WO2012085217A1 - A diagnostic method for dementias - Google Patents

Abstract

The invention relates to a method of determining whether a subject has or is at risk of developing a dementia, such as Alzheimer's disease (AD), a mild cognitive impairment (MCI), fronto-temporal dementia (FTD), or dementia with Lewis bodies (DLB), which method comprises detecting the presence of a mutation, deletion or insertion of one or several nucleotides in the CTGF gene locus in a biological sample of said subject.

Description

A diagnostic method for dementias

FIELD OF THE INVENTION The present invention relates generally to the fields of genetics and medicine. The present invention discloses in particular the identification of a dementia susceptibility gene, which can be used for the diagnosis or prognosis of a dementia or for the detection of predisposition to a dementia. BACKGROUND OF THE INVENTION

Diagnosing a dementia can often be difficult, especially in the early stages, because many of the symptoms of the disease mirror the natural signs of aging. In some situations, a definitive diagnosis may not be possible until the patient has died and an autopsy can be performed.

There are also several forms of dementia, many of which involve a loss and/or cognitive dysfunction. Azheimer's disease (AD), Mild cognitive impairement (MCI), fronto-temporal dementia (FTD), and dementia with Lewis bodies (DLB), are typical examples of such disorders.

Alzheimer's disease (AD) is a neurodegenerative disease of the central nervous system associated with progressive memory loss resulting in dementia. The prevalence of AD is expected to reach 8.64 million by the year 2047. AD is characterized by extracellular plaques and intracellular tangles in the hippocampus, cerebral cortex, and other areas of the brain essential for cognitive function. Plaques are formed mostly from the deposition of amyloid beta ("Αβ"), a peptide derived from amyloid precursor protein ("APP"). Filamentous tangles are formed from paired helical filaments composed of neurofilament and hyperphosphorylated tau protein, a microtubule-associated protein. It is not clear, however, whether these two pathological changes are only associated with the disease or truly involved in the degenerative process.

Two main forms of the disease coexist. First of all, the early onset form (<65 years old) mainly characterized by a strong familial clustering and Mendelian disease transmission. Early onset familial AD accounts for only 5% of all AD cases. Secondly, a late onset form (>65 years old) showing no obvious familial aggregation.

Genetic studies have identified three genes that cause autosomal dominant, early onset AD, amyloid precursor protein ("APP"), presenilin 1 ("PS 1"), and presenilin 2 ("PS2"). A fourth gene, apolipoprotein E ("ApoE"), is the strongest and most common genetic risk factor for AD, but does not necessarily cause it. All mutations associated with APP and PS proteins can lead to an increase in the production of Αβ peptides, specifically the more amyloidogenic form, Αβ 42. In addition to genetic influences on amyloid plaque and intracellular tangle formation, environmental factors (e.g., cytokines, neurotoxins, etc.) may also play important roles in the development and progression of AD.

The main clinical feature of AD is a progressive cognitive decline leading to memory loss. Language impairments are a prominent part of AD. Non-cognitive or behavioural symptoms are also common in AD. Personality changes are commonly reported and range from progressive passivity to marked agitation. Patients may exhibit changes such as decreased expressions of affection. Depressive symptoms are present in up to 40%. A similar rate for anxiety has also been recognized. Psychosis occurs in 25%.

Currently, the primary method of diagnosing AD in living patients involves taking detailed patient histories, administering memory and psychological tests, and ruling out other explanations for memory loss, including temporary (e.g., depression or vitamin B 12 deficiency) or permanent (e.g., stroke) conditions. These clinical diagnostic methods, however, are not foolproof. Given the magnitude of the public health problem posed by AD, considerable research efforts have been undertaken to elucidate the etiology of AD as well as to identify biomarkers (secreted proteins or metabolites) that can be used to diagnose and/or predict whether a person is likely to develop AD. With regards to biomarkers, the proteins Αβ and tau are probably the most well characterized.

A number of U.S. patents and applications have been published relating to methods for diagnosing AD. For example, US 4,728,605 shows that AD can be diagnosed by testing of peripheral blood immunocytes (especially T cells), US 5,874,312 describes a method for measuring a cytokine production level of peripheral mononuclear cells to diagnose and determine the disease stage of AD, US 6,027,896 discloses an AD diagnostic method detecting the presence or absence of an apolipoprotein E type 4 in a subject. Additionally, a number of reports in the scientific literature relate to certain biochemical markers and their correlation/association with AD, including Brain Derived Neurotrophic Factor (BDNF) (Fahnestock et al., 2002, J. Neural. Transm. Suppl. 2002(62):241-52); Platelet-Derived Growth Factor (PDGF) (Masliah et al., 1995, Neurobiol. Aging 16(4):549-56); leptin, which could play a role in AD-associated weight loss (Power et al., 2001, Dement. Geriatr. Cogn. Disord. 12(2): 167-70).

Mild cognitive impairment (MCI) is a general term most commonly defined as subtle but measurable memory disorder. As MCI is a very early sign of dementia, subjects with MCI are becoming the focus of many prediction studies and early intervention trials. Hence, MCI has emerged as the most convenient avenue to identify AD as early as possible, since most MCI patients will progress to AD (Chetelat and al, 2003, Brain, 2003 (126): 1955-1967).

Fronto-temporal dementia (FTD) is one of the most common early onset dementia after AD. FTD is a clinical syndrome caused by degeneration of the frontal lobe of the brain and may extend back to the temporal lobe. Pick's disease is the most known FTD. As AD, FTD is associated with a tau pathology. Symptoms can be classified into two groups which underlie the functions of the frontal lobe: behavioral and cognitive like planning and organizing skills.

Dementia with Lewy bodies (DLB) is an illness that presents with progressive dementia or psychosis. Parkinsonian signs, which may be absent or mild at the onset, eventually become common and rigidity is usually severe. Lewy bodies are found profusely in the brainstem, basal forebrain, hypothalamic nuclei and neocortex. DLB is characterized by the relative absence of tangles and hyperphosphorylated tau in the brain. Parkinson's disease is a type of Lewy Body diseases occurring in the middle or late life, with very gradual progression and a prolonged course. The genetic component of such dementias remains largely unknown. Exhaustive knowledge of the diseases' genetic determinants will lead to the development of tools to aid early diagnosis.

As CTGF gene strongly promotes extracellular matrix gene expression, it was hypothesized that CTGF gene would be involved in postlesional restructuring of central nervous system tissue, especially in glial scar formation (Hertel and al., Eur J Neurosci. 2000 Jan ; 12(1) :376- 80). CTGF gene was then shown to be highly expressed in neurons associated with plaques and in neuritic components of plaques in AD's patients (Ueberham and al., Neuroscience 116 (2003) 1-6). Zhong Zhao and al (The FASEB Journal, 19(14):2081-2, September 26, 2005) further disclosed an increased CTGF expression in the brain possibly correlated to the clinical progression of AD.

However, using CTGF as a gene marker in the diagnosis or prognosis of dementia or for the detection of predisposition to dementia, has never been suggested.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a new genetic approach for dementia diagnostic. The present invention now discloses the identification of a human dementia susceptibility gene locus, the CTGF gene locus (CCN2), which can be used for detecting predisposition to, for diagnosis and prognosis of a dementia, especially dementia with memory and/or cognitive dysfunction, such as MCI, AD, FTD and DLB. The invention resides, in particular, in a method which comprises detecting in a sample from the subject the presence of a mutation, an insertion or a deletion of one or more bases in the CTGF gene locus (CCN2), the presence of said mutation, insertion or deletion being indicative of the presence or predisposition to a dementia.

