WO2021142417A2 - Systems for detecting alzheimer's disease - Google Patents

Abstract

Described herein are methods and apparatuses (e.g., systems, devices, kits, etc.) for collecting and analyzing genetic data that may useful for diagnosing and treating a patient. In particular, described herein are methods for processing blood samples to determine risk or severity of Alzheimer's disease.

Description

SYSTEMS FOR DETECTING ALZHEIMER'S DISEASE

CROSS REFERENCE TO RELATED APPLICATIONS [0001] This patent application claims priority to U.S. provisional patent application no. 62/959,803, filed on January 10, 2020, titled “METHODS FOR DETECTING ALZHEIMER'S DISEASE” and herein incorporated by reference in its entirety.

INCORPORATION BY REFERENCE

[0002] All publications and patent applications mentioned in this specification are herein incorporated by reference in their entirety to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

FIELD

[0003] Described herein are methods and apparatuses (e.g., systems, devices, kits, etc.) for collecting and analyzing genetic data that may useful for diagnosing and treating a patient. In particular, described herein are methods for processing blood samples to determine risk or severity of Alzheimer's disease.

BACKGROUND

[0004] As populations age, neurodegenerative diseases become a significant concern for health systems, particularly the economic burdens associated with medications, hospitalization, and potential high-risk groups. Among such neurodegenerative diseases, Alzheimer's disease (AD) is the main type of dementia in the elderly, and the main cause of death in aging people. Epidemiological studies indicate that around 10% of the population over 65 years and 50% of individuals over 90 years suffer from AD.

[0005] One particular challenge associated with AD is reliable and efficient diagnosis. Currently, the definitive diagnosis of AD can only be made by post-mortem histopathological analysis of brain tissue. These analyses show the presence of two characteristic pathological morphologies that correspond to extracellular and intracellular protein aggregates, respectively: i) amyloid plaques composed of a small peptide of 39-43 amino acids, the b-amyloid peptide (Ab), derived from proteolytic processing of the amyloid precursor protein (APP); and ii) the neurofibrillary tangles (NFT) formed by the aggregation of the hyperphosphorylated form of the Tau protein.

[0006] Today there is not a single test to diagnose AD. Rather, diagnosis is mainly based on: i) a clinical diagnosis whose central axis is the analysis of the mental, neuropsychological and functional state of the patient; and ii) the analysis of the accepted biomarkers of AD using methodologies that are too technically sophisticated and expensive (brain imaging techniques) or invasive (lumbar puncture). Accordingly, there is a growing research oriented to the identification of certain genes that may increase the risk of developing AD. However, although some genetic tests are available for these genes, they are not recommended by health professionals and have no value as a diagnostic tool.

[0007] Evaluation of a patient's cognitive status is based on assessments of memory, the ability to solve simple problems, and other cognitive skills. For example, such tests may seek to know if a patient is: i) aware of the symptoms, ii) aware of the date, time, and their surroundings, and iii) capable of remembering a short list of words, following instructions, and doing simple calculations. Most of these tests consist of a set of 20-30 questions to analyze basic cognitive functions, requiring basic information provided by the patient (name, address, etc.), answers to simple questions of common knowledge (name, address, etc.), and remembering simple lists of different words, names or addresses. Among the many cognitive scales to assess the patient's condition the Mini-Mental State Examination (MMSE) and Alzheimer's disease Assessment Scale-Cognitive (ADAS-Cog) scales are each widely used.

[0008] The MMSE test is one of the most used cognitive analysis scales and can be applied by medical personnel or researchers. It is used to measure a patient's cognitive status in clinical practice throughout the world. Time to administer the test is about 10 minutes, analyzing the cognitive function in areas of orientation, memory, attention, mathematical calculations, language, and visual construction. The score assigned to the patient has a range of 0-30 points and has a cut-off of 23/24 points. Such scores are generally accepted as indicating cognitive impairment and are associated with the diagnosis of dementia in 79% and 95% of cases, respectively. Despite its wide use, its sensitivity decreases in such a way that it is not effective in the management of patients with mild cognitive impairment or a psychiatric disorder. There is evidence that its sensitivity also decreases when assessing the cognitive deterioration of patients with a high educational level. Other confounding factors include the age and the cultural and socioeconomic background of the evaluated person.

[0009] The ADAS-Cog scale is the de facto scale in the cognitive assessment of patients with suspected AD and is also considered the most useful scale in the evaluation of cognitive changes in clinical studies of drugs to combat this disease. It was designed for the analysis of specific syndromes of cognitive and non-cognitive AD and has become an important tool in the measurement of the therapeutic efficacy of drugs and interventions on cognition. It consists of 11 items that evaluate functions related to memory, praxis and language. The ADAS-Cog best cut- off score is > or = 12 with sensitivity and specificity values of 89.19 % and 88.53 % respectively. Although compared with the MMSE scale it seems to be less sensitive to the educational level of the patient, it has the disadvantage that its long duration, of approximately 40-45 minutes, makes it impractical in clinical evaluation. A further and important disadvantage of the use of ADAS- Cog is that it also fails to evaluate some central deficits of AD such as: i) attention, ii) processing, and iii) information retrieval.

[0010] As mentioned hereinabove, the evaluation of neuropsychiatric symptoms is useful in the diagnosis of AD. This methodology also allows for excluding other pathologies such as depression, Vascular Dementia (VD), Frontotemporal Dementia (FTD), Multiple system atrophy (MSA), Mild cognitive impairment (MCI), or other types of dementia that may affect cognitive functions. The Geriatric Depression Scale (GDS), Hachinski Ischemic Score (HIS) and Free and Cued Selective Reminding Test (FCSRT) are used in the differential clinical diagnosis of AD to rule out cases of depression, DV and FTD. Because depression is a neuropsychiatric symptom very common in the elderly, there are a number of scales that focus on its identification. The GDS long form questionnaire consists of 30 “yes or no” questions related to the patient's mood, whereas the Short Form variant is 15 questions. Of the Short Form questions, 10 indicate depression when they are answered affirmatively and the rest indicate depression when they are answered negatively. The GDS has been observed to have a sensitivity value of 92% and specificity value of 89%. However, its main disadvantage is that it has only been validated for people with mild dementia, not for those with moderate to severe dementia who will have difficulty understanding the test questions.

[0011] The HIS scale is used to identify VD once there is a diagnosis of dementia for a given patient. It allows for differential diagnosis of AD by ruling out VD cases with a sensitivity of 89% and specificity of 89%. Yet the HIS scale is not, itself, a diagnostic tool. One of the great difficulties in the differential diagnosis of AD with respect to VD is that cases in which VD is presented as an individual pathology are rare. The most usual scenario is that it coexists with the development of amyloid plaques and other AD-associated neuropathologies.

[0012] The FCSRT scale has been suggested by the International Working Group (IWG) as a test for the differential diagnosis of AD with respect to other pathologies, its usefulness being seen in the discarding of other types of dementia such as Hippocampal Amnestic Syndrome and FTD.

[0013] With respect to the patient's functional abilities, the Katz Index of Independence in Activities of the Daily Living Scale (Katz ADF) is the most appropriate tool for the analysis of how a patient performs activities of daily living. This evaluation is made based on the analysis of 6 functions: i) bathing, ii) dressing, iii) going to the bathroom, iv) transferring, (v) continence and vi) feeding. The patient gives a yes/no answer to these 6 items. A score of 6 indicates complete function, 4 -moderate function, and 2 or less indicates severe functional damage. The Lawton Instrumental Activities of Daily Living (Lawton IADL) test measures more complex functional abilities than the Katz index. The skills evaluated are those that are required to live in society such as: i) shopping, ii) cooking, iii) using transportation and iv) managing finances and medication administration. However, such tests suffer from a lack of sensitivity to slight improvements over time, making them ineffective in determining the patient's evolution in the development or decline of their activities.

[0014] Imaging techniques can provide immediate functional and structural details of the brain. In the case of AD, these techniques are very useful in predicting and monitoring progression of disease, providing a visualization of the structure/physiology of the brain, and they allow the detection of proteins and protein aggregates associated with AD manifestation. Imaging techniques including Computer Tomography (CT), Positron Emission Tomography (PET), Single Photon Emission Computed Tomography (SPECT) and Magnetic Resonance Imaging (MRI), have been positioned in recent years as methodologies to support the clinical diagnosis of AD. Among these, PET and MRI techniques are the most used in the diagnosis of this disease.

[0015] Many studies have shown that alterations in cerebral metabolism precede clinical manifestations of AD. Accordingly, PET techniques are used to assess the changes in cerebral glucose metabolism and, in this way, act as an evaluation of neuronal activity, further distinguishing AD from other types of dementia according to the different patterns of glucose consumption in the brain. FDG-PET, named for the use of the fluorodeoxy glucose marker (FDG), is a marker of neurodegeneration and has been useful in detecting regional hypometabolism characteristic of glucose in AD. The use of FDG-PET allows for distinguishing healthy control individuals (HC) from patients with AD with a sensitivity value of 99% and a specificity that is in a range of 73-98% when compared to post-mortem histopathological analysis. Another variant of PET, 1 lC-PiB-PET, named for the use of the marker [11C] N- methyl [11C] 2- (4'-methylaminophenyl)-6-hydroxy-benzothiazole (llC-PiB, from Pittsburgh compound B), detects brain deposits of Ab peptide allowing the differentiation of individuals with AD from healthy control patients, presenting a good correlation with biopsies performed post mortem. These high levels of amyloid deposits have also been observed in patients with Mild Cognitive Impairment (MCI) who progress to AD and has shown to be useful in the differential diagnosis of this disease with respect to other pathologies of Parkinson's disease (PD), FTD, and Lewy Body Dementia (LBD). It has been observed that llC-PiB-PET shows a sensitivity of 94% and a specificity of 52.6% in predicting the passage of MCI to AD. Today, 18F-based markers such as 18F-florbetapir (2012), 18F-flutemetamol (2013), and 18F- florbetaben (2014) have been approved by the FDA as markers for Ab peptide, however, these have presented a lower half-life, certain non- specificity and poor clinical use compared to markers based on 11C. The values of sensitivity and specificity for the biomarkers based on 18F are in a range of 89-97% and 63-93%, respectively.

