their respective associations with the A-species. Neurogranin trunc P75 correlated with both A.sub.1-40 and A.sub.1-38 in MCI as well as AD, but in neither case with A.sub.1-42. On the other hand, tau showed an albeit weak, yet significant correlation with A.sub.1-42 in MCI (0.383, P<0.05), while BACE1 correlated with A.sub.1-42 in AD (0.408, P<0.01). When considering the ratios of CSF A.sub.1-42/A.sub.1-40 and CSF A.sub.1-42/A.sub.1-38 more differences were found (Table 6). In the MCI population, both neurogranin trunc P75 and BACE1 were correlated with both ratios, whereas tau was not.

1.9 Relationship of CSF Biomarkers with MMSE Decline

[0097] In addition to their diagnostic performance, the inventors evaluated the analytes as possible progression biomarkers by assessing their relationship with cognitive decline. Therefore, the inventors considered the CSF levels of all analytes, as single parameters as well as ratios, in a subset of MCI (n=36) and AD (n=45) patients where MMSE was monitored. Adjusted for age and gender, a linear mixed model analysis, with the longitudinal MMSE measurements as outcome, demonstrated that none of the single analytes at baseline, i.e. neurogranin trunc P75, BACE1 nor the classic CSF AD biomarkers, were associated with the rate of cognitive decline (Table 7). Also the majority of biomarker ratios was not correlated, although, there was one specific ratio of CSF biomarkers showing a consistently significant impact on the longitudinal decline in MMSE in both MCI and AD patients, i.e. CSF neurogranin trunc P75/BACE1 (=0.018; P<0.05 for MCI; =0.051; P<0.01 for AD). Statistical significance remained unchanged in case two possible confounders, CSF levels of tau and A.sub.1-42, were added to the model as fixed effects. To illustrate this correlation between the neurogranin trunc P75/BACE1 ratio and cognitive deterioration, the subset of patients with MMSE at and follow-up after LP was partitioned into three tertiles according to the ratio levels (FIG. 5). In both the MCI and AD group, individuals (each represented by thin lines) with the highest ratio levels, i.e. in the highest tertile, had the steepest slope or the biggest change in MMSE. This effect was more pronounced in AD patients. There was one additional ratio that was also significantly correlated with cognitive decline, albeit only in AD patients: CSF neurogranin trunc P.sub.75/A.sub.1-42 (=3.589; P<0.05).

Results

2.1 Generation of Neurogranin Monoclonal Antibodies

[0098] Combining ADNGCI2 as capture antibody and ADNGCT1 as biotinylated detector antibody resulted in the research sandwich ELISA specific for neurogranin truncated at P75 (FIG. 1c).

2.2 Age Dependence of CSF Neurogranin, CSF BACE1 and the Classic CSF AD Biomarkers

[0099] In preparation of the clinical study, the inventors first verified the neurogranin trunc P75 assay as a proof-of-concept in a large set of CSF samples. Specifically, the control group for the clinical study (n=20) in addition to a cohort of diagnostically undefined CSF samples (n=161) (Biomnis, France) was analyzed. The latter group of samples within a wide range of age and with a similar number of male and female subjects was selected. In total, 181 samples (104 female/77 male subjects) were analyzed with a median age of 50.3 (years) (range 7.0-92.1). Apart from neurogranin trunc P75, we analyzed the CSF levels of BACE1, tau, A.sub.1-42, A.sub.1-40 and A.sub.1-38. Table 1 summarizes the concentrations of all analytes. As depicted, only in a very minor number of samples neurogranin trunc P75 levels could not be determined, i.e. 2.8%, which was highly comparable to the other five analytes, i.e. 1.7% for total-tau, A.sub.1-42 and A.sub.1-40, 1.1% for A.sub.1-38, and 0% for BACE1. In addition, all CSF analytes showed an association with age, albeit very weak in case of A.sub.1-42: neurogranin trunc P75 (=0.253; P=0.0007), BACE1 (=0.412; P<0.0001), tau (=0.366; P<0.0001), A.sub.1-42 (=0.147; P=0.050), A.sub.1-40 (=0.305; P<0.0001) and A.sub.1-38 (=0.326; P<0.0001). FIG. 3, which illustrates this relationship between the several CSF analytes and age, also confirms the accurate selection of the age-matched control group for the clinical study. Finally, there was a strong correlation between neurogranin trunc P75, BACE1 and tau on the one hand, and a solid relationship between the three A species on the other hand (Table 2).