A particular object of this invention resides in an in vitro method of detecting predisposition to or diagnosis and/or prognosis of a dementia occurring in a subject, the method comprising detecting the presence of a mutation, an insertion or a deletion of one or more bases in the CTGF gene locus in a sample from the subject, the presence of said mutation, insertion or deletion being indicative of the presence of a dementia or the predisposition to a dementia. In a preferred embodiment, said dementia is a dementia with memory and/or cognitive dysfunction. In another preferred embodiment, said dementia is a senile dementia. Preferably it affects humans over 60 or 65 years old.

The method of the invention allows for detection and prognosis of a dementia, especially a dementia with memory and/or cognitive dysfunction selected from mild cognitive impairment, Alzheimer's disease, fronto-temporal dementia and dementia with Lewis bodies.

In a preferred embodiment, said mutation, insertion or deletion is located within 20 kb, upstream the start codon of the CTGF gene and 20 kb, downstream the 3 'UTR of the CTGF gene. Preferably, the mutation, insertion or deletion lies in the surrounding sequences of 15 kb region, upstream the starting codon of the CTGF gene and 15 kb region, downstream the untranslated region (3'UTR). More preferably, the mutation, insertion or deletion lies in the surrounding sequences of 10 kb region, upstream the starting codon of the CTGF gene and 10 kb region, downstream the untranslated region (3'UTR). Even more preferably, the mutation, insertion or deletion lies in the surrounding sequences of 5 kb region, upstream the starting codon of the CTGF gene and 5 kb region, downstream the untranslated region (3'UTR).

In another preferred embodiment, said mutation, insertion or deletion is one or several single nucleotide polymorphism(s) SNP(s) or a haplotype of SNPs associated with a dementia, especially a dementia with memory and/or cognitive dysfunction.

In a more preferred embodiment, said single nucleotide polymorphisms are SNPs flanking CTGF gene, which are allelic variants lying close to the CTGF gene.

Preferably, the alteration in the CTGF gene locus is determined by performing a selective hydridization assay, a sequencing assay, a micro sequencing assay, and/or an allele -specific amplification assay. In another aspect of the invention, said alteration in the CTGF gene is determined by restriction enzyme digestion, the detection of at least one said SNP being an indication of a dementia, especially a dementia with memory and/or cognitive dysfunction.

DETAILED DESCRIPTION OF THE INVENTION

This invention provides valuable genetic markers to detect a predisposition or to predict a diagnosis or a prognosis of a dementia, especially a dementia with memory and/or cognitive dysfunction, such as mild cognitive impairment, Alzheimer's disease, fronto-temporal dementia and dementia with Lewy bodies.

Early detection of a dementia and regular monitoring of a dementia, would allow for initiation of therapies capable of halting and even reversing the dementia process. This early detection would prevent progression of dementia diseases, especially of mild cognitive impairment, Alzheimer's disease, fronto-temporal dementia and Lewy body dementia, and the important symptoms these conditions entail. The development of these various techniques for early detection of a dementia bodes well for the future care of patients with loss of cognitive functions.

The inventors have now identified a major gene associated with human dementia. They have shown that AD and MCI in Caucasian cohorts, are markedly dependent on allelic variants lying in the CTGF gene locus.

Various nucleic acid samples from individuals with AD or MCI were submitted to a particular analysis process. Three hundred and eighty AD patients were recruited at the Saint Marguerite Hospital, Marseille, France. The clinical diagnosis of AD was established using NINCDS- ADRDA criteria for AD (McKhann et al., Neurology, 1984, 34:939). Four hundred and twenty six MCI patients were also recruited at the Saint Marguerite Hospital, Marseille, France. All MCI patients met criteria for an amnestic form of MCI (Peterson et al., Arch Neurol, 1999, 56:303) as all has memory complaint (usually verified by an informant), performed poorly on neuropsychological tests of verbal memory, and had normal general cognitive function. Finally 90 DLB and 99 FTD patients were included. Healthy controls (N=376), without any neurological pathology, were recruited at the Blood Bank Center. Some local volunteer were also enrolled

This analysis process led to the identification of particular single nucleotide polymorphism(s) (SNP(s)) in said populations.

The present invention is not limited to the particular SNPs that were found significantly correlated with dementias in all tested populations. Indeed, several reasons could account for the failure in identifying other significant SNPs in some populations, including an insufficient cohort, the incomplete assessment of confounding variables, a lower frequency of the SNPs in said populations, etc.

Definitions

Within the context of this invention, "Dementia with memory and/or cognitive dysfunction" refers to a progressive and abnormally accelerated deterioration of mental faculties and emotional stability. It designates any disease resulting in a general loss of cognitive abilities, including impairment of memory as well as one or more of the following: aphasia, apraxia, agnosia, or disturbed planning, organizing, and abstract thinking abilities. It generally affects older persons, usually over the age of 65. Within the context of this invention, "Alzheimer's patient" and "AD patient" refer to an individual who has been diagnosed with AD or has been given a probable diagnosis of AD.

Within the context of this invention, "Mild cognitive impairment's patient" and "MCI patient" refer to an individual who has been diagnosed with MCI or has given a probable diagnosis of MCI.

Within the context of this invention, "Fronto-temporal dementia's patient" and "FTD patient" refer to an individual who has been diagnosed with FTD or has given a probable diagnosis of FTD. Within the context of this invention, "Dementia with Lewy bodies' patient" and "DLB patient" refer to an individual who has been diagnosed with DLB or has given a probable diagnosis of DLB. Within the context of this invention, "the CTGF gene locus" (Connection Tissue Growth Factor), also called CCN2 gene locus, designates all sequences or products in a cell or organism, including CTGF coding sequences, CTGF non-coding sequences (e.g., introns), CTGF regulatory sequences controlling transcription and/or translation (e.g., promoter, enhancer, terminator, etc.), all corresponding expression products, such as CTGF RNAs (e.g., mRNAs) and CTGF polypeptides (e.g., a pre-protein and a mature protein); as well as surrounding sequences of 20 kb region, preferably 15 kb region, more preferably 10 kb region and even more preferably 5 kb region upstream the starting codon of the CTGF gene and 20 kb region, preferably 15 kb region, more preferably 10 kb region and even more preferably 5 kb region downstream the untranslated region (3'UTR).

For example, the CTGF locus comprises surrounding sequences comprising the SNPs of Table 1A (such as rs9399005, rs4897554 and rs6918698) identified by sequencing. For instance, SNP rs6918698 is located into the CTGF promoter whereas rs9399005 and rs4897554 are in the 3' region. Within the context of the present invention, the term "prognosis" includes the detection, monitoring, dosing, comparison, etc., at various stages, including early, pre- symptomatic stages, and late stages, in human. Prognosis typically includes the assessment (prediction) of the progression of a dementia, especially a dementia with memory and/or cognitive dysfunction, and the characterization of a subject to define most appropriate treatment (pharmaco-genetics), etc. The present invention provides prognostic methods to determine the speed of the progression of a dementia, especially a dementia with memory and/or cognitive dysfunction, resulting from a mutation or a polymorphism in the CTGF gene locus.