[0016] Images obtained by MRI have many applications and are used in different modalities. The most used MRI in the diagnosis of AD are functional magnetic resonance imaging (fMRI), volumetric magnetic resonance imaging (vMRI), and hippocampal volumetry. Regarding its use in the diagnosis of AD, MRI provides visualization of atrophied brain regions in early stages of the disease and distinguish it from that of aging. Moreover, such imaging techniques distinguish HC from individuals suffering from AD with a high sensitivity and specificity, as well as providing evidence of their progression from a stage of MCI to AD. For example, in distinguishing HC from AD patients MRI techniques have shown a sensitivity of a range of 41- 95% and a specificity of 67-85%; and a sensitivity of 58-85% together with a specificity of 81- 95% in distinguishing AD from other dementias.

[0017] The incorporation of biomarkers represents an important change in the diagnostic criteria of AD. Research has been aimed at detecting the onset of disease or tracking its progression. Such biomarkers can be grouped into four categories: i) cerebrospinal fluid (CSF) biomarkers, ii) brain imaging biomarkers, iii) genetic biomarkers and iv) plasma biomarkers. Today, only CSF and brain imaging biomarkers are currently accepted within the diagnostic criteria of AD.

[0018] Three CSF biomarkers have been established and recognized internationally for the diagnosis of AD. These correspond to the peptide biomarkers Aβ_1-42 and Tau, the latter being measured as total Tau (T-Tau) and phosphorylated Tau (¹⁸¹Thr P-tau). This set of biomarkers is currently accepted as in vivo evidence of AD and has been incorporated into the diagnostic criteria of the National Institute of Neurological and Communicative Disorders and the Stroke- Alzheimer's disease and Related Disorders Association (NINCDS-ADRDA) for both the diagnosis of AD and its preclinical stages. Regarding the levels of these biomarkers in the CSF, several studies have shown that the levels of Aβ_1-42 peptide decreased while the levels of T-Tau and 181Thr P-tau increased in patients affected by AD, with respect to HC individuals. Together, these biomarkers increase the validity for the diagnosis of AD with a sensitivity of 95% and a specificity of 85%.

[0019] Still, the use of CSF biomarkers and brain imaging biomarkers as a diagnostic tool in AD has important disadvantages from a clinical and economic point of view. Even though Aβ_{1- 42,} T-tau and ¹⁸¹Thr P-tau levels in CSF are accepted diagnostic criterion for AD, the main disadvantage lies in the method of obtaining these biomarkers, i.e., the lumbar puncture, one of the most invasive procedures performed in clinical medicine. The procedure consists of the insertion of a thin needle between the lumbar vertebrae L3 and L4 of the patient in order to extract a sample of CSF that will be subsequently evaluated for the biomarkers. Lumbar puncture is not practiced routinely as it requires qualified personnel, reagents, advanced equipment, and is considered invasive and of certain risk for the patient. Although the sample takes only 15 minutes to collect, once the puncture is done, the patient must remain in the hospital for a period of 1-4 hours depending on the amount of CSF removed and whether or not the patient is capable of travel. Moreover, within a few hours or even days of taking the sample, the patient may experience headaches, nausea, accelerated heart rate or low blood pressure, Such side effects can last up to 5 days. In view of the adverse effects, the method of lumbar puncture prevents proper follow-up of patients and its use as a routine diagnostic tool becomes very difficult. Regarding brain imaging techniques, the PET technique is expensive and requires both advanced equipment and specialized personnel, limiting its availability. In addition, the use of intravenous radiolabels is both invasive and implies exposure of the patient to radioactivity. Among the other disadvantages associated with PET are limited sensitivity, spatial resolution and quantification of the target. The use of MRI, on the other hand, presents limitations related to the fact that the cerebral atrophy observed in AD is not necessarily the product of a neuronal loss. Moreover, the use of MRI is associated with feelings of claustrophobia in the patient, and the common occurrence of metal implants or medical devices among elderly patients can limit its use.

[0020] As noted hereinabove, the diagnosis of AD is subject to the use of scales of cognitive assessment that rely on variables unrelated to the development of the disease (MMSE), are not entirely satisfactory in the analysis of cognitive impairment of a patient (ADAS-Cog), do not constitute a diagnostic tool in themselves (HIS), or are very insensitive (IADL and Kast ADL). As a result, diagnosis of AD has sensitivity and specificity values the range widely (37-100% and 41-100%, respectively), which implies the delivery of an erroneous diagnosis in many cases. Unfortunately, the use of biomarkers currently accepted in the diagnosis of AD are subject to invasive methods of collection and expensive techniques that cannot be used routinely in clinical diagnosis, thus preventing proper follow-up of patients. The classic proteinaceous biomarkers of AD, peptide Aβ_1-42, T-tau and ¹⁸¹Thr P-tau, have not been shown to be very robust in their relationship to AD and have very low concentrations in the blood compared to CSF. For example, the concentration of the Tau neuronal protein is 2-300 pg/mL, while its levels in the plasma are approximately 100 times lower. The low blood concentration of these proteins makes it difficult to use them as biomarkers from this alternative, and less invasive source. In this context, it is necessary to identify new types of non-protein biomarkers that could be useful in the diagnosis of AD. Accordingly, it is necessary to search and identify new biomarkers, which are sensitive, specific, and less invasive.

[0021] It would also be beneficial to identify methods and apparatuses for performing accurate and inexpensive blood tests (assays) for both genetic background (genotype) and expression (phenotype). Furthermore, it would be beneficial to provide apparatuses and methods for analyzing genetic background and expression.

SUMMARY OF THE DISCLOSURE

[0022] Described herein are methods and apparatuses (e.g., devices, systems, kits, etc.) for performing blood tests that may determining an individual's genetic background (e.g., genotype, including polymorphism, such as SNPs), and also the individual's genetic expression (e.g., microRNA expression) levels. In addition, described herein are methods and apparatuses for analyzing combined genetic background and microRNA expression levels; these methods and apparatuses may include predictive modeling based on both the genetic background (e.g., SNPs) and microRNA expression. These methods and apparatuses may generally be applied to the analysis of any condition or disorder, including in particular neurological disorders such as Alzheimer's disease. Specifically, described herein are methods and apparatuses for the non- invasive detection and diagnosis of Alzheimer's disease (AD). In some embodiments, diagnosis is based on the identification of a panel of miRNAs and genetic biomarkers. In some such embodiments, the genetic biomarkers comprise single nucleotide polymorphisms (SNPs) in three genes of interest (APOE, PICALM and CR1), for which there is evidence as potential biomarkers for the diagnosis of AD. In preferred embodiments, a comparison of the levels of miRNA found in the blood of patients affected by AD is compared to samples from HC. In some such embodiments, the genotype of said patients, regarding the polymorphic variants (e.g.,

SNPs) of the genes of interest are determined. In certain embodiments of the invention described herein, collected biomarker data is analyzed and/or assessed using a predictive modeling system to provide a diagnosis of, or to determine the risk of developing, AD.

[0023] For example, described herein are blood tests that include taking one or more sample of a subject's plasma and identifying both one or more SNPs and a plurality of microRNAs. The blood test may be used in particular with both a spike-in control and a “housekeeping” microRNA control and may result in an improved assay. Data from the assay, including the presence or absence of one or more SNPs and the expression levels of each of the microRNAs may be analyzed using a predictive modeling network (e.g., neural network) that is trained to identify status, condition or risk that a subject is susceptible to a particular disorder or condition. [0024] In certain aspects, provided herein are diagnostic tools based on peripheral miRNA biomarkers that are i) cost effective, ii) non-invasive and iii) sensitive/specific and that will incorporate genetic biomarkers based on SNPs in three genes whose variants are strongly related to the risk of developing AD. Together, such a panel of biomarkers, when analyzed with predictive modeling systems, provide sensitive and specific tools for the diagnosis, assessment, and/or monitoring of AD.

[0025] For example, described herein are method of classifying a sample of a patient suffering from or at risk of developing a disorder. In particular, described herein are methods and apparatuses for classifying a sample of a patient suffering from or at risk of developing Alzheimer's disease. These methods may include: determining in said sample an expression level of at least seven miRNA selected from the group set forth in Table 1, or combinations thereof; b) assessing the pattern of expression level(s) determined in step a) by comparison with one or several pattern(s) of expression levels from a control sample; and c) classifying the sample of said patient from the outcome of the comparison in step b) into one of at least two classes.

[0026] Any of these methods may include diagnosing a disorder. For example described herein are methods of diagnosing Alzheimer's disease, predicting risk of developing Alzheimer's disease, or predicting an outcome of Alzheimer's disease in a patient suffering from or at risk of developing Alzheimer's disease, said method comprising the steps of: a) determining in a sample from said patient, the expression level of at least one miRNA selected from the group set forth in Table 1, or combinations thereof; b) assessing the pattern of expression level(s) determined in step a) by comparison with one or several pattern(s) of expression levels from a control sample; and c) diagnosing Alzheimer's disease, predicting a risk of developing Alzheimer's' disease, or predicting an outcome of Alzheimer's disease from the outcome of the comparison in step b). [0027] Any of these methods may include determining the expression levels of each of the miRNAs set forth in Table 1.

[0028] Also described herein are methods for identifying a subject at risk of developing a disorder. For example, described herein are methods for identifying a subject at risk of developing Alzheimer's disease, comprising: obtaining a first dataset associated with a sample obtained from the patient, wherein the first dataset comprises the expression level for at least Four miRNA selected from the group set forth in Table 1, or combinations thereof; and analyzing the first dataset to assess the expression level of the miRNA, wherein the expression level of the miRNA positively or negatively correlates with an increased risk of Alzheimer's disease in the subject. The analysis may further comprise comparing the first dataset to a second dataset associated with a control sample, wherein the second dataset comprises quantitative data for a control expression level of the miRNA, and wherein a statistically significant difference between the miRNA expression level of the first data set and the control expression level of the second data set indicates an increased risk of Alzheimer' disease in the patient. The datasets may comprise the expression levels for each of the miRNAs set forth in Table 1.

[0029] Also described herein are methods for determining Alzheimer' s disease risk in a patient, comprising: obtaining a sample from the patient, wherein the sample comprises at least one miRNA set forth in Table 1, or combinations thereof; contacting the sample with a reagent; generating a complex between the reagent and the miRNA; detecting the complex to obtain a first dataset associated with the sample, wherein the first dataset comprises quantitative expression data for the miRNA; and analyzing the first dataset to assess the expression level of the miRNA, wherein the expression level of the miRNA positively or negatively correlates with an increased risk of AD in the patient. The sample may comprise each of the miRNAs set forth in Table 1.