2.3 CSF Levels of Neurogranin Trunc 75, BACE1 and the Classic AD Biomarkers in MCI and AD Patients

[0100] The inventors investigated the levels of neurogranin trunc P75 and BACE1 in the CSF of MCI (n=38) and AD (n=50) patients versus cognitively normal participants (CTRL) (n=20). In addition, total-tau, A.sub.1-42, A.sub.1-40 and A.sub.1-38 were quantified in the samples, which were tested blinded from clinical diagnosis. Table 3 sums up the levels of every analyte, combined with the demographic and clinical data of the population.

[0101] Compared to the CTRL group, CSF neurogranin trunc P75 was significantly increased in MCI patients (P<0.01), but was not significantly different in AD patients (Table 3, FIG. 4). Likewise, no significant differences were noted between the MCI and AD groups. On the contrary, CSF BACE1 levels were higher in the MCI group compared to the AD patients (P<0.01), whereas no significant differences were seen between CTRL and MCI, nor between CTRL and AD. For both proteins, there was a trend of increasing group levels when progressing from cognitively normal to MCI, followed by a decrease in case of progression to AD. These findings are reflected by Receiver Operating Characteristics (ROC) analyses, comparing the CTRL group versus both groups of patients (Table 4). CSF neurogranin trunc P75, as single analyte, performed well in discriminating MCI from cognitively healthy with an area under the ROC curve (AUC) of 0.741 (P=0.003), but failed to distinguish AD from control (AUC=0.615, ns). Regarding BACE1, no significant discriminating power was noted for CSF BACE1 as single analyte, for MCI (AUC=0.620, ns), nor for AD (AUC=0.554, ns). As expected, the ROC statistics demonstrated strong diagnostic value of the classic CSF biomarkers total-tau and A.sub.1-42. AUC values for CSF total-tau were 0.842 (P<0.0001) in case of MCI and 0.854 (P<0.0001) for AD, whereas for CSF A.sub.1-42, AUC values were 0.765 (P=0.001) in case of MCI and 0.849 (P<0.0001) for AD. Combining the biomarkers into ratios clearly resulted in higher AUC values as well as increased statistical strength (Table 4). The AUC value for the ratio of CSF neurogranin trunc P75/A.sub.1-42 was 0.874 (P<0.0001) for discriminating MCI from CTRL and 0.903 (P<0.0001) in case of AD versus CTRL, paralleling the largest discriminating performance of the ratio CSF A.sub.1-42/total-tau (AUC(MCI)=0.871, P<0.0001; AUC(AD)=0.923, P<0.0001). The combination of both synaptic proteins neurogranin trunc P75 and BACE1 also resulted in an AUC value that was superior to the values of both analytes as single parameter. The AUC value for CSF neurogranin trunc P75/BACE1 was 0.715 (P=0.008) regarding MCI patients and 0.806 (P.sub.<0.0001) for the AD group.

[0102] Also, the inventors investigated the relationship between all CSF analytes in the clinical groups, and similar to the proof-of-concept analyses described above, there was a strong correlation between neurogranin trunc P75 and BACE1, in MCI (0.793, P<0.0001) as well as AD patients (0.711, P<0.0001) (Table 5). Moderately strong correlations between these synaptic analytes and total-tau were observed as well, i.e. neurogranin trunc P75 and tau were both correlated in MCI (0.635, P<0.0001), and AD (0.617, P<0.0001), as were BACE1 and tau in MCI (0.588, P<0.001), and AD (0.497, P<0.001). Even though neurogranin trunc P75, BACE1 and tau appeared closely associated, subtle differences became apparent when studying their respective associations with the A-species. Neurogranin trunc P75 correlated with both A.sub.1-40 and A.sub.1-38 in MCI as well as AD, but in neither case with A.sub.1-42. On the other hand, tau showed an albeit weak, yet significant correlation with A.sub.1-42 in MCI (0.383, P<0.05), while BACE1 correlated with A.sub.1-42 in AD (0.408, P<0.01). When considering the ratios of CSF A.sub.1-42/A.sub.1-40 and CSF A.sub.1-42/A.sub.1-38 more differences were found (Table 6). In the MCI population, both neurogranin trunc P75 and BACE1 were correlated with both ratios, whereas tau was not.