Mutation, deletion or insertion

The mutation, deletion or insertion may be determined at the level of the CTGF DNA, RNA or polypeptide. Optionally, the detection is performed by sequencing all or part of the CTGF gene locus or by selective hybridization or amplification of all or part of the CTGF gene locus. More preferably a CTGF gene locus specific amplification is carried out before the mutation, deletion or insertion identification step. The mutation, deletion or insertion may be in the coding and/or non-coding region of the locus, alone or in various combination(s). Mutations more specifically include point mutations. Deletions may encompass any region of two or more residues in a coding or non-coding portion of the gene locus, such as from two residues up to the entire gene or locus. Typical deletions affect smaller regions, such as domains (introns) or repeated sequences or fragments of less than about 50 consecutive base pairs, although larger deletions may occur as well. Insertions may encompass the addition of one or several residues in a coding or non-coding portion of the gene locus. Insertions may typically comprise an addition of between 1 and 50 base pairs in the gene locus. The CTGF gene locus mutation, deletion or insertion may result in the creation of stop codons, frameshift mutations, amino acid substitutions, particular RNA splicing or processing, product instability, truncated polypeptide production, etc. The mutation, deletion or insertion may result in the production of a CTGF polypeptide with altered function, stability, targeting or structure. The mutation, deletion or insertion may also cause a reduction in protein expression or, alternatively, an increase in said production.

A particular object of this invention is a method of detecting predisposition to and/or prognosis of a dementia, especially a dementia with memory and/or cognitive dysfunction, the method comprising detecting the presence of one or several mutation, deletion or insertion or SNPs selected from the SNPs listed Table 1A, Table IB, or Table 1C.

Table 1A: Dementia - associated mutation(s), deletion(s) or insertion(s) in the CTGF gene locus

Nucleotide Genotype

position in SNP reference Polymorphism correlated with Sequence genomic dementia reference sequence of

chromosome

132273257 rs6918698 C/G GC or GG SEQ ID NO: 1

132268964 rs9399005 C/T CC and CT SEQ ID NO: 2

132304684 rs4897554 A/G GG SEQ ID NO: 3 rs 1029122 C/T TT and CC SEQ ID NO: 67 rsl 1154653 G/T TG and TT SEQ ID NO: 68 rs 10293697 (or C/T SEQ ID NO: 69 rs7645481 )

rsl931001 A/C AA or CC SEQ ID NO: 70 rs9321315 A/T TT or TA SEQ ID NO: 71 rs9483364 A/G GG or GA SEQ ID NO: 72

Association with dementia was shown, by comparing the distribution of genotypes between the affected patients and the control groups.

In particular, the presence of at least one allele C at position rs9399005 (which means that the subject is homozygote CC or heterozygote CT) and/or two alleles G at position rs4897554 (which means that the subject is homozygote GG) were shown to be deleterious, i.e. strongly correlated with a dementia, especially a dementia with memory and/or cognitive dysfunction as AD and MCI. The results show that SNP rs9399005 is more particularly associated with mild cognitive impairment (MCI) and Alzheimer's disease, which means that it correlates with the development of a dementia.

Preferably, the invention thus provides a method of determining whether a subject has or is at risk of developing an early form of dementia such as a mild cognitive impairment (MCI), which method comprises detecting the presence of a SNP at position rs9399005 or rs4897554 in the CTGF gene locus in a biological sample of said subject.

Preferably, the invention further provides a method of determining whether a subject has or is at risk of developing an advanced form of dementia such as Alzheimer's disease (AD) which method comprises detecting the presence of a SNP at position rs9399005 in the CTGF gene locus in a biological sample of said subject.

Linkage Disequilibirum

Once a first SNP has been identified in a genomic region of interest, more particularly in CTGF gene locus, other additional SNPs in linkage disequilibrium with this first SNP can be identified. Indeed, any SNP in linkage disequilibrium with a first SNP associated with a dementia, especially a dementia with memory and/or cognitive dysfunction will be associated with this trait. Therefore, once the association has been demonstrated between a given SNP and a dementia, especially a dementia with memory and/or cognitive dysfunction, the discovery of additional SNPs associated with this trait can be of great interest in order to increase the density of SNPs in this particular region. Identification of additional SNPs in linkage disequilibrium with a given SNP involves: (a) amplifying a fragment from the genomic region comprising or surrounding a first SNP from a plurality of individuals; (b) identifying of second SNP in the genomic region harboring or surrounding said first SNP; (c) conducting a linkage disequilibrium analysis between said first SNP and second SNP; and (d) selecting said second SNP as being in linkage disequilibrium with said first marker. Subcombinations comprising steps (b) and (c) are also contemplated. These SNPs in linkage disequilibrium can also be used in the diagnostic methods according to the present invention.

The inventors have already identified several markers in linkage disequilibrium (LD) with rs9399005 or rs4897554. These additional markers are listed in Table IB. Any of these SNPs may be used in addition or in lieu of rs9399005 or rs4897554.

Table IB: Markers in linkage disequilibrium (LD)

Position Position Sequence from start from the reference

SNP LD with LD with codon of end of

# reference Position rs4897554 rs9399005 the gene the gene

14 rs9493142 132286336 yes 27875 24676 SEQ ID NO: 4

16 rs6929832 132286364 yes yes 27847 24648 SEQ ID NO: 5

25 rs6914218 132292595 yes yes 21616 1841 7 SEQ ID NO: 6

27 rs12194675 132293721 yes yes 20490 17291 SEQ ID NO: 7

SEQ ID NO: 8

32 rs1321 1 178 132295966 yes yes 18245 15046

SEQ ID NO: 9

33 rs1220021 1 132296478 yes yes 17733 14534

SEQ ID NO: 10

35 rs12214123 13229701 1 yes yes 17200 14001

SEQ ID NO: 11

36 rs12192391 132297943 yes yes 16268 13069

SEQ ID NO: 12

42 rs1931004 132304007 yes yes 10204 7005

SEQ ID NO: 13

45 rs12526196 1323051 69 yes yes 9042 5843

SEQ ID NO: 14

48 rs6917644 132307358 yes yes 6853 3654 SEQ ID NO: 15 rs9399005 132310657 yes yes 3554 355

SEQ ID NO: 16 rs6918698 132314950 yes yes -739 -3938

SEQ ID NO: 17 rs9493150 132315684 yes -1473 -4672

SEQ ID NO: 18 rs928501 132317130 yes -2919 -61 18

SEQ ID NO: 19 rs7748518 13231891 1 yes -4700 -7899

SEQ ID NO: 20 rs9388949 132319155 yes yes -4944 -8143

SEQ ID NO: 21 rs10872387 132324004 yes -9793 -12992

SEQ ID NO: 22 rs12192108 132325558 yes -1 1347 -14546

SEQ ID NO: 23 rs7747551 132325860 yes -1 1649 -14848

SEQ ID NO: 24 rs12663962 132329042 yes -14831 -18030

SEQ ID NO: 25 rs9402376 132329677 yes -15466 -18665

SEQ ID NO: 26 rs9493157 132330850 yes -16639 -19838

SEQ ID NO: 27 rs1 1 154654 132333478 yes -19267 -22466

SEQ ID NO: 28 rs1930994 132333962 yes -19751 -22950

SEQ ID NO: 29 rs1930995 132334269 yes -20058 -23257

SEQ ID NO: 30 rs7767139 132334474 yes -20263 -23462

SEQ ID NO: 31 rs7767582 132334584 yes -20373 -23572

SEQ ID NO: 32 rs12055428 132335258 yes -21047 -24246

SEQ ID NO: 33 rs9483365 132335602 yes -21391 -24590

SEQ ID NO: 34 rs9402377 132336015 yes -21804 -25003

SEQ ID NO: 35 rs975445 132337814 yes -23603 -26802

SEQ ID NO: 36 rs9493160 132338560 yes -24349 -27548

SEQ ID NO: 37 rs2027034 132338658 yes -24447 -27646

SEQ ID NO: 38 rs9388953 132339202 yes -24991 -28190

SEQ ID NO: 39 rs928505 132339834 yes yes -25623 -28822

SEQ ID NO: 40 rs928504 132339928 yes -25717 -28916

SEQ ID NO: 41 rs4452668 132341441 yes -27230 -30429

SEQ ID NO: 42 rs9375854 132343988 yes -29777 -32976

SEQ ID NO: 43 rs2151532 132316423 yes -2212 -541 1

SEQ ID NO: 44 rs7768619 132319026 yes -4815 -8014

SEQ ID NO: 45 rs1931002 132320175 yes -5964 -9163 SEQ ID NO: 46

59 rs7763896 132321497 yes -7286 -10485

SEQ ID NO: 47

60 rs12527379 132321533 yes -7322 -10521

SEQ ID NO: 48

63 rs2095252 132325281 yes -1 1070 -14269

SEQ ID NO: 49

71 rs4897555 132330159 yes -15948 -19147

SEQ ID NO: 50

73 rs1931000 1323311 17 yes -16906 -20105

The experiments of the Example section, shows 25 SNPs associated with Azheimer's disease (AD), Mild cognitive impairement (MCI), fronto-temporal dementia (FTD), or dementia with Lewis bodies (DLB), These SNPs are listed in Table 1C:

Table 1C: Markers associated with dementias

AD MCI DLB FTD

rsl029122

rslll54653

rsl2192108

rsl931001

rs2095252 rs2095252

rsl0872387

rsl2192391

rsl2194675

rsl2200211

rsl2214123

rsl2527379 rsl2527379

rsl3211178

rsl2663962

rsl931000 rsl931000 rsl931000

rsl931001

rs4897554 rs4897554 rs4897554 rs4897554

rs6918698 rs6918698

rs7747551

rs9321315

rs7768619

rs9388949 rs9388949

rs9399005 rs9399005 rs9399005

rs9483364

rs9493150 rs9493150

rs9493157 A particular object of this invention is a method of determining whether a subject has or is at risk of developing a dementia, which method comprises detecting the presence of a SNP or combination of SNPs as listed in Table 1C, i.e. selected from the group consisting of rsl2192108, rs2095252, rsl0872387, rsl2527379, rsl2663962, rsl931000, rsl931001, rs4897554, rs6918698, rs7747551, rs9388949, rs9399005, rs9483364, rs9493150, rs9493157, rsl931001, rsl2192391, rsl2194675, rsl2200211, rsl2214123, rsl3211178, rs7768619, rsl029122, rsl 1154653, and rs9321315, especially wherein said SNP is preferably selected from the group consisting of rs4897554, rs9399005, rsl931001, rsl931000, rs9388949, rsl3211178, rsl029122, rs9321315, and rsl 1154653.

More particularly, it is provided a method of determining whether a subject has or is at risk of developing AD, which method comprises detecting the presence of a SNP or combination of SNPs, selected from the group consisting of rsl2192108, rs2095252, rsl0872387, rsl2527379, rsl2663962, rsl931000, rsl931001, rs4897554, rs6918698, rs7747551, rs9388949, rs9399005, rs9483364, rs9493150 and rs9493157, preferably rs7747551, rsl931001, rsl2663962, rsl0872387, and/or rsl2192108.

It is provided a method of determining whether a subject has or is at risk of developing MCI, which method comprises detecting the presence of a SNP or combination of SNPs, selected from the group consisting of rsl 931001, rs2095252, rs 12192391, rs 12194675, rsl 2200211, rsl2214123, rsl2527379, rsl3211178, rsl931000, rs4897554, rs6918698, rs7768619, rs9399005, rs9493150, preferably rsl931001, rsl2192391, rsl2194675, rsl2200211, rsl2214123, rsl3211178, and/or rs7768619. It is also provided a method of determining whether a subject has or is at risk of developing DLB, which method comprises detecting the presence of a SNP or combination of SNPs, selected from the group consisting of rsl029122, rs4897554, rs9388949, preferably rsl029122. It is also provided a method of determining whether a subject has or is at risk of developing FTD, which method comprises detecting the presence of a SNP or combination of SNPs, selected from the group consisting of rsl l l54653, rsl931000 , rs4897554, rs9321315, rs9399005, preferably rsl 1154653 and/or rs9321315.

In a particular aspect of the invention, the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing, Alzheimer's disease :

genotype CC or TC with respect to rs9399005

genotype GG with respect to rs4897554

genotype GG or GC with respect to rs9493150

genotype GG or GA with respect to rs9483364

genotype A A with respect to rsl 0872387

genotype GG with respect to rsl 2527379

genotype GG or GC with respect to rs6918698

genotype CC with respect to rs 1931001

genotype TT with respect to rs 12192108

genotype GG with respect to rs 12663962

genotype A A or AG with respect to rs 1931000

genotype AA or AG with respect to rs2095252

genotype AA with respect to rs7747551

genotype TT or TC with respect to rs9388949; and/or

genotype CC or TC with respect to rs9493157.

More particularly the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, Alzheimer's disease, wherein said risk alleles are : genotype GG with respect to rs4897554,

genotype CC or TC with respect to rs9399005,

genotype CC with respect to rsl 931001,

genotype AA or AG with respect to rsl931000, and

genotype TT or TC with respect to rs9388949. In another particular aspect of the invention, the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing, mild cognitive impairment:

genotype CC or TC with respect to rs9399005

genotype GG with respect to rs4897554

genotype GG or GC with respect to rs9493150

genotype AA or AG with respect to rs 12527379

genotype GG or GC with respect to rs6918698:

genotype CC with respect to rs 1931001

genotype AA with respect to rsl2192391

genotype AA with respect to rs 12194675

genotype A A with respect to rs 12200211

genotype GG with respect to rs 12214123

genotype GG with respect to rsl3211178

genotype A A or AG with respect to rsl931000

genotype AA or AG with respect to rs2095252, and/or

genotype TT with respect to rs7768619.

More particularly the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, mild cognitive impairment, wherein said risk alleles are

genotype GG with respect to rs4897554

genotype CC or TC with respect to rs9399005

genotype CC with respect to rsl931001, and

genotype GG with respect to rsl3211178.

In still another particular aspect of the invention, the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing, dementia with Lewis bodies:

genotype GG with respect to rs4897554

genotype CT with respect to rs 1029122, and/or

genotype TC with respect to rs9388949. More particularly the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, dementia with Lewis bodies, wherein said risk alleles are

genotype GG with respect to rs4897554 and

genotype CT with respect to rs 1029122.

In a further particular aspect of the invention, the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing, fronto-temporal dementia:

genotype CC or TC with respect to rs9399005

genotype GG with respect to rs4897554

genotype TG with respect to rsl 1154653

genotype AA with respect to rs9321315, and/or

genotype AA or AG with respect to rs 1931000.

More particularly, the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, fronto-temporal dementia, wherein said risk alleles are

genotype GG with respect to rs4897554

genotype AA with respect to rs9321315, and

genotype TG with respect to rsl 1154653.