[0030] Any of the methods described herein may include comprising assessing a single nucleotide polymorphism marker (SNP) in the patient; and combining said assessment with the assessment of the expression level of the miRNA(s) to identify risk of Alzheimer's disease in the patient. These methods may include assessing at least one of the SNPs set forth in Table 2, or combinations thereof. Any of these methods may include assessing each of the SNPs set forth in Table 2. In some variations, these methods may include assessing the cognitive and/or neuropsychiatric status of the patient.

[0031] In general, the sample may be a blood sample. The control sample may be associated with a control subject or with a control population. The control sample may be obtained from the patient prior to manifestation of Alzheimer's. The control sample may be associated with a control subject or a control population characterized by analysis of cerebrospinal fluid (CSF) for Aβ1-42, total Tau (T-Tau), and phosphorylated Tau (181Thr P-tau) levels.

[0032] The expression of the miRNA may be significantly decreased compared to expression of the control miRNA. In some variations, the expression of the miRNA is significantly increased compared to expression of the control miRNA. For example, the expression level of the miRNA marker may positively correlate with an increased risk of Alzheimer's disease in the subject. The expression level of the miRNA marker may negatively correlate with an increased risk of Alzheimer's disease in the subject.

[0033] Also described herein are apparatuses for performing any of the methods described herein. For example, described herein are apparatuses, e.g., including a memory including instructions that, when performed, may execute any of the methods described herein. Thus, any of these methods may be implemented, at least in part, on one or more computers. [0034] As mentioned above, any of these methods and apparatuses may include predictive modeling (e.g., the apparatus may include or be configured as a predictive modeling system).

The first dataset may be stored on a storage memory. For example, obtaining the first dataset associated with the sample may comprise obtaining the sample and processing the sample to experimentally assess the sample. Obtaining the first dataset associated with the sample may comprise receiving the first dataset directly or indirectly from a third party that has processed the sample to experimentally assess the sample.

[0035] The statistically significant difference may be determined, at least in part, using a predictive modeling system. The expression levels may be obtained from a nucleotide-based assay. The expression levels may be obtained from an RT-PCR assay, a sequencing-based assay, a microarray assay, or a combination thereof.

[0036] In general the subject may be a human or non-human subject (e.g., an animal, including mammalian, subject). The subject may be a patient.

[0037] Also described herein are computer-implemented methods for identifying a subject at risk of a disease. For example, computer-implemented methods for identifying a subject at risk of Alzheimer's disease, comprising: storing, in a storage memory, a first dataset associated with a sample obtained from the subject, wherein the first dataset comprises quantitative expression data for a miRNA selected from the group set forth in Table 1; and analyzing, by a computer processor, the first dataset to determine the expression level of the miRNA, wherein the expression level of the miRNA marker positively or negatively correlates with an increased risk of Alzheimer's disease in the patient. The first dataset may comprise quantitative expression data for each of the miRNAs set forth in Table 1.

[0038] Any of these methods may include storing, in a storage memory, at least a second dataset associated with the sample obtained from the subject, wherein the second dataset comprises quantitative data for a SNP selected from the group set forth in Table 2; and analyzing, by a computer processor, the second dataset to determine the SNP, wherein the presence of the SNP positively or negatively correlates with an increased risk of Alzheimer's disease in the patient. The second dataset may comprise quantitative data for each of the SNPs set forth in Table 2. Any of these methods may include combining the analysis of the first dataset with the analysis of the second data set to diagnose Alzheimer's disease, predict risk of developing Alzheimer's disease, or predict an outcome of Alzheimer' s disease in a patient; wherein combining the analysis of the first dataset with the analysis of the second data set comprises implementation of a predictive modeling system.

[0039] For example, a computer program product useful for performing the method according to any one of the preceding claims, comprising: a) means for receiving data representing an expression level of at least one miRNA in a patient blood sample selected from the group set forth in Table 1, or combinations thereof; b) means for receiving data representing at least one control pattern of expression levels for comparing with the expression level of the at least one miRNA from said sample; c) means for comparing said data representing the expression level of the at least one miRNA in a patient sample; and d) means for determining a diagnosis of Alzheimer's disease, a prediction of a risk of developing Alzheimer's disease, or a prediction of an outcome of Alzheimer's disease from the outcome of the comparison in step b); wherein the means of step d) comprises a predictive modeling system.

[0040] Also described herein are kits for use in quantifying Alzheimer's disease risk in a patient, comprising: a set of reagents comprising a plurality of reagents for determining from a blood sample obtained from the patient quantitative expression data for a miRNA selected from the group set forth in Table 1, or combinations thereof; instructions for using the plurality of reagents to determine quantitative expression data from the sample for a first dataset, and analyzing said first dataset by comparing the first dataset to determine the expression level of the miRNA marker, wherein the expression level of the miRNA marker positively or negatively correlates with an increased risk of AD in the subject. For example, a kit for use in quantifying Alzheimer's disease risk in a patient may include: a set of reagents consisting essentially of a plurality of reagents for determining from a blood sample obtained from the patient quantitative expression data for a miRNA selected from the group set forth in Table 1; and instructions for using the plurality of reagents to determine quantitative expression data from the blood sample. [0041] For example, described herein are systems comprising: a plurality of miRNA target detection polynucleotides comprising at least four miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51 (shown in Table 1), one or more processors; and a memory coupled to the one or more processors, the memory storing computer-program instructions, that, when executed by the one or more processors, perform a computer- implemented method comprising: receiving an expression level for each of the miRNA target detection polynucleotide in the plurality of miRNA target detection polynucleotides from a patient sample; analyzing, in the one or more processors, the expression level for each of the miRNA target detection polynucleotide in the plurality of miRNA target detection polynucleotides from the patient sample using a trained neural network, wherein the trained neural network is trained on a dataset including the expression each of the miRNA target detection polynucleotide in the plurality of miRNA target detection polynucleotides, so that the trained neural network determines a risk score; and outputting the risk score from the trained neural network, wherein the risk score indicates a risk of Alzheimer' s disease in the patient. [0042] In some examples the plurality of miRNA target detection polynucleotides includes at least five miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 6 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 7 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 8 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 9 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 10 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 11 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 12 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 13 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 14 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 15 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 16 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 17 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 18 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 19 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 20 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 21 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 22 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 23 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 24 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 25 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 26 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 27 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 28 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 29 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 30 miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. In some examples the plurality of miRNA target detection polynucleotides includes at least 31 or more (e.g., at least 32, 33, 34, 35, 36, 37, 38. 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 or 50) miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51. [0043] In some examples, the plurality of miRNA target detection polynucleotides comprises all of the miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51.

[0044] In some examples the computer-implemented method further comprises classifying the patient into one of at least two classes based on the risk score, as described above. The system may output a progression score indicating progression of Alzheimer’s disease in the patient.

[0045] Any of the systems described herein may include plurality of SNP detection polynucleotides comprising SEQ ID Nos: 52-55. The computer-implemented method may also include receiving an expression level for each of the SNP detection polynucleotides from a patient sample, wherein analyzing, further comprises analyzing the expression level for each of the SNP detection polynucleotides from the patient sample using the trained neural network so that the trained neural network uses the expression pattern of the SNP detection polynucleotides from a patient sample in determining the risk score. BRIEF DESCRIPTION OF THE DRAWINGS [0046] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0047] The novel features of the invention are set forth with particularity in the claims that follow. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:

[0048] FIG. 1 illustrates the study design for assessing the diagnosis of Alzheimer's disease using biomarkers found in patient blood samples. Briefly, of 104 patients assessed for eligibility, 70 patients (35 with AD and 35 healthy controls) were selected for analysis via the proposed diagnostic blood test.

[0049] FIG. 2 depicts a flow diagram illustrating the SNP genotyping assay performed on whole blood samples from patients.

[0050] FIG. 3 illustrates a Taqman™ assay, which consists of a pair of specific non-labeled “starters” and a Taqman™ probe which has, at its 5ΌH end, a fluorophore which can be, for example, FAM (e.g., 6-fluorescein amidite) or VIC^® fluorescent dye. At its 3'OH end, the Taqman™ probe has a non-fluorescent quencher (NFQ) that represses the emission of fluorescence when the Taqman™ probe is intact and specifically linked to the target sequence. When the PCR reaction occurs, the exonuclease activity of the Taq Polymerase enzyme cuts only the Taqman™ probe that is hybridized to the NFQ and producing the emission of fluorescence only if the target sequence is complementary to the probe. In this way, the PCR reaction produces an increase in the emission of fluorescence and depending on the type of signal emitted according to the probe is linked to a FAM or VIC^® it can be discriminated which allele is present in the sample.

[0051] FIG. 4 illustrates the process of expression analysis of miRNA from patient blood samples.

[0052] FIGS. 5A-5H illustrate the TaqMan MicroRNA Assay. Specifically, in 5A and 5B sequence of mature miRNA is hybridized to a specific loop-stem RT -primer. The hybridization between the miRNA and the RT -primer stem-loop is due to a six-nucleotide overhang. The process of reverse transcription (RT) illustrated in 5B and 5C generates a first strand of complementary cDNA. The process of denaturation shown in 5D allows the hybridization of a first direction (forward primer; shown in 5E) that allows the extension of a second strand of cDNA (shown in 5F) to which the Taqman probe (black; shown in 5G) and a first antisense (yellow; also shown in 5G) hybridize. The extension of the first sense and antisense during the qPCR reaction in 5H allows the amplification of the target sequence, hydrolysis of the probe, and consequent signal emission, with a principle similar to that illustrated in FIG. 3.

DETAILED DESCRIPTION

[0053] The central diagnosis of AD comprises a clinical analysis for evaluation of the cognitive damage in a patient and inclusion in the diagnostic criteria of one or two biomarkers, i.e., i) CSF biomarkers and/or ii) cerebral imaging biomarkers. Such diagnostic methodology includes significant and inherent disadvantages, i.e., i) obtaining CSF biomarkers is highly invasive and can eventually deliver an irreproducible diagnosis due to the difficult conditions of storage and transport of samples, and ii) biomarkers of cerebral imaging involve the use of highly specialized and expensive equipment; the requirement of personnel highly trained in specialized techniques and unsatisfactory sensitivity and specificity. Accordingly, it is necessary to search for new biological biomarkers that are less invasive and easily obtainable (e.g., in other sources than the CSF), and that can be more sensitive and specific.

