CSF MICRORNA MARKERS OF ALZHEIMER'S DISEASE

CROSS-REFERENCE TO RELATED APPLICATION

Benefit is claimed to U.S. Provisional Application No. 62/377,196 filed August 19, 2016.

FIELD

The disclosed methods are for identifying persons at risk of Alzheimer's disease, or confirming a diagnosis of Alzheimer's disease. Also disclosed are methods of selecting people for interventions designed to lower risk factors associated with Alzheimer's disease. More

specifically, the field involves miRNA based tests that are indicative of Alzheimer's disease.

ACKNOWLEDGEMENT OF GOVERNMENT SUPPORT

The work resulting in the inventions disclosed herein was supported by the United States government under the terms of the following grants: NCATS UH2-TR000903, NIA P30-AG08017, and NCATS UL1-TR000128, all of which were awarded by the National Institutes of Health. The United States government has certain rights to this invention.

BACKGROUND

Alzheimer's disease (AD) is the most common form of dementia, accounting for 60-80% of cases, and is the sixth leading cause of death in the United States (U.S.). The greatest known risk factor for AD is increasing age— the majority of people with AD are age 65 or older. AD is a progressive disease, and dementia symptoms gradually worsen over several years. Today in the U.S., 5.4 million Americans have AD, and that number is estimated to climb to as high as 16 million by 2050, barring the development of medical breakthroughs to prevent or cure the disease (Alzheimer's Association, http://www.alz.org/) Alzheimer's Association, 2015 Alzheimer's disease facts and figures. Alzheimers Dement 11, 332-384 (2015); incorporated by reference herein)

In "The Global Impact of Dementia 2013-2050" policy brief, it is estimated that 46.8 million people worldwide are living with dementia in 2015, and that this number will almost double every 20 years, reaching 131.5 million in 2050 (Alzheimer's Disease International,

http://www.alz.co.uk/G8policybrief). The current cost of AD in the U.S. is estimated to be $100 billion each year and is among the most costly disease in this country. The estimated cost of health care, long-term care and hospice for people with AD and other dementias in the U.S. is estimated to be $236 billion in 2016. These costs increase with dementia severity and the presence of behavioral disturbances that result in increased caregiving time required for physical care. Therefore, any treatment that slows cognitive decline, delays institutionalization, or reduces caregivers' hours will have tremendous economic and social benefits. Further, while current AD treatments cannot stop disease progression, they can temporarily reduce the worsening of dementia symptoms and improve quality of life for those with AD and their caregivers. Thus, a preclinical tool that can be used to diagnose AD earlier in stage and allow treatments to be initiated earlier in the disease would be a great asset. Several PET ligands for imaging cerebral amyloid pathology have been developed for this purpose, but the high cost of PET and the lack of information regarding non- amyloid disease mechanisms limit the usefulness of these imaging studies.

Cerebrospinal fluid (CSF) is the biofluid that directly bathes the brain and spinal cord, and is most informative regarding any changes that are occurring in disease states. Thus, CSF serves as an excellent candidate for biomarker studies in neuropathological brain diseases such as AD (Ghidoni R et al, Neurodegener Dis 8, 413-420 (2011); incorporated by reference herein). The most extensively studied CSF protein biomarkers of AD include Abeta ₄₂ , tau, and phospho-tau, which have received intense study because they are associated with the "classical" AD pathology of plaques and tangles. While these CSF biomarkers have some diagnostic utility, they have not performed very well as outcome measures in clinical trials (Quinn JF, J Alzheimers Dis 33 Suppl 1, S371-376 (2013); incorporated by reference herein). Perhaps more importantly, the bias towards plaque and tangle pathology has limited the ability to identify other pathogenic mechanisms, which might be plausible with a more open-ended approach to CSF biomarker development. The existence of extracellular RNAs (exRNAs) in biofluids represents a fertile molecular landscape from which diagnostic and prognostic biomarkers may be accessed, characterized, and exploited. Accordingly, the identification of exRNAs in CSF provides an opportunity to define important biomarkers that characterize and differentiate central nervous system (CNS) diseases (Rao P et al, Front Mol Neurosci 6, 39 (2013); incorporated by reference herein). MiRNAs are the most well studied exRNA species as they are found in virtually all biofluids including CSF, saliva, plasma, serum, and urine. Thus, their persistence and altered abundance in the extracellular fluid or CSF may play a role in the spreading of the disease throughout the brain, such as occurs in AD.

MiRNAs are members of the non-protein-coding family of RNAs that serve as regulators of post-transcriptional gene expression (Chekulaeva M and Filipowicz W, Curr Opin Cell Biol 21, 452-460 (2009); incorporated by reference herein). MiRNAs are small, -20-24 nucleotide, genomically encoded RNAs that regulate messenger RNA (mRNA) translation and/or stability by base-pairing to sequences in the mRNA, usually in the 3' untranslated region (3'UTR) (Bartel DP Cell 136, 215-233 (2009); incorporated by reference herein). Complete complementarity between the entire nucleotide length of the miRNA and mRNA leads to degradation, hence 'silencing' of the transcript. In contrast, partial base pairing, specifically between nucleotides 2-8 of the miRNA (its seed sequence) and mRNA leads primarily to translation suppression and some degradation (Baek D et al, Nature 455, 64-71 (2008); incorporated by reference herein). Importantly, miRNAs are stable in circulating fluids, presumably because they are contained within ribonucleoprotein complexes or membrane vesicles that afford them protection against nuclease digestion.

SUMMARY

Disclosed is a method of assessing the presence of Alzheimer's disease in a subject. The method involves obtaining a cerebrospinal fluid sample from a subject. It further involves extracting total RNA from the sample and measuring a first expression level of a first miRNA, a first expression level of a second miRNA and a first expression level of a third miRNA in the sample via reverse transcription polymerase chain reaction. Each of the first, second, and third miRNAs comprise a sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8; SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31 SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, or SEQ ID NO: 36. The method further comprises measuring a second expression level of the first miRNA, a second expression level of the second miRNA and a second expression level of the third miRNA via reverse transcription polymerase chain reaction of a negative control. For a first miRNA, second miRNA or third miRNA

comprising a sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, or SEQ ID NO: 8 a result showing that the first expression level is greater than the second expression level indicates that the subject has

Alzheimer's disease. For a first miRNA second miRNA or third miRNA comprising a sequence selected from SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31 SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, or SEQ ID NO: 36, a result showing that the first expression level is less than the second expression level indicates that the subject has Alzheimer's disease. The first, second and third miRNAs have different sequences. In some examples, the method involves measuring the expression level of at least three of SEQ ID NO: 19, SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 10, SEQ ID NO: 3, SEQ ID NO:

I, SEQ ID NO: 7, SEQ ID NO: 6, SEQ ID NO: 15, SEQ ID NO: 33, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 29, SEQ ID NO: 22; SEQ ID NO: 25, SEQ ID NO: 8, SEQ ID NO: 12, SEQ ID NO: 30, SEQ ID NO: 2, SEQ ID NO: 36, SEQ ID NO: 4, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 27, SEQ ID NO: 31, SEQ ID NO: 18, SEQ ID NO: 21, SEQ ID NO: 26, SEQ ID NO: 28, or SEQ ID NO: 34 in any combination including one or more of the following combinations: SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 6; SEQ ID NO: 3, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 14; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 13; SEQ ID NO: 10, SEQ ID NO: 14, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 6; SEQ ID NO: 10, SEQ ID NO: 34, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 25; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 7; SEQ ID NO: 15, SEQ ID NO: 19, SEQ ID NO: 1; SEQ ID NO: 1 1, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 17, SEQ ID NO: 33, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 15, SEQ ID NO: 12; SEQ ID NO: 17, SEQ ID NO: 33, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 15, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 15, SEQ ID NO: 14; SEQ ID NO: 17, SEQ ID NO: 19, SEQ ID NO: 1; SEQ ID NO: 22, SEQ ID NO: 16, SEQ ID NO: 6; SEQ ID NO: 19, SEQ ID NO: 1, SEQ ID NO: 4; SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 17; SEQ ID NO: 3, SEQ ID NO: 1, SEQ ID NO: 2; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 29, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 33; SEQ ID NO: 29, SEQ ID NO: 19, SEQ ID NO: 20; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 17; SEQ ID NO:

I I, SEQ ID NO: 10, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 1; SEQ ID NO: 3, SEQ ID NO: 25, SEQ ID NO: 19; SEQ ID NO: 17, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 10, SEQ ID NO: 24, SEQ ID NO: 19; SEQ ID NO: 33, SEQ ID NO: 12, SEQ ID NO: 19; SEQ ID NO: 14, SEQ ID NO: 29, SEQ ID NO: 19; SEQ ID NO: 10, SEQ ID NO: 30, SEQ ID NO: 6; SEQ ID NO: 10, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 17, SEQ ID NO: 14, SEQ ID NO: 7; SEQ ID NO: 17, SEQ ID NO: 19, SEQ ID NO: 6; SEQ ID NO: 17, SEQ ID NO: 6, SEQ ID NO: 7; SEQ ID NO: 14, SEQ ID NO: 19, SEQ ID NO: 1; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 1; SEQ ID NO: 30, SEQ ID NO: 8, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 16, SEQ ID NO: 1; SEQ ID NO: 22, SEQ ID NO: 17, SEQ ID NO: 6; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 36; SEQ ID NO: 22, SEQ ID NO: 3, SEQ ID NO: 2; SEQ ID NO: 10, SEQ ID NO: 19, SEQ ID NO: 1; SEQ ID NO: 3, SEQ ID NO: 25, SEQ ID NO: 1; SEQ ID NO: 11, SEQ ID NO: 19, SEQ ID NO: 4; SEQ ID NO: 13, SEQ ID NO: 17, SEQ ID NO: 19; SEQ ID NO: 33, SEQ ID NO: 14, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 17, SEQ ID NO: 19; SEQ ID NO: 10, SEQ ID NO: 14, SEQ ID NO: 20; SEQ ID NO: 10, SEQ ID NO: 19, SEQ ID NO: 20; SEQ ID NO: 17, SEQ ID NO: 30, SEQ ID NO: 8; SEQ ID NO: 16, SEQ ID NO: 36, SEQ ID NO: 1; SEQ ID NO: 14, SEQ ID NO: 25, SEQ ID NO: 7; SEQ ID NO: 22, SEQ ID NO: 11, SEQ ID NO: 10; SEQ ID NO: 10, SEQ ID NO: 12, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 1; SEQ ID NO: 23, SEQ ID NO: 16, SEQ ID NO: 6; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 7; SEQ ID NO: 13, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 19; SEQ ID NO: 17, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 6; SEQ ID NO: 15, SEQ ID NO: 31, SEQ ID NO: 19; SEQ ID NO: 16, SEQ ID NO: 1, SEQ ID NO: 20; SEQ ID NO: 3, SEQ ID NO: 29, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 15, SEQ ID NO: 1; SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 33; SEQ ID NO: 17, SEQ ID NO: 23, SEQ ID NO: 6; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 29; SEQ ID NO: 27, SEQ ID NO: 33, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 19; SEQ ID NO: 14, SEQ ID NO: 19, SEQ ID NO: 6; SEQ ID NO: 3, SEQ ID NO: 24, SEQ ID NO: 2; SEQ ID NO: 10, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 14, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 15, SEQ ID NO: 36; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 8; SEQ ID NO: 11, SEQ ID NO: 14, SEQ ID NO: 25; SEQ ID NO: 3, SEQ ID NO: 15, SEQ ID NO: 20; SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 20; SEQ ID NO: 22, SEQ ID NO: 16, SEQ ID NO: 19; SEQ ID NO: 15, SEQ ID NO: 12, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 10; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 18; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 21; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 27; SEQ ID NO: 11, SEQ ID NO: 33, SEQ ID NO: 8; SEQ ID NO: 17, SEQ ID NO: 14, SEQ ID NO: 6; SEQ ID NO: 33, SEQ ID NO: 19, SEQ ID NO: 20; SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 19; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 14; SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 26; SEQ ID NO: 3, or SEQ ID NO: 15, SEQ ID NO: 31.

In still further examples the methods involve measuring a first expression level of a fourth miRNA comprising a sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8; SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31 SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, and SEQ ID NO: 36 in the sample via reverse transcription polymerase chain reaction and a second expression level of the fourth miRNA via reverse transcription polymerase chain reaction of the negative control, provided that the fourth miRNA has a different sequence from the first, second, and third miRNAs.

In still further examples, the first, second, third and fourth miRNAs are selected from SEQ ID NO: 7, SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 14 SEQ ID NO: 30, SEQ ID NO: 1, SEQ ID NO: 19, SEQ ID NO: 22, SEQ ID NO: 24, SEQ ID NO: 26, SEQ ID NO: 23, SEQ ID NO: 27, SEQ ID NO: 20; SEQ ID NO: 6, SEQ ID NO: 12, SEQ ID NO: 18, SEQ ID NO: 35, and SEQ ID NO: 36 in any combination including one or more of the following combinations: SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 17, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 22, SEQ ID NO: 3, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 27, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 17, SEQ ID NO: 23, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 26, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 1, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 24, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 22, SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 14, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 19, SEQ ID NO: 1, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 7, SEQ ID NO: 20; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 7, SEQ ID NO: 20; SEQ ID NO: 22, SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 24, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 1 1, SEQ ID NO: 27, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 19, SEQ ID NO: 1; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 18, SEQ ID NO: 7; SEQ ID NO: 22, SEQ ID NO: 17, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 22, SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 17; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 24, SEQ ID NO: 17, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 14, SEQ ID NO: 30, SEQ ID NO: 1, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 27, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 26, SEQ ID NO: 17, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 17, SEQ ID NO: 23, SEQ ID NO: 7; SEQ ID NO: 22, SEQ ID NO: 3, SEQ ID NO: 14, SEQ ID NO: 7; SEQ ID NO: 30, SEQ ID NO: 19, SEQ ID NO: 1, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 7; SEQ ID NO: 11, SEQ ID NO: 14, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 14, SEQ ID NO: 30, SEQ ID NO: 19, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 24, SEQ ID NO: 7; SEQ ID NO: 17, SEQ ID NO: 23, SEQ ID NO: 30, SEQ ID NO: 7; SEQ ID NO: 3, SEQ ID NO: 27, SEQ ID NO: 14, SEQ ID NO: 19; SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 26, SEQ ID NO: 7; and SEQ ID NO: 3, SEQ ID NO: 11, SEQ ID NO: 1, SEQ ID NO: 20.

In yet other examples, the first, second, third and fourth miRNAs are selected from SEQ ID NO: 22, SEQ ID NO: 20, SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 34 SEQ ID NO: 27, SEQ ID NO: 26, SEQ ID NO: 32, SEQ ID NO: 24, SEQ ID NO: 15, SEQ ID NO: 31, SEQ ID NO: 17, SEQ ID NO: 33, SEQ ID NO: 14; SEQ ID NO: 25, SEQ ID NO: 30, SEQ ID NO: 16, SEQ ID NO: 12, SEQ ID NO: 29, SEQ ID NO: 36, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 21 SEQ ID NO: 9, and SEQ ID NO: 7 in any combination including one or more of the following

combinations: SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 7, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 32;

SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 32; SEQ ID NO: 15, SEQ ID NO: 1, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 33, SEQ ID NO: 32; SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 32, SEQ ID NO: 29; and SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 27, SEQ ID NO: 32.

In yet other embodiments, the methods of the invention involve measuring a first expression level of a fifth miRNA comprising a sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8; SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO:

26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31 SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, and SEQ ID NO: 36 in the sample via reverse transcription polymerase chain reaction and a second expression level of the fifth miRNA via reverse transcription polymerase chain reaction of the negative control, provided that the fifth miRNA has a different sequence from the first, second, third and fourth miRNAs.

In some embodiments, the first, second, third, fourth and fifth miRNAs are selected from SEQ ID NO: 22, SEQ ID NO: 20, SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 34 SEQ ID NO:

27, SEQ ID NO: 26, SEQ ID NO: 32, SEQ ID NO: 24, SEQ ID NO: 15, SEQ ID NO: 31, SEQ ID NO: 17, SEQ ID NO: 33, SEQ ID NO: 14; SEQ ID NO: 25, SEQ ID NO: 30, SEQ ID NO: 16, SEQ ID NO: 12, SEQ ID NO: 29, SEQ ID NO: 36, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 21 SEQ ID NO: 9, and SEQ ID NO: 7 in any combination including one or more of the following combinations: SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 33, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 33, SEQ ID NO: 32; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 33, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 15,

SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36; and SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 32, SEQ ID NO: 29.

In other embodiments, the methods of the invention involve measuring a first expression level of a sixth miRNA comprising a sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8; SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31 SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, and SEQ ID NO: 36 in the sample via reverse transcription polymerase chain reaction and a second expression level of the sixth miRNA via reverse transcription polymerase chain reaction of the negative control, provided that the sixth miRNA has a different sequence from the first, second, third, fourth and fifth miRNAs.