Causal Mutation

Mutations in the CTGF gene locus which are responsible for a dementia, especially a dementia with memory and/or cognitive dysfunction may be identified by comparing the sequences of the CTGF gene locus from patients presenting a dementia, especially a dementia with memory and/or cognitive dysfunction and control individuals. Based on the identified association of SNPs of CTGF, the identified locus can be scanned for mutations. In a preferred embodiment, functional regions such as exons and splice sites, promoters and other regulatory regions of the CTGF gene locus are scanned for mutations. Preferably, patients presenting a dementia, especially a dementia with memory and/or cognitive dysfunction carry the mutation shown to be associated with a dementia, especially a dementia with memory and/or cognitive dysfunction and control individuals do not carry the mutation or allele associated with a dementia, especially a dementia with memory and/or cognitive dysfunction. It might also be possible that patients presenting a dementia, especially a dementia with memory and/or cognitive dysfunction carry the mutation shown to be associated with a dementia, especially a dementia with memory and/or cognitive dysfunction with a higher frequency than control individuals. The method used to detect such mutations generally comprises the following steps: amplification of a region of the CTGF gene locus comprising a SNP or a group of SNPs associated with a dementia, especially a dementia with memory and/or cognitive dysfunction from DNA samples of the CTGF gene locus from patients presenting a dementia, especially a dementia with memory and/or cognitive dysfunction and control individuals; sequencing of the amplified region; comparison of DNA sequences of the CTGF gene from patients presenting a dementia, especially a dementia with memory and/or cognitive dysfunction and control individuals; determination of mutations specific to patients presenting a dementia, especially a dementia with memory and/or cognitive dysfunction.

Sequencing

Sequencing can be carried out using techniques well known in the art, using automatic sequencers. The sequencing may be performed on the complete CTGF gene locus or, more preferably, on specific domains thereof, typically those known or suspected to carry deleterious mutations or other alterations.

Amplification

Amplification is based on the formation of specific hybrids between complementary nucleic acid sequences that serve to initiate nucleic acid reproduction. Amplification may be performed according to various techniques known in the art, such as by polymerase chain reaction (PCR), ligase chain reaction (LCR), strand displacement amplification (SDA) and nucleic acid sequence based amplification (NASBA). These techniques can be performed using commercially available reagents and protocols. Preferred techniques use allele- specific PCR or PCR-SSCP. Amplification usually requires the use of specific nucleic acid primers, to initiate the reaction. Nucleic acid primers useful for amplifying sequences from the CTGF gene locus are able to specifically hybridize with a portion of the CTGF gene locus that flank a target region of said locus, said target region being altered in certain subjects having dementia especially a dementia with memory and/or cognitive dysfunction. Examples of such target regions are provided in Table 2. Another particular object of this invention resides in a nucleic acid primer useful for amplifying sequences from the CTGF gene or locus including surrounding regions. Such primers are preferably complementary to, and hybridize specifically to nucleic acid sequences in the CTGF gene locus. Particular primers are able to specifically hybridize with a portion of the CTGF gene locus that flank a target region of said locus, said target region being altered in certain subjects having a dementia, especially a dementia with memory and/or cognitive dysfunction. Primers that can be used to amplify CTGF target region comprising SNPs as identified in Table 2 may be designed based on their sequence or on the genomic sequence of CTGF.

Selective hybridization

Hybridization detection methods are based on the formation of specific hybrids between complementary nucleic acid sequences that serve to detect nucleic acid sequence mutation(s), deletion(s) or insertion(s). A particular detection technique involves the use of a nucleic acid probe specific for wild-type or altered CTGF gene or RNA, followed by the detection of the presence of a hybrid. The probe may be in suspension or immobilized on a substrate or support (as in nucleic acid array or chips technologies). The probe is typically labeled to facilitate detection of hybrids. In this regard, a particular embodiment of this invention comprises contacting the sample from the subject with a nucleic acid probe specific for an altered CTGF gene locus, and assessing the formation of a hybrid. In a particular preferred embodiment, the method comprises contacting simultaneously the sample with a set of probes that are specific, respectively, for wild type CTGF gene locus and for various altered forms thereof. In this embodiment, it is possible to detect directly the presence of mutation(s), deletion(s) or insertion(s) in the CTGF gene locus in the sample. Also, various samples from various subjects may be treated in parallel.

Within the context of this invention, a probe refers to a polynucleotide sequence which is complementary to and capable of specific hybridization with a (target portion of a) CTGF gene or RNA, and which is suitable for detecting polynucleotide polymorphisms associated with CTGF alleles which predispose to or are associated with a dementia, especially a dementia with memory and/or cognitive dysfunction. Probes are preferably perfectly complementary to the CTGF gene, RNA, or target portion thereof. Probes typically comprise single- stranded nucleic acids of between 8 to 1000 nucleotides in length, for instance of between 10 and 800, more preferably of between 15 and 700, typically of between 20 and 500. It should be understood that longer probes may be used as well. A preferred probe of this invention is a single stranded nucleic acid molecule of between 8 to 500 nucleotides in length, which can specifically hybridize to a region of a CTGF gene locus or RNA that carries a mutation, a deletion, or an insertion.

The method of the invention employs a nucleic acid probe specific for an altered (e.g., a mutated) CTGF gene or RNA, i.e., a nucleic acid probe that specifically hybridizes to said altered CTGF gene or RNA and essentially does not hybridize to a CTGF gene or RNA lacking said mutation, deletion or insertion. Specificity indicates that hybridization to the target sequence generates a specific signal which can be distinguished from the signal generated through non-specific hybridization. Perfectly complementary sequences are preferred to design probes according to this invention. It should be understood, however, that certain mismatch may be tolerated, as long as the specific signal may be distinguished from non-specific hybridization.

Particular examples of such probes are nucleic acid sequences complementary to a target portion of the genomic region including the CTGF gene locus or RNA carrying a point mutation as listed in Table 1A or IB above. More particularly, the probes can comprise a sequence selected from the group consisting of SEQ ID NO 1 to 52 or a fragment thereof comprising the SNP or a complementary sequence thereof.

The sequence of the probes can be derived from the sequences of the CTGF gene and RNA as provided in the present application. Nucleotide substitutions may be performed, as well as chemical modifications of the probe. Such chemical modifications may be accomplished to increase the stability of hybrids (e.g., intercalating groups) or to label the probe. Typical examples of labels include, without limitation, radioactivity, fluorescence, luminescence, enzymatic labeling, etc. The invention also concerns the use of a nucleic acid probe as described above in a method of detecting the presence of or predisposition to a dementia, especially a dementia with memory and/or cognitive dysfunction in a subject or in a method of assessing the response of a subject to a treatment of a dementia, especially a dementia with memory and/or cognitive dysfunction. Further aspects and advantages of the present invention will be disclosed in the following experimental section, which should be regarded as illustrative and not limiting the scope of the present application. MATERIALS AND METHODS

Subjects

Three groups of subjects were recruited.

AD patients (N=380) were recruited at the Saint Marguerite Hospital, Marseille, France. The patients were Caucasians originating from South-East of France. The clinical diagnosis of AD was established using NINCDS-ADRDA criteria for AD (McKhann and al., 1984). This AD group includes 36.2% of male patients and 63.8% of female patients. The age mean is 80 +/- 8.0 years for this group, 78.3 +/- 8.1 years for males and 81.4 +/- 8.0 years for females.

MCI patients (N=426) were recruited at the Saint Marguerite Hospital, Marseille, France. All MCI patients met criteria for an amnesic form of MCI (Peterson and al., 1999) as all had memory complaint (usually verified by an informant), performed poorly on neuropsychological tests of verbal memory, and had normal general cognitive function. This MCI group includes 39.8% males and 60.2% females. The age mean is 70.5 +/- 12.3 years for this group, 70.0 +/- 12.8 years for males and 71 +/- 11.8 years for females.

The gender and age distribution is not significantly different between these two patient groups (respectively p=0.39 and p=0.67).

DLB (N=90) AND FTD patients (N=99) were recruited at the Saint Marguerite Hospital, Marseille, France.

Healthy controls (N=376), without any neurological pathology, were recruited at the Blood Bank Center (volunteer blood donors). This control group includes 68.0% males and 32.0% females. The median age is 41.8 +/- 10.6 years for this group, 42.8 +/- 9.6 years for males and 40.2 +/- 11.9 years for females.

