Novel diagnostic method

Abstract

The present invention relates to methods of measuring the presence and / or levels of a combination of five biomarkers in a sample, said biomarkers being neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, pTau-231 and / or p-Tau217). The methods can be used to predict the subject's risk or likelihood of having and / or developing Alzheimer's disease, thus also provided is a method of predicting a subject's risk or likelihood of having and / or developing Alzheimer's disease, said method comprising measuring the presence of and / or the levels of the combination of five biomarkers. Further provided is a method of prognosing or diagnosing Alzheimer's disease in a subject comprising the methods of measuring the presence of and / or the levels of the combination of five biomarkers, and a method of treating Alzheimer's disease comprising predicting a subject's risk or likelihood of developing or having Alzheimer's disease using the methods of measuring the presence and / or levels of the combination of five biomarkers and administering a treatment if the risk or likelihood exceeds a threshold value.

Description

[0001]NOVEL DIAGNOSTIC METHOD FIELD OF THE INVENTION The present invention relates to methods of measuring the presence and / or levels of a combination of five biomarkers in a sample, said biomarkers being neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau231, p-Tau181 and / or p-Tau217). The methods can be used to predict the subject’s risk or likelihood of having and / or developing Alzheimer’s disease, thus also provided is a method of predicting a subject’s risk or likelihood of having and / or developing Alzheimer’s disease, said method comprising measuring the presence of and / or the levels of the combination of five biomarkers. Further provided is a method of prognosing or diagnosing Alzheimer’s disease in a subject comprising the methods of measuring the presence of and / or the levels of the combination of five biomarkers, and a method of treating Alzheimer’s disease comprising predicting a subject’s risk or likelihood of developing or having Alzheimer’s disease using the methods of measuring the presence and / or levels of the combination of five biomarkers and administering a treatment if the risk or likelihood exceeds a threshold value. BACKGROUND TO THE INVENTION Alzheimer's disease (AD) is a multifactorial neurodegenerative disorder where the accumulation of several proteins including amyloid β (Aβ) and tau constitutes the pathological hallmark of the disease (Bateman et al. (2012) N. Engl. J. Med.,367(9):795-804, doi: https: / / doi.org / 10.1056 / nejmoa1202753). While memory loss is the initial symptom, AD can eventually cause devastating damage to higher cognitive functions. Unfortunately, irreversible brain damage often occurs before clinical symptoms manifest, making early detection crucial for potential clinical benefit. The complexity and severity of the disease requires a multitude of tests in the evaluation of patients for AD pathology, and with no single test currently available, diagnosis must be based on an individual’s history, physical examination and cognitive testing. Furthermore, AD may manifest with atypical non-amnestic presentations, including a language presentation (i.e. logopenic variant primary progressive aphasia), a visuospatial presentation (i.e. posterior cortical atrophy), a dysexecutive presentation (mimicking behavioral variant of frontotemporal dementia) and corticobasal syndrome (McKhann et al. (2011) Alzheimer’s Dement., 7(3):263-269, doi: https: / / doi.org / 10.1016%2Fj.jalz.2011.03.005; and Dubois et al. (2014) Lancet Neurol., 13(6):614-29, doi: https: / / doi.org / 10.1016 / s1474-4422(14)70090-0). Thus, the use of clinical criteria to define AD results in the misclassification of AD patients as non-AD in cases of atypical manifestations and vice versa. With the current standard of care (SOC), Aβ and tau pathology are detected by positron emission tomography (PET) scans or through altered biomarker levels in the cerebrospinal fluid (CSF) (Scheltens et al. (2021) Lancet, 397(10284):1577-1590, doi: https: / / doi.org / 10.1016 / s0140-6736(20)32205-4; and Toledo et al. (2015) Brain, 138(9):2701- 2715, doi: https: / / doi.org / 10.1093 / brain / awv199). Recently, the National Institute of Aging Alzheimer’s Association (NIA-AA) and the International Working Group 2 introduced a framework for the diagnosis of Alzheimer’s disease in clinical research settings that proposed the use of an “A / T / N” system where “A” refers to amyloid (amyloid PET or CSF Aβ1-42), “T” refers to a tau biomarker (CSF phosphorylated tau (p-Tau) or tau PET) and “N” refers to biomarkers of neurodegeneration or neuronal injury ([18F]-fluorodeoxyglucose–PET, structural MRI or CSF total tau) (Jack et al. (2016) Neurology, 87(5):539-547, doi: https: / / doi.org / 10.1212 / wnl.0000000000002923; and Alcolea et al. (2021) J. Neurol. Neurosurg. Psychiatry, 92(11):1206-1214, doi: https: / / doi.org / 10.1136 / jnnp-2021-326603). However, the high cost and inherent risk of using radiotracers, as well as the high cost, invasiveness and risks of collecting CSF samples, necessitate the development of alternative approaches that use plasma biomarkers for the “A / T / N” classification of neurodegenerative dementias (Alcolea et al. (2021)). There are no FDA-approved blood tests available to accurately detect multiple AD-related analytes from a single test. Although over recent years a large amount of clinical evidence has been published that supports the utility of individual assays for evaluating AD from plasma (Jack et al. (2016); Chatterjee et al. (2021) Translational Psychiatry, 11:27, doi: https: / / doi.org / 10.1038 / s41398-020-01137-1; Cicognola et al. (2021) Alzheimer’s Res. Ther., 13(1):68, doi: https: / / doi.org / 10.1186 / s13195-021-00804-9; Chouliaras et al. (2022) J. Neurol. Neurosurg. Psychiatry, 93(6):651-658, doi: https: / / doi.org / 10.1136 / jnnp-2021-327788; Verberk et al. (2020) Alzheimer’s Res. Ther., 12(1):118, doi: https: / / doi.org / 10.1186 / s13195-020-00682-7; Karikari et al. (2020) Lancet Neurol., 19(5):422-433, doi: https: / / doi.org / 10.1016 / s1474- 4422(20)30071-5, Nakamura et al. (2018) Nature, 554(7691):249-254, doi: https: / / doi.org / 10.1038 / nature25456; Ashton et al. (2021) Acta Neuropathol., 141(5):709-724, doi: https: / / doi.org / 10.1007 / s00401-021-02275-6; and Ashton et al. (2021) Nature Comms., 12:3400, doi: https: / / doi.org / 10.1038 / s41467-021-23620-z). Several strategies have been developed to accurately measure Amyloid (A), Tau (Total and phosphorylated) (T) and Neurodegeneration (N) biomarkers from plasma (Koychev et al. (2021) J. Alzheimer's Dis., 79(1):177-195, doi: https: / / doi.org / 10.3233 / jad-200900; and Cullen et al. (2021) Nat. Commun., 12(1):3555, doi: https: / / doi.org / 10.1038 / s41467-021-23746-0). Among these biomarkers, the ratio of Aβ1-42 / Aβ1-40, p-Tau181, p-Tau217, NfL and GFAP (marker for astroglial activation and astrocytosis) are the most promising indicators to aid in the diagnosis of AD. While individual biomarkers perform reasonably well, they do not provide enough diagnostic accuracy to detect AD nor differentiate it from other commonly occurring dementias. For example, p-Tau181 has high sensitivity but lacks specificity, whereas GFAP has high specificity but much lower sensitivity (Chouliaras et al. (2022)). The current gold-standard to diagnose AD and distinguish from non-AD pathologies and dementias is the amyloid positron emission tomography (PET) scan, which has been approved for clinical use by the U.S. Food and Drug Administration (FDA), the European Medicines Agency and other regulatory agencies around the world. Amyloid PET is a crucial tool for the diagnosis of Alzheimer disease, as it allows the non-invasive detection of amyloid plaques, a core neuropathologic feature that defines the disease. However, the method is costly and is unlikely to become widespread for possible preventative screening of subjects who may be at risk of developing AD, or early-stage diagnosis. Furthermore, it is not widely available throughout the world as it requires specialist technology and training. There is therefore a great need to develop blood-based diagnostic assays for the accurate detection of AD and also for the specific and selective distinguishing of AD from non-AD pathologies and dementias which may have similar clinical manifestations. Such assays will provide more effective and earlier identification of AD, providing earlier access to important new medications. Early identification of the disease will also help individuals with timely diagnosis and reduce suffering due to AD. SUMMARY OF THE INVENTION According to a first aspect of the invention, there is provided a method for measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from a subject, comprising the steps of: (a) obtaining the sample from the subject; and (b) measuring the presence and / or levels of the biomarkers: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217) in the sample. In particular embodiments, the presence and / or level of the combination of five biomarkers is used to predict the subject’s risk or likelihood of having and / or developing Alzheimer’s disease. Thus, in a further aspect of the invention there is provided a method of predicting a subject’s risk or likelihood of having and / or developing Alzheimer’s disease, said method comprising measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from the subject, wherein the biomarkers are: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217). In some embodiments, a ratio of the level of any one or more of the biomarkers are measured, such as the ratio of Aβ1-42 / Aβ1-40 levels, the ratio of Aβ1-40 / Aβ1-42 levels, the ratio of p-Tau181 / Aβ1-42 levels and / or the ratio of Aβ1-42 / p-Tau181 levels are measured. In further embodiments, the method further comprises measuring the ionic index of lysosomes in peripheral blood mononuclear cells (PBMCs) obtained from the subject. In yet further embodiments, the prediction additionally comprises using information on the gender and / or age of the subject. In still further embodiments, the prediction additionally comprises using information on the ApoE status of the subject. In certain embodiments, the presence and / or levels of the biomarkers is used to generate a likelihood score for the subject having and / or developing Alzheimer’s disease. According to another aspect of the invention, there is provided a method of prognosing or diagnosing Alzheimer’s disease in a subject at risk of and / or suspected of having Alzheimer’s disease, said prognosis or diagnosis comprising the method described herein. In a yet further aspect, there is provided a method of treating Alzheimer’s disease in a subject at risk of developing or suspected of having Alzheimer’s disease, said method comprising the steps of: (a) predicting the subject’s risk or likelihood of developing or having Alzheimer’s disease by the method described herein; and (b) administering a treatment for Alzheimer’s disease to the subject if the risk or likelihood exceeds one or more threshold value. BRIEF DESCRIPTION OF THE FIGURES Figure 1: Normalised levels (z-score normalised values) of different plasma biomarkers across different study groups. These biomarkers represented ‘A / T / N’ framework in blood for diagnosing AD. The mean difference for pairwise comparisons is shown as a Cumming estimation plot (Ho et al. (2019)). Upper panels: z-score normalised values for each sample is represented as a dot. Lower panels: each mean difference is plotted as a 5000 bootstrap sampling distribution. Mean differences are depicted as black dots; 95% confidence intervals are indicated by the vertical error bars. Permutation t-test. *** p < 0.001, ** p < 0.01, * p < 0.05, ns = not significant. Figure 2: Normalised levels (z-score normalised values) of additional plasma biomarkers present in Esya multi-analyte assay across different study groups. The mean difference for pairwise comparisons is shown as a Cumming estimation plot (Ho et al. (2019)). Upper panels: z-score normalised values for each sample is represented as a dot. Lower panels: each mean difference is plotted as a 5000 bootstrap sampling distribution. Mean differences are depicted as black dots; 95% confidence intervals are indicated by the vertical error bars. Permutation t-test. *** p < 0.001, ** p < 0.01, * p < 0.05, ns = not significant. Figure 3: Development of an optimal model for AD likelihood score calculation and evaluating AD status from control and AD samples. A) Scatter-box-whisker plots of individual AD likelihood scores (0.0-1.0) from 80 samples (38 control and 42 AD). Each dot represents the AD likelihood score for one sample. The AD likelihood scores are calculated using a logistic regression equation of age, gender and different combinations of plasma markers (NfL, GFAP, Aβ1-42 / Aβ1-40 ratio and p-Tau181). Each dot represents one individual. The dashed horizontal line indicates the optimal cut-off for identification (0.445) for Esya multi- analyte assay. B) Table showing combinations tested in A, including AD diagnosis sensitivity, specificity and accuracy values for each. Figure 4: Diagnostic performance metrics for the multi-analyte assay. Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained from a logistic regression equation of age, gender and different combinations of the plasma markers (NfL, GFAP, Aβ1-42 and Aβ1-40 (used as an Aβ1-42 / Aβ1-40 ratio), p-Tau181 and p-Tau181 / Aβ1-42 is shown for different equations. AUC and corresponding 95% CI for each equation is shown in right along with the Upper Control Limit (UCL) and the Lower Control Limit (LCL). Figure 5: Diagnostic performance of the multi-analyte assay. AD likelihood score for evaluating AD status from control or other non-AD dementias. Each dot represents one individual. The dashed horizontal line indicates the optimal cut-off for identification (0.465). Figure 6: Schematic describing the working principle of the Esya analytical tool. Each analyte z-score reports the number of standard deviations from the mean of the control cohort. The AD likelihood score quantifies how likely this sample corresponds to a subject with Alzheimer's disease. The score ranges from 0 to 1. The closer the value is to 0, the more likely the sample corresponds to control. The closer the value is to 1, the more likely the sample corresponds to a subject with Alzheimer's disease (as determined by correlation of the test with PET and CSF clinical diagnosis). The cut-off value, shown as a red line, determines the classification of the sample as cognitively normal or Alzheimer's disease. Figure 7: Development of an optimal model for AD likelihood score calculation by addition of genetic risk factor ApoE4 status to the Esya multi-analyte assay. A) Scatter-box-whisker plots of individual AD likelihood scores (0.0-1.0) from 53 samples (24 control and 29 AD). Each dot represents the AD likelihood score for one sample.  The AD likelihood scores are calculated using a logistic regression equation of age, gender and different combinations of plasma markers (NfL, GFAP, Aβ1-42 / Aβ1-40 ratio and p-Tau181) and ApoE4 status. Each dot represents one individual. The dashed horizontal line indicates the optimal cut-off for identification (0.429) for Esya multi-analyte assay. B) Table showing combinations tested in A, including AD diagnosis sensitivity, specificity and accuracy values for each. Figure 8: Diagnostic performance metrics for the Esya multi-analyte assay. Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained from a logistic regression equation of age, gender and different combinations of the plasma markers (NfL, GFAP, Aβ1-42 and Aβ1-40 (used as an Aβ1-42 / Aβ1-40 ratio), p-Tau181, p-Tau181 / Aβ1-42 and ApoE4 status is shown for different equations. AUC and corresponding 95% CI for each equation is shown on the right along with the Upper Control Limit (UCL) and the Lower Control Limit (LCL). Figure 9: Mean lysosomal ion levels and ionic index across control and AD study groups. The mean difference for pairwise comparisons is shown as a Cumming estimation plot (Ho et al.2019). Upper panels: values for each sample is represented as a dot. Lower panels: each mean difference is plotted as a 5000 bootstrap sampling distribution. Mean differences are depicted as black dots; 95% confidence intervals are indicated by the vertical error bars. Permutation t-test. *** p < 0.001, ** p < 0.01, * p < 0.05, ns = not significant. Figure 10: Development of an optimal model for AD likelihood score calculation by addition of lysosomal ionic status to the Esya multi-analyte assay (Esya multi-modal assay). A) Scatter-box-whisker plots of individual AD likelihood scores (0.0-1.0) from 33 samples (20 control and 13 AD). Each dot represents the AD likelihood score for one sample.  