[0054] The measurement of biomarkers in blood is a valid alternative in the search for new biomarkers for AD. In addition to being relatively easy to obtain from patients, biomarkers identified in blood would allow adequate monitoring of patients over time. In line with this need, multiple studies have identified different plasma proteins whose levels of expression are deregulated in patients with AD compared to healthy individuals. Among these proteins, α2- macroglobulin (α2M), complement factor H (CFH), α1-antitrypsin and al-antichymotrypsin have shown increased levels in the plasma/serum of patients affected by this disease. Conversely, decreased levels of Apolipoprotein A1 in the blood of patients affected by AD have been found. Although the evidence indicates that these proteins could reflect pathological processes of AD and differentiate sick subjects from healthy patients, these differences have not yet been able to be sensitive, specific and reproducible.

[0055] Micro RNAs (miRNAs) have been found in several biological fluids (e.g., such as plasma, serum, saliva, milk and CSF) and, in recent years, circulating miRNAs have emerged as potential candidates for AD biomarkers. They have important advantages over currently established biomarkers, i.e., i) relative ease of sampling, ii) non-invasiveness, iii) stability, iv) sensitivity and specificity, and v) cost-effectiveness.

[0056] Briefly, miRNAs are a class of non-coding RNA that is 20-23 nucleotides in length, whose biological function is the post-transcriptional regulation of gene expression by binding to complementary sites in the 3'UTR region of a specific target mRNA, in such a way that results in cleavage of the mRNA, destabilization of the mRNA (e.g., through shortening of its poly(A) tail), and/or less efficient translation of the mRNA into proteins by ribosomes.

[0057] Biologically, these molecules have demonstrated their role in a wide range of processes such as: i) metabolism, ii) apoptosis, iii) cell proliferation, iv) division of stem cells, v) muscle differentiation and vi) brain morphogenesis. Due to its importance in the regulation of these physiological and developmental functions, deregulations in its expression have been related to a wide range of diseases such as: i) Cancer, ii) autoimmune diseases, iii) heart diseases, iv) gastrointestinal diseases and (v) lung diseases. These miRNAs have a high stability and can be released into the extracellular environment by traveling in body fluids bound to proteins or inside exosomes or microvesicles.

[0058] Moreover, researchers have studied the genetic background associated with AD, which can be classified into two types according to the age of the patient in which the disease begins to manifest: i) early onset AD (EOAD, < 65 age), and ii) late onset AD (LOAD, > 65 age).

[0059] Of the two, LOAD is the most common and complex epidemiological variant of AD, its development implies a relationship with risk factors not only genetic, but also epigenetic and environmental. In relation to the search for genetic biomarkers associated with LOAD, studies of genome-wide association studies (GWAS) and next generation sequencing studies (NGS) have found evidence of genes that are involved in various cellular functions such as: i) metabolism of cholesterol, ii) cell adhesion and endocytosis, and iii) immune response. Among these, APOE (cholesterol metabolism), PICALM (endocytosis) and CR1 (immune response) are among the top 10 genetic risk factors associated with LOAD.

[0060] Among the genetic risk factors associated with LOAD, the APOE gene, particularly through its e4 allele, is the first discovered and most established risk factor for LOAD in different populations. The % of individuals that carry the ε4 allele is approximately 50% in patients with LOAD compared to HC individuals, where it is present only in 20-25%. In comparison with control individuals, the risk of developing AD is 4 times higher in individuals who have a copy of the ε4 allele and 12 times higher in individuals who have two copies of the allele. Although the ε4 allele is strongly linked to AD, the allelic variation of the APOE gene only allows predicting the risk of developing AD by less than 20%. Although there is a blood test for the determination of the ε4 allele of APOE, this test is used mainly in clinical trials to assess the risk of developing AD and has no impact on the diagnosis of this disease.

[0061] PICALM is a protein whose cellular function is related to the endocytic pathway. It is a ubiquitous expression protein which regulates the formation of the clathrin coat during endocytic processes. This gene was identified in one of the first GWAS and its relationship as a risk factor for LOAD has been validated in European and Asian cohorts.

[0062] CR1 (complement 1 receptor), whose gene is located on chromosome 1q32, is a multifunctional protein expressed in microglia and blood cells, such as erythrocytes. CR1 is a cell surface receptor that binds to complement factors C3b and C4b. Two SNPs for this gene (rs6656401 and rs3818361) have been linked to LOAD in Caucasian and Canadian population cohorts.

[0063] However, the genetic analysis associated with EOAD has allowed the identification of high penetrance mutations in three genes that are currently linked to this variant of AD: i) the APP gene, and the genes of ii) Presenilin 1 (PSEN1) and iii) Presenilin 2 (PSEN2). The mutations of these genes have effects on the processing of the APP protein and the production of the Ab peptide favoring the manifestation of the disease. The EOAD is a variant of the disease largely genetically determined and is heritable in a range of 92-100%.

[0064] In certain embodiments, the panel of genetic biomarkers includes the SNPs Rs:6656401 (CR1) and/or Rs:3851179 (PICALM) that have been associated with LOAD.

Table 1: Sequence and access number of miRBase of miRNA from the panel of biomarkers.

Table 2: ID number and SNP ofAPOE, PICALM and CR1 genes.

[0065] Definitions

[0066] Unless otherwise defined herein, scientific and technical terms used in this application shall have the meanings that are commonly understood by those of ordinary skill in the art. Generally, nomenclature used in connection with, and techniques of, chemistry, cell and tissue culture, molecular biology, cell and cancer biology, neurobiology, neurochemistry, virology, immunology, microbiology, pharmacology, genetics and protein and nucleic acid chemistry, described herein, are those well-known and commonly used in the art.

[0067] The methods and techniques of the present disclosure are generally performed, unless otherwise indicated, according to conventional methods well known in the art and as described in various general and more specific references that are cited and discussed throughout this specification. See, e.g. “Principles of Neural Science”, McGraw-Hill Medical, New York, N.Y. (2000); Motulsky, “Intuitive Biostatistics”, Oxford University Press, Inc. (1995); Lodish et al., “Molecular Cell Biology, 4th ed.”, W. H. Freeman & Co., New York (2000); Griffiths et al., “Introduction to Genetic Analysis, 7th ed.”, W. H. Freeman & Co., N.Y. (1999); and Gilbert et al., “Developmental Biology, 6th ed.”, Sinauer Associates, Inc., Sunderland, MA (2000).

[0068] All of the above, and any other publications, patents and published patent applications referred to in this application are specifically incorporated by reference herein. In case of conflict, the present specification, including its specific definitions, will control.

[0069] The term “agent” is used herein to denote a chemical compound (such as an organic or inorganic compound, a mixture of chemical compounds), a biological macromolecule (such as a nucleic acid, an antibody, including parts thereof as well as humanized, chimeric and human antibodies and monoclonal antibodies, a protein or portion thereof, e.g., a peptide, a lipid, a carbohydrate), or an extract made from biological materials such as bacteria, plants, fungi, or animal (particularly mammalian) cells or tissues. Agents include, for example, agents whose structure is known, and those whose structure is not known.

[0070] A “patient,” “subject,” or “individual” are used interchangeably and refer to either a human or a non-human animal. These terms include mammals, such as humans, primates, livestock animals (including bovines, porcines, etc.), companion animals (e.g., canines, felines, etc.) and rodents (e.g., mice and rats).

[0071] The term “modulate” as used herein includes the inhibition or suppression of a function or activity (such as cell proliferation) as well as the enhancement of a function or activity.

[0072] The term “formula,” “algorithm,” or “model” as used herein includes any mathematical equation, algorithmic, analytical or programmed process, or statistical technique that takes one or more continuous or categorical input variables and calculates an output value, sometimes referred to as a “predicted value.” Non-limiting examples of “formulas” include sums, ratios, and regression operators, such as coefficients or exponents, biomarker value transformations and normalizations (including, without limitation, those normalization schemes based on clinical parameters, such as gender, age, or ethnicity), rules and guidelines, statistical classification models, and neural networks trained on historical populations. Of particular use in combining Alzheimer's disease markers and other biomarkers are linear and non-linear equations and statistical classification analyses to determine the relationship between levels of Alzheimer's disease markers detected in a subject sample. For example and without limitation to any specific methodology, such algorithms, and methods of risk index construction, may utilize pattern recognition features, including established techniques such as cross-correlation, Principal Components Analysis (PCA), factor rotation, Logistic Regression (LogReg), Linear Discriminant Analysis (LDA), Linear Discriminant Analysis (LDA), Support Vector Machines (SVM), Random Forest (RF), Gradient Boosting Machines(GBM), Partial Least Squares, Sparse Partial Least Squares, Flexible Discriminant Analysis, Recursive Partitioning Tree (RPART), as well as other related decision tree classification techniques, Nearest Shrunken Centroids (SC)", stepwise model selection procedures, Kth-Nearest Neighbor, Boosting or Boosted Tree, Decision Trees, Neural Networks, Bayesian Networks, Support Vector Machines, and Hidden Markov Models, among others.

[0073] As used herein, the term "random forest" refers to a machine learning ensemble classifier developed by Leo Breiman and Adele Cutler, consisting of multiple single classification trees. (L. Breiman, Random Forests, MACHINE LEARNING 45 (1): 5-32.

(2001)).