In some embodiments, the first, second, third, fourth, fifth and sixth miRNAs are selected from SEQ ID NO: 22, SEQ ID NO: 20, SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 34 SEQ ID NO: 27, SEQ ID NO: 26, SEQ ID NO: 32, SEQ ID NO: 24, SEQ ID NO: 15, SEQ ID NO: 31, SEQ ID NO: 17, SEQ ID NO: 33, SEQ ID NO: 14; SEQ ID NO: 25, SEQ ID NO: 30, SEQ ID NO: 16, SEQ ID NO: 12, SEQ ID NO: 29, SEQ ID NO: 36, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 21 SEQ ID NO: 9, and SEQ ID NO: 7 in any combination including one or more of the following combinations: SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 33, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO:

16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO:

17, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ

ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36; and SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29.

In yet other embodiments, the methods of the invention involve measuring a first expression level of a seventh miRNA comprising a sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8; SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25,

SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31 SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, and SEQ ID NO: 36 in the sample via reverse transcription polymerase chain reaction and a second expression level of the seventh miRNA via reverse transcription polymerase chain reaction of the negative control, provided that the seventh miRNA has a different sequence from the first, second, third, fourth, fifth and sixth miRNAs.

In some embodiments, the first, second, third, fourth, fifth, sixth and seventh miRNAs are selected from SEQ ID NO: 22, SEQ ID NO: 20, SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 34 SEQ ID NO: 27, SEQ ID NO: 26, SEQ ID NO: 32, SEQ ID NO: 24, SEQ ID NO: 15, SEQ ID NO:

31, SEQ ID NO: 17, SEQ ID NO: 33, SEQ ID NO: 14; SEQ ID NO: 25, SEQ ID NO: 30, SEQ ID NO: 16, SEQ ID NO: 12, SEQ ID NO: 29, SEQ ID NO: 36, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 21 SEQ ID NO: 9, and SEQ ID NO: 7 in any combination including one or more of the following combinations: SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 33, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 33, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 8, SEQ ID NO: 33, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO:

32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 1, SEQ ID NO: 33, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 27, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 15, SEQ ID NO: 7, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 36; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 26, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 20, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36; SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 27, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29; and SEQ ID NO: 17, SEQ ID NO: 9, SEQ ID NO: 8, SEQ ID NO: 16, SEQ ID NO: 32, SEQ ID NO: 36, SEQ ID NO: 29.

The methods can further involve testing the ApoE genotype of the subject, including testing whether the subject is heterozygous or homozygous for ApoE4. Testing the ApoE genotype of the subject can occur at any time including prior to the obtaining the cerebrospinal fluid from the subject.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

Figure 1 is a plot showing that linear combinations of candidate AD miRNAs in CSF increase sensitivity and specificity. 2-marker models result in an AUC of 0.75 while 3-marker models increase AUC to 0.80 and the 4-marker models increase the AUC to 0.82.

Figure 2 is a plot showing that the addition of the ApoE genotype to miRNA increases sensitivity and specificity. Although ApoE alone has an AUC of 0.73 in the subjects measured and the 3-miRNA marker models result in an AUC of 0.80, the 3 miRNA model in combination with ApoE genotype increases the AUC to 0.84. Figure 3 is a diagram showing the flow of original analysis and reanalysis of candidate CSF miRNA markers. The box labeled A shows that of the 754 probes on the array, 338 miRNA candidates were included in the original analysis. The box labeled B shows that 46 of those were selected as potential AD biomarkers and candidates for the verification studies by the original analysis. Of these, 20 were verified and 26 were not. The box labeled C shows that of the 754 RNA array probes, 151 miRNA candidates were included in the reanalysis. The box labeled D shows that 36 miRNA targets (SEQ ID NO: 1 - SEQ ID NO: 36 herein) were selected as candidates based on the reanalysis. Of these, the same 20 were verified from the original analysis and 16 new candidates identified.

Figure 4 shows the projections for both miRNA-only models and miRNA + ApoE4 models based on 26 top candidates (left panel), and the results of applying the same nearest-neighbor classifier to the cooperative cohort using the same miRNA markers (right panel). The AUC values for the two model types applied to the cohort are larger than projected, and the effect of adding ApoE4 to miRNA models is the same magnitude in both cohorts, showing that the overall classification ability of the candidate miRNAs and their complementary relationship with ApoE genotype is well supported by the current experimental evidence.

Figure 5 shows the effect of the most effective miRNAs biomarkers.

SEQUENCE LISTING

SEQ ID NO: 1 refers to human miR-378a-3p.

SEQ ID NO: 2 refers to human miR-520b.

SEQ ID NO: 3 refers to human miR-1291.

SEQ ID NO: 4 refers to human miR-603.

SEQ ID NO: 5 refers to human miR-202-3p.

SEQ ID NO: 6 refers to human miR-519b-3p.

SEQ ID NO: 7 refers to human miR-597-5p.

SEQ ID NO: 8 refers to human miR-484.

SEQ ID NO: 9 refers to human miR-584-5p.

SEQ ID NO: 10 refers to human miR-143-3p.

SEQ ID NO: 11 refers to human miR-142-3p.

SEQ ID NO: 12 refers to human miR-328-3p.

SEQ ID NO: 13 refers to human miR-145-3p.

SEQ ID NO: 14 refers to human miR-24-3p.

SEQ ID NO: 15 refers to human miR-193a-5p SEQ ID NO: 16 refers to human miR-■30a-3p.

SEQ ID NO: 17 refers to human miR- ■19b-3p.

SEQ ID NO: 18 refers to human miR- ■30d-5p.

SEQ ID NO: 19 refers to human miR- ■340-5p.

SEQ ID NO: 20 refers to human miR- ^■ 140-5p.

SEQ ID NO: 21 refers to human miR- ■532-5p.

SEQ ID NO: 22 refers to human miR- ■125b-5p

SEQ ID NO: 23 refers to human miR- ■26b-5p.

SEQ ID NO: 24 refers to human miR- ■16-5p.

SEQ ID NO: 25 refers to human miR- ■28-3p.

SEQ ID NO: 26 refers to human miR- ■146b-5p

SEQ ID NO: 27 refers to human miR- ■146a-5p.

SEQ ID NO: 28 refers to human miR- ■27b-3p.

SEQ ID NO: 29 refers to human miR- ■331-3p.

SEQ ID NO: 30 refers to human miR- ■29a-3p.

SEQ ID NO: 31 refers to human miR- ■195-5p.

SEQ ID NO: 32 refers to human miR- ■15b-5p.

SEQ ID NO: 33 refers to human miR- ■223-3p.

SEQ ID NO: 34 refers to human miR- ■145-5p.

SEQ ID NO: 35 refers to human miR- ■590-5p.

SEQ ID NO: 36 refers to human miR- ■365a-3p.

DETAILED DESCRIPTION

The following explanations of terms and methods are provided to better describe the present disclosure and to guide those of ordinary skill in the art in the practice of the present disclosure.

Alzheimer's disease (AD): A progressive brain disorder that occurs gradually and results in memory loss, behavioral and personality changes, and a decline in mental abilities. These losses are related to the death of brain cells and the breakdown of the connections between them. The course of this disease varies from person to person, as does the rate of decline. On average, AD patients live for 8 to 10 years after they are diagnosed, though the disease can last up to 20 years. AD advances by stages, from early, mild forgetfulness to a severe loss of mental function. At first, AD destroys neurons in parts of the brain that control memory, especially in the hippocampus and related structures. As nerve cells in the hippocampus stop functioning properly, short-term memory fails. AD also attacks the cerebral cortex, particularly the areas responsible for language and reasoning.

Dementias of all types including, but not limited to, AD result in progressive deterioration in the functioning of the subject and result in steadily worsening behavioral problems that coincide with the deterioration in cognitive functioning and are part of the same disease process. Typical behavioral problems shown by subjects with AD include, but are not limited to, depression, psychosis, delusions, sleep disturbance, wandering, anger outbursts, aggression, agitation, apathy, anxiety, suspiciousness, fearfulness and paranoia. In the final stages of most forms of AD, including AD, victims are bedridden, lose urinary and bowel control and suffer epileptic attacks. Death is usually due to pneumonia or urinary tract infection.

The clinical manifestations of AD are fairly characteristic, memory disturbance occurs early in the disease; subjects have difficulty learning and remembering new material. Spatial and temporal disorientation also may occur early, with subjects becoming lost in familiar surroundings. Aphasia, apraxia and acalculia develop as the disease progresses, and apathy or paranoia may occur. Subjects often have delusions of theft and spousal infidelity. Subjects may wander, pace, open and close drawers repeatedly, and repeat the same questions. Sleep-wake cycle abnormalities may become evident; for example, a subject may be awake at night but think that it is daytime. Activities of daily living decline throughout the illness. Subjects lose the ability to eat and groom themselves and have difficulty dressing. In the terminal stages of the disease, subjects exhibit cognitive decline in virtually all intellectual spheres, motor abnormalities become evident and both urinary and fecal incontinence develops.