The AD likelihood scores are calculated using a logistic regression equation calculated from log10 values of age, gender, lysosomal ionic index and combinations of plasma markers (NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, and p-Tau181). Each dot represents one individual. B) Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained using the Esya multi-modal assay in A. The dashed diagonal line indicates the optimal cut-off for identification of AD (0.417). C) Table showing combinations tested in B, including AD diagnosis sensitivity, specificity and accuracy values for each. Figure 11: Addition of lysosomal ionic status to the Esya multi-analyte assay (Esya multi-modal assay) using z-score values. A) As in Figure 10 with AD likelihood scores calculated using a logistic regression equation calculated from z-scores of age, gender, lysosomal ionic index and combinations of plasma markers (NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, and p-Tau181). Each dot represents one individual. B) Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained using the Esya multi-modal assay in A. Figure 12: Diagnostic performance metrics for the Esya multi-analyte assay with p-Tau217. Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained from a logistic regression equation of age, gender and different combinations of the plasma markers (NfL, GFAP, Aβ1-42, Aβ1-42 and Aβ1-40 (used as an Aβ1-42 / Aβ1-40 ratio), p-Tau217, p-Tau217 / Ab1-42 is shown for different equations. AUC and corresponding 95% CI for each equation is shown in right along with the Upper Control Limit (UCL) and the Lower Control Limit (LCL). N = 100, 46 Control and 54 AD patients. Figure 13: Diagnostic performance metrics for the Esya multi-analyte assay with p-Tau217, NfL / Aβ1-42, and GFAP / Aβ1-42. Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained from a logistic regression equation of age, gender and different combinations of the plasma markers (NfL / Aβ1-42, GFAP / Aβ1-42, and p-Tau217 is shown for different equations. AUC and corresponding 95% CI for each equation is shown in right along with the Upper Control Limit (UCL) and the Lower Control Limit (LCL). N = 100, 46 Control and 54 AD patients. Figure 14: Development of an optimal model for AD likelihood score calculation by addition of lysosomal ionic status to the NfL, GFAP and p-Tau217. A) Scatter-box- whisker plots of individual AD likelihood scores (0.0–1.0) from 38 samples (24 control and 14 AD). Each dot represents the AD likelihood score for one sample.  The AD likelihood scores are calculated using a logistic regression equation of age, gender, lysosomal ionic index and combinations of plasma markers (NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, and p-tau217). Each dot represents one individual. B) Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained using the Esya multi-modal assay in A. The dashed horizontal line indicates the optimal cut-off for identification of AD (0.5) for the assay with lysosomal ionic index and p-Tau217. C) Table showing combinations tested in B, including AD diagnosis sensitivity, specificity and accuracy values for each equations. Figure 15: Diagnostic performance metrics for the Esya multi-analyte assay with p-Tau231, p-Tau181 and p-Tau217. Receiver Operating Characteristic (ROC) curve using the AD likelihood score obtained from a logistic regression equation of age, gender and different combinations of the plasma markers (NfL, GFAP, p-Tau231, p-Tau181, p-Tau217 / Aβ1-42 is shown for different equations. AUC and corresponding 95% CI for each equation is shown in right along with the Upper Control Limit (UCL) and the Lower Control Limit (LCL). N = 100, 46 Control and 54 AD patients. DETAILED DESCRIPTION OF THE INVENTION The present invention is based on a novel method of combining the protein concentration levels of neurofilament light (NfL), glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217) in human plasma samples to calculate a likelihood score which reflects / predicts a subject’s risk or likelihood of having and / or developing Alzheimer’s disease, or the likelihood that a subject will be identified as positive on an amyloid positron emission tomography (PET) scan. Thus, according to a first aspect of the invention there is provided a method for measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from a subject, comprising the steps of: (a) obtaining the sample from the subject; and (b) measuring the presence and / or levels of the biomarkers: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40),and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217) in the sample. The term “biomarker” is used herein in its normal meaning and context in the field of disease, disease prediction and diagnostics / prognostics. They are measurable indicators of a biological state or condition, such as a disease state, and can also in certain circumstances be used to predict or determine a subject’s likelihood of responding to a particular treatment. Each of the individual biomarkers measured according to the methods herein are known for their association with Alzheimer’s disease (AD), however the present inventors have demonstrated herein the efficacy of combining the measurement of these five biomarkers and / or their ratios to predict a subject’s risk or likelihood of having and / or developing AD, such as to generate a likelihood score as described herein. For example, neurofilament light polypeptide (NfL polypeptide; also known as neurofilament light chain) is a major structural protein in neurons, particularly concentrated in large projection axons which are sensitive to mechanical and metabolic compromise and thus the degeneration of these cells is a significant problem in many neurological disorders. As such, the detection of neurofilament subunits in cerebrospinal fluid (CSF) and blood / plasma has become widely used as a biomarker of ongoing axonal compromise in a several neurological disorders. In particular, it is a useful marker for disease monitoring in amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS), AD, Huntington's disease, as well as in the follow-up of patients with brain tumours. Glial fibrillary acidic protein (GFAP) is another structural protein involved in the function of the cytoskeleton in astrocytes in the central nervous system (CNS). Upregulated GFAP is, at least partially, involved in glial scarring which is a consequence of several neurodegenerative conditions, as well as injury that severs neural material. It’s generally high abundance in the CNS has led to interest in the use of GFAP as a blood biomarker of acute brain and spinal cord injury, such as in traumatic brain injury and cerebrovascular disease, as well as for neuroinflammatory diseases, such as MS and neuromyelitis optica which targets astrocytes. β-Amyloid 1-42 (Aβ1-42; also referred to as Aβ42) and β-Amyloid 1-40 (Aβ1-40; also referred to as Aβ40) are forms of amyloid β peptide, with 42 or 40 amino acids respectively, that are the main component of the amyloid plaques found in the brains of people with AD. While the normal function of amyloid β is not fully understood, both the Aβ1-42 and Aβ1-40 forms are found in AD patients and both make up amyloid plaques, although Aβ1-42 has been found to be the major component of cerebrovascular amyloid deposits, while Aβ1-40 is more soluble and represents total elevated amyloid β in the circulation of AD patients. Due to its more hydrophobic nature, Aβ1-42 is the most amyloidogenic form of the peptide. The "amyloid hypothesis" that amyloid plaques in the brain are responsible for the pathology of AD, is widely accepted, but an alternative hypothesis is that amyloid oligomers rather than plaques may be responsible for the disease. Tau proteins (abbreviated from tubulin associated unit) have roles primarily in maintaining the stability of microtubules in axons and are abundant in the neurons of the CNS. Hyperphosphorylation of tau protein (tau inclusions, p-Tau) leads to the self- assembly of insoluble tangles of paired helical filaments and straight filaments called neurofibrillary tangles and is associated with several pathologies and dementias of the nervous system such as AD, Parkinson's disease, frontotemporal dementia and other tauopathies. All of the six tau isoforms are present in an often hyperphosphorylated state in paired helical filaments in the brain of AD patients, and thus may be used interchangeably herein. For example, plasma levels of phosphorylated tau at threonine 181 (p-Tau181) has in particular been shown to be able to differentiate between clinically diagnosed or autopsy confirmed AD and Frontotemporal Lobar Degeneration (FTLD; Thijssen et al. (2020) Nat. Med., 26(3):387-397, doi: https: / / doi.org / 10.1038%2Fs41591-020-0762-2). p-Tau217 has also been shown to be able to differentiate clinically diagnosed AD from cognitively normal participants and patients with mild cognitive impairment or FTLD, demonstrating its potential equivalence for p-Tau181 or benefits in using p-Tau181 and p-Tau217 together (Thijssen et al. (2021) Lancet Neruol., 20(9):739-752, doi: 00214-3). The same has been shown for p-Tau231 with plasma levels showing strong association with Aβ positron emission tomography (PET) retention in early accumulating regions and association with longitudinal increases in Aβ PET uptake in individuals without overt Aβ pathology at baseline, with p-Tau217 (Milà-Alomà et al. (2022) Nat. Med.28(9):1797- 1801, doi: https: / / doi.org / 10.1038 / s41591-022-01925-w). The calculation of ratios between several of these biomarkers have previously been shown to be useful in the diagnosis of AD and the differentiation of AD with non-AD pathologies and dementias. For example, with respect to Aβ1-42 and Aβ1-40, a decrease in CSF levels of Aβ1-42 is characteristic of AD, but defining a cut-off or threshold level below which a subject is deemed to have AD (e.g. rather than a similar non-AD pathology) is difficult due to significant inter-subject variability (Hansson et al. (2019) Alzheimer’s Res. Ther., 11:34, doi: https: / / doi.org / 10.1186%2Fs13195-019-0485-0). Therefore, it has been suggested to incorporate the CSF level of Aβ1-40, which is a reflection of total CSF amyloid levels, by use of the Aβ1-42 / Aβ1-40 ratio (Amft et al. (2022) Alzheimer’s Res. Ther., 14:60, doi: and Niemantsverdriet et al. (2017) J. Alzheimer’s Dis., 60(2):561-576, doi: https: / / doi.org / 10.3233%2FJAD-170327). That this ratio is superior to using Aβ1-42 levels alone in the differentiation of AD from non-AD pathologies was shown by Constantinides et al. ((2023) Diagnostics (Basel), 13(4):783, doi: https: / / doi.org / 10.3390%2Fdiagnostics13040783). This study also demonstrated the general superiority of biomarker ratios for distinguishing AD, including the p-Tau181 / Aβ1-42 ratio and several ‘composite’ ratios of ratios. However, Constantinides et al. only ever tested and compared each ratio individually (i.e. the Aβ1-42 / Aβ1-40 ratio or the p-Tau181 / Aβ1-42 ratio to distinguish AD from non-AD pathologies). As mentioned hereinbefore, the measurement of several biomarkers in the blood of AD patients have previously been disclosed, with the ratio of Aβ1-42 / Aβ1-40 levels, p-Tau181, p-Tau231, p-Tau217, NfL and GFAP (marker for astroglial activation and astrocytosis) levels being the most promising indicators to aid in the diagnosis of AD. Thus, these markers could provide useful information about the “A / T / N” categories with a minimally invasive approach, facilitate early diagnosis, and discriminate AD from other pathophysiologies presenting similar symptoms. However, individually they do not provide sufficient diagnostic accuracy. Therefore, a combination of the presence and / or levels of these biomarkers in the blood or plasma is essential to develop a diagnostic test that can provide both high accuracy for AD detection, as well as high specificity to differentiate AD from other similar dementias. Thus, in particular embodiments of the present invention there is provided methods for measuring the presence and / or levels of a combination of five biomarkers in a sample from a subject, wherein said biomarkers are neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated tau (p-Tau181, p-Tau231 and / or p-Tau217). In particular, the presence and / or level of p-Tau181 may be measured. Further, p-Tau217 and / or p-Tau231 (which may be used in addition to or interchangeably with p-Tau181 or together herein) presence / levels are measured. As will be readily appreciated, the use of the term “combination” herein refers to the concomitant (i.e. simultaneous) and / or sequential measurement of (at least) all five biomarkers in a sample. Such measurement of each biomarker may therefore be in a single test sample, such as by a multi-plex assay. Alternatively, the measurement of each biomarker may be distinct from the measurement of one or more of the other biomarkers in the sample, for example wherein multiple samples are obtained or a single test sample is divided with e.g. one biomarker measured in each division of the sample. Still alternatively, both concomitant measurement (e.g. by a multi-plex assay) and sequential measurement (e.g. in different of divided samples) may be used, for example by measuring two, three or four biomarkers in a single test sample concomitantly and measuring the other of the one, two or three biomarkers sequentially. In one embodiment, the levels of NfL, GFAP, Aβ1-42, Aβ1-40 and p-Tau181 (and / or p-Tau231 and / or p-Tau217 together with or in place of p-Tau181 as described hereinbefore) are measured in a single test sample, such as concomitantly. In a particular embodiment, the levels of NfL, GFAP, Aβ1-42 and Aβ1-40 are measured concomitantly in a single assay (e.g. in a 4-plex assay), and the level of p-Tau181 (and / or p-Tau231 and / or p-Tau217 together with or in place of p-Tau181 as described hereinbefore) is measured sequentially / separately. In some embodiments, the levels of at least those biomarkers from which common ratios are derived are measured in a single test sample, i.e. concomitantly (e.g. the levels of Aβ1-42 and Aβ1-40). As demonstrated by the data presented herein, the combination of the five biomarkers as described herein provided good diagnostic performance to identify AD in subjects, with accuracy gradually improved from using a combination of 3 markers (ATN) to 4 biomarkers (ATN with GFAP), but achieving maximum accuracy, sensitivity and specificity by using an optimal combination of the five biomarkers. Further addition of biomarkers did not improve accuracy. Any method known in the art for measuring analyte levels may suitably be used in the methods described herein. In particular, methods for the quantitative detection of proteins and / or fragments thereof (i.e. the biomarkers herein) are used. Such methods include those which specifically measure the presence and levels of an analyte of interest (i.e. a biomarker), e.g. using antibodies or binding fragments that specifically recognise the analyte. Immunoassays are one example of specific analyte measurement methods, in which antibodies or fragments which specifically and / or selectively bind the analyte of interest are used and may be conjugated to enzymes or fluorescent markers to visualise and quantify binding. Thus, in one embodiment the measurement of the biomarkers is by immunoassay. In a further embodiment, the immunoassay is an enzyme-linked immunoassay. In a particular embodiment, the immunoassay comprises beads (e.g. magnetic beads, such as paramagnetic beads) which are coated with a ‘capture’ antibody that recognises and binds the analyte (i.e. biomarker). In another embodiment, the capture antibody may be immobilised on a surface, such as a surface of a 96-well plate or the like. In a yet other embodiment, the capture antibody is in solution and may comprise one half of a binding pair, which allows subsequent immobilisation using the second half of the binding pair. In a still other embodiment, the analyte is immobilised directly on a surface (e.g. by adsorption) without the need for a capture antibody. In a further embodiment, the immunoassay comprises a ‘detector’ antibody which also recognise and bind the analyte. The detector antibody may comprise an enzyme, such as OPD (o-phenylenediamine dihydrochloride), TMB (3,3',5,5'-tetramethylbenzidine), ABTS (2,2'-Azinobis [3-ethylbenzothiazoline-6-sulfonic acid]-diammonium salt) or PNPP (p- Nitrophenyl Phosphate, Disodium Salt). The presence of the enzyme (by virtue of binding of the detector antibody to the analyte, which is qualitative) is then used to drive a reaction which can be measured, usually by colour change (e.g. OPD turns to amber / yellow in the presence of horseradish peroxidase (HRP), TMB turns blue in the presence of HRP and turns yellow after the addition of sulfuric or phosphoric acid, ABTS turns green with HRP, and PNPP turns yellow in the presence of alkaline phosphatase). In a further embodiment, the detector antibody may comprise one half of a binding pair (e.g. biotin) to which a second half of the binding pair (e.g. streptavidin), conjugated to an enzyme, can bind. In a certain embodiment, the detector antibody comprises biotin, and binding of the detector antibody to the analyte is measured by addition of a streptavidin-β-galactosidase (SBG) conjugate. Following addition of SBG conjugate, a resorufin β-D-galactopyranoside (RGP) substrate solution is added which in the presence of β-galactosidase is hydrolysed into a fluorescent product that provides the signal to be measured. In a further embodiment, the immunoassay is performed in multiple wells of an e.g.96-well plate and is read-out ‘digitally’, such as wherein each well contains a single molecule (such as a single bead according to embodiments described hereinbefore) and either produces a signal (when the analyte is present) or does not (when the analyte is absent). The combined measurement of a number of these ‘digital’ wells provides quantitation of the amount of analyte in a sample. Thus, in one embodiment the immunoassay is a single- molecule array immunoassay. In another embodiment, the immunoassay comprises oligonucleotide linked antibodies. In a particular embodiment, the immunoassay comprises oligonucleotide-conjugated capture and detector antibodies. As demonstrated in Feng et al. (2023) Nature Comms., 14(1):7238 (doi: https: / / doi.org / 10.1038 / s41467-023-42834-x) such immunoassays comprising oligonucleotide-conjugated capture and detector antibodies (referred to therein as NUcleic acid Linked Immuno-Sandwich Assay; NULISATM) may comprise a polyA oligonucleotide which can be used to capture the bound analyte in complex with the antibody and a biotinylated oligonucleotide which can be used in a sequential capture step to capture analyte which is dual bound with both antibodies. Thus, in one embodiment the immunoassay comprises multiple capture and release steps. In a further embodiment, the immunoassay further comprises ligation of the antibody-conjugated oligonucleotides, followed by optional quantitative PCR (e.g. RT-PCR) or sequencing. NULISATMhas been shown to achieve sensitivity in the attomolar range and significantly reduce assay background. In particular embodiments, the immunoassay is performed in vitro. Thus, in a further embodiment the immunoassay is an in vitro single-molecule array immunoassay. In some embodiments, the method may further comprise measuring the ionic index of lysosomes in peripheral blood mononuclear cells (PBMCs) obtained from the subject. Lysosomes are membrane-bound organelles containing hydrolytic enzymes that can break down many kinds of biomolecules. They have a specific composition, of both in terms of membrane proteins and lumenal proteins. Lysosomes are acidic, with the lumen's acidic pH of around 4.5 to 5.0 (Ohkuma & Poole (1978) PNAS, 75(7):3327-3331, doi: https: / / doi.org / 10.1073 / pnas.75.7.3327) being maintained by pumping protons (H+ ions) in from the cytosol across the membrane via proton pumps and chloride ion channels (Mindell (2012) Annual Review of Physiology, 74(1):69-86, doi: 142317; and Ishida et al. (2013) J. of Gen. Phys., 141(6):705-720, doi: https: / / doi.org / 10.1085%2Fjgp.201210930). Recent studies have also indicated that lysosomes can act as a source of intracellular calcium (Medina et al. (2015) Nat. Cell Biol., 17(3):288-299, doi: https: / / doi.org / 10.1038%2Fncb3114). Thus, in one embodiment the ionic index comprises the mean calcium (Ca2+) levels of the lysosomes. In a further embodiment, the ionic index comprises the mean pH of lysosomes. In a yet further embodiment, the ionic index comprises the 2-ion index of lysosomes. The 2-ion index comprises the ratio of mean pH / mean Ca2+levels. In some embodiments, the ionic index comprises one or more of: the mean calcium (Ca2+) levels, the mean pH and / or the 2-ion index of lysosomes in the PBMCs. In a particular embodiment, the ionic index comprises all of: the mean calcium (Ca2+) levels, the mean pH and the 2-ion index of lysosomes in the PBMCs. As demonstrated by the data presented herein, mean calcium levels in lysosomes of PBMCs from AD patients is reduced compared to non-AD control subjects, while the mean pH and 2-ion index measured in these lysosomes is elevated in AD patients compared to control subjects. Thus, as also demonstrated herein, additionally measuring the ionic index of lysosomes in PBMCs in the methods described herein further improves the accuracy, sensitivity and specificity of diagnosing AD. The term “sample” used herein refers to any sample (e.g. fluid or other tissue sample) obtained from a subject,and encompasses any such sample which may be readily obtained, either invasively or non-invasively, for use in a diagnostic, prognostic or monitoring assay. The sample may be taken at any time, e.g. prior to the onset of AD (such as when a subject is at risk of or suspected as being at risk of developing AD) or after the onset of AD symptoms (such as when the subject is predicted to be suffering from AD). Samples may include, without limitation whole blood, plasma, red blood cells, white blood cells (e.g. peripheral blood mononuclear cells; PBMCs), saliva, urine, stool, tears, sweat, sebum, synovial fluid, cerebrospinal fluid (CSF), lymph, fine needle aspirate, cell lysates, cellular secretion products and biopsies (e.g. tissue biopsies). Preferably the sample is plasma and / or PBMCs. In a particular embodiment, the presence and / or levels of the combination of five biomarkers described herein is measured in a plasma sample obtained from a subject. In another particular embodiment, the ionic index of lysosomes is measured in PBMCs from a subject. PBMCs and plasma may be obtained and / or isolated according to known methods in the art. The terms “subject”, “individual” and “patient” may be used interchangeably herein and refer to any subject for whom prognosis, diagnosis, treatment and / or monitoring is desired, or for whom an AD prediction or likelihood score is desired. The subject may be a mammal, such as a human or a clinical model animal, e.g. a non-human primate, mouse, rat, dog or rabbit. Preferably, the subject is a human. The subject may be asymptomatic for Alzheimer’s disease, such as wherein the subject is suspected as being at risk of AD or of developing AD. Alternatively, the subject may show some signs of AD, such as cognitive decline / impairment, which may be mild or moderate. Still alternatively, the subject may show significant signs of AD, such as severe cognitive impairment. In some embodiments, the subject shows signs of cognitive impairment and it is desired to distinguish whether the subject is suffering from AD or a non-AD pathology or dementia. Thus, in certain embodiments the methods described herein, in particular the likelihood score defined herein, assist the distinguishing of Alzheimer’s disease over non- Alzheimer’s neurodegenerative diseases or dementias. Such distinguishing may comprise a likelihood of the subject having AD over a non-AD disease or dementia. Data provided herein demonstrates the good diagnostic performance of the methods and likelihood score described herein for evaluating the AD status of patients from control non-dementia subjects, as well as from those suffering from non-AD dementias. In further embodiments, the methods described herein, in particular the subject’s risk or likelihood or the subject’s likelihood score defined herein, provide the likelihood of the subject being identified as positive on an amyloid positron emission tomography (PET) scan. As will be readily appreciated, a high risk / likelihood or high likelihood score indicates an increased likelihood that the subject will be identified as positive on an amyloid PET scan. Conversely, a low risk / likelihood or low likelihood score will indicate a decreased likelihood that the subject will be identified as positive on an amyloid PET scan, or an increased likelihood that the subject will be identified as negative. Thus, the methods herein provide an initial screening to help determine patients which should be further investigated using amyloid PET and would likely be identified as positive, distinguishing from those who would likely be identified as negative and for whom amyloid PET is not necessary and thus avoiding unnecessary scan referrals. Such initial screening is useful due to the relative high cost and low availability of amyloid PET scanning, compared to the obtaining of a sample (e.g. a blood / plasma sample) from a subject and measuring the presence / levels of the combination of biomarkers as described herein. By avoiding unnecessary referrals for amyloid PET scans, not only are costs reduced, but the ability to diagnose patients with AD is enhanced and accelerated due to the increased availability of equipment etc. for those subjects deemed likely to have or be at risk of developing AD. As well as being able to provide the likelihood of a subject being identified as positive on an amyloid PET scan, the methods herein have also been found to be in concordance with the likelihood of the subject being identified as positive for AD in an assay for Aβ1-42 / Aβ1-40 ratio, with autopsy and / or with any other clinical diagnoses for AD. Thus, in particular embodiments of the present invention the presence and / or level of the combination of five biomarkers is used to predict the subject’s risk or likelihood of having and / or developing Alzheimer’s disease. This risk or likelihood, i.e. prediction, may be presented as a ‘likelihood score’. As mentioned hereinbefore, this likelihood score will provide an indication as to the subject’s risk or likelihood of having or developing AD, as well as indicate the likelihood of a subject being identified as positive on an amyloid PET scan (or the likelihood of being identified as negative on said amyloid PET scan). Thus, the terms “likelihood”, “prediction / predicting”, “likelihood score” and the like may be used interchangeably herein. As such, in one embodiment the presence and / or levels of the biomarkers is used to generate a likelihood score for the subject having and / or developing Alzheimer’s disease. In a further aspect of the invention, there is provided a method of predicting a subject’s risk or likelihood of having and / or developing Alzheimer’s disease, said method comprising measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from the subject, wherein the biomarkers are: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217). Any mathematical method may be used to calculate the likelihood score from the values obtained from measuring the presence and / or levels of the combination of biomarkers described herein. Suitably however, ratios of the levels of certain biomarkers may be used. In particular, the ratio of Aβ1-42 in the context of total amyloid (i.e. Aβ1-40) levels may be used, i.e. the Aβ1-42 / Aβ1-40 ratio. As described hereinbefore, this ratio is superior to using a value for Aβ1-42 levels alone for the differentiation of AD from non-AD dementias. These previous findings are confirmed by the data presented herein. Since it is known that a decrease in CSF levels of Aβ1-42 is characteristic of AD, a subject may be deemed at risk of or likely to have AD as described herein when said Aβ1-42 / Aβ1-40 ratio is decreased / reduced. Said decrease / reduction is relative and / or compared to the ratio of these levels in a sample from a subject not having AD or having a non-AD neurodegenerative disease or dementia. Also in particular, a ratio of the levels of two hallmark biomarkers in AD, p-Tau181 and Aβ1-42 may be used, such as in the form p-Tau181 / Aβ1-42. This ratio, used alone, has also been described previously (Constantinides et al. ((2023)) for its efficacy in differentiating AD from non-AD dementias, and this is again confirmed herein. The p-Tau181 / Aβ1-42 ratio will be appreciated to be elevated / increased in subjects deemed at risk of, likely to have or have AD, since hyperphosphorylated p-Tau181 is known to be elevated and circulating / CSF Aβ1-42 levels are known to be decreased in AD. Again, said elevation / increase is relative / compared to the ratio of levels in a sample from a non-AD or non-AD dementia subject. As described hereinbefore, it will be readily appreciated that p-Tau217 and / or p-Tau231 may be used together with or in place of p-Tau181 herein. Thus, the p-Tau217 / Aβ1-42 ratio and / or p-Tau231 / Aβ1-42 ratio may be used. Alternatively, the reverse of these ratios may be used, i.e. the Aβ1-40 / Aβ1-42 ratio and / or the Aβ1-42 / p-Tau181 ratio (or the Aβ1-42 / p-Tau217 ratio when p-Tau217 is used in in place of p-Tau181 and / or the Aβ1-4 / p-Tau231 ratio when p-Tau231 is used). If said reversed ratios are used then the resultant values will be appreciated to be elevated / increased for the Aβ1-40 / Aβ1-42 ratio in an at risk or AD having subject, and decreased / reduced for the Aβ1-42 / p-Tau181 ratio (or Aβ1-42 / p-Tau217 and Aβ1-42 / p-Tau231 ratios) compared to a non-AD or non-AD dementia subject. Other ratios may be used in addition or alternatively to their respective individual biomarker levels, such as the NfL / Aβ1-42 and / or GFAP / Aβ1-42 ratios and respective reverse ratios. However, as demonstrated herein, incorporating the NfL / Aβ1-42 and / or GFAP / Aβ1-42 ratios into the present method does not provide an additional advantage for diagnosing AD although no decrease in diagnostic accuracy is observed. Therefore, in some embodiments the NfL / Aβ1-42 and / or GFAP / Aβ1-42 ratios are not use in the method described herein. Thus, in some embodiments a ratio of the level of any one or more of the biomarkers are measured. In a certain embodiment, the ratio of Aβ1-42 / Aβ1-40 levels, the ratio of Aβ1-40 / Aβ1-42 levels, the ratio of p-Tau181 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau181 levels, the ratio of NfL / Aβ1-42 and / or the ratio of Aβ1-42 / NfL are measured. In particular embodiments, the ratio of Aβ1-42 / Aβ1-40 levels and the ratio of p-Tau181 / Aβ1-42 levels are measured. Alternatively or additionally, the Aβ1-42 / p-Tau217 ratio or the p-Tau217 / Aβ1-42 ratio may be used and / or the Aβ1-42 / p-Tau231 ratio or the p-Tau231 / Aβ1-42 ratio. In further embodiments, the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease when: the Aβ1-42 / Aβ1-40 ratio is decreased; the Aβ1-40 / Aβ1-42 ratio is elevated; the level of p-Tau181, p-Tau231 and / or p-Tau217 is elevated; the level of NfL is elevated; the level of GFAP is elevated; the p-Tau181 / Aβ1-42 ratio, the p-Tau217 / Aβ1-42 ratio and / or the p-Tau231 / Aβ1-42 ratio is elevated; the Aβ1-42 / p-Tau181 ratio, the Aβ1-42 / p-Tau217 ratio and / or the Aβ1-41 / p-Tau231 ratio is decreased; the NfL / Aβ1-42 ratio is elevated; and / or the Aβ1-42 / NfL ratio is decreased, in the sample compared to a subject not having Alzheimer’s disease or compared to a subject having a non-Alzheimer’s neurodegenerative disease or dementia. In a particular embodiment, the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease when: the Aβ1-42 / Aβ1-40 ratio is decreased; the level of p-Tau181, p-Tau231 and / or p-Tau217 is elevated; the level of NfL is elevated; the level of GFAP is elevated; and / or the p-Tau181 / Aβ1-42 ratio, the p-Tau217 / Aβ1-42 ratio and / or the p-Tau231 / Aβ1-42 ratio is elevated, in the sample compared to a subject not having Alzheimer’s disease or compared to a subject having a non-Alzheimer’s neurodegenerative disease or dementia. Gender is a risk factor in Alzheimer’s disease, with females more likely to develop AD than males. Therefore, in one embodiment the predictions / likelihood score generated herein additionally comprise using information on the gender of the subject, in particular the encoded gender (i.e. those born with female chromosomes and hormones). References herein to “gender” refer to subjects born with female chromosomes and hormones, i.e. their encoded gender. Due to this increased risk, in certain embodiments the value for the encoded gender of a female subject used in the predictions and likelihood score described herein may be 1, while the value for the encoded gender of a male subject may be 0. However, it is believed that the main reason for females being at greater risk of developing AD is because women live longer than men and old age is the biggest risk factor for the disease. Thus, in a further embodiment the predictions / likelihood score generated herein additionally comprise using information on the age of the subject. In a further embodiment, prediction additionally comprises using information on the gender and / or age of the subject. In a particular embodiment, prediction comprises using information on both the gender and age of the subject. As demonstrated by the data presented herein, further using information on the encoded gender and age of a subject in the prediction and likelihood calculation methods provides further improved evaluation of the AD status of the subject / diagnosis of AD, as well as distinguishing AD from non-AD