[0074] MicroRNAs comprise one class biomarkers assessed via methods of the invention. MicroRNAs, also referred to herein as miRNAs, are short RNA strands approximately 20-23 nucleotides in length. MiRNAs are encoded by genes that are transcribed from DNA but are not translated into protein and thus comprise non-coding RNA. The miRNAs are processed from primary transcripts known as pri- miRNA to short stem-loop structures called pre -miRNA and finally to the resulting single strand miRNA. The pre-miRNA typically forms a structure that folds back on itself in self-complementary regions. These structures are then processed by the nuclease Dicer in animals or DCL1 in plants. Mature miRNA molecules are partially complementary to one or more messenger RNA (mRNA) molecules and can function to regulate translation of proteins. Identified sequences of miRNA can be accessed at publicly available databases such as, and without limitation, deepBase, miRBase, microRNA.org, and miRGen 2.0. 0075] miRNAs and their corresponding stem-loop sequences described herein may be found in searchable databases of miRNA sequences and annotation, found on the world wide web (e.g., at microrna.sanger.ac.uk, www.microRNA.org, www.mirbase.org, or www.mirz.unibas.ch/cgi/miRNA.cgi. Entries in the miRBase Sequence database represent a predicted hairpin portion of a miRNA transcript (the stem-loop), with information on the location and sequence of the mature miRNA sequence. The miRNA stem-loop sequences in the database are not strictly precursor miRNAs (pre-miRNAs), and may in some instances include the pre-miRNA and some flanking sequence from the presumed primary transcript. The miRNA nucleobase sequences described herein encompass any version of the miRNA, including the sequences described in Release 10.0 of the miRBase sequence database and sequences described in any earlier Release of the miRBase sequence database. A sequence database release may result in the re-naming of certain miRNAs. A sequence database release may result in a variation of a mature miRNA sequence.

[0076] miRNAs are generally assigned a number according to the naming convention "mir- [number]." The number of a miRNA is assigned according to its order of discovery relative to previously identified miRNA species. For example, if the last published miRNA was mir-121, the next discovered miRNA will be named mir-122, etc. When a miRNA is discovered that is homologous to a known miRNA from a different organism, the name can be given an optional organism identifier, of the form [organism identifier]- mir- [number]. Identifiers include hsa for Homo sapiens and mmu for Mus Musculus. For example, a human homolog to mir-121 might be referred to as hsa-mir-121 whereas the mouse homolog can be referred to as mmu-mir-121. [0077] Mature microRNA is commonly designated with the prefix "miR" whereas the gene or precursor miRNA is designated with the prefix "mir." For example, mir-121 is a precursor for miR- 121. When differing miRNA genes or precursors are processed into identical mature miRNAs, the genes/precursors can be delineated by a numbered suffix. For example, mir-121-1 and mir-121-2 can refer to distinct genes or precursors that are processed into miR- 121. Fettered suffixes are used to indicate closely related mature sequences. For example, mir-121a and mir- 121b can be processed to closely related miRNAs miR-121a and miR-121b, respectively. In the context of the invention, any microRNA (miRNA or miR) designated herein with the prefix mir- * or miR-* is understood to encompass both the precursor and/or mature species, unless otherwise explicitly stated otherwise.

[0078] It may be observed that two mature miRNA sequences originate from the same precursor. When one of the sequences is more abundant than the other, a "*" suffix can be used to designate the less common variant. For example, miR-121 may be the predominant product whereas miR-121 * is the less common variant found on the opposite arm of the precursor. If the predominant variant is not identified, the miRs can be distinguished by the suffix "5p" for the variant from the 5' arm of the precursor and the suffix "3p" for the variant from the 3 ' arm. For example, miR-121-5p originates from the 5' arm of the precursor whereas miR- 121-3p originates from the 3 ' arm. Less commonly, the 5p and 3p variants are referred to as the sense ("s") and anti-sense ("as") forms, respectively. For example, miR-121-5p may be referred to as miR-121-s whereas miR- 121 -3p may be referred to as miR- 121 -as.

[0079] The above naming conventions have evolved over time and are general guidelines rather than absolute rules. For example, the let- and lin- families of miRNAs continue to be referred to by these monikers. The mir/miR convention for precursor/mature forms is also a guideline and context should be taken into account to determine which form is referred to. Further details of miR naming can be found at www.mirbase.org or Ambros et al., A uniform system for microRNA annotation, RNA 9:277-279 (2003).

[0080] In some cases, miRNAs can interrupt translation by binding to regulatory sites embedded in the 3'-UTRs of their target mRNAs, leading to the repression of translation. Target recognition involves complementary base pairing of the target site with the miRNA's seed region (positions 2-8 at the miRNA's 5' end), although the exact extent of seed complementarity is not precisely determined and can be modified by 3' pairing. In other cases, miRNAs function like small interfering RNAs (siRNA) and bind to perfectly complementary mRNA sequences to destroy the target transcript.

[0081] The miRNA database available at miRBase (www.mirbase.org) comprises a searchable database of published miRNA sequences and annotation. Further information about miRBase can be found in the following articles, each of which is incorporated by reference in its entirety herein: Griffiths- Jones et al., miRBase: tools for microRNA genomics. NAR 2008 36(Database Issue):D154-D158; Griffiths- Jones et al., miRBase: microRNA sequences, targets and gene nomenclature. NAR 200634(Database Issue):D140-D144; and Griffiths-Jones, S. The microRNA Registry. NAR 2004 32(Database Issue):D109-Dl 11 . Representative miRNAs contained in Release 16 of miRBase, made available September 2010.

[0082] In some embodiments, nucleic acid biomarkers, including nucleic acid payload within a vesicle, is assessed for nucleotide variants. The nucleic acid biomarker may comprise one or more RNA species, e.g., mRNA, miRNA, snoRNA, snRNA, rRNAs, tRNAs, siRNA, hnRNA, shRNA, enhancer RNA (eRNA), or a combination thereof. Techniques to isolate and characterize vesicles and miRNAs are known to those of skill in the art. Similarly, DNA payload can be assessed.

[0083] In certain embodiments, the methods provided herein may include the presence or absence, expression level, mutational state, genetic variant state, or any modification (such as epigenetic modification, or post-translation modification) of a biomarker (e.g. any one or more biomarker listed in Tables 1 and 2). The expression level of a biomarker can be compared to a control or reference, to determine the overexpression or underexpression (or upregulation or downregulation) of a biomarker in a sample. In some embodiments, the control or reference level comprises the amount of a same biomarker, such as a miRNA, in a control sample from a subject that does not have or exhibit the condition or disease. In another embodiment, the control of reference levels comprises that of a housekeeping marker whose level is minimally affected, if at all, in different biological settings such as diseased versus non-diseased states. In yet another embodiment, the control or reference level comprises that of the level of the same marker in the same subject but in a sample taken at a different time point. Other types of controls are described herein.

[0084] Nucleic acid biomarkers include various RNA or DNA species. For example, the biomarker can be single or double- stranded mRNA, microRNA (miRNA), small nucleolar RNAs (snoRNA), small nuclear RNAs (snRNA), ribosomal RNAs (rRNA), heterogeneous nuclear RNA (hnRNA), ribosomal RNA (rRNA), siRNA, transfer RNAs (tRNA), or shRNA. The DNA can be double- stranded DNA (dsDNA), single stranded DNA (ssDNA), complementary DNA (cDNA), or noncoding DNA. miRNAs are short ribonucleic acid (RNA) molecules which average about 22 nucleotides long. miRNAs act as post-transcriptional regulators that bind to complementary sequences in the three prime untranslated regions (3' UTRs) of target messenger RNA transcripts (mRNAs), which can result in gene silencing. One miRNA may act upon 1000s of mRNAs. miRNAs play multiple roles in negative regulation, e.g., transcript degradation and sequestering, translational suppression, and may also have a role in positive regulation, e.g., transcriptional and translational activation. By affecting gene regulation, miRNAs can influence many biologic processes. Different sets of expressed miRNAs are found in different cell types and tissues.

[0085] Biomarkers for use with the invention may further include peptides, polypeptides, or proteins, which terms are used interchangeably throughout unless otherwise noted. In some embodiments, the protein biomarker comprises its modification state, truncations, mutations, expression level (such as overexpression or under expression as compared to a reference level), and/or post-translational modifications, such as described above. In a non-limiting example, a biosignature for a disease can include a protein having a certain post- translational modification that is more prevalent in a sample associated with the disease than without.

[0086] The methods provided herein may include a number of the same type of biomarkers (e.g., two or more different microRNA or mRNA species) or one or more of different types of biomarkers (e.g. mRNAs, miRNAs, proteins, peptides, ligands, and antigens). [0087] Biomarkers that can be derived and analyzed from a vesicle include miRNA (miR), miRNA*nonsense (miR*), and other RNAs (including, but not limited to, mRNA, preRNA, priRNA, hnRNA, snRNA, siRNA, shRNA). A miRNA biomarker can include not only its miRNA and microRNA* nonsense, but its precursor molecules: pri-microRNAs (pri-miRs) and pre-microRNAs (pre-miRs). The sequence of a miRNA can be obtained from publicly available databases such as http://www.mirbase.org/, http://www.microrna.org/, or any others available. Unless noted, the terms miR, miRNA and microRNA are used interchangeably throughout unless noted. In some embodiments, the methods of the invention comprise isolating vesicles, and assessing the miRNA payload within the isolated vesicles. The biomarker can also be a nucleic acid molecule (e.g. DNA), protein, or peptide. The presence or absence, expression level, mutations (for example genetic mutations, such as deletions, translocations, duplications, nucleotide or amino acid substitutions, and the like) can be determined for the biomarker. Any epigenetic modulation or copy number variation of a biomarker can also be analyzed. The one or more biomarkers analyzed can be indicative of a particular tissue or cell of origin, disease, or physiological state. Furthermore, the presence, absence or expression level of one or more of the biomarkers described herein can be correlated to a phenotype of a subject, including a disease, condition, prognosis or drug efficacy.

[0088] Examples

[0089] The invention now being generally described, it will be more readily understood by reference to the following examples, which are included merely for purposes of illustration of certain aspects and embodiments of the present invention, and are not intended to limit the invention.

Diagnosis of Alzheimer's disease based on circulating miRNAs and genetic SNP biomarkers from blood samples.

[0090] The non-invasive diagnostic test for the detection of AD is based on the identification of a panel of circulating miRNAs and on genetic (SNP) biomarkers of three genes of interest (APOE, PICALM and CR1). The levels of miRNA found in the blood of patients affected by AD (relative to those of healthy control individuals), as well as the genotyping of these patients (i.e. identifying polymorphic variants of the genes of interest) were applied to a predictive modeling system. Such predictive modeling systems may include k-Nearest Neighbors algorithms,

Random Forest algorithms, Naive Bayes algorithms (e.g., Gaussian Naive Bayes), and/or Logistic Regression algorithms. Thus, diagnosis or assessment of risk for AD could be made. [0091] Briefly, of 104 patients assessed for eligibility, 70 patients (35 with AD and 35 healthy controls) were selected for analysis via the proposed diagnostic blood test (see FIG. 1). Demographic and clinical characteristics of the patient population are presented in Table 3. Table 3 Demographic and clinical characteristics of patient cohort

[0092] The method of diagnosis was based on the use of quantitative-type polymerase chain reaction (qPCR) which provides a real-time fluorescence emission record in each amplification cycle allowing quantitative analysis (miRNA expression analysis) and/or qualitative (genotype determination). The qPCR reaction was carried out in a device called a thermocycler, which cyclically generates temperature changes in order to amplify the genetic material in analysis. Extraction of blood sample

[0093] The following procedure describes obtaining a blood sample from a patient affected by AD or a healthy patient, as a control. The protocol for the extraction of venous blood is described below.