2.4 Relationship of CSF Biomarkers with MMSE Decline

[0103] The inventors considered the CSF levels of all analytes, as single parameters as well as ratios, in a subset of MCI (n=36) and AD (n=41) patients where MMSE was monitored longitudinally post-LP, for at least one year. Adjusted for age and gender, a linear mixed model analysis, with the longitudinal MMSE measurements as outcome, demonstrated that none of the single analytes at baseline, i.e. neurogranin trunc P75, BACE1 nor the classic CSF AD biomarkers, were associated with the rate of cognitive decline. However, there was one specific ratio of CSF biomarkers showing a consistently significant impact on the longitudinal decline in MMSE across both the MCI and AD group, i.e. CSF neurogranin trunc P75/BACE1 (=0.018; P<0.05 for MCI; =0.051; P<0.01 for AD) (FIG. 5). Statistical significance remained unchanged in case two possible confounders, CSF levels of tau and A.sub.1-42, were added to the model as fixed effects. There was one additional ratio that was also significantly correlated with cognitive decline, albeit only in AD patients: CSF neurogranin trunc P75/A.sub.1-42 (=3.589; P<0.05).

SUMMARY

[0104] In the sample set of the current study, the inventors have shown that the ratio of CSF Neurogranin trunc P75/BACE1 correlates with ongoing and future cognitive decline in MCI due to AD as well as dementia due to AD, in contrast to neurogranin trunc P75 and BACE1 as single parameter, or in contrast to the current AD CSF biomarkers total-tau, A.sub.42 or A.sub.40.

TABLE-US-00001 TABLE 1 Overview of all quantified analytes and their correlation with age, in the cohort of unselected CSF samples and the age-matched control group for the clinical study Correlation with age P-value of age Analytes % N Median (p25-p75) (Spearman's p-value) dependency CSF neurogranin trunc P75 (pg/mL) 2.8 122 (89-191) 0.253 0.0007 CSF BACE1 (pg/mL) 0 934 (702-1304) 0.412 <0.0001 CSF total-tau (pg/mL) 1.7 213 (163-301) 0.366 <0.0001 CSF A.sub.1-42 (pg/mL) 1.7 373 (211-557) 0.147 0.050 CSF A.sub.1-40 (pg/mL) 1.7 4872 (3177-6772) 0.305 <0.0001 CSF A.sub.1-38 (pg/mL) 1.1 1229 (869-1761) 0.326 <0.0001 % N = percentage of samples not quantifiable (out of 181)

TABLE-US-00002 TABLE 2 Spearman's correlation analysis on the CSF biomarkers in the age-matched controls from the clinical study and the cohort of diagnostically unselected samples. Each studied relationship between analytes, represented by Spearman's -values in the table, was statistically significant (P < 0.0001). neurogranin total- trunc P75 BACE1 tau A.sub.1-42 A.sub.1-40 A.sub.1-38 neurogranin trunc P75 BACE1 0.746 total-tau 0.668 0.550 A.sub.1-42 0.448 0.474 0.306 A.sub.1-40 0.595 0.625 0.488 0.888 A.sub.1-38 0.658 0.688 0.519 0.806 0.948