dementias. The subject’s genetic ApoE status is also a major risk factor in in AD, in particular in late-onset sporadic AD. In particular, the ε4 variant correlates strongly with AD with 40-65% of AD patients having the APOE4 allele, and those harbouring the allele can be 10-30 times more likely to develop AD, albeit not in all ethnic groups (Sadigh-Eteghad et al. (2012) Neurosciences, 17(4):321-326, PMID: 23022896; and Sepehrnia et al. (1989) Am. J. Hum. Genet., 45(4):586-591, PMID: 2491016). The ApoE protein belongs to a family of fat-binding proteins and in astrocytes in the CNS is involved in the transport of cholesterol to neurons (Wang et al. (2021) PNAS, 118(33):2020.06.18.159632, doi: https: / / doi.org / 10.1073%2Fpnas.2102191118). It is the principal cholesterol carrier in the brain (Puglielli et al. (2003) Nature Neurosci., 6(4):345-351, doi: https: / / doi.org / 10.1038%2Fnn0403-345). In AD, ApoE enhances the proteolytic breakdown of amyloid, and the ε4 variant has been shown to be less effective as others in this process (Wisniewski & Frangione et al. (1992) Neurosci Lett., 135(2):235-238, doi: https: / / doi.org / 10.1016%2F0304-3940%2892%2990444-C; and Jiang et al. (2008) Neuron, 58(5):681-693, doi: https: / / doi.org / 10.1016%2Fj.neuron.2008.04.010). Even in the absence of wherein amyloid plaques are thought to drive disease, the involvement of the ε4 variant can still be explained by its interaction with ApoER2, a neuronal reelin receptor, and resulting blocking of reelin signalling which is a key process involved in AD (Kovács (2021) Int. J. Mol. Sci., 23(1):462, doi: https: / / doi.org / 10.3390%2Fijms23010462). Thus, in certain embodiments the predictions / likelihood score generated herein additionally comprises using information on the ApoE status of the subject, in particular the presence of an APOE2, APOE3 or APOE4 allele in the subject. For example, wherein the subject has the ε4 variant they are more likely to be deemed at risk of developing or having AD. In certain embodiments, the value for the presence of the APOE4 allele in the subject used in the predictions and likelihood score described herein may be 1 or 2. Other APOE alleles do not associate with disease (e.g. APOE2 and APOE3) and may therefore be assigned a value of 0 in the predictions and likelihood score calculations described herein. ApoE status may be referred to as “ApoE score” herein. As demonstrated by the data herein, further using information on the ApoE status of a subject in the prediction and likelihood calculation methods provides further improved evaluation of the AD status of the subject / diagnosis of AD, as well as distinguishing AD from non-AD dementias. Calculation of the likelihood score described herein can suitably be based on the log10converted levels of the biomarkers. Furthermore, to account for any cohort specific heterogeneity in biomarker levels, the z-score standardisation method on the log10transformed analyte levels may be used. In particular, z-scores may be calculated with the ^^^^^^^^^^^^^^ formula ^ = and using the mean and standard deviation (SD) from control samples within a cohort, with “Y” representing the calculated z-score. Thus, in one embodiment the presence and / or levels of the biomarkers are normalised by calculating z- score values. As can be seen from the data presented herein, when the z-scores for each biomarker level and ratio of levels is plotted (with z-score on y-axis), significant alterations in the Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231, p-Tau217 and NfL levels in AD samples are revealed, compared to samples from non-AD controls. Furthermore, the Aβ1-42 / Aβ1-40 ratio is notably reduced in non-AD dementia and people with mixed dementia. Among the samples with non-AD dementia, no change in p-Tau181 and NfL levels were observed, while both of these biomarkers were significantly affected in mixed dementia. Additionally, while GFAP levels are elevated in AD, they alone showed no alteration in non-AD dementias, and the p-Tau181 / Aβ1-42 ratio is significantly altered in AD and mixed dementia, with no changes shown in non-AD dementia samples. In further embodiments, the likelihood score described herein is generated using a logistic regression of the presence and / or levels of the biomarkers, and / or of the z-score values. In a particular embodiment, the logistic regression is of the z-score values for the biomarker levels as described herein. In a yet further embodiment, the logistic regression is additionally of the gender, age and / or ApoE status of the subject. According to this embodiment, the value assigned for the gender and / or ApoE status of the subject is as described herein (e.g. female = 1, male = 0, APOE2 / 3 = 0, APOE4 = 0, 1 or 2) and the age is the z-score of the age if the subject in number of years. In a still further embodiment, the logistic regression is additionally of the measurement of the ionic index of lysosomes in PBMCs obtained from the subject as described herein. In a particular embodiment, the logistic regression is of log10values for the ionic index. In another embodiment, the log10values of the ionic index are log10values of the lysosomal features that make up the ionic index as described hereinbefore, i.e. one or more of: the mean calcium (Ca2+) levels, the mean pH and / or the 2-ion index of lysosomes in the PBMCs from the subject. In another particular embodiment, the logistic regression is of z-score values for the ionic index. In a further embodiment, the z-score values of the ionic index are z-scores of the lysosomal features that make up the ionic index. Thus, in certain embodiments of the present invention the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, and p-Tau181 / Aβ1-42 ratio. In a further embodiment, the likelihood score may be generated additionally using the ionic index of lysosomes in PBMCs obtained from the subject and / or the ApoE status of the subject. Thus, in one embodiment the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau181 / Aβ1-42 ratio, and the ionic index of lysosomes in PBMCs obtained from the subject. In another embodiment, the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau181 / Aβ1-42 ratio, and the ApoE status of the subject. In a still other embodiment, the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau181 / Aβ1-42 ratio, the ionic index of lysosomes in PBMCs obtained from the subject and the ApoE status of the subject. According to any of these embodiments, p-Tau217 and / or p-Tau231 levels may be used in place of or together with p-Tau181 as described hereinbefore. Thus, in some embodiments the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau217 and / or p-Tau231, and the p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio. In another embodiment, the likelihood score may be generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio. In another embodiment, the likelihood score may be generated using the levels and / or values for: NfL, GFAP, Aβ1-42, p-Tau181, p-Tau231and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio. In another embodiment, the likelihood score may be generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau217 and / or p-Tau231, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, and the ionic index of lysosomes in PBMCs obtained from the subject. In a further embodiment, the likelihood score may be generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, and the ionic index of lysosomes in PBMCs obtained from the subject. In another embodiment, the likelihood score may be generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau217 and / or p-Tau231, p-Tau217 / Aβ1-42 ratio, and the ApoE status of the subject. In another embodiment, the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, and the ApoE status of the subject. In a still other embodiment, the likelihood score may be generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau217 and / or p-Tau231, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, the ionic index of lysosomes in PBMCs obtained from the subject and the ApoE status of the subject. In a further embodiment, the likelihood score may be generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, the ionic index of lysosomes in PBMCs obtained from the subject and the ApoE status of the subject. In a specific embodiment, the likelihood score is generated using values for the age and encoded gender of the subject, and the z-score values for plasma levels of NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio as independent variables, wherein the encoded gender for a female subject is assigned a value of 0 or 1 and for a male subject is assigned a value of the other of 0 or 1. As will be appreciated according to this specific embodiment, wherein the encoded gender for a female subject is assigned the value of 0, a male subject will be assigned a value of 1 for encoded gender. Alternatively, wherein a female subject is assigned the value of 1 for encoded gender, a male subject will be assigned 0. In a further embodiment, the likelihood score is generated using values for the age and encoded gender of the subject, and the log10 values for plasma levels of NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio as independent variables, and the log10 value for the ionic index of lysosomes in PBMCs obtained from the subject. In a yet further embodiment, the likelihood score is generated using values for the age and encoded gender of the subject, and the log10 values for plasma levels of NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio as independent variables, and a value for the ApoE status of the subject. According to this yet further embodiment, the ApoE status of the subject is assigned a value of 0, 1 or 2 as described hereinbefore and is thus not a log10value as used for the biomarker levels (or a z-score value as described herein), although such a value may be used for the ApoE status of the subject if appropriate. In particular, wherein the subject has an APOE2 or APOE3 allele they are assigned a value of 0, or if they have an APOE4 allele (i.e. the ε4 variant) they are assigned a value of 1 or 2. In a still further embodiment, the likelihood score is generated using values for the age and encoded gender of the subject, and the z- score values for plasma levels of NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio as independent variables, the z-score value for the ionic index of lysosomes in PBMCs obtained from the subject, and a value for the ApoE status of the subject. In further specific embodiments, the likelihood score may be calculated according to the hereinbefore mentioned methods (i.e. using log10or z-score values and a logistic regression thereof) using any one of the following formulae: Formula (I): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))) Formula (II): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))) Formula (III): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (h*gender)))) Formula (IV): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))) Formula (V): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j*ApoE score)))) Formula (VI): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j*ApoE score)))) Formula (VII): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[ Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))) Formula (VIII): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (h*gender) + (j*ApoE score)))) Formula (IX): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))) Formula (X): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))) Formula (XI): 1 / (1 + e^(-(a + (b*age) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))) Formula (XII): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (k*lysosomal ionic index)))) Formula (XIII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XIV): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))) Formula (XV): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))) Formula (XVI): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XVII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XVIII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (n*[Aβ1-42]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XIX): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (n*[Aβ1-42]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))) Formula (XX): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender)))) Formula (XXI): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XXII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (l*[GFAP / Aβ1-42 ratio]) + (f*[p-Tau217]) + (h*gender)))) Formula (XXIII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (l*[GFAP / Aβ1-42 ratio]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))) Formula (XXIV): 1 / (1 + e^(-(-a + (b*age) + (m*[NfL / Aβ1-42 ratio]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender)))) Formula (XXV): 1 / (1 + e^(-(-a + (b*age) + (m*[NfL / Aβ1-42 ratio]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))) Formula (XXVI): 1 / (1 + e^(-(-a + (b*age) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))) Formula (XXVII): 1 / (1 + e^(-(-a + (b*age) + (d*[GFAP]) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))) Formula (XXVIII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))) Formula (XXIX): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XXX): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j*ApoE score)))) Formula (XXXI): 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j*ApoE score)+(n*[Aβ1-42])))) Formula (XXXII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau231]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) Formula (XXXIII): 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau231]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))) According to the above formulae, a, b c, f, g, h and i are positive or negative numbers, and d, j, k, l, m and n are positive numbers. The same integer repeated in a particular formula does not preclude them from differing in their numerical value. These formulae may also be referred to as “equations” and / or “models” herein and throughout the data. As can be seen from the data presented herein, all of formulae (I) to (XXXIII) provide good accuracy, specificity and selectivity for diagnosing AD or distinguishing between AD and non-AD dementias. The hereinbefore described methods, predictions and formulae can be used to generate a likelihood score that can be compared to a threshold value, such as a likelihood score generated from values of biomarker measurements from a sample obtained from a non-AD control subject (e.g. a ‘healthy’ subject). It will be readily appreciated the comparison of the generated likelihood score to a threshold value is desired in order to determine whether the subject has AD, is at risk of developing AD, is likely to develop AD, or has a non-AD dementia. In one embodiment, the generated likelihood score is compared to one or more threshold value, above which the subject is deemed at risk of, is likely to have and / or has AD. Multiple threshold values may be used to stratify subjects according to their risk of AD or AD severity, for example a first threshold value can be used to determine a subject’s risk of AD, while a second higher threshold value can be used to determine that the subject is likely to have AD and a still higher third threshold value can be used to determine that the subject has AD. Alternatively or in addition, different threshold values can be used for different biomarkers or groups of biomarkers. For example, a first threshold value can be used to evaluate the level of the combination of five biomarkers described herein, while a second threshold value can be used to evaluate the ionic index. In a further embodiment, this one or more threshold value is also the value above which it is deemed that the subject has AD rather than any other non- AD dementia. In another embodiment, the likelihood score is compared to one or more threshold value above which the subject is deemed at risk of, is likely to have and / or has AD, and is compared to another one or more threshold value above which the subject is deemed to have or likely to have AD rather than any other non-AD dementia. In a yet further embodiment, wherein the subject’s likelihood value is greater than one or more threshold value it is deemed likely that the subject will be identified as positive on an amyloid positron emission tomography (PET) scan. Thus, in a particular embodiment the subject’s risk or likelihood or the likelihood score is compared to one or more threshold value, over which the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease, and deems it likely that the subject will be identified as positive on an amyloid positron emission tomography (PET) scan. In further embodiments, the subject’s likelihood value deems it likely that the subject will be identified as positive for AD in an assay for Aβ1-42 / Aβ1-40 ratio, with autopsy and / or with any other clinical diagnoses for