Statistical analysis

Univariate and multivariate logistic regressions were used to analyse the relationship between the probability of an individual to develop a dementia Univariate analyses were performed to compare genotype frequencies between cases and controls. Multivariate analyses were performed on cases and controls including all the genotyped Tag SNPs as covariates. The statistical SPSS software (version 10.0) was used for these analyses.

DNA extraction

Aliquots of 3 to 5 ml of blood were collected on sodium citrate and kept at -20°C. DNA was extracted using the standard salting out method. Some subjects refused blood collection. In this case, mouth cell samples were collected using foam-tipped applicators and applied to indicating FTA1 cards following the protocol described by Whatman ( htt p :./, \v \v . \v h at man .co.uk/). TagSNPs

TagSNPs were selected according to HapMap Data on the Caucasian reference population. This selection was done on a region covering the CTGF gene + 10 kb. This selection was performed on markers characterized by a MAF>15% and the r cut off value was fixed to 0.8. Eight Tag SNPs were selected: rsl0872387, rs9483364, rsl2527379, rsl931002, rs9493150, rs6918698, rs9399005 and rs4897554. Six SNPs (rsl0872387, rs9483364, rsl2527379, rsl931002, rs9493150, rs6918698) were located into the CTGF promoter whereas the others (rs9399005 and rs4897554) were located in the 3' region.

DNA amplification

Based on the CTGF gene reference sequence (Ensembl reference population ENSG00000118523), the sets of primers used to amplify the human CTGF gene were designed with primer3 software. Polymerase chain reactions were carried out on a robocycler gradient 96 (Stratagene, La Jolla, USA) in 30 ml reactions containing ImM of each primer, lOOng DNA, lOmM Tris-HCl pH=9, 0,1% Triton X-100, 50mM KC1, 0.2mg/ml BSA, 1.5mM MgC12, ImM of dNTP and 1.5U Taq polymerase. Following the initial denaturation step at 94°C for 5 min, the samples were subjected to 35 cycles consisting of 94°C for 1 min, annealing temperature for 45 sec and 72°C for 45 sec.

Sequencing

Purified PCR products were sequenced using ABI Prism BigDye Terminator cycle sequencing system (PE Applied Biosystems, Foster City, U.S.A.) on ABI Prism automatic sequencer. Sequencing reactions were performed on both strands Sequencing by GATC biotech (GATC, Marseille France).

Polymorphism genotyping by PCR and restriction enzyme digestion

Polymorphisms (rsl0872387, rs9483364, rsl2527379, rsl931002, rs9493150, rs6918698, rs9399005 and rs4897554) were genotyped by restriction enzyme analysis under standard conditions described by the enzyme manufacturers (New England Biolabs, Beverly, USA). Polymerase chain reaction (PCR) amplifications were carried out on a robocycler gradient 96 (Stratagene, La Jolla, U.S.A.) according to standard protocol. Each digestion was resolved on a 12% acrylamide gel, stained with ethidium bromide (lmg/ml) and visualized by UV. Primers are described in Table 2. In order to facilitate detection of the mutation by restriction enzyme digestion: a T has been changed to G (underlined base) in SEQ ID NO: 51, AC has been changed to TG (underlined base) in SEQ ID NO: 60 and AA has been changed to CT (underlined base) in SEQ ID NO: 66.

Polymorphism genotyping by PCR with specific TaqMan probes

Half of the DNA samples were genotyped a second time in a double-blind protocol by using TaqMan system (Applied Biosystems, Courtaboeuf, France. Each reaction contained 12.5 ng of genomic DNA, TaqMan Universal PCR Master Mix (Applied Biosystems, Courtaboeuf, France), 900 nM of each primer and 200 nM of each fluorescently-labelled hybridisation probe in a total volume of 5 μΐ. PCR was conducted in an ABI Prism Sequence Detection System 7900 (Applied Biosystems, Courtaboeuf, France) using the following conditions: 50°C for 2 min, 95°C for 10 min and 40 cycles of amplification (95°C denaturation for 15 s, 60°C annealing/extension for 1 min).

Additional markers were genotyped by another technology.

The genotyping was done with the Illumina's GoldenGate Assay (standard used during the International HapMap Project). Custom GoldenGate Genotyping Kits for VeraCode make it possible to simultaneously genotype 384 loci in a single well of a standard 96-well microplate. This technology involves using the holographically inscribed VeraCode beads and two-color detection instrument that identifies individual bead types and detects their assay hybridization signals. First of all, the DNA sample used in this assay is activated for binding to paramagnetic particles. Then, assay oligonucleotides are hybridized to the activated DNA. Three oligonucleotides are designed to identify the allele at each SNP locus in the experiment. Two oligonucleotides are specific to each allele of the tested SNP site. Each of them, contains a region of genomic complementarity but has a different nucleotide at the 3' end, which corresponds to the complement of each possible allele in the genomic DNA. Moreover each oligonucleotide also contains a unique universal PCR primer sequence. A third oligonucleotide hybridizes several bases downstream from the SNP site. This oligonucleotide also contains a region of genomic complementarity and a universal PCR primer site. This primer is characterized by an additional unique address sequence that targets a particular bead type for assay readout later in the procedure. During the primer hybridization process, the assay oligonucleotides hybridize to the activated genomic DNA sample bound to paramagnetic particles. Following hybridization, several wash steps are performed, reducing noise by removing excess and non- specifically hybridized oligonucleotides. During the next step, a genomic DNA amplification, allele specific, is performed using a DNA polymerase. The generated products provide a template for PCR using universal PCR primers Cy3 and Cy5 labeled. Finally, the dye-labeled single-stranded DNA is hybridized to complementary VeraCode bead types. It allows for the separation of the assay products in solution onto a solid surface for individual SNP genotype readout. After hybridization, the BeadXpress Reader analyzes Cy3 and Cy5 fluorescence signals on each bead in addition to the inscribed holographic element of each bead that identifies the locus of interest. The data were then analyzed with BeadStudio Data Analysis Software for automated genotype calling. This approach is an unmatched combination of sensitivity, high call rates, reproducibility, and accuracy.

The markers identified through this analysis were rsl029122, rsl 1154653, rsl0293697, rsl931001 and rs9321315

Markers in LD with rs4897554 and rs9399005 were genotyped with the Illumina's GoldenGate Assay;

The genotyping was done with the Illumina's GoldenGate Assay for the S Ps in Linkage disequilibrium with rs4897554. These markers are rsl2192391, rsl2194675, rs 12200211, rsl2214123, rsl3211178, rs6914218, rs6929832, rsl2526196, rs6917644, rs6918698, rs7748518, rs928501, rs9399005, rs9493150, rs9388949, rsl l l54654, rsl2055428, rsl930994, rs9402377, rsl0872387, rsl2192108, rs2027034, rs7747551, rs975445, rsl2663962, rs9375854, rs4452668, rs9402376, rsl931004, rs928505, rs9388953, rs9493157, rs7767139, rs928504, rsl930995, rs7767582, rs9483365, rs9493160, rs9493142. Same approach was done for the markers in LD with rs9399005. These markers are rs7768619, rsl2192391, rsl2194675, rsl2200211, rsl2214123, rsl2526196, rsl3211178, rs4897554, rs6914218, rs6917644, rs6929832, rs6918698, rsl931004, rs9388949, rsl2527379, rs2095252, rsl931000, rs928505, rs2151532, rsl931002, rs7763896, rs4897555.

Table 2: Primers for genotyping by restriction enzyme analysis.