TABLE-US-00003 TABLE 3 Summary of the demographic, clinical and biomarker data of the clinical groups, where data on the analytes are summarized as median values with 25th and 75th quartiles. Median differences between the age-matched control group (CTRL) and the MCI and AD groups were tested using a Kruskal-Wallis test. CTRL MCI AD Demographic Gender, N (F/M) 20 (10/10) 38 (23/15) 50 (27/23) Age at LP (yr) 74 (69-76) 73 (69-79) 75 (68-78) Clinical MMSE at LP (/30) 27 (24-30) 25 (23-27) 18 (12-23) Biochemical CSF neurogranin trunc P75 (pg/mL) 159 (92-205) 214 (161-256)** 172 (141-230) CSF BACE1 (pg/mL) 1472 (1012-2121) 1777 (1291-2276) 1378 (1031-1679) CSF total-tau (pg/mL) 349 (242-434) 567 (446-706)**** 535 (454-711)**** CSF A.sub.1-42 (pg/mL) 503 (267-687) 283 (219-350)* 208 (166-287)**** CSF A.sub.1-40 (pg/mL) 7352 (4884-9514) 7685 (6604-9881) 5662 (4307-7413) CSF A.sub.1-38 (pg/mL) 1728 (1206-2169) 1802 (1539-2385) 1316 (1075-1659) CSF A.sub.1-42/total-tau 1.539 (0.911-1.916) 0.502 (0.380-0.622)*** 0.354 (0.290-0.470)**** CSF A.sub.1-42/A.sub.1-40 0.070 (0.052-0.080) 0.035 (0.028-0.040)**** 0.038 (0.029-0.045)**** CSF A.sub.1-42/A.sub.1-38 0.286 (0.225-0.364) 0.154 (0.111-0.179)**** 0.162 (0.115-0.191)**** CSF neurogranin trunc P75/BACE1 0.095 (0.088-0.111) 0.120 (0.101-0.142)* 0.136 (0.114-0.157)*** CSF neurogranin trunc P75/A.sub.1-42 0.297 (0.228-0.492) 0.767 (0.560-1.026)**** 0.843 (0.626-1.135)**** AD = Alzheimer's disease; CTRL = cognitively healthy; F = female; LP = lumbar puncture; M = Male; MCI = Mild Cognitive Impairment with high probability for AD; MMSE = Mini-Mental State Examination; N = number of samples; yr = years The statistical significance (versus CTRL) is represented by *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

TABLE-US-00004 TABLE 4 ROC-statistics on the CSF biomarkers as single analytes or as ratio, where the set of data of the CTRL group was aligned with the MCI or AD group. The P-values <0.05 are emphasized in bolt font. MCI vs CTRL (n = 38) AD vs CTRL (n = 50) AUC P-value AUC P-value Single analyte neurogranin trunc P75 0.741 0.003 0.615 0.137 BACE1 0.620 0.137 0.554 0.487 total-tau 0.842 <0.0001 0.854 <0.0001 A.sub.1-42 0.765 0.001 0.849 <0.0001 A.sub.1-40 0.559 0.467 0.660 0.038 A.sub.1-38 0.622 0.130 0.645 0.059 Ratio of analytes A.sub.1-42/A.sub.1-40 0.904 <0.0001 0.887 <0.0001 A.sub.1-42/A.sub.1-38 0.894 <0.0001 0.876 <0.0001 A.sub.1-42/total-tau 0.871 <0.0001 0.923 <0.0001 neurogranin trunc P75/A.sub.1-42 0.874 <0.0001 0.903 <0.0001 neurogranin trunc P75/BACE1 0.715 0.008 0.806 <0.0001 AD = Alzheimer's disease; AUC = Area Under the ROC; CTRL = cognitively healthy; F = female; M = Male; MCI = Mild Cognitive Impairment with high probability for AD; MMSE = Mini-Mental State Examination; N = number of samples; ROC = Receiver Operating Curve; yr = years

TABLE-US-00005 TABLE 5 Spearman's correlation analysis on the CSF biomarkers in MCI (above diagonal) and AD patients (below diagonal). MCI (n = 38) neurogranin trunc P75 BACE1 tau A.sub.1-42 A.sub.1-40 A.sub.1-38 AD (n = 50) neurogranin 0.793**** 0.635**** 0.048 0.544*** 0.731**** trunc P75 BACE1 0.711**** 0.588*** 0.238 0.619**** 0.763**** tau 0.617**** 0.497*** 0.383* 0.649**** 0.631**** A.sub.1-42 0.196 0.408** 0.277 0.649**** 0.320 A.sub.1-40 0.565**** 0.679**** 0.522*** 0.781**** 0.770**** A.sub.1-38 0.706**** 0.726**** 0.478*** 0.526**** 0.819**** AD = Alzheimer's disease; MCI = Mild Cognitive Impairment with high probability for AD. The statistical significance is represented by *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