AD. Thus, according to these further embodiments the subject’s risk or likelihood score is compared to one or more threshold value, over which it is deemed likely that the subject will be identified as positive for AD in an assay for Aβ1-42 / Aβ1-40 ratio, with autopsy and / or with any other clinical diagnoses for AD. As demonstrated by the data presented herein, a suitable threshold value is a likelihood score of about 0.2 or above. Thus, in one embodiment the threshold value is about 0.2 or above. In further embodiments, the threshold value may be a likelihood score of about 0.3 or above. In yet further embodiments, the threshold value may be a likelihood score of about 0.4 or above. In still further embodiments, the threshold value may be between about 0.4 and 0.5, such as about 0.42, about 0.43, about 0.44 or about 0.45. In a further embodiment, the threshold value is a likelihood score of about 0.42, about 0.43, about 0.44, about 0.45, about 0.46 or about 0.47, in particular 0.429, 0.445 or 0.465. Thus, in some embodiments the threshold value is between about 0.42 and about 0.43. In further embodiments, the threshold value may be between about 0.43 and about 0.44. In yet further embodiments, the threshold value is between about 0.44 and about 0.45. In still further embodiments, the threshold value may be between about 0.45 and about 0.46. In further embodiments, the threshold value is between about 0.46 and about 0.47. In a preferred embodiment, the threshold value is a likelihood score of 0.429, such as wherein calculating the likelihood score comprises z-score values of the levels of the combination of five biomarkers described herein in addition to the age, encoded gender and ApoE status of the subject. In another preferred embodiment, the threshold value is a likelihood score of 0.445, such as wherein calculating the likelihood score comprises z-score values of the levels of the combination of five biomarkers described herein. In a further preferred embodiment, the threshold value is a likelihood score of 0.465, such as wherein the likelihood score is used to distinguish AD from other non-AD dementias. In a yet further preferred embodiment, the threshold value is a likelihood score of 0.417, such as wherein calculating the likelihood score comprises measuring the ionic index of lysosomes in peripheral blood mononuclear cells (PBMCs) obtained from the subject. Methods of Diagnosis and Treatment As will be readily appreciated by the disclosures herein, the present methods, predictions and likelihood score may be used to diagnose and / or prognose Alzheimer’s disease in a subject. Thus, in one aspect of the invention there is provided a method of prognosing or diagnosing Alzheimer’s disease in a subject at risk of and / or suspected of having Alzheimer’s disease, said prognosis or diagnosis comprising the methods described herein. In a particular embodiment, the likelihood score is used in the prognosis or diagnosis of Alzheimer’s disease. For example, wherein a subject’s likelihood score is above a threshold as described hereinbefore, the subject is diagnosed or prognosed as having AD. Additionally or alternatively, wherein the subject’s likelihood score is above a threshold value, the subject is diagnosed with AD rather than a non-AD dementia. Thus, the present methods, predictions and likelihood score may facilitate in determining an appropriate treatment regimen for the subject. In particular, to facilitate determining a subject’s likelihood of responding to a particular treatment, such as an AD treatment when they are deemed to be at risk of, is likely to have and / or has AD rather than a non-AD dementia for which AD treatment is not effective. Thus, in a further aspect there is provided a method of treating Alzheimer’s disease in a subject at risk of developing or suspected of having Alzheimer’s disease, said method comprising the steps of: (a) predicting the subject’s risk or likelihood of developing or having Alzheimer’s disease by the methods described herein; and (b) administering a treatment for Alzheimer’s disease to the subject if the risk or likelihood exceeds a threshold value. Suitable AD treatments will be readily recognised by the skilled person, and include those in development or that are yet to be developed but which will treat or ameliorate symptoms of AD diagnosed by the methods, predictions and likelihood score herein. In particular, where no true treatments for AD are known, suitable treatments include those that improve outcome and / or symptoms for patients. For example, the treatment being administered may be a cholinesterase inhibitor (e.g. galantamine, rivastigmine or donepezil) which improve cognitive function, an N-methyl-D-aspartate (NMDA) antagonist (e.g. memantine) which symptoms of cognitive impairment, an orexin antagonist (e.g. suvorexant) used to treat insomnia, some antipsychotics (e.g. brexpiprazole), or may be an immunotherapy which targets amyloid β to reduce amyloid plaques (e.g. lecanemab or aducanumab). As described hereinbefore, the present methods, predictions and likelihood score may be used in combination with a confirmatory method, such as amyloid PET. Thus, in one embodiment step (a) of the methods of treatment described herein may additionally comprise identifying the subject as positive on an amyloid positron emission tomography (PET) scan if the risk or likelihood exceeds a threshold value. In a further embodiment, the methods of treatment may additionally comprise identifying the subject as positive for AD in an assay for Aβ1-42 / Aβ1-40 ratio, with autopsy and / or with any other clinical diagnoses for AD if the risk or likelihood exceeds a threshold value. In particular embodiments, a likelihood score for the subject having and / or developing Alzheimer’s disease is generated in step (a) according to the methods described herein. In a further embodiment, the threshold value used in the present methods of treatment is a likelihood score of about 0.2 or above, about 0.3 or above or about 0.4 or above. In a yet further embodiment, the threshold value is a likelihood score between about 0.4 and about 0.5, between about 0.45 and about 0.46 or between about 0.46 and about 0.47. In a particular embodiment, the threshold value is a likelihood score of about 0.44, about 0.45, about 0.46 or about 0.47, such as 0.445 or 0.465. Preferably, the threshold value is a likelihood score of 0.429, such as wherein calculating the likelihood score comprises z-score values of the levels of the combination of five biomarkers described herein in addition to the age, encoded gender and ApoE status of the subject. Alternatively, the threshold value is a likelihood score of 0.445, such as wherein calculating the likelihood score comprises z-score values of the levels of the combination of five biomarkers described herein. Still alternatively, the threshold value is a likelihood score of 0.465, such as wherein the likelihood score is used to distinguish AD from other non-AD dementias. Further alternatively, the threshold value is a likelihood score of 0.417, such as wherein calculating the likelihood score comprises measuring the ionic index of lysosomes in peripheral blood mononuclear cells (PBMCs) obtained from the subject. Kits and Assays According to a yet further aspect, there is provided an immunoassay array for the measurement of a combination of five biomarkers as described herein. Suitable immunoassays are described hereinbefore and include enzyme-linked immunoassays, NUcleic acid Linked Immuno-Sandwich Assay (NULISA™) and / or in vitro single-molecule array immunoassays. Thus, in one embodiment the immunoassay is an enzyme-linked immunoassay. In a further embodiment, the immunoassay is an in vitro single-molecule array immunoassay. In a yet further embodiment, the immunoassay is an enzyme-linked single- molecule immunoassay. In another embodiment, the immunoassay comprises oligonucleotide linked antibodies. In a particular embodiment, the immunoassay comprises oligonucleotide-conjugated capture and detector antibodies. In a further embodiment, the immunoassay comprises multiple capture and release steps. In a yet further embodiment, the immunoassay further comprises ligation of the antibody-conjugated oligonucleotides, followed by optional quantitative PCR (e.g. RT-PCR) or sequencing. In a particular embodiment, the measurement is in a single sample / single test sample obtained from a subject. Such samples are described hereinbefore and include blood, in particular plasma and PBMCs from a subject. In another aspect, there is provided a kit for the measurement of a combination of five biomarkers as described herein, said kit comprising an immunoassay array or NUcleic acid Linked Immuno-Sandwich Assay kit as described herein. In one embodiment, the kit further comprises instructions for use, such as instructions for calculating a likelihood score using the methods described herein. In a further embodiment, the instructions may provide for the determining of a subject’s risk of having and / or risk of developing AD, such that they are referred to an amyloid PET scan, where if the risk / likelihood score is above a threshold value as described herein they are deemed likely to be identified as positive on said amyloid PET scan. If the subject’s risk / likelihood score is below the threshold value the subject may not be referred to an amyloid PET scan or will be deemed likely to be identified as negative on said scan. Unless defined otherwise, all technical and scientific terms used herein have the meaning commonly understood by a person skilled in the art to which this invention belongs. As may be used herein, the term “about” includes up to and including 10% greater and up to and including 10% lower than the value specified, suitably up to and including 5% greater and up to and including 5% lower than the value specified, especially the value specified. The term “between” as may be used herein includes the values of the specified boundaries. Throughout the specification and the claims which follow, unless the context requires otherwise, the word “comprise”, and variations thereof such as “comprises” and “comprising”, will be understood to imply the inclusion of a stated integer, step, group of integers or group of steps but not to the exclusion of any other integer, step, group of integers or group of steps. In addition, as used herein and the appended claims, the singular forms “a”, “an” and “the” include plural referents, and vice versa, unless the content clearly dictates otherwise. Thus, for example references to “a sample” may include two or more such samples (unless a single test sample is explicitly referred to). It will be understood that all embodiments described herein may be applied to all aspects of the invention and vice versa, and such combinations would be readily apparent from the description provided herein and to those skilled in the art. Other features and advantages of the present invention will be apparent from the description provided herein. It should be understood, however, that the description and the specific examples while indicating preferred embodiments of the invention are given by way of illustration only, since various changes and modifications will become apparent to those skilled in the art. CLAUSES A set of clauses defining the invention, its aspects and embodiments is as follows: 1. A method for measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from a subject, comprising the steps of: (a) obtaining the sample from the subject; and (b) measuring the presence and / or levels of the biomarkers: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, pTau231 and / or p-Tau217) in the sample. 2. The method of clause 1, wherein the presence and / or level of the combination of five biomarkers is used to predict the subject’s risk or likelihood of having and / or developing Alzheimer’s disease. 3. A method of predicting a subject’s risk or likelihood of having and / or developing Alzheimer’s disease, said method comprising measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from the subject, wherein the biomarkers are: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, pTau231 and / or p-Tau217). 4. The method of any one of clauses 1 to 3, wherein a ratio of the level of any one or more of the biomarkers are measured, such as the ratio of Aβ1-42 / Aβ1-40 levels, the ratio of Aβ1-40 / Aβ1-42 levels, the ratio of p-Tau181 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau181 levels, the ratio of p-Tau217 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau217 levels, the ratio of p-Tau231 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau231 levels, the ratio of NfL / Aβ1-42 and / or the ratio of Aβ1-42 / NfL are measured, in particular wherein the ratio of Aβ1-42 / Aβ1-40 levels and the ratio of p-Tau181 / Aβ1-42 levels are measured. 5. The method of any one of clauses 1 to 4, wherein the measurement of the biomarkers is by immunoassay, such as an enzyme-linked immunoassay. 6. The method of clause 5, wherein the immunoassay is an in vitro single-molecule array immunoassay. 7. The method of any one of clauses 1 to 6, wherein the method further comprises measuring the ionic index of lysosomes in peripheral blood mononuclear cells (PBMCs) obtained from the subject. 8. The method of clause 7, wherein the ionic index comprises one or more of: the mean calcium (Ca2+) levels, the mean pH and / or the 2-ion index of lysosomes in the PBMCs, in particular wherein the ionic index comprises all of: the mean calcium (Ca2+) levels, the mean pH and the 2-ion index of lysosomes in the PBMCs, wherein the 2-ion index comprises the ratio of mean pH / mean Ca2+levels. 9. The method of any one of clauses 2 to 8, wherein the prediction additionally comprises using information on the gender and / or age of the subject. 10. The method of any one of clauses 2 to 9, wherein the prediction additionally comprises using information on the ApoE status of the subject, in particular the presence of an APOE2, APOE3 or APOE4 allele in the subject. 11. The method of any one of clauses 1 to 10, wherein the sample is blood, serum or plasma. 12. The method of any one of clauses 1 to 11, wherein the presence and / or levels of the biomarkers is used to generate a likelihood score for the subject having and / or developing Alzheimer’s disease. 13. The method of any one of clauses 1 to 12, wherein the presence and / or levels of the biomarkers are normalised by calculating z-score values. 14. The method of clause 12 or clause 13, wherein the likelihood score is generated using a logistic regression of the presence and / or levels of the biomarkers, and / or of the z-score values. 15. The method of clause 14, wherein the logistic regression is additionally of the gender, age and / or ApoE status of the subject, and / or of the measurement of the ionic index of lysosomes in PBMCs obtained from the subject, optionally of normalised z-score values for the ionic index. 16. The method of any one of clauses 12 to 15, wherein the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, and optionally the ionic index of lysosomes in PBMCs obtained from the subject and / or the ApoE status of the subject. 17. The method of any one of clauses 12 to 16, wherein the likelihood score is generated using values for the age and encoded gender of the subject, and the z-score values for plasma levels of NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio as independent variables, and optionally the z-score value for the ionic index of lysosomes in PBMCs obtained from the subject and / or a value for the ApoE status of the subject, wherein the encoded gender for a female subject is assigned a value of 0 or 1 and for a male subject is assigned a value of the other of 0 or 1, optionally wherein the ApoE status of the subject is assigned a value of 0, 1 or 2. 