A feature of AD is the development of multiple cognitive deficits that include memory impairment and at least one of the following cognitive disturbances: aphasia, apraxia, agnosia or a disturbance in executive functioning. The cognitive deficits must be sufficiently severe to cause impairment in occupational or social functioning and must represent a decline from a previously higher level of functioning.

Memory impairment is required to make the diagnosis of AD and is a prominent early symptom. Individuals with AD become impaired in their ability to learn new material, or they forget previously learned material. Most individuals with AD have both forms of memory impairment, although it is sometimes difficult to demonstrate the loss of previously learned material early in the course of the disorder. They may lose valuables like wallets and keys, forget food cooking on the stove, and become lost in unfamiliar neighborhoods. In advanced stages of AD, memory impairment is so severe that the person forgets his or her occupation, schooling, birthday, family members and sometimes even name. Memory may be formally tested by asking the person to register, retain, recall and recognize information. The ability to learn new information may be assessed by asking the individual to learn a list of words. The individual is requested to repeat the words (registration), to recall the information after a delay of several minutes (retention, recall), and to recognize the words from a multiple list (recognition). Individuals with difficulty learning new information are not helped by clues or prompts, e.g., multiple-choice questions, because they did not learn the material initially. In contrast, individuals with primarily retrieval deficits can be helped by clues and prompts because their impairment is in the ability to access their memories. Remote memory may be tested by asking the individual to recall personal information or past material that the individual found of interest, e.g., politics, sports, entertainment. It is also useful to determine (from the individual and informants) the impact of the memory disturbances on the individual's functioning, e.g., ability to work, shop, cook, pay bills, return home without getting lost.

Deterioration of language function (aphasia) may be manifested by difficulty producing the names of individuals and objects. The speech of individuals with aphasia may become vague or empty, with long circumlocutory phrases and excessive use of terms of indefinite reference, such as "thing" and "it". Comprehension of spoken and written language and repetition of language may also be compromised. In the advanced stages of AD, individuals may be mute or have a deteriorated speech pattern characterized by echolalia, i.e., echoing what is heard; or palilalia, i.e., repeating sounds or words over and over. Language is tested by asking the individual to name objects in the room, e.g., tie, dress, desk, lamp; or body parts, e.g., nose, chin, shoulder, follow commands, e.g., "point at the door and then at the table"; or repeat phrases, e.g., "no ifs, ands or buts".

Amplifying a nucleic acid molecule: To increase the number of copies of a nucleic acid molecule, such as a cerebrospinal fluid miRNA. The resulting products are called amplification products. An example of in vitro amplification is the polymerase chain reaction (PCR). Other examples of in vitro amplification techniques include quantitative real-time PCR.

A commonly used method for real-time quantitative polymerase chain reaction involves the use of a double stranded DNA dye (such as SYBR Green I dye). For example, as the amount of PCR product increases, more SYBR Green I dye binds to DNA, resulting in a steady increase in fluorescence. SYBR green binds to double stranded DNA, but not to single stranded DNA. In addition, SYBR green fluoresces strongly at a wavelength of 497 nm when it is bound to double stranded DNA, but does not fluoresce when it is not bound to double stranded DNA. As a result, the intensity of fluorescence at 497 nm may be correlated with the amount of amplification product present at any time during the reaction. The rate of amplification may in turn be correlated with the amount of template sequence present in the initial sample. Generally, Ct values are calculated similarly to those calculated using the TaqMan® system below. Because the probe is absent, amplification of the proper sequence may be checked by any of a number of techniques. One such technique involves running the amplification products on an agarose or other gel appropriate for resolving nucleic acid fragments and comparing the amplification products from the quantitative real time PCR reaction with control DNA fragments of known size.

Another commonly used method is real-time quantitative TaqMan® PCR (Applied

Biosystems). This type of PCR has reduced the variability traditionally associated with quantitative PCR, thus allowing the routine and reliable quantification of PCR products to produce sensitive, accurate, and reproducible measurements of levels of gene expression. The PCR step can use any of a number of thermostable DNA-dependent DNA polymerases, it typically employs a Taq DNA polymerase, which has a 5'-3' nuclease activity but lacks a 3'-5' proofreading endonuclease activity. Thus, TaqMan® PCR typically utilizes the 5'-nuclease activity of Taq or Tth polymerase to hydrolyze a hybridization probe bound to its target amplicon, but any enzyme with equivalent 5' nuclease activity can be used.

Two oligonucleotide primers are used to generate an amplicon typical of a PCR reaction. A third oligonucleotide, or probe, is designed to detect nucleotide sequence located between the two PCR primers. The probe is nonextendible by Taq DNA polymerase enzyme, and is labeled with a reporter fluorescent dye and a quencher fluorescent dye. Any laser-induced emission from the reporter dye is quenched by the quenching dye when the two dyes are located close together as they are on the probe. During the amplification reaction, the Taq DNA polymerase enzyme cleaves the probe in a template-dependent manner. The resultant probe fragments disassociate in solution, and signal from the released reporter dye is free from the quenching effect of the second fluorophore. One molecule of reporter dye is liberated for each new molecule synthesized, and detection of the unquenched reporter dye provides the basis for quantitative interpretation of the data.

Examples of fluorescent labels that may be used in quantitative PCR include but need not be limited to: HEX, TET, 6-FAM, JOE, Cy3, Cy5, ROX TAMRA, and Texas Red. Examples of quenchers that may be used in quantitative PCR include, but need not be limited to TAMRA (which may be used as a quencher with HEX, TET, or 6-FAM), BHQ1, BHQ2, or DABCYL. TAQMAN® RT-PCR can be performed using commercially available equipment, such as, for example, ABI PRISM 7700® Sequence Detection System (Perkin-Elmer- Applied Biosystems), or Lightcycler (Roche Molecular Biochemicals).

In one embodiment, the 5' nuclease procedure is run on a real-time quantitative PCR device such as the ABI PRISM 7700® Sequence Detection System. The system includes a thermocycler, laser, charge-coupled device (CCD), camera and computer. The system amplifies samples in a 96- well format on a thermocycler. During amplification, laser-induced fluorescent signal is collected in real time through fiber optic cables for all 96 wells, and detected at the CCD. The system includes software for running the instrument and for analyzing the data.

In some examples, 5'-nuclease assay data are initially expressed as Ct, or the threshold cycle. As discussed above, fluorescence values are recorded during every cycle and represent the amount of product amplified to that point in the amplification reaction. The point when the fluorescent signal is first recorded as statistically significant is the threshold cycle (Ct).

To minimize errors and the effect of sample-to-sample variation, RT-PCR can be performed using an internal standard. The ideal internal standard is expressed at a constant level among different tissues, and is unaffected by the experimental treatment. RNAs most frequently used to normalize patterns of gene expression are the mRNA products of housekeeping genes.

ApoE: ApoE plays a potential role in the pathogenesis of AD. ApoE performs various functions as a protein constituent of plasma lipoproteins, including its role in cholesterol metabolism. It was first identified as a constituent of liver-synthesized very low density

lipoproteins (VLDL) that transport triglycerides from the liver to peripheral tissues.

Polymorphism analysis of the ApoE gene can be used to indicate whether or not a subject has or will develop AD (Scarmeas et al, Neurology 58, 1182-1188 (2002); incorporated by reference herein). The ApoE gene contains three major isoforms, ApoE2, ApoE3 and ApoE4, which differ from one another only by single amino acid substitutions.

ApoE is a plasma protein involved in cholesterol transport and is encoded by a gene on chromosome 19. It is synthesized primarily in the liver and is thought to be involved in repair of the nervous systems after injury. ApoE genotype is an important contributor to susceptibility to AD but is found in several other neurodegenerative diseases, such as dementia pugilistica. Three common alleles, E2, E3 and E4, correspond to six phenotypes. The ε4 allele has been identified as a risk factor for AD, with the attributable risk estimated to be 45-60%. E4 homozygotes are at a greater risk than E4 heterozygotes. ApoE4 is present in plaques and may facilitate amyloid accumulation in the brain.

The association of the ApoE4 allele with a higher risk and an earlier onset of AD is probably due to its higher affinity for β-amyloid protein compared to other isoforms, which results in the reduced clearance of β-amyloid protein (Selkoe, Nature, Vol. 399, A23-A31 (1999);

incorporated by reference herein)

There are three major isoforms of ApoE, referred to as ApoE2, ApoE3 and ApoE4 which are products of three alleles at a single gene locus. Three homozygous phenotypes (ApoE2/2, E3/3 and E4/4) and three heterozygous phenotypes (ApoE3/2, E4/3 and E4/2) arise from the expression of any two of the three alleles. The most common phenotype is ApoE3/3 and the most common allele is E3 (Mahley, Science, 240, 622-630 (1988); incorporated by reference herein).