[0094] 1.-The extraction of venous blood is performed in a 3 ml container comprising EDTA as an anticoagulant agent (e.g., Vacutainer^® Venous Blood Collection Tubes, Becton, Dickinson and Company, New Jersey, USA). The blood sample is stored at room temperature or at 4°C and processed within 1 hour.

[0095] 2. -The blood sample is centrifuged for 10 minutes at a speed of 9000 x g at 4°C.

[0096] 3. -The supernatant of the centrifugation (i.e., plasma) is recovered and poured into a new conical bottom tube. The cell fraction obtained (sediment) is stored at -80°C.

[0097] 4. -The plasma sample is centrifuged for 10 minutes at 16000 x g, at 4°C and the supernatant is stored at -80°C until its later use. Nucleic acid purification

[0098] Genomic DNA (gDNA) and miRNA are collected for further use in the determination of the genotype of patients affected by AD and the expression analysis of biomarkers, respectively. Genomic DNA and miRNA were obtained using commercially available kits PureLink™ Genomic DNA Mini Kit (Invitrogen™) and miRNeasy^® Serum / Plasma kit (Qiagen^®, Dusseldorf, Germany). Such procedures allowed obtaining gDNA from the cellular fraction of a blood sample from a patient affected by AD and its corresponding healthy control, as outlined in FIG. 2.

[0099] gDNA collected using the PureLink™ Genomic DNA Mini Kit was carried out according to the manufacturer's instructions as follows.

[0100] 1.- To a sterile microcentrifuge tube, 200 μL of blood sample were added, followed by 20 μL of Proteinase K.

[0101] 2.- Then, 20 μL of RNase A was mixed well by brief vortexing with the blood sample, and incubated at room temperature for 2 minutes.

[0102] 3.- 200 μL of PureLink Genomic Lysis/Binding Buffer was added and mixed well by vortexing to obtain a homogenous solution. It was left incubating at 55°C for 10 minutes to promote protein digestion.

[0103] 4.- 200 μL of 96-100% ethanol was added to the lysate and mixed well by vortexing for 5 seconds to yield a homogenous solution.

[0104] 5.- The lysate was transferred to the PureLink Spin Column and centrifuged at 10,000 x g for 1 minute at room temperature. The collection tube was discarded and the spin column placed into a clean PureLink Collection Tube.

[0105] 6.- 500 μL of Wash Buffer 1 prepared with ethanol was added to the column. And centrifuged at room temperature at 10,000 x g for 1 minute. The collection tube was discarded and the spin column placed into a clean PureLink collection tube.

[0106] 7.- 500 μL of Wash Buffer 2 prepared with ethanol was added to the column and centrifuged at maximum speed for 3 minutes at room temperature. Once again, the collection tube was discarded.

[0107] 8.- The spin column was placed in a sterile 1.5 mL microcentrifuge tube. And then,

25-200 μL of PureLink Genomic Elution Buffer was added. For 1 minute, the solution was incubated at room temperature.

[0108] 9.- The column was centrifuged at maximum speed for 1 minute at room temperature. This tube contained purified genomic DNA. Once obtained, the gDNA was stored at -20°C until its use. miRNA purification

[0109] Extraction of miRNA from patient plasma samples was carried out with the miRNeasy^® Serum / Plasma advanced kit as follows, in accordance with the manufacturer's instructions.

[0110] 1. -Serum or plasma were prepared, or thawed from frozen samples.

[0111] 2.- 200-250^1* μl of plasma were transfered into a 2 ml reaction vessel.

[0112] 3.- 60 mΐ of buffer RPL* was added and mixed vigorously by vortexing for over 20 seconds and left at room temperature for 3 minutes.

[0113] 4.- To the homogenate, 3.5 mΐ miRNeasy^® Serum/Plasma Spike-In Control was added

(at 1.6 x 10⁸ copies/μl).

[0114] 5.- 20 μl of buffer RPP* was added and mixed vigorously by vortexing for over 20 seconds and incubated at room temperature for 3 minutes.

[0115] 6.- Samples were then centrifuged at 13500 rpm for 3 minutes at room temperature to pellet the precipitate.

[0116] 7.- The supernatant was transferred to a new reaction tube and then added 1 volume of isopropanol. The resulting solution was mixed well by vortexing and transferred to an RNeasy UCP MiniElute column. The entire sample was centrifuge for 15 seconds at 11000 rpm and the flow-through discarded.

[0117] 8.- 700 μl of buffer RWT was added to the RNeasy UCP MinElute spin column.

Then, centrifuged for 15 seconds at 11000 rpm. The flow-through discarded.

[0118] 9.- 500 μl of buffer RPE was added to the RNeasy UCP MinElute spin column. Then, centrifuged for 15 seconds at 11000 rpm. The flow-through discarded.

[0119] 10.- 500 μl of 80% ethanol was added to the RNeasy UCP MinElute spin column.

Then, centrifuged for 2 minutes at 11000 rpm. The flow-through and the collection tube discarded.

[0120] 11- The RNeasy UCP MinElute spin column was placed in a new 2 ml collection tube and centrifuged at full speed for 5 min to dry the membrane. Any flow-through was discarded.

[0121] 12.- The RNeasy^® MinElute^® spin column was placed in a new 1.5 ml collection tube. Directly to the center of the spin column membrane, 14 μl RNase-free water was added. The lid was gently closed, and the spin column/ collection tube centrifuged for 1 min at full speed to elute the RNA. The eluted miRNA was stored at -80°C until its use. 1_* buffer volume was adjusted according to plasma volume added to the reaction vessel. The quantities described in the present patent correspond to 200 uL of plasma. [0122] Once the purification of gDNA and miRNA was completed, the genotype was determined by identifying SNPs in the genes of interest and analysis of miRNA expression in the biomarkers of interest, respectively.

Analysis of genotype determination in patients affected by AD

[0123] As mentioned above, the analysis of genotypes was based on the identification of SNPs in three genes of interest.

[0124] The identification of each of the SNPs shown in Table 2 was carried out using a predesigned test, i.e., a TaqMan^® SNP Genotyping assay, provided by ThermoFisher Scientific (Massachusetts, USA). The use of this assay allowed the detection and amplification of specific alleles in gDNA using the qPCR technique.

[0125] Each pre-designed test consists of two Taqman^® probes that have different fluorophores and a pair of “starters” to detect a specific SNP target. Each Taqman^® probe and the pair of provided primers can only be joined and amplified to the allele of interest. The test was provided in the form of a 20x solution containing the Taqman^® probes and the specific starters for the target SNPs. As shown in the outline of the process in FIG. 3, the Taqman^® assay consists of a pair of specific non-labeled starters and a Taqman probe which has at its 5'OH end a fluorophore which can be FAM or VIC^® fluorescent dye. At its 3'OH end, the Taqman probe has a non-fluorescent quencher that represses the emission of fluorescence when the Taqman probe is intact and specifically linked to the target sequence. When the PCR reaction occurs, the exonuclease activity of the Taq Polymerase enzyme cuts only the Taqman probe that is hybridized to the NFQ and produces the emission of fluorescence only if the target sequence is complementary to the probe. Thusly, the PCR reaction produces an increase in the emission of fluorescence and, depending on the type of signal emitted according to the probe that is linked to FAM or VIC, it can be discriminated which allele is present in the sample.

[0126] The TaqMan SNP Genotyping Assay protocol for genotyping based on the detection of SNPs on genes of interest was conducted as follows:

[0127] 1.-The number of necessary test reactions, including the corresponding controls, were calculated.

[0128] 2. -The following reagents were mixed according to the number of reactions required for the assay as indicated in Table 4, which shows the volume of reagents required for 1 reaction. (An excess of 10% of the indicated volume was used, as shown in Table 4). Table 4: Volume of reagents required for 1 reaction (rxn).

* See step 4

[0129] 3. -The reactions were briefly centrifuged to bring the reaction mixture to the bottom of the wells and eliminate air bubbles. [0130] 4. -Each sample of genomic DNA, including controls, was diluted in nuclease-free water to a concentration of 1-20 ng/μl. The diluted DNA sample must have a volume of 11.25 μl. [0131] 5. -The appropriate volume of reaction mix was added to each well to achieve a total volume of 25 μl (i.e., the sum of the reaction mix (13.75 μl) and the diluted DNA sample (11.25 μl))· [0132] 6.- The reactions were, again, centrifuged briefly to bring the reaction mixture to the bottom of the wells and eliminate air bubbles.

[0133] 7.- The thermocycler was programmed with the qPCR parameters indicated in Table

5, the reaction plate was loaded and the qPCR program was initiated. Table 5: qPCR program parameters

Analysis ofmiRNA expression in patients affected by AD

[0134] The analysis of miRNA expression in patients affected by AD and their respective healthy control was performed using qPCR, such as the TaqMan^® MicroRNA Assays predesigned assay provided by ThermoFisher Scientific (Massachusetts, USA). The use of this assay allowed the analysis of expression of selected miRNAs, as outlined in FIG. 4.

[0135] Each pre-designed assay consists of a specific RT -primer for the miRNA under study, a pair of specific nucleic acid primers, and a probe. A schematic of this process is shown in FIGS. 5A-5H.

[0136] The protocol of the MicroRNA Assay and qPCR for the analysis of miRNA expression in patients was as follows.

[0137] 1.-RNA template (i.e., the miRNA isolated from the patient) and the 5X RT -primer were thawed on ice. Before use, RT -primer tubes were mixed by vortexing then centrifuge briefly.

[0138] 2.- Reagents from the TaqMan^® MicroRNA Reverse Transcription Kit were mixed as indicated in Table 6, in accordance with the number of reactions required for each assay.