18. The method of any one of clauses 12 to 17, wherein the likelihood score is calculated using any one of formulae (I) to (XXXIII): (I) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))); (II) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))); (III) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (h*gender)))); (IV) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))); (V) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j*ApoE score)))); (VI) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j*ApoE score)))); (VII) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))); (VIII) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (h*gender) + (j*ApoE score)))); (IX) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender) + (j*ApoE score)))); (X) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))); (XI) 1 / (1 + e^(-(a + (b*age) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))); (XII) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (k*lysosomal ionic index)))); (XIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XIV) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))); (XV) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))); (XVI) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XVII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XVIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (n*[Aβ1-42]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) (XIX) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (n*[Aβ1-42]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))) (XX) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender)))); (XXI) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j* ApoE score)))); (XXII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (l*[GFAP / Aβ1-42 ratio]) + (f*[p-Tau217]) + (h*gender)))); (XXIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (l*[GFAP / Aβ1-42 ratio]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))); (XXIV) 1 / (1 + e^(-(-a + (b*age) + (m*[NfL / Aβ1-42 ratio]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender)))); (XXV) 1 / (1 + e^(-(-a + (b*age) + (m*[NfL / Aβ1-42 ratio]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))); (XXVI) 1 / (1 + e^(-(-a + (b*age) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))); (XXVII) 1 / (1 + e^(-(-a + (b*age) + (d*[GFAP]) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))); (XXVIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))); (XXIX) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XXX) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j*ApoE score)))); (XXXI) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j*ApoE score)+(n*[Aβ1-42])))); (XXXII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau231]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XXXIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau231]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))), wherein a, b c, f, g, h and i are positive or negative numbers, and d, j, k, l, m and n are positive numbers. 19. The method of any one of clauses 2 to 18, wherein the subject’s risk or likelihood or the likelihood score provides the likelihood of the subject being identified as positive on an amyloid positron emission tomography (PET) scan. 20. The method of any one of clauses 1 to 19, wherein the method assists the distinguishing of Alzheimer’s disease over non-Alzheimer’s neurodegenerative diseases or dementias. 21. The method of clause 20, wherein the likelihood score provides the likelihood of the subject having Alzheimer’s disease over non-Alzheimer’s neurodegenerative diseases or dementias. 22. The method of any one of clauses 2 to 21, wherein the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease when: the Aβ1-42 / Aβ1-40 ratio is decreased; the Aβ1-40 / Aβ1-42 ratio is elevated; the level of p-Tau181, p-Tau231 and / or p-Tau217 is elevated; the level of NfL is elevated; the level of GFAP is elevated; the p-Tau181 / Aβ1-42 ratio, the p-Tau217 / Aβ1-42 ratio and / or the p-Tau231 / Aβ1-42 ratio is elevated; and / or the Aβ1-42 / p-Tau181 ratio, the Aβ1-42 / p-Tau217 ratio and / or the Aβ1-42 / p-Tau231 ratio is decreased, in the sample compared to a subject not having Alzheimer’s disease or compared to a subject having a non-Alzheimer’s neurodegenerative disease or dementia. 23. The method of any one of clauses 2 to 22, wherein the subject’s risk or likelihood or the likelihood score is compared to one or more threshold value, over which the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease, and / or deems likely that the subject will be identified as positive on an amyloid positron emission tomography (PET) scan. 24. The method of clause 23, wherein the threshold value is a likelihood score of about 0.2 or above, such as about 0.3 or above or about 0.4 or above. 25. The method of clause 23 or clause 24, wherein the threshold value is a likelihood score of between about 0.4 and about 0.5, such as about 0.45. 26. The method of any one of clauses 23 to 25, wherein the threshold value is a likelihood score of about 0.42, about 0.43, about 0.44, about 0.45, about 0.46 or about 0.47, such as 0.429, 0.445 or 0.465. 27. A method of prognosing or diagnosing Alzheimer’s disease in a subject at risk of and / or suspected of having Alzheimer’s disease, said prognosis or diagnosis comprising the method of any one of clauses 1 to 27. 28. The method of clause 27, wherein the likelihood score is used in the prognosis or diagnosis of Alzheimer’s disease. 29. A method of treating Alzheimer’s disease in a subject at risk of developing or suspected of having Alzheimer’s disease, said method comprising the steps of: (a) predicting the subject’s risk or likelihood of developing or having Alzheimer’s disease by the method of any one of clauses 2 to 28; and (b) administering a treatment for Alzheimer’s disease to the subject if the risk or likelihood exceeds a threshold value. 30. The method of clauses 29, wherein step (a) additionally comprises identifying the subject as positive on an amyloid positron emission tomography (PET) scan if the risk or likelihood exceeds a threshold value. 31. The method of clause 29 or clause 30, wherein a likelihood score for the subject having and / or developing Alzheimer’s disease is generated according to the methods of any one of clauses 12 to 18, optionally wherein the threshold value is a likelihood score of about 0.2 or above, about 0.3 or above or about 0.4 or above, such as between about 0.4 and about 0.5, preferably about 0.42, 0.429, about 0.43, about 0.44, 0.445, about 0.45, about 0.46, 0.465 or about 0.47. 32. An immunoassay array for the measurement of a combination of five biomarkers according to clause 1, optionally wherein the measurement is in a single test sample obtained from a subject. 33. The immunoassay array of clause 32, wherein the immunoassay is an enzyme-linked immunoassay and / or wherein the immunoassay is an in vitro single-molecule array immunoassay. 34. The immunoassay array of clause 32 or clause 33, wherein the immunoassay comprises oligonucleotide linked antibodies, optionally comprising oligonucleotide-conjugated capture and detector antibodies, comprising multiple capture and release steps, and / or further comprising ligation of the antibody-conjugated oligonucleotides, followed by optional quantitative PCR or sequencing. 35. A kit for the measurement of a combination of five biomarkers according to clause 1, said kit comprising the immunoassay array of clause 32 or clause 33. The invention will now be described using the following, non-limiting examples: EXAMPLES Materials and Methods Sample Collection Blood samples were obtained in summary, these blood samples were collected in K2-EDTA tubes, centrifuged to isolate plasma, aliquoted in 1 mL tubes and stored at −80°C until used for analysis. Development of the multi-analyte assay is performed on the data collected from 97 samples across seven study cohorts comprised of: (i) 32 samples (control: 18 and AD: 14) were purchased from a Clinical Research Organization (CRO in California) consisting of 14 FDG-PET positive AD samples. This sample set represents the U.S. cohort; (ii) 20 non-demented controls consisting of 10 autopsy confirmed and 10 CSF Aβ ratio negative controls; (iii) 28 with AD consisting of 15 autopsy confirmed and 13 CSF Aβ ratio positive; (iv) 2 FTD consisting of 1 clinically and 1 autopsy confirmed samples; (v) 4 samples with LBD consisting of 3 clinically and 1 autopsy confirmed samples; (vi) 6 vascular dementia (VAD) including 1 autopsy confirmed and 5 clinically diagnosed samples; and (vii) 5 autopsy confirmed mixed dementia samples; among them 3 having comorbidity of AD and VAD, while 2 subjects had AD and LBD. Sample Processing and Measurement of Combination of 5 Biomarkers a) Measurement of 4 Biomarkers Using Neuro 4-Plex E Ninety-seven human plasma samples from patients were tested for the quantitative determination of biomarkers Neurofilament-Light (NfL), Glial Fibrillary Acidic Protein (GFAP), Abeta 1-40 (Aβ1-40), Abeta 1-42 (Aβ1-42) at Quanterix, (Billerica, MA, USA) using their Simoa® Neuro 4-Plex E Advantage kit (Product #103670; Thijssen et al. (2022)). Commercially available Simoa® Neuro 4-Plex E Advantage kits were used according to manufacturer’s instructions. The assays were performed on the Simoa HD-X analyzer using Single Molecule Array (Simoa) technology (Rissin et al. (2010); and Wilson et al. (2016)). Plasma samples were diluted 4x and ran in duplicate. A four-parameter logistic curve fit data reduction method was used to generate a calibration curve (5-parameter for Aβ1-42 and Aβ1-40) for each analyte. Two control samples of known concentration of the protein of interest (high-control and low-control) were included as quality control. The mean of replicates for each sample was calculated and represented graphically. Results were included in the analysis if the coefficient of variation across replicates was <25%. b) Measurement of p-Tau181 Plasma samples from 97 patients were tested for quantitative determination of p-Tau181 at Quanterix, (Billerica, MA, USA) using their Simoa® Human p-Tau181 Advantage V2 assay kit (Product #103714). Commercially available Simoa® p-Tau181 Advantage kits were used according to manufacturer’s instructions. The assays were performed on the Simoa HD-X analyzer using Single Molecule Array (Simoa) technology. Plasma samples were diluted 4x and ran in duplicate. The mean of replicates for each sample was calculated and represented graphically. Results were included in the analysis if the coefficient of variation across replicates was <25%. NULISA plasma biomarker analysis Human plasma samples stored at -80°C were thawed and centrifuged at 10,000g for 10 minutes. Then, 30µL supernatant samples were plated in 96-well plates and analysed with Alamar’s NULISAseq™ CNS Disease Panel 120. The NULISAseq workflow involved immunocomplex formation using DNA-barcoded capture and detection antibodies, followed by capturing and washing the immunocomplexes on paramagnetic oligo-dT beads. Subsequently, the immunocomplexes were released into a low-salt buffer, captured and washed on streptavidin beads. Finally, the proximal ends of the DNA strands on each immunocomplex were ligated to generate a DNA reporter molecule containing both target- specific and sample-specific barcodes. The DNA reporter molecules were pooled and amplified by PCR, purified, and sequenced on Illumina NextSeq 2000. The sequencing data were processed using the NULISAseq algorithm from Alamar Biosciences. The data were rescaled and log2 transformed to obtain NULISA Protein Quantification (NPQ) units for downstream statistical analysis. Statistical Methods Statistical analyses were performed in Origin Pro2022 and python, using the scikit-learn module. A logistic regression equation (Log-likelihood) or a bias-reduced binary logistic regression (maximum penalized likelihood, used for equations involving ApoE status; Firth (1993)) was performed using diagnosis (control or AD) as the dependent variable and using age, encoded gender (female = 1; male = 0), and the z-score values of plasma NfL, GFAP, Aβ1-42 and Aβ1-40 (used as an Aβ1-42 / Aβ1-40 ratio), p-Tau181, p-Tau181 / Aβ1-42 as independent variables and ApoE status (E2 / E3 = 0 and E4 = 0,1 or 2). The AD likelihood score was calculated using the formulae as described herein, or specifically: Equation 1: 1 / (1+EXP(-(-11.4 + (0.114*F) + (1.42*N) + (2.64*Q) + (1.54*Z) + (0.845*AD) + (0.63*AH) + (-1.51*H)))) Equation 2: 1 / (1+EXP(-(0.666 + (-0.0228*F) + (-0.522*N) + (1.92*Q) + (0.843*Z) + (0.0244*AD) + (0.401*AH) + (-1.65*H)))) Equation 3: 1 / (1+EXP(-(-0.695 + (-0.0113*F) + (-1.63*N) + (3.3*Q) + (1.74*Z) + (0.109*AD) + (-0.505*AH) + (-2.43*H)))) Equation 4: 1 / (1+EXP(-(0.714 + (-0.0286*F) + (-0.601*N) + (2.01*Q) + (1.03*Z) + (-0.087*AD) + (0.978*AH) + (-0.923*H)))) Equation 5: 1 / (1+EXP(-(-0.284 + (-0.0129*F) + (-0.0536*N) + (1.68*Q) + (0.514*Z) + (-0.0639*AD) + (0.644*AH) + (-0.972*H)))) Equation 6: 1 / (1+EXP(-(7.85 + (-0.134*F) + (0.298*N) + (3.06*Q) + (-0.00175*Z) + (-0.524*AD) + (1.31*AH) + (-2.94*H)))) Equation 7: 1 / ( 1+EXP(-(-1.84 + (-0.00509*F) + (0.738*N) + (1.37*Q) + (1.01*Z) + (0.392*AD) + (0.825*AH) + (0.0402*H)))) Equation 8: 1 / (1+EXP(-(-0.0145 + (-0.0158*F) + (0.169*N) + (1.69*Q) + (0.989*AH) + (-1.04*H)))) Equation 9: 1 / (1+EXP(-(-0.142 + (-0.0152*F) + (-0.106*N) + (1.62*Q) + (0.994*Z) + (-0.936*H)))) Equation 10: 1 / (1+EXP(-(-0.394 + (-0.0114*F) + (-0.0265*N) + (1.69*Q) + (0.481*Z) + (0.695*AH) + (-0.965*H)))) Equation 11: 1 / (1+EXP(-(-2.33 + (0.00854*F) + (1.13*N) + (1.34*Z) + (0.189*AD) + (-0.961*H) + (4.94*AE)))) Equation 12: 1 / (1+EXP(-(-0.81 + (-0.0125*F) + (0.73*N) + (0.567*Q) + (1.32*Z) + (0.158*AD) + (-0.875*H) + (4.37*AE)))) Equation 13: 1 / (1+EXP(-(-2.85 + (0.0118*F) + (1.11*N) + (0.581*Q) + (0.876*Z) + (0.476*AD) + (0.905*AH) + (-0.876*H) + (4.55*AE)))) Equation 14: 1 / (1+EXP(-(-0.296 + (-0.0202*F) + (0.6683*N) + (0.598*Q) + (1.43*Z) + (-0.927*H) + (4.71*AE)))) Equation 15: 1 / (1+EXP(-(-0.0989 + (-0.0174*F) + (1.08*N) + (0.618*Q) + (0.807*AH) + (-1.15*H) + (3.75*AE)))) Equation 16: 1 / (1+EXP(-(-1.6 + (0.00394*F) + (1.46*N) + (0.864*AH) + (-1.21*H) + (4.41*AE)))) Equation 17: 1 / (1+EXP(-(1.69 + (-0.036*F) + (1.22*Q) + (0.836*AH) + (-1.11*H) + (2.88*AE)))) Equation 18: 1 / (1+EXP(-(-2.36 + (0.00978*F) + (1.4*N) + (0.587*Q) + (0.477*AD) + (1.11*AH) + (-1.14*H) + (3.83*AE)))) Equation 18b: 1 / (1+EXP(-(-0.121 + (-0.185*F) + (2.374*N) + (4.447*Q) + (16.622*Z) + (-1.776*AD) + (0.287*H) + (10.332*AF)))) Equation 19: 1 / (1+EXP(-(-1.641 + (0.004*F) + (-0.228*N) + (1.315*Q) + (0.106*AD) + (1.275*AM) + (-1.172*H)))) Equation 20: 1 / (1+EXP(-(-1.572 + (0.004*F) + (-0.239*N) + (1.335*Q) + (1.226*AM) + (-1.207*H)))) Equation 21: 1 / (1+EXP(-(-2.144+ (0.009*F) + (-0.119*N) + (1.150*Q) + (0.184*AD) + (1.174*AM) + (-1.114*H) + (1.115*AE)))) Equation 22: 1 / (1+EXP(-(-2.011+ (0.008*F) + (-0.150*N) + (1.196*Q) + (1.098*AM) + (-1.175*H) +(1.047AE)))) Equation 23: 1 / (1+EXP(-(-2.088+ (0.008*F) + (-0.309*N) + (1.071*Q) + (0.376*AL) + (1.282*AM) + (-1.158*H) + (1.165*AE)))) Equation 24: 1 / (1+EXP(-(-1.551 + (0.003*F) + (-0.355*N) + (1.256*Q) + (0.277*AL) + (1.368*AM) + (- 1.187*H)))) Equation 25: 1 / (1+EXP(-(0.086 + (-0.016*F) + (-0.285*N) + (0.686*Q) + (1.185*AG) + (0.219*AD) + (- 0.495*H) ))) Equation 26: 1 / (1+EXP(-(-0.804 + (-0.007*F) + (-0.166*N) + (0.632*Q) + (1.043*AG) + (0.279*AD) + (-0.664*H) +(1.636*AE)))) Equation 27: 1 / (1+EXP(-(-2.585 + (0.019*F) + (-0.239*N) + (1.414*Q) + (1.040*AH) + (-0.193*AD) + (-1.146*H) +(1.549*AE)))) Equation 28: 1 / (1+EXP(-(-0.720 + (-0.008*F) + (-0.620*N) + (1.248*AJ) + (1.305*AG) + (-1.237*H)))) Equation 29: 1 / (1+EXP(-(-1.401 + (-0.000*F) + (-0.499*N) + (1.126*AJ) + (1.184*AG) + (-1.304*H) +(1.209*AE)))) Equation 30: 1 / (1+EXP(-(-0.076 + (-0.016*F) + (0.631*AK) + (0.593*Q) + (0.976*AG) + (-0.900*H)))) Equation 31: 1 / (1+EXP(-(-1.097+ (-0.005*F) + (0.586*AK) + (0.553*Q) + (0.880*AG) + (-1.031*H) +(1.621*AE)))) Equation 32: 1 / (1+EXP(-(-3.530 + (-0.016*F) + (2.480*AF) + (3.299*AG) + (-1.637*H)))) Equation 33: 1 / (1+EXP(-(-0.103+ (-0.019*F) + (0.139*N) + (0.158*Q) + (1.025*AG) + (-0.901*H)))) Equation 34: 1 / (1+EXP(-(-3.427+ (-0.040*F) + (0.739*Q) + (3.208*AF) + (2.358*AG) + (0.559*H)))) Equation 35: 1 / (1+EXP(-(5.245+ (-0.134*F) + (1.081*N) + (1.582*AD) + (1.139*Q) + (2.265*AG) + (-1.594*H)))) Equation 36: 1 / (1+EXP(-(-3.624+ (-0.028*F) + (0.887*N) + (0.275*Q) + (2.741*AF) + (1.767*AG) + (1.079*H)))) Equation 37: 1 / (1+EXP(-(-2.423+ (0.014*F) + (-0.302*N) + (1.156*Q) + (0.256*AN) + (1.031*AM) + (-1.134*H)+( 1.052*AE)))) Equation 38: 1 / (1+EXP(-(-2.417+ (0.014*F) + (-0.266*N) + (1.168*Q) + (0.198*Z) + (1.034*AM) + (-1.150*H) + (1.114*AE)))) wherein F = age, N = NfL, Q = GFAP, Z = p-Tau181, AD = Aβ1-42 / Aβ1-40 ratio, AH = p-Tau181 / Aβ1-42 ratio, H = encoded gender, AE = ApoE status, AF = lysosomal ionic index, AG = p-Tau217, AM = p-Tau217 / Aβ1-42 ratio, AJ = GFAP / Aβ1-42 ratio, AK = NfL / Aβ1-42 ratio, AL = Aβ1-42, AN = p-Tau231. ApoE Genotyping Cultured fibroblasts (2-5 x 105) were trypsinized and pelleted by centrifugation at 500g for 5 mins. Cell pellets were washed with PBS and frozen at -80ºC till further processing. Genomic DNA was isolated using Purelink Genomic DNA Mini Kit (Thermo Scientific, Cat no: K182001) according to manufacturer’s instructions. ApoE genotyping was performed according to published literature (Pantelidis et al. (2003)). Briefly, the primer combinations detailed in Table 1 were used, and a touchdown PCR was performed using Taq PCR Core Kit (Qiagen, Cat No: 201223). Table 1 – PCR Primers Used for ApoE Genotyping Primer Name Sequence Comment SEQ ID NO MDL-1  CGGACATGGAGGACGTGT 1 Combine with primer MDL-1 to 2 MDL-2 CTGGTACACTGCCAGGCG identify E3 allele Combine with primer MDL-1 to 3 MDL-3 CTGGTACACTGCCAGGCA identify E2 allele Combine with primer MDL-2 to 4 MDL-4 CGGACATGGAGGACGTGC  identify E4 allele The PCR products were analyzed on a 2% agarose gel in TAE buffer. The allelic combination was determined by presence or absence of amplification in the appropriate primer combination. Monocyte Isolation PBMCs were processed from whole blood within 2-6 hours post-collection. The Cell Preparation Tubes (CPT) containing blood samples were centrifuged at room temperature (18- 25°C) in a horizontal rotor at 2000xg for 20 minutes. After centrifugation, approximately 3- 4mL of plasma were aspirated and discarded, leaving 1-2mL of plasma above the PBMC layer. The remaining plasma and the entire PBMC layer from the CPT were transferred into two separate conical tubes. Tube containing cells were washed in Hank’s Balanced Salt Solution (HBSS) and centrifuged at 1500 rpm for 15 minutes. Cell pellets were collected and resuspended in HF (95% HBSS and + 5% Fetal Bovine Serum), followed by another centrifugation at 1500 rpm for 10 minutes. Excess HF were removed without disturbing the cell pellet and cells were resuspended in HBSS. Monocytes were prepared from PBMCs using Dynabeads™ Untouched™ Human Monocytes Kit according to manufacturer’s protocol. Monocytes were cryopreserved in 1ml of Freezing Medium and aliquoted into cryovials, and immediately stored upright in a StrataCooler followed by transferring in a -80°C freezer. Cryopreserved monocytes from PBMCs were transferred to liquid nitrogen after a few days. Propagation of Monocytes To propagate monocytes, the cells were removed from liquid nitrogen and thawed quickly using 1mL of blood cell growth medium and transferred into a 50mL falcon tube containing 19 mL of blood cell growth medium to dilute DMSO present in the freezing media. The tubes were centrifuged at 600 RCF to pellet the cells, and they were resuspended in 1mL of blood cell growth medium. The cells were transferred to a T-25 flask containing 5mL of blood cell growth medium and kept in the incubator for 4 hours to allow for recovery. The medium was gently aspirated and discharged to break apart cell aggregates and a further incubation of 14 hours will be performed in blood cell growth media. pH and Ca2+Sensor Incubation in PBMCs 1mL of cells from T25 flasks were centrifuged at 600 RCF and resuspended in 200µl of monocyte attachment medium. 