The amino acid sequences of the three types differ only slightly. ApoE4 differs from ApoE3 in that in ApoE4 arginine is substituted for the normally occurring cysteine at amino acid residue 112. The most common form of ApoE2 differs from ApoE3 at residue 158, where cysteine is substituted for the normally occurring arginine. ApoE phenotypes and genotypes are well-described and known in the art as described above. The established nomenclature system as well as the phenotypes and genotypes for ApoE, are described in, for example, Zannis et al, J. Lipid Res., 23, 911-914 (1982); incorporated by reference herein.

Subjects with the ApoE4/4 genotype are as much as eight times as likely to be affected by AD as subjects with the ApoE2/3 or ApoE3/3 genotypes. Further, the average age of onset of AD and the average age of survival is lower for those having one ApoE4 allele, and lowest for those having two ApoE4 alleles (see U.S. Pat. No. 5,508, 167, incorporated by reference herein in its entirety).

Methods of testing the ApoE4 genotype of a subject are well known in the art, including for example, DNA sequencing methods, nucleic amplification systems such as the amplification refractory mutation system, as well as hybridization methods involving microarrays.

Array: An arrangement of molecules, such as biological macromolecules (such as peptides or nucleic acid molecules) or biological samples (such as tissue sections), in addressable locations on or in a substrate. A "microarray" is an array that is miniaturized. In certain example arrays, one or more molecules (such as reagents to perform TaqMan® polymerase chain reaction on a specific miRNA) will occur on the array a plurality of times. The number of addressable locations on the array can vary, for example from at least one, to at least 2, to at least 3, at least 4, at least 5, at least 6, at least 10, at least 20, at least 30, at least 50, at least 75, at least 100, at least 150, at least 200, at least 300, at least 500, least 550, at least 600, at least 800, at least 1000, at least 10,000, or more. In some examples, arrays include positive and/or negative controls, such as addressable locations that amplify housekeeping genes as internal controls. In an example, the array is a commercially available array such as a TaqMan ^® Array Human MicroRNA set (ThermoFisher).

Within an array, each arrayed sample is addressable, in that its location can be reliably and consistently determined within at least two dimensions of the array. The feature application location on an array can assume different shapes. For example, the array can be regular (such as arranged in uniform rows and columns) or irregular. Thus, in ordered arrays the location of each sample is assigned to the sample at the time when it is applied to the array, and a key may be provided in order to correlate each location with the appropriate target or feature position. Often, ordered arrays are arranged in a symmetrical grid pattern, but samples could be arranged in other patterns (such as in radially distributed lines, spiral lines, or ordered clusters). Addressable arrays may be computer readable, in that a computer can be programmed to correlate a particular address on the array with information about the sample at that position (such as TaqMan probe binding data, including for instance signal intensity). In some examples of computer readable formats, the individual features in the array are arranged regularly, for instance in a Cartesian grid pattern, which can be correlated to address information by a computer.

Control: A reference standard. The control can comprise a negative control which indicates the expression levels of the disclosed miRNA in a subject that does not have Alzheimer's disease. Such a negative control can be a CSF sample from one or more patients known not to have Alzheimer's disease. The negative control can be a CSF sample from a subject with another form of dementia such as Parkinson's dementia or vascular dementia, or from a subject with no form of dementia. The negative control can be a pooled sample from several individuals. The negative control can also be an artificial CSF fluid spiked with miRNA at levels similar to that observed in normal patients. A control can also comprise a positive control from a subject known to have Alzheimer's disease or an artificial CSF fluid spiked with miRNA at levels similar to that observed in Alzheimer's patients.

Expression: The process by which the coded information of a gene is converted into an operational, non-operational, or structural part of a cell, such as the synthesis of an RNA such as a microRNA. As disclosed herein, expression involves the expression of a particular set of microRNAs in a patient sample being higher or lower than that of a negative control.

Indicating: Performance of the disclosed testing methods provides an indication that a subject may or is likely to have a particular disease or condition. In particular, the disclosed tests provide an indication that the subject may have or is likely to have Alzheimer's disease. A test that provides an indication does not provide a certain diagnosis of a disease, such as Alzheimer's disease. Instead, such a test provides a piece of information that a subject is at least more likely than not to have Alzheimer's disease. Such tests can be combined with other tests and clinical observations to provide a more definitive prediction.

Testing methods differ in their sensitivity and specificity. The "sensitivity" of a testing method is the percentage of diseased individuals who test positive (percent of true positives). The "specificity" of a diagnostic assay is 1 minus the false positive rate, where the false positive rate is defined as the proportion of those without the disease who test positive. Measuring the expression level: As disclosed herein, measuring the expression level of a miRNA in a sample or a control involves using reverse transcription polymerase chain reaction. This includes the use of an array such as a TaqMan® low density array.

MicroRNA: MicroRNAs are a major class of biomolecules involved in control of gene expression. For example, in human heart, liver or brain, miRNAs play a role in tissue specification or cell lineage decisions. In addition, miRNAs influence a variety of processes, including early development, cell proliferation and cell death, and apoptosis and fat metabolism. The large number of miRNA genes, the diverse expression patterns and the abundance of potential miRNA targets suggest that miRNAs may be a significant source of genetic diversity.

A mature miRNA is typically an 18-25 nucleotide non-coding RNA that regulates expression of an mRNA including sequences complementary to the miRNA. These small RNA molecules are known to control gene expression by regulating the stability and/or translation of mRNAs. For example, miRNAs bind to the 3' UTR of target mRNAs and suppress translation. MiRNAs may also bind to target mRNAs and mediate gene silencing through the RNAi pathway. MiRNAs may also regulate gene expression by causing chromatin condensation.

A miRNA silences translation of one or more specific mRNA molecules by binding to a miRNA recognition element (MRE,) which is defined as any sequence that directly base pairs with and interacts with the miRNA somewhere on the mRNA transcript. Often, the MRE is present in the 3' untranslated region (UTR) of the mRNA, but it may also be present in the coding sequence or in the 5' UTR. MREs are not necessarily perfect complements to miRNAs, usually having only a few bases of complementarity to the miRNA and often containing one or more mismatches within those bases of complementarity. The MRE may be any sequence capable of being bound by a miRNA sufficiently that the translation of a gene to which the MRE is operably linked (such as a CMV gene that is essential or augmenting for growth in vivo is repressed by a miRNA silencing mechanism such as the RISC. A microRNA can interchangeably be abbreviated to 'miRNA' or 'miR' .

Nucleic acid molecules: A deoxyribonucleotide or ribonucleotide polymer including, without limitation, cDNA, mRNA, genomic DNA, and synthetic (such as chemically synthesized) DNA. The nucleic acid molecule can be double-stranded or single-stranded. Where single- stranded, the nucleic acid molecule can be the sense strand or the antisense strand. In addition, nucleic acid molecule can be circular or linear.

Nucleotide sequences or nucleic acid sequences: The terms "nucleotide sequences" and "nucleic acid sequences" refer to deoxyribonucleic acid (DNA) or ribonucleic acid (RNA) sequences, including, without limitation, messenger RNA (mRNA), DNA/RNA hybrids, or synthetic nucleic acids. The nucleic acid can be single-stranded, or partially or completely double stranded (duplex). Duplex nucleic acids can be homoduplex or heteroduplex.

Sample (or biological sample): A biological specimen containing microRNA that is obtained from a subject. As disclosed herein, a sample comprises cerebrospinal fluid collected by, for example, a spinal tap.

Subject: Living multi -cellular vertebrate organisms, a category that includes human and non-human mammals, such as mice. In some examples a subject is a human patient, including a patient with some form of dementia. Clinical examples described herein involve human subjects, Subject interventions: In some embodiments, the methods of the invention further comprise lowering the risk of developing Alzheimer's Disease, delaying the onset of Alzheimer's Disease, or slowing the progression of Alzheimer's Disease in a subject in need thereof. Lowering the risk of developing Alzheimer's Disease, delaying the onset of Alzheimer's Disease, or slowing the progression of Alzheimer's Disease in a subject in need thereof may be performed by any means, including, but not limited to, treating the subject with antibodies to beta-amyloid, treating or preventing a cardiovascular disease, such as high blood pressure, diabetes and high cholesterol, preventing head trauma (with head protection), increasing a regimen of physical exercise for the subject, and modifying the diet or eating habits of the subject. In particular examples, the person identified as being at increased risk of developing Alzheimer's Disease, or having incipient or established Alzheimer's Disease, is advised to adopt behaviors or undergo treatments that lower cardiovascular risk (such as regular exercise, reducing obesity, treating hypertension, lowering blood glucose levels, lowering blood lipids such as LDL and triglycerides, or improving ratios of HDL to LDL, and/or lipoprotein particle size). These interventions can lower the risk of cardiovascular disease that has been associated with brain amyloid deposition and dementia.