TABLE-US-00006 TABLE 6 Spearman's correlation analysis on CSF neurogranin trunc P75, BACE1 and total-tau with the ratios CSF A.sub.1-42/A.sub.1-40 and CSF A.sub.1-42/A.sub.1-38 in CTRL as well as MCI and AD. A.sub.1-42/A.sub.1-40 A.sub.1-42/A.sub.1-38 CTRL MCI CTRL MCI AD (n = 20) (n = 38) AD (n = 50) (n = 20) (n = 38) (n = 50) neurogranin trunc 0.371 0.608**** 0.651**** 0.430 0.662**** 0.365** P75 BACE1 0.063 0.485** 0.497*** 0.140 0.551*** 0.167 total-tau 0.092 0.275 0.465*** 0.012 0.293 0.080 AD = Alzheimer's disease; CTRL = cognitively healthy; MCI = Mild Cognitive Impairment with high probability for AD. The corresponding statistical significance is represented by **P < 0.01; ***P < 0.001; ****P < 0.0001.

TABLE-US-00007 TABLE 7 Statistical significance of the association between cognitive decline and the single analytes and ratios, based on linear mixed model analysis (MCI n = 36; AD n = 45), adjusted for age and gender. The P-values <0.05 are underlined and emphasized in bolt font. MCI (n = 36) AD (n = 45) Single analyte P-value P-value total-tau 0.684 0.245 A.sub.1-40 0.534 0.927 A.sub.1-42 0.192 0.490 A.sub.1-38 0.888 0.493 BACE1 0.524 0.996 Neurogranin trunc P75 0.766 0.203 Ratio of analytes A.sub.1-42/A.sub.1-40 0.434 0.605 A.sub.1-42/total-tau 0.122 0.092 Neurogranin trunc P75/A.sub.1-42 0.286 0.023 Neurogranin trunc P75/total-tau 0.999 0.615 BACE1/A.sub.1-42 0.888 0.458 BACE1/total-tau 0.099 0.098 Neurogranin trunc P75/BACE1 0.036 0.005