100µl of cells were plated in duplicate in a 96-well plate. After adhering for 1hr, cells were treated with 3µl of CalipHluor (20µM) (final concentration of approx. 600nM) and incubated for 30 min. Cells were gently washed twice with 100µl of monocyte attachment medium to remove CalipHluor excess and 200µl of attachment medium were added. After 60min of incubation, the cells were imaged using high content imaging platform. Imaging and Image Analysis Cells were imaged in Green (G, 488nm, for pH), Orange (O, 550nm, for Ca2+), and Red (R, 650 nm, for Normalizing Dye) using the In Cell Analyzer 6000 high-throughput confocal microscope. Image analysis for quantification of pH and calcium in a single lysosome was performed using software built in-house. An in vitro calibration curve was obtained from the bead images acquired at different pH values in low and high Ca2+buffers. pH and Ca2+values from individual lysosomes were calculated by comparing the values obtained from cells with the calibration curve. Example 1: Biomarker Levels in Plasma are Differentially Affected in Subjects with Various Neurocognitive Impairments Figure 1 shows changes in different plasma biomarkers in controls (non-demented), AD, other non-AD dementia (Frontotemporal dementia (FTD), Lewy-body dementia (LBD) and Vascular dementia (VD)). To ensure that the concentration of each of the five analytes in plasma were normally distributed, log10 converted values were used to develop the formulae and validate the results. To account for cohort specific heterogeneity in analyte levels across the different banked samples, the z-score standardisation method was used on the log10 transformed analyte levels. The mean and standard deviation (SD) was calculated from the control samples within each cohort, and the z-scores for the individuals were computed with the ^^^^^^^^^^^^^^ formula ^ = ^^^^^^^^^ . Z-score values are plotted between the study groups for each biomarker. As can be seen in Figure 1, significant alterations in the Aβ1-42 / Aβ1-40 ratio (‘A’), p-Tau181 (‘T’) and NfL (‘N’) levels were revealed in AD samples. Additionally, the Aβ1-42 / Aβ1-40 ratio is also notably reduced in non-AD dementia and people with mixed dementia. Among the samples with non-AD dementia, no change in p-Tau181 and NfL levels was observed, while both of these biomarkers were significantly affected in mixed dementia. These findings strongly suggest that the 'A / T / N' status from plasma is differentially affected in people with various neurocognitive impairments. The observed changes in these biomarkers may therefore serve as potential indicators to distinguish between AD from control, non-AD dementia, and mixed dementia cases. Figure 2 illustrates changes in two further plasma biomarkers (GFAP and p-Tau181 / Aβ1-42 ratio) among individuals with non-demented conditions, AD, and other non-AD dementia. GFAP is a reliable indicator of inflammation and astrocyte reactivity, and is significantly affected in AD and mixed dementia. However, non-AD dementia did not show any alteration in GFAP levels. Similarly, p-Tau181 / Aβ1-42 ratio is correlated with the amyloid PET status of subjects and is significantly altered in AD and mixed dementia, while non-AD dementia samples did not show any changes in this ratio. Example 2: Diagnostic Performance of Different Biomarker Level Combinations Figure 3 demonstrates the diagnostic performance of different combinations of NfL, GFAP, Aβ1-42, Aβ1-40, and p-Tau181. AD likelihood scores were calculated by combining: age + encoded gender + Aβ1-42 / Aβ1-40 + p-Tau181 + NfL (“ATN”); ATN + GFAP (ATN+I); ATN + I + p-Tau181 / Aβ1-42 (“Esya multi-analyte assay”); and ATN + I + p-Tau181 / Aβ1-42 + Aβ1-42 (“Esya multi-analyte assay+Aβ42”), and were plotted for 38 control and 42 AD individuals. According to these assay results, the accuracy gradually improved from using a combination of 3 markers (“ATN”) to 4 biomarkers (ATN with GFAP). However, the maximum accuracy of 86.25% with 88.09% sensitivity and 84.21% specificity was achieved by using an optimal combination of 5 biomarkers (Aβ1-42 / Aβ1-40 + p-Tau181+ NfL + GFAP + p-Tau181 / Aβ1-42) along with age and gender. Further addition of biomarkers or features did not improve the accuracy of the test. Therefore, it is concluded that the optimal model for the multi-analyte assay is to use 5 biomarkers along with age and gender, with the greatest accuracy, sensitivity and selectivity to generate an AD likelihood score produced. Example 4: Various Formulae / Equations Using the Optimal Model Yield Good Diagnostic Performance The multi-analyte test's diagnostic performance is shown in Figure 4. Logistic regression equations were developed by adjusting the coefficient of each biomarker, including age and gender. The equations include both positive and negative integers for NfL, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau181 / Aβ1-42, age and gender. GFAP had only positive integers in all equations. Table 2 provides detailed variations of these integers with respect to a defined equation. The coefficients and complete equations are mentioned hereinbefore. Table 2 – Coefficients Used in Different Plasma Biomarkers During Development of the Multi-Analyte Assay Eqn Eqn Eqn Eqn Eqn Eqn Eqn Eqn Eqn Eqn Features 1 2 3 4 5 6 7 8 9 10 Intercept (a) - + - + - + - - - - Age (b) + - - - - - - - - - NfL (c) + - - - - + + + - - GFAP (d) + + + + + + + + + + p-Tau181 (f) + + + + + - + n / a + + Aβ1-42 / + + + - - - + n / a n / a n / a Aβ1-40 (g) p-Tau181 / + + - + + + + + n / a + Aβ1-42 (i) Gender (h) - - - - - - + - - - Example 5: Diagnostic Performance of Optimal Model / Multi-Analyte Test Figure 5 illustrates the accuracy of the multi-analyte test in diagnosing AD and differentiating it from non-AD dementia. Equation 1 was utilized to demonstrate the assay's ability to differentiate between AD and non-AD dementia. The comparison included 12 non-AD dementia samples (2 FTD, 4 BD and 6 VD), 5 samples with mixed dementia (3 with AD+VD and 2 with AD+LBD), 38 control samples and 42 AD samples. The assay correctly identified 10 out of 12 non-AD dementia samples with an AD likelihood score below the cut-off, resulting in a specificity of 83.33%. 36 out of 42 samples were also identified with an AD likelihood score above the cut-off, resulting in a sensitivity of 85.71%. Furthermore, the assay detected 4 out of 5 samples with mixed dementia and an AD likelihood score above the cut-off, indicating an 80% sensitivity in identifying AD within mixed dementia. Example 6: Automated Calculation of the AD Likelihood Score Figure 6 describes automated calculation of multi-analyte assay results. The final results of the assay are calculated through an automated analytical tool that combines age, gender and cohort of the subject under investigation with the biomarker levels (obtained from HD-X analyzer) of NfL, GFAP, Aβ1-42 and Aβ1-40 (used as an Aβ1-42 / Aβ1-40 ratio) and p-Tau181. The analytical tool is a precision application developed by the inventors that performs (a) z- score standardisation to correct for any preanalytical / analytical variability, (b) uses these z- score values to compute an AD likelihood score, and (c) provides an output that graphically represents the results of the multi-analyte assay. In the graphical output, the reference ranges of the biomarkers (95% CI) from cohort-matched controls are shown as solid vertical grey lines and the biomarker levels of the subject under investigation are represented as green dots. The three steps to operate the assay analytical tool are demonstrated as follows: In Step 1, launch the Esya multi-analyte assay analytical tool. In Step 2, input the values of the biomarkers (as obtained from HD-X analyzer) of NfL, GFAP, Aβ1-42, Aβ1-40, and p- Tau181, along with age, gender and cohort of the subject under investigation. In Step 3, press the calculate button to display the biomarker z-score values and the AD likelihood score. Example 7: Diagnostic Performance of Different Biomarker Level Combinations in further Combination with ApoE Status Figure 7 describes diagnostic performance of different combinations of NfL, GFAP, Aβ1-42, Aβ1-40 and p-Tau181, along with genetic risk factor ApoE4. AD likelihood score were calculated by combining: age + gender + Aβ1-42 / Aβ1-40 + p-Tau181 + NfL + ApoE status (“ATN + ApoE”); ATN + GFAP + ApoE status (“ATN+I+ApoE”); and ATN+ I + p-Tau181 / Aβ1-42 + ApoE status (“Esya multi-analyte assay + ApoE”), and plotted for 24 control and 29 AD individuals. Addition of ApoE status into the model showed significant increase in accuracy of the ATN model (77.5 % for ATN against 92.15% for ATN+ApoE). The accuracy marginally improved from using a combination of 4 markers (ATN + ApoE) to 6 biomarkers (Esya multi- analyte assay + ApoE). The maximum accuracy of 94.11% with 96.29% sensitivity and 91.66% specificity was achieved by using an optimal combination of 6 biomarkers (Aβ1-42 / Aβ1-40 + p-Tau181 + NfL + GFAP + p-Tau181 / Aβ1-42 + ApoE status) along with age and gender. Therefore, it is concluded that plasma biomarkers along with ApoE status provide the best results for diagnosing AD. The multi-analyte test's diagnostic performance after addition of ApoE status is shown in Figure 8. The equations include different combinations NfL, Aβ1-42 / Aβ1-40 ratio, p-Tau181 and p-Tau181 / Aβ1-42, along with ApoE status, age and gender. Table 3 provides detailed variations of different coefficients associated with these equations. The coefficients and complete equations are mentioned hereinbefore. Table 3 – Coefficients Used in Different Plasma Biomarkers During Development of the Multi-Analyte Assay Features Eqn Eqn Eqn Eqn Eqn Eqn Eqn Eqn 18 11 12 13 14 15 16 17 Intercept (a) - - - - - - + - Age (b) + - + - - + - + NfL (c) + + + + + + n / a + GFAP (d) n / a + + + + n / a + + p-tau181 (f) + + + + n / a n / a n / a n / a Ab1-42 / Ab1-40(g) + + + n / a n / a n / a n / a + p-tau181 / Ab1-42n / a n / a + n / a + + + + (i) Gender (h) - - - - - - - - ApoE4 status + + + + + + + + (j) Example 8: Lysosomal Ionic Index Adds an Additional Modality Figure 9 demonstrates the changes in lysosomal ion levels (Ca2+and pH) and lysosomal ionic ^^^^ ^^^^^^^^^ ^^ index (^^^^^^^^^ ^^^^^ ^^^^^ = ^^^^ ^^^^^^^^^ ^^^^^^^) observed among control individuals (healthy) and Alzheimer's disease (AD) patients. The mean lysosomal Ca2+, is decreased in AD (although statistically not significant). Similarly, the mean lysosomal pH and lysosomal ionic index, are both significantly increased in AD compared to healthy controls. Example 9: Adding Organellar Ionic Index Improves the Multi-Analyte Test Figure 10 describes diagnostic performance of the combination of NfL, GFAP, Aβ1-42, Aβ1-40, and p-Tau181 along with lysosomal ionic index. AD likelihood score calculated by combining age + gender + Aβ1-42 / Aβ1-40 + p-Tau181 + NfL + GFAP (“Esya multi-analyte assay”), Esya multi-analyte assay + lysosomal ionic index (“Esya multi-modal assay”) and plotted for 20 control and 13 AD individuals. The addition of lysosomal ionic index status into the model provided significant increase in accuracy of the only plasma multi-analyte assay model (84.8% for multi-analyte assay against 93.9% for Esya multi-modal assay). Therefore, it is concluded that plasma biomarkers along with lysosomal ionic index provide the best results for diagnosing AD. Example 10: p-Tau217 Plasma Levels can be Used Additionally or Alternatively to the Plasma Biomarker p-Tau181 The multi-analyte test's diagnostic performance after addition of p-Tau217 status is described in Figure 12. The equations include different combinations NfL, A ^1-42, A ^1-42 / Ab1-40 ratio, p-Tau217, p-Tau217 / A ^1-42along with ApoE4 status, age, and gender. Table 4 provides detailed variations of different coefficients associated with these equations. The coefficients and complete equations are mentioned hereinbefore. Table 4 – Coefficients Used in Different Plasma Biomarkers During Development of the Multi-Analyte Assay Features Eqn Eqn Eqn Eqn Eqn Eqn Eqn Eqn Eqn 19 20 21 22 23 24 25 26 27 Intercept (a) - - - - - - + - - Age (b) + + + + + + - - + NfL (c) - - - - - - - - - GFAP (d) + + + + + + + + + p-Tau217 (f) n / a n / a n / a n / a n / a n / a + + n / a A ^1-42(n) n / a n / a n / a n / a + + n / a n / a n / a A ^1-42 / A ^1-40+ n / a + n / a n / a n / a + + - (g) p-Tau181 n / a n / a n / a n / a n / a n / a n / a n / a + / A ^1-42(h) p-Tau217 / + + + + + + n / a n / a n / a A ^1-42 (i) Gender (h) - - - - - - - - - ApoE4 status n / a n / a + + + n / a n / a + + (j) The addition of p-Tau217 status into the plasma combination model showed increase in accuracy of the model (AUC: 0.945, Specificity = 89.13% and Sensitivity = 83.33%, Eqn 21) compared to model that incorporates p-Tau181 (AUC: 0.932, and Specificity = 86.95% and Sensitivity = 79.63%, Eqn 27). Therefore, it is concluded that plasma biomarker combinations that incorporates p-Tau217 (or p-Tau217 / A ^1-42) provide the best results for diagnosing AD. Further addition of p-Tau231 or p-Tau181 into Eqn 21 do not improve the test accuracy (Eqn 37 and Eqn 38 respectively). Example 11: Diagnostic Performance of Different Biomarker Level Combinations in further Combination with p-Tau217 Levels or the NfL / Aβ1-42 and GFAP / Aβ1-42 Ratios Figure 13 describes the diagnostic performance of plasma combination after replacing NfL and GFAP with NfL / A ^1-42 or GFAP / A ^1-42 ratio respectively. The equations include different combinations NfL / A ^1-42, p-tau217, GFAP / Aβ1-42 along with ApoE4 status, age, and gender. The replacement of NfL by NfL / A ^1-42 status or replacing GFAP by GFAP / A ^1-42 into the plasma combination model do not increase diagnostic accuracy (Specificity = 89.13% and Sensitivity = 83.33%, remain same in Eqn 22, Eqn 29 and Eqn 31). Therefore, it is concluded that plasma biomarker combination that incorporates NfL / A ^1-42 or GFAP / A ^1-42 ratio may not provide additional advantage for diagnosing AD, although no decrease in diagnostic accuracy is observed. Example 12: Diagnostic Performance of Different Biomarker Level Combinations with Addition of Lysosomal Ionic Index to NfL, GFAP and p-Tau217 Levels Figure 14 describes diagnostic performance of the combination of NfL, GFAP, Aβ1-42, Aβ1-40, and p-Tau217 levels along with lysosomal ionic index. AD likelihood score calculated by combining Age + Gender + p-Tau217 + NfL + GFAP (Eqn 33, Specificity = 91.66% Sensitivity = 78.57%, Age + Gender + p-tau217 + NfL + GFAP + Lysosomal ionic index (Eqn 36 = Specificity = 95.8% Sensitivity = 100% was plotted for 24 control and 14 AD individuals. The addition of p-Tau217 into the model showed increase in accuracy of the Age + Gender + p-Tau217 + NfL + GFAP model (86.84 %) against Age + Gender + p-Tau217 + NfL + GFAP + Lysosomal ionic index model (97.36%). Therefore, it is concluded that plasma biomarkers along with lysosomal ionic index and p-Tau217 so far provide the best results for diagnosing AD.