Table 6: Volume of reagents required for 1 RT rxn.

* See step 4

[0139] 3.- the RT reactions were mixed and centrifuged to carry the reaction mixture to the bottom of the tube. The reaction tubes were then left on ice.

[0140] 4.- For each RT reaction (15 μl final volume), 10 μl of the RT mix, 2 μl of quenched

RNA (1-10 ng), and 3 μl of loop-stem primer RT 5X were added. The RT reactions were then mixed and centrifuged briefly, and incubated on ice for 5 minutes. [0141] 5. -The thermocycler was programmed with the RT parameter indicated in Table 7, the reaction plate was loaded and the RT program was initiated. If PCR amplification could not be run immediately after the RT run, the reactions were stored at -15 to -25 °C. Table 7: RT reaction parameters

[0142] 6.- For the subsequent qPCR reaction, the reaction volume was adjusted to 20 μL.

The reaction plate for each qPCR was prepared as shown in Table 8. Table 8: Volume of reagents required for 1 qPCR reaction

* This volume has a built-in 20% excess included due to pipetting losses.

[0143] 7. -The appropriate volume of reaction mix was added to each well (20 μL) and the qPCR program, having the parameters noted Table 9, was initiated. Table 9 qPCR parameters for amplification of RT reaction product

Machine learning and predictive modeling based on miRNA and SNP expression data [0144] A predictive model for diagnosis of AD in a patient is generated by applying a “random forest” or “random decision forest” ensemble-learning algorithm. Random forests give an estimate of how well individuals in a new data set can be classified into existing groups by creating a set of classification trees based on continual sampling of the experimental input variables. Each observation is classified based on the majority votes from all the classification trees. Input variables may include patient's age, miRNA expression levels, and a coded version of the SNP1 (ApoE) gene.

[0145] Briefly, a decision tree is used to create a model that predicts the value of a target variable based on several input variables. Tree models where the target variable can take a discrete set of values are called classification trees. A classification tree splits training data (i.e., a set of examples used to fit the parameters of the model) into disjointed regions of the predictor space. To classify a new sample, the predicted value will be the mean of the response value for the training observations that lie in the same region as the sample. The regions are built in a top- down, greedy fashion (e.g., top-down induction of decision trees or TDIDT); meaning that for each step, the data is successively split by one of the input variables, the best split is made at that particular step (rather than looking ahead and picking a split that will lead to a better tree in some future step), and a cut point which minimizes the residual sum of squares that result from applying the predictive strategy above described to all of the training samples. This is done until a stopping criterion is reached.

[0146] The random forests learning method then applies bootstrapping (“bootstrap aggregating” or “bagging”) to construct a multitude of classification trees (hence a forest), each of which are fitted to a randomly chosen subset of the predictor variables. Tree “bagging” consists essentially of sampling subsets of the training set, fitting a decision tree to each, and aggregating their result. After each tree is fitted, the prediction for a new sample is taken by evaluating every tree in the forest with the new dataset (e.g. miRNA levels associated with the new sample) and taking the mode of the multiplicity of trees as the final prediction.

Results

[0147] In addition to being less invasive, the blood sample analysis disclosed herein also provides significant and reliable improvements in both sensitivity and specificity relative to existing diagnostic methods.

Table 10: Preliminary comparison of blood sample analysis to current AD diagnostic tests

[0148] The fact that such biomarker analysis can be obtained from blood samples provides advantages in terms of ease and abundance for collecting samples, reduced costs relative to current diagnostics, and improved patient compliance and follow-up. Otherwise, the values of specificity and sensitivity shown in Table 10 demonstrate that the blood sample analysis disclosed herein already reach values of sensitivity and specificity that equal, and in some cases exceed, the values observed in conventional diagnostic methods. Greater sensitivity and specificity can be expected with the inclusion of further miRNA targets.

[0149] While specific embodiments of the subject invention have been discussed, the above specification is illustrative and not restrictive. Many variations of the invention will become apparent to those skilled in the art upon review of this specification and the claims below. The full scope of the invention should be determined by reference to the claims, along with their full scope of equivalents, and the specification, along with such variations.

[0150] Any of the methods (including user interfaces) described herein may be implemented as software, hardware or firmware, and may be described as a non-transitory computer-readable storage medium storing a set of instructions capable of being executed by a processor (e.g., computer, tablet, smartphone, etc.), that when executed by the processor causes the processor to control perform any of the steps, including but not limited to: displaying, communicating with the user, analyzing, modifying parameters (including timing, frequency, intensity, etc.), determining, alerting, or the like.

[0151] Terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. For example, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and/or groups thereof. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items and may be abbreviated as "/".

[0152] Although the terms “first” and “second” may be used herein to describe various features/elements (including steps), these features/elements should not be limited by these terms, unless the context indicates otherwise. These terms may be used to distinguish one feature/element from another feature/element. Thus, a first feature/element discussed below could be termed a second feature/element, and similarly, a second feature/element discussed below could be termed a first feature/element without departing from the teachings of the present invention.

[0153] Throughout this specification and the claims which follow, unless the context requires otherwise, the word “comprise”, and variations such as “comprises” and “comprising” means various components can be co-jointly employed in the methods and articles (e.g., compositions and apparatuses including device and methods). For example, the term “comprising” will be understood to imply the inclusion of any stated elements or steps but not the exclusion of any other elements or steps.

[0154] In general, any of the apparatuses and methods described herein should be understood to be inclusive, but all or a sub-set of the components and/or steps may alternatively be exclusive, and may be expressed as “consisting of’ or alternatively “consisting essentially of’ the various components, steps, sub-components or sub-steps.

[0155] As used herein in the specification and claims, including as used in the examples and unless otherwise expressly specified, all numbers may be read as if prefaced by the word "about" or “approximately,” even if the term does not expressly appear. The phrase “about” or “approximately” may be used when describing magnitude and/or position to indicate that the value and/or position described is within a reasonable expected range of values and/or positions. For example, a numeric value may have a value that is +/- 0.1% of the stated value (or range of values), +/- 1% of the stated value (or range of values), +/- 2% of the stated value (or range of values), +/- 5% of the stated value (or range of values), +/- 10% of the stated value (or range of values), etc. Any numerical values given herein should also be understood to include about or approximately that value, unless the context indicates otherwise. For example, if the value "10" is disclosed, then "about 10" is also disclosed. Any numerical range recited herein is intended to include all sub-ranges subsumed therein. It is also understood that when a value is disclosed that "less than or equal to" the value, "greater than or equal to the value" and possible ranges between values are also disclosed, as appropriately understood by the skilled artisan. For example, if the value "X" is disclosed the "less than or equal to X" as well as "greater than or equal to X" (e.g., where X is a numerical value) is also disclosed. It is also understood that the throughout the application, data is provided in a number of different formats, and that this data, represents endpoints and starting points, and ranges for any combination of the data points. For example, if a particular data point “10” and a particular data point “15” are disclosed, it is understood that greater than, greater than or equal to, less than, less than or equal to, and equal to 10 and 15 are considered disclosed as well as between 10 and 15. It is also understood that each unit between two particular units are also disclosed. For example, if 10 and 15 are disclosed, then 11, 12, 13, and 14 are also disclosed.

[0156] Although various illustrative embodiments are described above, any of a number of changes may be made to various embodiments without departing from the scope of the invention as described by the claims. For example, the order in which various described method steps are performed may often be changed in alternative embodiments, and in other alternative embodiments one or more method steps may be skipped altogether. Optional features of various device and system embodiments may be included in some embodiments and not in others. Therefore, the foregoing description is provided primarily for exemplary purposes and should not be interpreted to limit the scope of the invention as it is set forth in the claims.

[0157] The examples and illustrations included herein show, by way of illustration and not of limitation, specific embodiments in which the subject matter may be practiced. As mentioned, other embodiments may be utilized and derived there from, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure.

Such embodiments of the inventive subject matter may be referred to herein individually or collectively by the term “invention” merely for convenience and without intending to voluntarily limit the scope of this application to any single invention or inventive concept, if more than one is, in fact, disclosed. Thus, although specific embodiments have been illustrated and described herein, any arrangement calculated to achieve the same purpose may be substituted for the specific embodiments shown. This disclosure is intended to cover any and all adaptations or variations of various embodiments. Combinations of the above embodiments, and other embodiments not specifically described herein, will be apparent to those of skill in the art upon reviewing the above description.

Claims

CLAIMS What is claimed is:

1. A system comprising: a plurality of miRNA target detection polynucleotides comprising at least four miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51:

CAGUUAUCACAGUGCUGAUGCU (SEQ ID NO. 1),

U AGC ACC AU CU G A A AUCGGUU A (SEQ ID NO. 2),

U AGC ACC AUUU G A A AU C AGU GUU (SEQ ID NO. 3),

U AGC ACC AUUU G A A AUCGGUU A (SEQ ID NO. 4),

CGU GUUC AC AGCGG ACCUU GAU (SEQ ID NO. 5), GCUAUUUCACGACACCAGGGUU (SEQ ID NO. 6),

C AAC ACC AGUCG AU GGGCU GU (SEQ ID NO. 7),

AGGC AGU GU AGUU AGCU G AUU GC (SEQ ID NO. 8),

U AGC AGC ACGU A A AU AUU GGCG (SEQ ID NO. 9),

U C ACCGGGU GU A A AU C AGCUU G (SEQ ID NO. 10), UCAAGUCACUAGUGGUUCCGUUUAG (SEQ ID NO. 11), AGU GGC AC AU GUUU GUU GU GAG (SEQ ID NO. 12), CGGGAGCUGGGGUCU GC AGGU (SEQ ID NO. 13),

C AU A A AGU AG A A AGC ACU ACU (SEQ ID NO. 14),

AU CU GU CUCG AUU GUUU CC AG (SEQ ID NO. 15),

AAU GC ACCCGGGC AAGGAUU CU (SEQ ID NO. 16),

UCU G AC AU C AGU G AUU CUCCUG (SEQ ID NO. 17),

GUU CC AC ACU G AC ACU GC AG A AGU (SEQ ID NO. 18), UCGU GGCCU GGU CU CC AUU AU (SEQ ID NO. 19), UAGUACUGUGCAUAUCAUCUAU (SEQ ID NO. 20), CAGGUCACGUCUCUGCAGUUAC (SEQ ID NO. 21),