REFERENCES

[0105] Albert et al. (2011) The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 7(3), 270-279. [0106] McKhann et al. (2011) The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 7(3), 263-269. [0107] Jack et al. (2013) Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol 12(2), 207-216. [0108] Duits et al. (2014) The cerebrospinal fluid Alzheimer profile: easily said, but what does it mean? Alzheimers Dement 10(6), 713-723. [0109] Lewczuk et al. (2015) Amyloid- 42/40 cerebrospinal fluid concentration ratio in the diagnostics of Alzheimer's disease: validation of two novel assays. J Alzheimers Dis 43(1), 183-191. [0110] Sauve et al. (2014) Additional use of A.sub.42/A.sub.40 ratio with cerebrospinal fluid biomarkers P-tau and A.sub.42 increases the level of evidence of Alzheimer's disease pathophysiological process in routine practice. J Alzheimers Dis 41(2), 377-386. [0111] Reddy et al. (2005) Differential loss of synaptic proteins in Alzheimer's disease: implications for synaptic dysfunction. J Alzheimers Dis 7(2), 103-117; discussion 173-180. [0112] Selkoe D J (2002) Alzheimer's disease is a synaptic failure. Science 298(5594), 789-791. [0113] Terry R D, Masliah E, Salmon D P, Butters N, DeTeresa R, Hill R, Hansen L A, Katzman R (1991) Physical basis of cognitive alterations in Alzheimer's disease: synapse loss is the major correlate of cognitive impairment. Ann Neurol 30(4), 572-580. [0114] Diez-Guerra (2010) Neurogranin, a link between calcium/calmodulin and protein kinase C signaling in synaptic plasticity. IUBMB Life 62(8), 597-606. [0115] Represa et al. (1990) Neurogranin: immunocytochemical localization of a brain-specific protein kinase C substrate. J Neurosci 10(12), 3782-3792. [0116] Watson et al. (1992) Localization of the protein kinase C phosphorylation/calmodulin-binding substrate RC3 in dendritic spines of neostriatal neurons. Proc Natl Acad Sci USA 89(18), 8581-8585. [0117] Zhong et al. (2009) Neurogranin enhances synaptic strength through its interaction with calmodulin. EMBO J 28(19), 3027-3039. [0118] Baudier et al. (1991) Purification and characterization of a brain-specific protein kinase C substrate, neurogranin (p17). Identification of a consensus amino acid sequence between neurogranin and neuromodulin (GAP43) that corresponds to the protein kinase C phosphorylation site and the calmodulin-binding domain. J Biol Chem 266(1), 229-237. [0119] Sjstedt et al. (2015) Defining the Human Brain Proteome Using Transcriptomics and Antibody-Based Profiling with a Focus on the Cerebral Cortex. PLoS One 10(6), e0130028. doi:10.1371/journal.pone.0130028. [0120] Kvartsberg et al. (2015) Cerebrospinal fluid levels of the synaptic protein neurogranin correlates with cognitive decline in prodromal Alzheimer's disease. Alzheimers Dement 11(10), 1180-1190. [0121] De Vos et al. (2015) C-terminal neurogranin is increased in cerebrospinal fluid but unchanged in plasma in Alzheimer's disease. Alzheimers Dement 11(12), 1461-1469. [0122] Janelidze et al. (2015) Cerebrospinal fluid neurogranin and YKL-40 as biomarkers of Alzheimer's disease. Ann Clin Transl Neurol 3(1), 12-20. doi: 10.1002/acn3.266. eCollection 2016 January [0123] Kester et al. (2015) Neurogranin as a Cerebrospinal Fluid Biomarker for Synaptic Loss in Symptomatic Alzheimer Disease. JAMA Neurol 72(11), 1275-1280. doi:10.1001/jamaneurol.2015.1867. [0124] Thorsell et al. (2010) Neurogranin in cerebrospinal fluid as a marker of synaptic degeneration in Alzheimer's disease. Brain Res 1362, 13-22. doi:10.1016/j.brainres.2010.09.073. [0125] Portelius et al. (2015) Cerebrospinal fluid neurogranin: relation to cognition and neurodegeneration in Alzheimer's disease. Brain 138 (Pt 11), 3373-3385. doi:10.1093/brain/awv267. [0126] Kvartsberg et al. (2015) Characterization of the postsynaptic protein neurogranin in paired cerebrospinal fluid and plasma samples from Alzheimer's disease patients and healthy controls. Alzheimers Res Ther 7(1), 40. doi:10.1186/s13195-015-0124-3. [0127] Kamenetz et al. (2003) APP processing and synaptic function. Neuron 37(6), 925-937. [0128] Kandalepas et al. (2013) The Alzheimer's -secretase BACE1 localizes to normal presynaptic terminals and to dystrophic presynaptic terminals surrounding amyloid plaques. Acta Neuropathol 126(3), 329-352. [0129] Vassar et al. (1999) Beta-secretase cleavage of Alzheimer's amyloid precursor protein by the transmembrane aspartic protease BACE. Science 286(5440), 735-741. [0130] Ahmed et al. (2010) BACE1 and