Claims

CLAIMS 1. A method for measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from a subject, comprising the steps of: (a) obtaining the sample from the subject; and (b) measuring the presence and / or levels of the biomarkers: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217) in the sample.

2. The method of claim 1, wherein the presence and / or level of the combination of five biomarkers is used to predict the subject’s risk or likelihood of having and / or developing Alzheimer’s disease.

3. A method of predicting a subject’s risk or likelihood of having and / or developing Alzheimer’s disease, said method comprising measuring the presence of and / or the levels of a combination of five biomarkers in a sample obtained from the subject, wherein the biomarkers are: neurofilament light (NfL) polypeptide, glial fibrillary acidic protein (GFAP), β-Amyloid 1-42 (Aβ1-42), β-Amyloid 1-40 (Aβ1-40), and phosphorylated Tau (p-Tau181, p-Tau231 and / or p-Tau217).

4. The method of any one of claims 1 to 3, wherein a ratio of the level of any one or more of the biomarkers are measured, such as the ratio of Aβ1-42 / Aβ1-40 levels, the ratio of Aβ1-40 / Aβ1-42 levels, the ratio of p-Tau181 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau181 levels, the ratio of p-Tau217 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau217 levels, the ratio of p-Tau231 / Aβ1-42 levels, the ratio of Aβ1-42 / p-Tau231 levels, the ratio of NfL / Aβ1-42 and / or the ratio of Aβ1-42 / NfL are measured, in particular wherein the ratio of Aβ1-42 / Aβ1-40 levels and the ratio of p-Tau181 / Aβ1-42 levels are measured.

5. The method of any one of claims 1 to 4, wherein the measurement of the biomarkers is by immunoassay, such as an enzyme-linked immunoassay, optionally wherein the immunoassay is an in vitro single-molecule array immunoassay.

6. The method of any one of claims 1 to 5, wherein the method further comprises measuring the ionic index of lysosomes in peripheral blood mononuclear cells (PBMCs) obtained from the subject, optionally wherein the ionic index comprises one or more of: the mean calcium (Ca2+) levels, the mean pH and / or the 2-ion index of lysosomes in the PBMCs, in particular whereinthe ionic index comprises all of: the mean calcium (Ca2+) levels, the mean pH and the 2-ion index of lysosomes in the PBMCs, wherein the 2-ion index comprises the ratio of mean pH / mean Ca2+levels.

7. The method of any one of claims 2 to 6, wherein the prediction additionally comprises using information on the gender and / or age of the subject, and / or wherein the prediction additionally comprises using information on the ApoE status of the subject, in particular the presence of an APOE2, APOE3 or APOE4 allele in the subject.

8. The method of any one of claims 1 to 7, wherein the sample is blood, serum or plasma.

9. The method of any one of claims 1 to 8, wherein the presence and / or levels of the biomarkers is used to generate a likelihood score for the subject having and / or developing Alzheimer’s disease, and / or wherein the presence and / or levels of the biomarkers are normalised by calculating z-score values.

10. The method of claim 9, wherein the likelihood score is generated using a logistic regression of the presence and / or levels of the biomarkers, and / or of the z-score values, optionally wherein the logistic regression is additionally of the gender, age and / or ApoE status of the subject, and / or of the measurement of the ionic index of lysosomes in PBMCs obtained from the subject, optionally of normalised z-score values for the ionic index.

11. The method of claim 9 or claim 10, wherein the likelihood score is generated using the levels and / or values for: the age and gender of the subject, NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and p-Tau181 / Aβ1-42 ratio, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio, and optionally the ionic index of lysosomes in PBMCs obtained from the subject and / or the ApoE status of the subject, and / or wherein the likelihood score is generated using values for the age and encoded gender of the subject, and the z-score values for plasma levels of NfL, GFAP, Aβ1-42 / Aβ1-40 ratio, p-Tau181, p-Tau231 and / or p-Tau217, and the p-Tau181 / Aβ1-42, p-Tau217 / Aβ1-42 and / or p-Tau231 / Aβ1-42 ratio as independent variables, and optionally the z-score value for the ionic index of lysosomes in PBMCs obtained from the subject and / or a value for the ApoE status of the subject, wherein the encoded gender for a female subject is assigned a value of 0 or 1 and for a male subject is assigned a value of the other of 0 or 1, optionally wherein the ApoE status of the subject is assigned a value of 0, 1 or 2.

12. The method of any one of claims 9 to 11, wherein the likelihood score is calculated using any one of formulae (I) to (XXXIII): (I) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))); (II) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))); (III) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (h*gender)))); (IV) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender)))); (V) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j*ApoE score)))); (VI) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j*ApoE score)))); (VII) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))); (VIII) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (h*gender) + (j*ApoE score)))); (IX) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender) + (j*ApoE score)))); (X) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))); (XI) 1 / (1 + e^(-(a + (b*age) + (d*[GFAP]) + (i*[p-Tau181 / Aβ1-42 ratio] + (h*gender) + (j*ApoE score)))); (XII) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (k*lysosomal ionic index)))); (XIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau181 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XIV) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))); (XV) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))); (XVI) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XVII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score))));(XVIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (n*[Aβ1-42]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))) (XIX) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (n*[Aβ1-42]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender)))) (XX) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender)))); (XXI) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (g*[Aβ1-42 / Aβ1-40 ratio]) + (h*gender) + (j* ApoE score)))); (XXII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (l*[GFAP / Aβ1-42 ratio]) + (f*[p-Tau217]) + (h*gender)))); (XXIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (l*[GFAP / Aβ1-42 ratio]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))); (XXIV) 1 / (1 + e^(-(-a + (b*age) + (m*[NfL / Aβ1-42 ratio]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender)))); (XXV) 1 / (1 + e^(-(-a + (b*age) + (m*[NfL / Aβ1-42 ratio]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j* ApoE score)))); (XXVI) 1 / (1 + e^(-(-a + (b*age) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))); (XXVII) 1 / (1 + e^(-(-a + (b*age) + (d*[GFAP]) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))); (XXVIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (k*lysosomal ionic index) + (f*[p-Tau217]) + (h*gender)))); (XXIX) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau181]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); (XXX) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j*ApoE score)))); (XXXI) 1 / (1 + e^(-(a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau217]) + (h*gender) + (j*ApoE score)+(n*[Aβ1-42])))); (XXXII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau231]) + (i*[p-Tau217 / Aβ1-42 ratio]) + (h*gender) + (j* ApoE score)))); or (XXXIII) 1 / (1 + e^(-(-a + (b*age) + (c*[NfL]) + (d*[GFAP]) + (f*[p-Tau231]) + (f*[p- Tau217]) + (h*gender) + (j* ApoE score)))), wherein a, b c, f, g, h and i are positive or negative numbers, and d, j, k, l, m and n are positive numbers.

13. The method of any one of claims 2 to 12, wherein the subject’s risk or likelihood or the likelihood score provides the likelihood of the subject being identified as positive on an amyloid positron emission tomography (PET) scan.

14. The method of any one of claims 1 to 13, wherein the method assists the distinguishing of Alzheimer’s disease over non-Alzheimer’s neurodegenerative diseases or dementias, optionally wherein the likelihood score provides the likelihood of the subject having Alzheimer’s disease over non-Alzheimer’s neurodegenerative diseases or dementias.

15. The method of any one of claims 2 to 14, wherein the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease when: the Aβ1-42 / Aβ1-40 ratio is decreased; the Aβ1-40 / Aβ1-42 ratio is elevated; the level of p-Tau181, p-Tau231 and / or p-Tau217 is elevated; the level of NfL is elevated; the level of GFAP is elevated; the p-Tau181 / Aβ1-42 ratio, the p-Tau217 / Aβ1-42 ratio and / or the p-Tau231 / Aβ1-42 ratio is elevated; the Aβ1-42 / p-Tau181 ratio, the Aβ1-42 / p-Tau217 ratio and / or the Aβ1-42 / p-Tau231 ratio is decreased; the NfL / Aβ1-42 ratio is elevated; and / or the Aβ1-42 / NfL ratio is decreased, in the sample compared to a subject not having Alzheimer’s disease or compared to a subject having a non-Alzheimer’s neurodegenerative disease or dementia.

16. The method of any one of claims 2 to 15, wherein the subject’s risk or likelihood or the likelihood score is compared to one or more threshold value, over which the subject is deemed at risk of, is likely to have and / or has Alzheimer’s disease, and / or deems likely that the subject will be identified as positive on an amyloid positron emission tomography (PET) scan.

17. A method of prognosing or diagnosing Alzheimer’s disease in a subject at risk of and / or suspected of having Alzheimer’s disease, said prognosis or diagnosis comprising the method of any one of claims 1 to 16, optionally wherein the likelihood score is used in the prognosis or diagnosis of Alzheimer’s disease.

18. A method of treating Alzheimer’s disease in a subject at risk of developing or suspected of having Alzheimer’s disease, said method comprising the steps of: (a) predicting the subject’s risk or likelihood of developing or having Alzheimer’s disease by the method of any one of claims 2 to 17; and (b) administering a treatment for Alzheimer’s disease to the subject if the risk or likelihood exceeds one or more threshold value.

19. The method of claim 18, wherein step (a) additionally comprises identifying the subject as positive on an amyloid positron emission tomography (PET) scan if the risk or likelihood exceeds a threshold value.

20. The method of claim 18 or claim 19, wherein a likelihood score for the subject having and / or developing Alzheimer’s disease is generated according to the methods of any one of claims 9 to 17.

21. An immunoassay array for the measurement of a combination of five biomarkers according to claim 1, optionally wherein the measurement is in a single test sample obtained from a subject.

22. The immunoassay array of claim 21, wherein the immunoassay is an enzyme-linked immunoassay, wherein the immunoassay is an in vitro single-molecule array immunoassay, and / or wherein the immunoassay comprises oligonucleotide linked antibodies, optionally wherein the immunoassay comprises oligonucleotide-conjugated capture and detector antibodies, comprises multiple capture and release steps, and / or further comprises ligation of the antibody-conjugated oligonucleotides, followed by optional quantitative PCR or sequencing.

23. A kit for the measurement of a combination of five biomarkers according to claim 1, said kit comprising the immunoassay array of claim 21 or claim 22.

Images