Gottesman et al., Association Between Midlife Vascular Risk Factors and Estimated Brain

Amyloid Deposition, JAMA 317(14): 1443 (2017).

For example, if the subject's total cholesterol is elevated (for example greater than 200) the subject is treated with diet or a pharmaceutical agent (such as a statin) to lower the total cholesterol below 200, and/or lower LDL below 130, 120, 100 or less. If the subject's triglycerides are elevated (for example greater 200), the subject is treated with diet or a pharmaceutical agent to lower triglycerides. In another example, if the triglyceride to HDL ratio is greater than 2: 1 that ratio is lowered toward or below 2: 1. In another example, if the ratio of total cholesterol/HDL is greater than 3.5 to 1, interventions are advised or adopted that lower that ratio toward to below 3.5 to 1. In another example, the subject is administered vitamins associated with improved neurological function, such as vitamins B6, B12 and/or folate. In particular examples, the subject is administered methyl cobalamin and/or methyl-tetrahydrofolate. In some examples, the vitamins are administered at dosages that lower blood homocysteine levels of the subject to less than 12, for example less than 10. In other examples, the subject is administered (alone or in combination with the vitamins) an agent that increases omega-3 fatty acid levels, such as fish oil or krill oil. If the subject has blood pressure that is elevated above normal levels, an anti-hypertensive drug is administered to lower the blood pressure. Examples of such anti-hypertensive drugs include an angiotensin converting enzyme (ACE) inhibitor, a thiazide diuretic, a beta blocker, an angiotensin II receptor blocker (ARB), a calcium channel blocker (CCB), and renin inhibitors. If a subject smokes tobacco or drinks unhealthy amounts of alcohol (such as more than 2 drinks a day for a male or 1 drink per day for a woman), the subject is advised to cease those behaviors or is treated with a pharmaceutical agent to improve tobacco-cessation (such as bupropion or a nicotine patch), or encourage alcohol abstinence (such as naltrexone, disulfiram, topiramate or baclofen). miRNAs in Alzheimer's disease

Several studies have demonstrated altered expression of miRNAs in AD (Cogswell JP et al, J Alzheimers Dis 14, 27-41 (2008); Heber SS et al, Proc Natl Acad Sci U S A 105, 6415-6420 (2008); Geekiyanage H and Chan C, J Neurosci 31, 14820-14830 (2011); Geekiyanage H et al, Exp Neurol 235, 491-496 (2012); Kumar P et al, PLoS One 8, e69807 (2013); Burgos K et al, PLoS One 9, e94839 (2014); Kiko T et al, J Alzheimers Dis 39,253-259 (2014); Dorval V et al, Front Mol Neurosci 6, 24 (2013); and Kim DH et al, Gene 545, 185-193 (2014); all of which are incorporated by reference herein). However, many such studies were not performed using CSF and many of those done with CSF collected postmortem. The miRNAs disclosed in the present application were observed in CSF samples obtained from living donors. TaqMan ^® human arrays were used to quantify miRNA expression in CSF from AD patients and controls, and a panel of miRNAs was identified that can discriminate AD from a negative control. Also disclosed are linear combinations of multiple candidate miRNAs that improve the sensitivity and specificity of biomarker performance beyond that of the ApoE genotype. Further segregating subjects by ApoE in combination with the linear miRNA combinations further improved the predictive power of the disclosed tests. Our studies showed decreased expression of miR-146a-p and miR-125b-5p, and increased expression of miR-29a-3p in CSF, in AD versus control (Table 2). The top 20 miRNAs identified in the original analysis did hold up as candidate miRNAs for the validation studies and were consistent under the stringent reanalysis. The reanalysis added an additional 16 candidate miRNAs that were not detected originally. EXAMPLES

Example 1 - Methods

Subjects: the Institutional Review Board of Oregon Health & Science University (OHSU) approved all of the donor procedures (TRB 6845); all subjects provided written informed consent. Participants underwent detailed clinical and laboratory evaluation, including cognitive testing and interview with a collateral historian. The samples were banked at the Oregon Alzheimer's Disease Center (OADC), the core program of the Layton Aging & Alzheimer's Disease Center, supported by the National institute on Aging.

CSF Sample collection: The OADC has standardized their CSF collection protocol to correspond to that used in other AD centers engaged in CSF biomarker research (Shi M et al, Ann Neurol 69, 570-580 (2011); incorporated by reference herein). All CSF examinations are done in the AM under fasting conditions, in the lateral decubitus position, with a 24-gauge Sprotte spinal needle that minimizes the discomfort of the procedure and reduces the incidence of lumbar puncture headache. The first 3 mL of CSF collected are sent to the clinical lab for cell count, and determination of glucose and total protein levels, and the remaining sample is snap-frozen and stored in a standardized fashion. Next, serial syringes of 5 mL CSF are collected, mixed, transferred to polypropylene tubes in 0.5 mL aliquots, and the tubes numbered to account for any gradient effect in subsequent experiments. All CSF tubes have an OADC subject number, but no other identifying information, in order to facilitate later collaborations and sharing of samples. Immediately after aliquots are transferred, the CSF is frozen on dry ice and stored in a -80°C freezer.

Apolipoprotein E genotyping: DNA was isolated from blood and amplified by Touchdown PCR with 250 μΜ dNTPs, 1 Unit Taq DNA Polymerase, buffer, IX Q-solution (Qiagen), and 0.5 μΜ forward and reverse primers for ApoE exon 4 (E4 allele). A product size of 443 nucleotides identified on a 1% agarose IX TBE gel was excised, cleaned with ExoSAP-IT reagent

(Affymetrix), and sequenced on a model 377 automated fluorescence sequencer (Applied

Biosystems). Chromatogram traces were examined and nucleotide sequences determined using FinchTV (Geospiza, Inc.).

RNA isolation and amplification: Total RNA was extracted from 0.5 mL of each CSF sample using the mirVana™ PARIS™ RNA and Native Protein Purification Kit (ThermoFisher Scientific), modified to include 2 aqueous extractions during the organic phase extraction steps in order to maximize RNA recovery (Burgos 2014 supra). RNA samples were concentrated using the RNA Clean & Concentrator™-5 Kit (Zymo Research) and eluted in 9 μΙ_, RNAse/DNase-free water. RNA concentrations were initially measured on a set of test CSF samples using the Quant- iT™ RiboGreen® RNA Assay Kit (ThermoFisher Scientific). The average concentration for the test group was 133 pg/μΐ _^ with a total RNA recovery of approximately 1 ng/mL CSF. The concentrated RNA samples were converted to cDNA and pre-amplified using a T-100 thermocycler (Bio-Rad, Hercules, CA) with Megaplex™ RT Primers, Human Pool Set v3.0, as per the manufacturer's protocol ("Megaplex Pools For microRNA Expression Analysis"), following instructions for detection of miRNA with pre-amplification. The pre-amplification products were diluted into a prescribed final volume of 100 μΐ _^ and stored at -20°C until ready for the final detection PCR reactions. Real-time PCR reactions followed the manufacturer's protocol, using 18μΙ. of diluted pre-amplification product.

MicroRNA qRT-PCR arrays: The expression profile of miRNAs in CSF samples was determined using the TaqMan® Array Human MicroRNA A + B Cards Set v3.0 (ThermoFisher). The arrays consist of two cards (A and B), each containing a total of 384 TaqMan® MicroRNA Assays per card, including potential endogenous control RNAs (U6 snRNA, RNU44, RNU48). For these arrays, there is an n=l technical replicate for each RNA probe. The qRT-PCR amplifications and data acquisition were performed on a QuantStudio™ 12K Flex Real-Time PCR System

(ThermoFisher) using automated baseline and threshold values determined by Expression Suite™ software (vl .2.2). Amplification data was imported into Expression Suite and cycle time (Ct) values were calculated using ExpressionSuite software (V.1.0.4, Life Technologies). The expression data was analyzed in separate batches according to the TLDA card lot number. The Ct value for each well was reported along with the amplification score (AmpScore) metric. Quality control filtering of the Ct values consisted of the following steps: i) samples were considered below the detection threshold and censored if Ct > 36 or if Expression Suite reported the Ct value as "Undetected"; ii) samples were excluded if AmpScore < 0.9; Hi) all other Ct values were accepted as reported by Expression Suite. RNU44 and RNU48 detection was inconsistent across all cards and samples, and excluded from further analysis. U6 snRNA detection was consistent within cards, across cards, and across samples, making it a strong candidate for quality assessment and normalization strategies.