Polymorphism Primer sequences(forward and reverse) PCR Restriction Expected profiles product Enzyme

size

rsl0872387 GACCTAATATTATAACACGGAC 190 bp aVall A/A genotype:

(SEQ ID NO: 51) 190 bp

AAAGCTGACCAATTATCTGCAA

(SEQ ID NO: 52) C/C genotype:

19+171 bp rsl9483364 TGTCATCAAATTGCCACAGG 304 bp SfaNI A/A genotype:

(SEQ ID NO: 53) 304 bp

TTTCTCCCTTCTCCTACAGCA (SEQ

ID NO: 54) G/G genotype:

120+184 bp rsl2527379 CCATCAGCACTGAAGCAATG (SEQ 331 bp Tsp509I A/A genotype:

ID NO: 55) 30+66+235 bp

CACACACATGTGCAGGTCAA G/G genotype: (SEQ ID NO: 56) 30+301 bp

A/G genotype:

19+66+235+301 bp

rs 1931002 TGATTCCCATAGGCATGGTT 264 bp Taql A/A genotype:

(SEQ ID NO: 57) 264 bp

TTCTTGGAAACCAAGGTGGTA

(SEQ ID NO: 58) G/G genotype:

70+194 bp rs9493150 TGTTAATACCTCCAGTTCCCATTT 186 bp Pstl C/C genotype: (SEQ ID NO: 59) 16+170 bp

TAAAGCCGATCTTTGCTGCA

(SEQ ID NO: 60) G/G genotype:

186 bp

rs6918698 CGGTTTTGGGACAAGAAAGA 262 bp Mnll C/C genotype:

(SEQ ID NO: 61) 262 bp CTCCTAGGTGAACCCCCTTT

(SEQ ID NO: 62) G/G genotype:

122+139 bp rs9399005 AAGTTCAGAAACAGACCTAGAGCA 271 bp Mscl C/C genotype:

(SEQ ID NO: 63) 271 bp CTTGCAGGCATACACACCAC

(SEQ ID NO: 64) T/T

genotype : 113+ 158 bp

rs4897554 TGCAAATTTTCCCTCACAAA 182 bp Nhel A/A genotype:

(SEQ ID NO: 65) 182 bp

CATCCATCCAAAAACTGCTAG

(SEQ ID NO: 66) G/G genotype:

21+161 bp

EXAMPLES

EXAMPLE 1

Association between SNPs rs9399005, rs4897554, rs9493150 rs9483364, rs 10872387, 12527379 and rs6918698 in the CTGF locus with Alzheimer disease in univariate analysis. The inventors have genotyped SNPs either by performing restriction analysis or by using the TaqMan system according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the AD group and on the control subjects are described in Table 3.

Table 3. Genotype distributions of the CTGF SNPs in the various study groups.

Polymorphism Genotype Alzheimer MCI DLB FTP Control rsl0872387 AA 239 (62.9%) 257 (60.5%) 53 (58.9%) ^ 205 (55.3%)

(bi.b /o)

AC 121 (31.8%) 146 (34.4%) 34 (37.8%) 33 144 (38.8%) (33.3%)

CC 20(5.3%) 22(5.2%) 3(3.3%) 3(3.0%) 22(5.9%) rs9483364 47

s ^β·"° GG 191(50.4%) 211 (49.9%) 50(55.6%) 179(48.9%)

(47.5%)

42

169(44.6%) 173 (40.9%) 29(32.2%) ... 3%)

(4 .4 Z 144(39.

% )

19(5.0%) 39(9.2%) 11(12.2%) 43(11.7%)

(lOAVo) rsl2527379 GG 103 (27.1%) 141 (33.1%) 32(35.6%) .I 151(40.5%)

(30.3%)

200(52.6%) 204(47.9%) 40(44.4%) 162(43.4%)

22

77(20.3%) 81 (19.0%) 18(20.0%) ... -ΓΓ 60(16.1%)

(ZZ.Z /o)

r«1 ^100? 7Q

GG 313(82.4%) 357 (84.0%) 79 (87.8%) , _n„ ' , 305 (82.2%)

(/y.oyoj

20

AG 64(16.8%) 64(15.1%) 11(12.2%) 64(17.3%)

(ZO.ZVo)

AA 3(0.8%) 4(0.9%) 0(0.0%) 0(0.0%) 2(0.5%) rs4897554 36

AA 119(32.3%) 151 (35.5%) 33 (36.7%) 57(21.8%)

(36.7%)

186 (50.5%) 195 (45.9%) 44(48.9%) 191(73.2%)

63(17.1%) 79(18.6%) 13(14.4%) 13(5.0%)

rs6918698 GG 103 (27.5%) 112(27.1%) 18(20.9%) „^ 96(25.9%)

(30.9%)

198 (52.8%) 216(52.3%) 47(54.7%) _/ (4_/1-o₀.yJ/^ 169(45.6%) o)

74 (19.7%) 85 (20.6%) 21 (24.4%) 106 (28.6%)

(ZO.ZVo)

rs9399005 57

CC 196(51.7%) 221 (52.2%) 48 (53.3%) ₍₅₇ ^ 183 (49.5%)

TC 162(42.7%) 185 (43.7%) 36(40.0%) .I⁹. 144(38.9%)

(jyAVo)

TT 21(5.5%) 17(4.0%) 6(6.7%) 3(3.0%) 43(11.6%) rs9493150 45

GG 191(50.5%) 202(48.3%) 38 (43.2%) 48.0%)

( ,46._η9Ζ%, 178 (

)

157(41.5%) 180(43.1%) 42(47.7%) ... * 143 (38.5%)

(4 . IVo)

30(7.9%) 36(8.6%) 8(9.1%) 50(13.5%)

(IU.4%)

A difference in CTGF genotype distribution between AD patients and controls was observed for seven polymorphisms (rs9399005, rs4897554, rs9493150, rs9483364, rs 10872387, 12527379 and rs6918698). In the AD subjects group, 358 (94.5%) subjects carry either the rs9399005 C/C or rs9399005C/T genotypes whereas 327 (88.4%) of the controls carry the same genotypes. This difference is significant (p=0.004; OR=0.668; 95% CI: 0.509-0.876).

For the rs4897554 polymorphism, 305 (82.9%) AD subjects carry the genotypes rs4897554A/A or rs4897554A/G versus 248 (95%) for the control. This difference is significant (p=l.510^"5; OR=1.985; 95% CI: 1.456-2.707).

Similarly, 30 (7.9%) AD patients carry the rs9493150C/C genotype whereas up to 50 (13.5%) of the controls carry this genotype. This difference is significant (p=0.015; OR=0.744; 95% CI: 0.586-0.944).

Similarly, 19 (5.0%) AD patients carry the rs9483364A/A genotype whereas up to 43 (11.7%) of the controls carry this genotype. This difference is significant (p=0.001 OR=0.630; 95% CI: 0.476-0.833).

Similarly, 239 (62.9%) AD patients carry the rsl0872387A/A genotype whereas up to 205 (55.3%) of the controls carry this genotype. This difference is significant (p=0.033; OR=1.172; 95% CI: 1.012-1.356).

Similarly, in AD patients, 103 (27.1%) carry the rsl2527379G/G genotype, 200 (52.6%) carry the rsl2527379A/G genotype and 77 (20.3%) carry the rsl2527379A/A genotype whereas into the controls group 151 (40.5%) carry the rsl2527379G/G genotype, 162 (43.4%) carry the rsl2527379A/G genotype and 60 (16.1%) carry the rsl2527379A/A genotype. This difference is significant (p=0.001; OR=1.431; 95% CI: 1.164-1.759).