BACE2 enzymatic activities in Alzheimer's disease. J Neurochem 112(4), 1045-1053. [0131] Fukumoto et al. (2002) Beta-secretase protein and activity are increased in the neocortex in Alzheimer disease. Arch Neurol 59(9), 1381-1389. [0132] Barao et al. (2013) BACE1 levels correlate with phospho-tau levels in human cerebrospinal fluid. Curr Alzheimer Res 10(7), 671-678. [0133] Pera et al. (2013) Distinct patterns of APP processing in the CNS in autosomal-dominant and sporadic Alzheimer disease. Acta Neuropathol 125(2), 201-213. [0134] Rosn et al. (2012) Cerebrospinal fluid profiles of amyloid -related biomarkers in Alzheimer's disease. Neuromolecular Med 14(1), 65-73. [0135] Savage et al. (2015) Soluble BACE-1 Activity and sAPP Concentrations in Alzheimer's Disease and Age-Matched Healthy Control Cerebrospinal Fluid from the Alzheimer's Disease Neuroimaging Initiative-1 Baseline Cohort. J Alzheimers Dis 46(2), 431-440. [0136] Zetterberg et al. (2008) Elevated cerebrospinal fluid BACE1 activity in incipient Alzheimer disease. Arch Neurol 65(8), 1102-1107. [0137] Zhong et al. (2007) Levels of beta-secretase (BACE1) in cerebrospinal fluid as a predictor of risk in mild cognitive impairment. Arch Gen Psychiatry 64(6), 718-726. [0138] Mulder et al. (2010) BACE1 activity in cerebrospinal fluid and its relation to markers of AD pathology. J Alzheimers Dis 20(1), 253-260. [0139] Van der Mussele et al. (2014) Depression in Mild Cognitive Impairment is associated with Progression to Alzheimer's Disease: A Longitudinal Study. J Alzheimers Dis 42(4), 1239-1250. [0140] Montine et al. (2012) National Institute on Aging-Alzheimer's Association guidelines for the neuropathologic assessment of Alzheimer's disease: a practical approach. Acta Neuropathol 123(1), 1-11. [0141] Braak & Braak E (1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82(4), 239-259. [0142] Thal D R, Rub U, Orantes M, Braak H (2002) Phases of A beta-deposition in the human brain and its relevance for the development of AD. Neurology 58(12), 1791-1800. [0143] Wayne P A (2000) Clinical and Laboratory Standards Institute. How to define and determine reference intervals in the clinical laboratory: approved guideline. CLSI document C28-A2. [0144] Teunissen et al. (2013) Consensus definitions and application guidelines for control groups in cerebrospinal fluid biomarker studies in multiple sclerosis. Mult Scler 19(13), 1802-1809. [0145] Slaets et al (2013) Cerebrospinal fluid A1-40 improves differential dementia diagnosis in patients with intermediate P-tau181P levels. J Alzheimers Dis 36(4), 759-767. [0146] Bateman et al. (2012) Clinical and biomarker changes in dominantly inherited Alzheimer's disease. N Engl J Med 367(9), 795-804. [0147] Palmqvist et al. (2015) Detailed comparison of amyloid PET and CSF biomarkers for identifying early Alzheimer disease. Neurology 85(14), 1240-1249. [0148] Brinkmalm et al. (2014) SNAP-25 is, a promising novel cerebrospinal fluid biomarker for synapse degeneration in Alzheimer's disease. Mol Neurodegener 9:53. doi: 10.1186/1750-1326-9-53. [0149] Coleman & Yao (2003) Synaptic slaughter in Alzheimer's disease. Neurobiol Aging 24(8), 1023-1027. [0150] Blennow et al. (1996) Synaptic pathology in Alzheimer's disease: relation to severity of dementia, but not to senile plaques, neurofibrillary tangles, or the ApoE4 allele. J Neural Transm (Vienna) 103(5), 603-618. [0151] Pannee et al. (2016) Round robin test on quantification of amyloid- 1-42 in cerebrospinal fluid by mass spectrometry. Alzheimers Dement 12(1), 55-59. [0152] Bjerke et al. (2015) Assessing the commutability of reference material formats for the harmonization of amyloid beta measurements. Clin Chem Lab Med doi: 10.1515/cclm-2015-0733. [0153] Zhabotinsky et al. (2006) Role of the Neurogranin Concentrated in Spines in the Induction of Long-term Potentiation. J NeuroSci 26(28), 7337-47. [0154] Alvarez-Bolado et al. (1996) Neurogranin in the development of the rat telencephalon. Neuroscience 73(2), 565-580. [0155] Fredolini et al. (2016) Immunocapture strategies in translational proteomics, Expt. Rev. Proteomics 13(1), 83-98 Ladenson et al. (2007) Markers for brain damage, EP 1 774 023 [0156] Zhou et al. (2011) Inhibition of beta-secretase in vivo via antibody binding to unique loops (D and F) of BACE1. J Biol Chem 286(10), 8677-8687. [0157] Rosseels et al. (2015) Tau monoclonal antibody generation based on humanized yeast models: impact on Tau oligomerization and diagnostics. J Biol Chem 290(7), 4059-4074.