Statistical analysis of miRNA expression in control and AD CSF: The analytical approach involved passing the complete miRNA data set through multiple quality and performance filters in order to hone in candidate AD miRNAs that: i) were frequently expressed within the dynamic range of qRT-PCR measurements for CSF samples; ii) whose Ct values trend strongly with the AD condition; Hi) were able to discriminate AD from control at the patient level. Test (i) was addressed by thorough data filtering using ExpressionSuite error flags. Technical failures were excluded, as were miRNAs with low (<~20%) detection rates in samples. Although the amplification

experiments for this study were allowed to run to 40 cycles before termination, empirical data- quality metrics indicated that virtually no amplifications occurred at or after 36 cycles, so the raw Ct values were censored to a threshold of 36; thus, censored values may be viewed as miRNA expression below a lower limit of detection. Ct values were then normalized relative to the U6 snRNA reference and also transformed onto an expression scale by taking the multiplicative inverse (i.e. 1 divided by the value) and scaling by the maximum value so that higher values indicate relatively greater quantities of miRNA expression; censored values were assigned a value of zero on this expression scale. 151 candidate miRNAs passed the initial quality filters and proceeded on to further analysis.

To address tests {if) and (Hi) a four-step battery of inferential test procedures was performed. In order to mitigate inferential bias due to testing method, we imposed the rule that a candidate miRNA would not be carried forward unless it had passed at least two of the four steps. The testing was performed using Stata Release 14 (http://www.stata.com/) and R Version 3.2.3 (http s : //cran . r-proj ect . org/) software .

The first two steps employed log-rank testing of normalized Ct values and predictions from univariate logistic regression on the transformed expression values using models that accounted for censoring. The goal in these phases was to select candidate miRNAs according to their strength of association with AD status, and selection cutoffs were chosen to achieve an average false discovery rate (FDR) of -30% under correction for multiple comparisons. This FDR heuristic was chosen to reflect our anticipation that selected candidates would, on average, be about twice as likely to be associated as not to be associated with AD status.

The last two steps of testing were designed to find candidate miRNA markers that had expression profiles that would provide a reliably clear signal of AD or non-AD status across the observed range of total miRNA expression contexts; the goal in these phases was to gauge the relative predictive importance of each miRNA while accounting for interactions with the other miRNAs. To test the signal strength random groups of miRNAs were repeatedly sampled and their ability to classify the CSF samples into "AD" and "non-AD" bins assessed. If the classification performed worse when a certain miRNA was excluded from the mix, the certain miRNA was inferred to contribute at least some signal, and was viewed as relatively more important than other miRNAs. For the testing, two different random-forest classifiers (Breiman L, Machine Learning 45, 5-32 (2001); incorporated by reference herein) were used - one based on conditional inference trees (Hothorn T et al, J Comp Graph Stats 15, 651-674 (2006); incorporated by reference herein) and the other based on CHAID trees (Kass GV J Royal Statistical Soc Series C (Appl Stats) 29, 119- 127 (1980); incorporated by reference herein)— in order to make the selection process more robust to the choice of underlying classification method; selection in both phases was based on an appropriate method-specific 'importance score' that could be calculated for each classifier.

Replication miRNA studies: Custom TaqMan ^® miRNA assays were generated to examine the expression of candidate AD miRNAs in a subset of 16 AD and 16 control CSF samples that were analyzed in the original arrays. The custom arrays included probes for 48 candidate miRNAs, 6 miRNAs not changed between AD and control CSF, 4 miRNAs not detected in CSF, and U6 snRNA. All RNA probes were produced in triplicate, to allow for analysis of n=3 technical replicates/RNA in the replication studies. 20 of the 36 candidate miRNAs selected by the four-step statistical evaluation were included in this study.

Example 2— Donor characteristics

All donor characteristics for the 50 AD and 49 control subjects considered in this phase of the biomarker study, including sex, age at CSF collection, MMSE, and ApoE genotype, are shown in Table 1. The 49 control subjects were community volunteers in good health, with Mini -Mental State Examination (Folstein MF, Arch Gen Psych 40, 812 (1983); incorporated by reference herein) scores between 28 and 30 (29.22 +/- 1.28), Clinical Dementia Rating scores of 0, and no evidence or history of cognitive or functional decline (Peskind ER et al, Arch Neurol 63, 936-939 (2006) and Nation DA et al, Neurology 81, 2024-2027 (2013); incorporated by reference herein). The 50 AD participants were recruited from OHSU Neurology clinics and diagnosed with probable AD according to ADRD A-NIND S criteria McKhann G et al, Neurology 34, 939-944 (1984) and McKhann GM et al, Alzheimers Dement 7, 263-269 (2011); both of which are incorporated by reference herein) at a consensus conference of OADC clinicians, with Mini-Mental State

Examination scores between 18 and 19 (18.28 +/- 6.4), and Clinical Dementia Rating scores of 1-2. ApoE genotyping was done for the 90/99 CSF donors who also donated blood samples (44 control and 46 AD). The genotype revealed that of the control group, 54.55% had zero, 38.64% had one, and 6.82% had two ApoE4 alleles. In the AD group 17.39% had zero, 58.70% had one, and 23.91% had two ApoE4 alleles. This difference is reflected in the increase in the ratio of ApoE4 alleles in AD vs. control, which increases from a ratio of 0.33 for zero alleles to 1.59 for 1 allele, and to 3.67 for 2 alleles. Thus, the AD group had a prevalence towards 1-2 ApoE4 alleles relative to the controls, consistent with the correlation between ApoE4 and AD ApoE (Seshadri S et al, Arch Neurol 52, 1074-1079 (1995); incorporated by reference herein).

Table 1 : Donor characteristics

Example 3— Identification of top candidate miRNAs that discriminate AD from control CSF

The TaqMan® miRNA arrays were assessed for uniformity of miRNA expression within each group by correlating the expression levels for each replicate to the average expression for the group. The miRNA expression from one sample was compared to the miRNA expression for all samples in the same group to show the correlation between one individual replicate relative to the average of the total replicates sampled within each group for control and AD (data not shown). These data showed that miRNA expression is consistent within each group (control versus AD) and that regulation of miRNAs is not random in individual samples.

For a miRNA to be included in the statistical analysis, it had to be present in at least 20% of the samples, which resulted in 151 miRNA candidates chosen for analysis. The discriminating abilities of the 151 miRNA that passed PCR quality control filtering were assessed using four statistical testing steps. The log-rank and ROC AUC tests assess the group differences and discriminating ability of each miRNA individually, while the two random forest tests (CART and CHAID) evaluate the classification performance of each miRNA as a member of a group. The p- values for the log-rank tests, AUC values from the ROC tests, and the relative importance ranks from the CART and CHAID tests are presented in Table 2. We summarized these results in a "Multitest Score" for each miRNA that indicates the number of steps where the miRNA was selected as a strong performer. MiRNA that performed well in at least two statistical tests (Multitest score = 2, 3, or 4) were considered top biomarker candidates. We also include for each candidate the fraction of samples in Control and AD groups with detectable levels of that miRNA (%

Detected) along with the log fold change and 95% confidence interval (CI) of the detected values in AD vs. Control samples. These analyses identified 36 candidate miRNAs, with 21 significantly altered in AD relative to control at the 5% level: 5 with increased presence in AD and 16 with decreased presence (Table 2).

0000449 miR-146a-5p 0.10 0.62 11 57 2 100 94 -0.90 -1.78, -0.02

0000419 miR-27b-3p 0.09 0.63 21 92 2 48 32 -0.85 -1.92, 0.23

0000760 miR-331-3p 0.08 0.67 144 52 2 57 32 -0.83 -2.17, 0.50

0000086 miR-29a-3p 0.07 0.63 12 7 4 98 100 -0.77 -1.55, 0.02

0000461 miR-195-5p 0.05 0.61 106 150 2 91 76 -0.75 -1.69, 0.18

0000417 miR-15b-5p 0.05 0.61 15 1 4 40 20 -0.71 -1.92, 0.50

0000280 miR-223-3p <0.01 0.62 39 5 3 100 100 -0.66 -2.00, 0.69

0000437 miR-145-5p 0.06 0.64 20 15 3 63 39 -0.42 -1.81, 0.97

0003258 miR-590-5p 0.06 0.61 36 36 2 47 27 -0.33 -1.20, 0.53

0000710 miR-365a-3p 0.04 0.63 24 51 2 55 32 -0.29 -1.10, 0.53

Example 4— Linear combinations of miRNAs increase sensitivity/specificity

Linear combinations of subsets of the 36 selected miRNAs via best-subsets logistic regression (employing multiple imputation of values missing due to technical dropout) were assessed and AUC computed to assess multimarker classification performance. Top-performing linear combinations of 3 and 4 miRNAs attained an AUC of 0.80-0.82 (Figure 1). Addition of ApoE genotype status to the miRNA based test further increased sensitivity/specificity as an AD biomarker (Figure 2). ApoE alone has an AUC of 0.73 in this group of subjects, and 3 miRNAs have an AUC of 0.80. However, 3 miRNAs and ApoE genotype increase the AUC to 0.84. Thus, a combination of new (miRNA) and a reference (ApoE) biomarker increased classification

performance.