GU GUU A AUU A A ACCU CU AUUU AC (SEQ ID NO. 22), UCAGUCACAUAUCUAGUGUCUA (SEQ ID NO. 23),

AUU A AGG AC AUUU GU G AUU GAU (SEQ ID NO. 24), AGUU CUU C AGU GGC A AGCUUU A (SEQ ID NO. 25), ACAGGUCCUAAGAGACUGCAU (SEQ ID NO. 26), UUCACCCCUCUCACCUAAGCAG (SEQ ID NO. 27), AGGGGGAAAGUUCUAUAGUCC (SEQ ID NO. 28),

AU CU GU CUCG AUU GUUU CC AG (SEQ ID NO. 29), UUCCCAGCCAACGCACCA (SEQ ID NO. 30), CGGGGCAGCUCAGUACAGGAU (SEQ ID NO. 31),

A A A AGU AUUU GC GGGUUUU GU C (SEQ ID NO. 32),

AGU GU GGCUUU CUU AG AGC (SEQ ID NO. 33),

GAAGU GCUUCGAUUUU GGGGU GU (SEQ ID NO. 34), UAAAUCCCAUGGUGCCUUCUCCU (SEQ ID NO. 35),

GCU GC ACCGGAGACU GGGUAA (SEQ ID NO. 36),

U C AGU GC AU C AC AG A ACUUU GU (SEQ ID NO. 37),

A A AGGU A ACU GU G AUUUUU GCU (SEQ ID NO. 38),

ACCGU GGCUUUCG AUU GUU ACU (SEQ ID NO. 39),

U A AC AGU CU AC AGCC AU GGUCG (SEQ ID NO. 40), UAACAGUCUCCAGUCACGGCC (SEQ ID NO. 41),

U CUUU GGUU AU CU AGCU GU AU G A (SEQ ID NO. 42),

A AC AUU C A ACCU GUCGGU G AGU (SEQ ID NO. 43),

C A AU C ACU A ACU CC ACU GCC AU (SEQ ID NO. 44),

AGCAGCAUU GUACAGGGCUAUC A (SEQ ID NO. 45), UAGCAGCACAGAAAUAUUGGC (SEQ ID NO. 46),

U C AUUUUU GU G AU GUU GC AGCU (SEQ ID NO. 47),

U GAG A ACU G A AUU CC AU GGGUU (SEQ ID NO. 48),

U AC AGU ACU GU G AU A ACUG A A (SEQ ID NO. 49),

U AGCUU AUC AG ACU G AU GUU G A (SEQ ID NO. 50), and U GU AGU GUUUCCU ACUUU AU GG A (SEQ ID NO. 51); one or more processors; and a memory coupled to the one or more processors, the memory storing computer- program instructions, that, when executed by the one or more processors, perform a computer-implemented method comprising: receiving an expression level for each of the miRNA target detection polynucleotide in the plurality of miRNA target detection polynucleotides from a patient sample; analyzing, in the one or more processors, the expression level for each of the miRNA target detection polynucleotide in the plurality of miRNA target detection polynucleotides from the patient sample using a trained neural network, wherein the trained neural network is trained on a dataset including the expression each of the miRNA target detection polynucleotide in the plurality of miRNA target detection polynucleotides, so that the trained neural network determines a risk score; and outputting the risk score from the trained neural network, wherein the risk score indicates a risk of Alzheimer' s disease in the patient.

2. The system of claim 1, wherein the computer-implemented method further comprises classifying the patient into one of at least two classes based on the risk score.

3. The system of claim 1, wherein outputting the risk score further comprises outputting a progression score indicating progression of Alzheimer's disease in the patient.

4. The system of claim 1, wherein the plurality of miRNA target detection polynucleotides comprises all of the miRNA target detection polynucleotides having sequences selected from Seq. ID Nos. 1-51.

5. The system of claim 1, further comprising a plurality of SNP detection polynucleotides comprising:

GACGTG[C/T]GCGGCC (SEQ ID No. 52),

CAGAAG[C/T]GCCTGG (SEQ ID NO. 53),

AACGAT [ A/C] T AAGCG (SEQ ID NO. 54), and CTTCTC[A/G]TCGCCT (SEQ ID NO. 55).

6. The system of claim 5, wherein the computer-implemented method further comprises receiving an expression level for each of the SNP detection polynucleotides from a patient sample, and wherein analyzing, further comprises analyzing the expression level for each of the SNP detection polynucleotides from the patient sample using the trained neural network so that the trained neural network uses the expression pattern of the SNP detection polynucleotides from a patient sample in determining the risk score.

7. A method for determining Alzheimer's disease risk in a patient, comprising: obtaining a sample from the patient, wherein the sample comprises at least one miRNA set forth in Table 1, or combinations thereof; contacting the sample with a reagent; generating a complex between the reagent and the miRNA; detecting the complex to obtain a first dataset associated with the sample, wherein the first dataset comprises quantitative expression data for the miRNA; and analyzing the first dataset to assess the expression level of the miRNA, wherein the expression level of the miRNA positively or negatively correlates with an increased risk of AD in the patient.

8. The method of claim 7, wherein the sample comprises each of the miRNAs set forth in Table 1.

9. The method of any one of claims 1-8, further comprising assessing a single nucleotide polymorphism marker (SNP) in the patient; and combining said assessment with the assessment of the expression level of the miRNA(s) to identify risk of Alzheimer's disease in the patient.

10. The method of claim 9, comprising assessing at least one of the SNPs set forth in Table 2, or combinations thereof.

11. The method of claim 9, comprising assessing each of the SNPs set forth in Table 2.

12. The method of any one of claims 1 to 11, further comprising assessing the cognitive and/or neuropsychiatric status of the patient.

13. The method of any one of claims 1 to 12, wherein the sample is a blood sample.

14. The method of any one of claims 1 to 13, wherein the control sample is associated with a control subject or with a control population.

15. The method of claim 14, wherein the control sample is obtained from the patient prior to manifestation of Alzheimer's.

16. The method of claim 15, wherein the control sample is associated with a control subject or a control population characterized by analysis of cerebrospinal fluid (CSF) for Aβ_{1 -42.} total Tau (T-Tau), and phosphorylated Tau (¹⁸¹Thr P-tau) levels.

17. The method of any one of claims 1 to 16, wherein expression of the miRNA is significantly decreased compared to expression of the control miRNA.

18. The method of claim 17, wherein expression of the miRNA is significantly increased compared to expression of the control miRNA.

19. The method of any one of claims 17 or 18, wherein the expression level of the miRNA marker positively correlates with an increased risk of Alzheimer's disease in the subject.

20. The method of any one of claims 17 or 18, wherein the expression level of the miRNA marker negatively correlates with an increased risk of Alzheimer's disease in the subject.

21. The method of any one of claims 1 to 20, wherein the method is implemented, at least in part, on one or more computers.

22. The method of claim 21, wherein the assessment employs a predictive modeling system.

23. The method of any one of claims 4 to 22, wherein the first dataset is obtained stored on a storage memory.

24. The method of claim 23, wherein obtaining the first dataset associated with the sample comprises obtaining the sample and processing the sample to experimentally assess the sample.

25. The method of claim 23, wherein obtaining the first dataset associated with the sample comprises receiving the first dataset directly or indirectly from a third party that has processed the sample to experimentally assess the sample.

26. The method of claim 5, wherein the statistically significant difference is determined, at least in part, using a predictive modeling system.

27. The method of any one of claims 1 to 26, wherein the expression levels are obtained from a nucleotide-based assay.

28. The method of claim 27, wherein the expression levels are obtained from an RT-PCR assay, a sequencing-based assay, a microarray assay, or a combination thereof.

29. The method of any one of claims 1 to 28, wherein the subject is a human subject.

30. A computer-implemented method for identifying a subject at risk of Alzheimer's disease, comprising: storing, in a storage memory, a first dataset associated with a sample obtained from the subject, wherein the first dataset comprises quantitative expression data for a miRNA selected from the group set forth in Table 1; and analyzing, by a computer processor, the first dataset to determine the expression level of the miRNA, wherein the expression level of the miRNA marker positively or negatively correlates with an increased risk of Alzheimer' s disease in the patient.

31. The method of claim 30, wherein the first dataset comprises quantitative expression data for each of the miRNAs set forth in Table 1.

32. The method of any one of claims 30 or 31, further comprising storing, in a storage memory, at least a second dataset associated with the sample obtained from the subject, wherein the second dataset comprises quantitative data for a SNP selected from the group set forth in Table 2; and analyzing, by a computer processor, the second dataset to determine the SNP, wherein the presence of the SNP positively or negatively correlates with an increased risk of Alzheimer' s disease in the patient.

33. The method of claim 32, wherein the second dataset comprises quantitative data for each of the SNPs set forth in Table 2.

34. The method of any one of claims 32 to 33, further comprising combining the analysis of the first dataset with the analysis of the second data set to diagnose Alzheimer's disease, predict risk of developing Alzheimer' s disease, or predict an outcome of Alzheimer' s disease in a patient; wherein combining the analysis of the first dataset with the analysis of the second data set comprises implementation of a predictive modeling system.

35. A computer program product useful for performing the method according to any one of the preceding claims, comprising: a) means for receiving data representing an expression level of at least one miRNA in a patient blood sample selected from the group set forth in Table 1, or combinations thereof; b) means for receiving data representing at least one control pattern of expression levels for comparing with the expression level of the at least one miRNA from said sample; c) means for comparing said data representing the expression level of the at least one miRNA in a patient sample; and d) means for determining a diagnosis of Alzheimer's disease, a prediction of a risk of developing Alzheimer' s disease, or a prediction of an outcome of Alzheimer' s disease from the outcome of the comparison in step b); wherein the means of step d) comprises a predictive modeling system.

36. A kit for use in quantifying Alzheimer's disease risk in a patient, comprising: a set of reagents comprising a plurality of reagents for determining from a blood sample obtained from the patient quantitative expression data for a miRNA selected from the group set forth in Table 1, or combinations thereof; instructions for using the plurality of reagents to determine quantitative expression data from the sample for a first dataset, and analyzing said first dataset by comparing the first dataset to determine the expression level of the miRNA marker, wherein the expression level of the miRNA marker positively or negatively correlates with an increased risk of AD in the subject.