Example 5— Replication Studies

The original analysis of 754 miRNAs led to a selection of 46 candidates (only partially overlapping with the current set of 36) that were retested using custom TaqMan ^® arrays to verify the differential expression of miRNAs in AD versus control CSF. The original set of 99 CSF samples were run on TLDA's with an n=l technical replicate/RNA probe, while the custom TLDAs were designed with an n=3 technical replicates/RNA probe using 32 of the 99 CSF samples (16 AD vs. 16 Control). Although 20 of the original 46 candidates held up as expected, the remaining 26 did not. The entire miRNA set was then reanalyzed using the more rigorous pipeline described above. This pipeline also selected the same top 20 (in addition to 16 other promising candidates), and interestingly, rejected all of the failed 26, mostly on data quality grounds. This demonstrates the need for strong data quality filtering procedures in biomarker selection pipelines. A flowchart showing the overlap between the original analysis and the reanalysis (Figure 3) strongly supports the inclusion of the top 20 candidates in the validation phase: 17 of the 20 top targets held up as expected in the reanalysis and the remaining 3 were ambiguous in performance, but not

contradictory. Table 3 - combinations of three miRNA that indicate that a subject has Alzheimer's disease

Table 4 - combinations of four miRNA that indicate that a subject has Alzheimer's disease.

Example 6—Total RNA Isolation and Custom Taqman Low Density Arrays

Total RNA was extracted from 0.5 mL of each cerebrospinal fluid (CSF) sample using the mirVana™ PARIS™ RNA and Native Protein Purification Kit (ThermoFisher). RNA samples were concentrated using the RNA Clean & Concentrator-5 Kit (Zymo Research). The concentrated RNA samples were converted to cDNA with Megaplex™ RT Primers, Human Pool Set v3.0, as per the manufacturer's instructions on a Veriti thermocycler (Life Technologies, Foster City, CA). The resulting complementary DNAs were pre-amplified using Megaplex™ PreAmp Primers, Human Pool Set v3.0 as per the manufacturer's instructions (Life Technologies) and pre-amplification products diluted to a final volume of 100 μL and stored at -20°C. The RNAs were amplified on custom TaqMan ^® Arrays (ThermoFisher) comprised of 36 target miRNAs. The RT-qPCR amplifications is performed on a QuantStudio™ 12K Flex Real-Time PCR. Amplification data are imported into ExpressionSuite, and quality control filtering of the Ct values performed as previously described (Lusardi et.al, 2017). Tables 5 below shows the characteristics of the donors used in the evaluation. Table 6 shows the data for the miRNAs sequences.

Example 7 - Discovery of CSF MicroRNA Biomarkers for AD

The purpose of this study was to examine miRNAs in CSF from living donors, distinct from prior studies using CSF obtained post-mortem. The cerebrospinal fluid was examined in two well- characterized repositories of AD and control patients (Table 7): for UH2, the Oregon Alzheimer's Disease Center (OADC) at Oregon Health & Science University; and for UH3, the Shiley-Marcos Alzheimer's Disease Research Center (SMADRC) at University of California, San Diego. In both phases a moderately large sample of -100 patients each was obtained, and a rigorous analytic approach was used to examine the diagnostic utility of combinations of miRNAs.

In the UH2 discovery cohort, a broad panel of 754 miRNAs was measured, and a set of 36 miRNAs in CSF that discriminate AD patients from neurologically normal (control) subjects was identified, using a consensus-based statistical selection methodology designed to mitigate false positive discoveries. The fact that several miRNAs in CSF were consistently expressed in both AD and control CSF seems to indicate a specific disease-associated effect, rather than a global change, in AD CSF miRNAs expression. In addition to marker-specific screening, combinations of these 36 candidate miRNAs in multimarker models were evaluated and an analysis of the computed area under the ROC curve (AUC) was performed to ascertain classification performance, followed by averaging of ROC curves across same-sized subsets to avoid early reliance on any particular subset for performance projections. As shown in Figure 2, combinations of miRNAs increase performance over single markers or ApoE genotype information alone, and importantly, the addition of an indicator for the ApoE4+ genotype to the model further improved classification performance.

These results demonstrate that the miRNA information is not redundant with ApoE, but provides independent discriminatory power and may potentially point to novel mechanisms for the regulation of the disease.

In the UH3 phase, 26 of the 36 miRNAs discovered in the UH2 were validated as biomarkers for AD (Table 8).

To set expectations for verification testing of classification performance of the candidate miRNAs in the independent UH3 cohort, nonparametric nearest-neighbor classifier based on all candidate markers were cross-validated on the UH2 cohort to obtain conservative projections of the ROC curves in the UH3 cohort. The projected curves were then compared to actual curves based on UH3 data. Figure 4 shows the projections for both miRNA-only models and miRNA + ApoE4 models based on 26 top candidates (left panel), and the results of applying the same nearest- neighbor classifier to the UH3 cohort using the same miRNA markers (right panel). The close agreement between the projections and results is compelling: the AUC values for the two model types applied to the UH3 cohort are larger than projected, and the effect of adding ApoE4 to miRNA models is the same magnitude in both cohorts, showing that the overall classification ability of the candidate miRNAs and their complementary relationship with ApoE genotype is well supported by the current experimental evidence. The shape and separation of the UH3 miRNA-only ROC curve is similar to the UH2 projection, but the shape of the miRNA + ApoE4 curve is even more favorable to practical prediction than the UH2 projection would suggest, because it suggests that much of the boost from including ApoE4 information may occur in high- specificity ranges of the miRNA classifier. Example 8 SA UGSTAD-QUINN STUD Y OF MIRNA BIOMARKERS FOR ALZHEIMER 'S DISEASE

Additional studies were made to determine the most effective miRNAs biomarkers (Figure 5), and combinations of biomarkers.

Table 9 - Sub-combinations of four miRNA that indicate that a subject has Alzheimer's disease

Table 10 - combinations of five miRNA that indicate that a subject has Alzheimer's disease

Table 11 - combinations of six miRNA that indicate that a subject has Alzheimer's disease Table 12 - combinations of seven miRNA that indicate that a subject has Alzheimer's disease

Example 9— Interventions for Reducing risk of developing Alzheimer's Disease, delaying the onset of Alzheimer's Disease, and slowing the progression of Alzheimer's Disease in a subject in need thereof

A subject is identified as being at increased risk of Alzheimer's dementia using the methods described herein. Once identified, the subject is treated with pharmaceutical agents or lifestyle modifications that have been associated with a lowered risk of late onset Alzheimer's disease. In one example, the subject is treated to lower cardiovascular risk factors, such as high blood pressure, diabetes and high cholesterol. Such treatments can include advising beginning or improving a dietary regimen and adopting eating habits associated with lower cardiovascular and neurological risk. For example, the subject may be advised to adopt a plant-based (vegetarian) diet, or to avoid foods high in unhealthy saturated fats of sugar. In other examples, the subject is treated for hypertension, or advised to adopt behaviors or undergo treatments such as regular exercise, reducing obesity, treating hypertension, lowering blood glucose levels, lowering blood lipids such as LDL and triglycerides, or improving ratios of HDL to LDL, and/or lipoprotein particle size. In one example, LDL particle size is increased by lowering triglyceride levels. In one example, if the subject's total cholesterol is elevated (e.g. greater than 200) the subject is treated with diet or a pharmaceutical agent (such as a statin) to lower the total cholesterol below 200, and/or lower LDL below 130, 120, 100 or less. If the subject's triglycerides are elevated (for example greater 200), the subject is treated with diet or a pharmaceutical agent to lower triglycerides. In another example, if the triglyceride to HDL ratio is greater than 2: 1 it is lowered toward or below 2: 1. In another example, if the ratio of total cholesterol/HDL is greater than 3.5 to 1, interventions are advised or adopted that lower that ratio toward to below 3.5 to 1.

Alternatively, the subject is administered vitamins associated with improved neurological function, such as vitamins B6, B12 and/or folate. In particular examples, the subject is

administered methylcobalamin (for example 1000 meg/day) and/or methyl -tetrahydrofolate (for example 1-10 mg/day). In some examples, the vitamins are administered in sufficient amounts to lower blood homocysteine levels to less than 12, for example less than 10 or less than 8.

Alternatively, a subject with hypertension is administered an anti -hypertensive drug in a sufficient amount to lower the blood pressure to clinically desired levels.

If a subject smokes tobacco or drinks unhealthy amounts of alcohol (such as more than 2 drinks a day for a male or 1 drink per day for a woman), the subject is advised to cease those behaviors or is treated with a pharmaceutical agent to improve tobacco-cessation or encourage alcohol abstinence (such as naltrexone, disulfiram, topiramate or baclofen).