Similarly, 301 (80.3%) AD patients carry the rs6918698G/G or rs6918698G/C genotypes whereas up to 265 (71.5%) of the controls carry these two genotypes. This difference is significant (p=0.005; OR=0.784; 95% CI: 0.662-0.929). See Table 4. Table 4. Association between TAG SNPs in the CTGF locus with AD in univariate analysis.

Univariate Analysis : Alzheimer vs Control

rsl0872387 AA vs AC vs CC p=0.059; OR= 1.260; 95% CI: 0.992-1.601 (NS)

AA vs AC + CC p=0.033; OR=1.172; 95% CI: 1.012-1.356 (NS)

AA + CC vs AC p=0.046; OR=0.858; 95% CI: 0.739-0.997 (S)

AA + AC vs CC p=0.691; OR=0.939; 95% CI: 0.687-1.282 (NS) rs9483364 AA vs AG vs GG p=0.080; OR=0.817; 95% CI: 0.652-1.024 (NS)

AA vs AG + GG p=0.001 OR=0.630; 95% CI: 0.476-0.833 (S)

AA + GG vs AG p=0.147; OR=1.114; 95% CI: 0.963-1.289 (NS)

AA + AG vs GG p=0.685; OR= 1.030; 95% CI: 0.892-1.189 (NS) rs 12527379 AA vs AG vs GG p=0.001; OR= 1.431; 95% CI: 1.164-1.759 (S) AA vs AG + GG P= =0.138; OR=1.151 ; 95% CI 0.956-1.387 (NS)

AA + GG vs AG P= =0.012; OR= 1.203; 95% CI 1.042-1.389(S)

AA + AG vs GG P= = 1.2 10^"4; OR=0.739; 95% CI: 0.634-0.862 (S) rs 1931002 AA vs AG vs GG P= =0.975; OR=1.006; 95% CI 0.706-1.433 (NS)

AA vs AG + GG P= =0.675; OR=1.212; 95% CI 0.494-2.973 (NS)

AA + GG vs AG P= =0.882; OR=0.986; 95% CI 0.815-1.192 (NS)

AA + AG vs GG P= =0.955; OR=1.005; 95% CI 0.834-1.213 (NS) rs4897554 AA vs AG vs GG P= =0.741; OR=0.957; 95% CI 0.739-1.240 (NS)

AA vs AG + GG P= =0.004; OR=1.308; 95% CI 1.089-1.570 (S)

AA + GG vs AG P= = 1.710^"8; OR=0.612; 95% CI: 0.516-0.726 (S)

AA + AG vs GG P= = 1.510^"5; OR=1.985; 95% CI: 1.456-2.707 (S)

rs6918698 CC vs CG vs GG P= =0.046; OR=0.814; 95% CI 0.664-0.996 (NS)

CC vs CG + GG P= =0.005; OR=0.784; 95% CI 0.662-0.929 (S)

CC + GG vs CG P= =0.048; OR=1.156; 95% CI 1.001 -1.335 (NS)

CC + CG vs GG P= =0.623; OR=1.042; 95% CI 0.885-1.225 (NS) rs9399005 TT vs TC vs CC P= =0.077; OR=0.817; 95% CI 0.654-1.022 (NS)

TT vs TC + CC P= =0.004; OR=0.668; 95% CI 0.509-0.876 (S)

TT + CC vs TC P= =0.287; OR= 1.082; 95% CI 0.936-1.252 (NS)

TT + TC vs CC P= =0.537; OR=1.046; 95% CI 0.907-1.207 (NS) rs9493150 CC vs CG vs GG P= =0.100; OR=0.835; 95% CI 0.674-1.035 (NS)

CC vs CG + GG P= =0.015; OR=0.744; 95% CI 0.586-0.944 (S)

CC + GG vs CG P= =0.404; OR= 1.064; 95% CI 0.919-1.232 (NS)

CC + CG vs GG P= =0.485; OR=1.052; 95% CI 0.912-1.215 (NS)

(S) significant ; (NS) non significant

EXAMPLE 2

Association between SNPs rs9399005, rs4897554 in the CTGF locus with Alzheimer disease in a multivariate analysis.

The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis (polymorphism were analyzed one by one). Similarly to the univariate analysis, the genotypes were considered as binary variables. In this analysis, we included rs9399005, rs4897554, rs9493150 rs9483364, rsl0872387, 12527379 and rs6918698 as covariates. In this analysis polymorphisms rs4897554 and rs9399005 were still significantly associated to AD (p=5.8 10^"5; OR=1.897; 95% CI: 1.389-2.591 and p=0.044; OR=0.739; 95% CI: 0.551-0.992 respectively) whereas the other covariates were excluded from the analysis. See Table 5.

Table 5. Association between TAG SNPs in the CTGF locus with AD in multivariate analysis. Multivariate Analysis : Alzheimer vs Control

rs4897554 AA + AG vs GG p=5.8 10^"s; OR=1.897; 95% CI: 1.389-2.591 rs9399005 TT vs TC + CC p=0.044; OR=0.739; 95% CI: 0.551-0.992 rs9483364 AA vs AG vs GG excluded

rs9493150 CC vs CG + GG excluded

rsl0872387 AA vs AC + CC excluded

rs 12527379 AA vs AG vs GG excluded

rs6918698 CC vs CG + GG excluded

EXAMPLE 3

Among the additional markers in the CTGF locus, rsl931001 is associated with Alzheimer disease in a univariate analysis.

The inventors have genotyped additional SNPs by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the AD group and on the control subjects are described in Table 6. One marker detects significant difference between cases and controls.

in AD patients, 229 (60.3%) carry the rsl931001C/C or rsl931001A/A genotypes, whereas into the controls group 183 (49.2%) carry these genotypes. This difference is significant (p=0.002; OR=0.799; 95% CI: 0.692-0.923). See Table 7.

Table 7. Association between additional markers in the CTGF locus with AD in univariate analysis.

Univariate Analysis : Alzheimer vs Control

rsl029122 TT vs TC vs CC p=0.459; OR= 1.081; 95% CI 0.880-1.329 (NS)

TT vs TC + CC p=0.878; OR=0.988; 95% CI 0.848-1.151(NS)

TT + CC vs TC p=0.188; OR=1.101; 95% CI 0.954-1.271 (NS)

TT + TC vs CC p=0.125; OR=0.863; 95% CI 0.715-1.042 (NS)

rsl 1154653 TT vs TG vs GG p=0.158; OR= 1.392; 95% CI 0.879-2.205 (NS)

TT vs TG + GG p=0.560; OR=0.700; 95% CI 0.210-2.328 (NS)

TT + GG vs TG p=0.076; OR= 1.254; 95% CI 0.977-1.609 (NS)

TT + TG vs GG p=0.105; OR=0.817; 95% CI 0.640-1.043 (NS)

rs 10293697 TT vs TC vs CC p=0.737; OR=0.959; 95% CI 0.753-1.223 (NS)

TT vs TC + CC p=0.645; OR= 1.035; 95% CI 0.894-1.198(NS)

TT + CC vs TC p=0.173; OR=0.900; 95% CI 0.774-1.047 (NS)

TT + TC vs CC p=0.057; OR=1.395; 95% CI 0.990-1.965 (NS)

rsl931001 AA vs AC vs CC p=0.023; OR=0.775; 95% CI 0.623-0.965 (S) AA vs AC + CC p=0.985; OR= 1.002; 95% CI: 0.794-1.265 (NS) AA + CC vs AC p=0.002; OR=0.799; 95% CI: 0.692-0.923 (S) AA + AC vs CC p=0.002; OR= 1.252; 95% CI: 1.083-1.447 (S) rs9321315 TT vs TA vs AA P= =0.455; OR= =0.889; 95% CI: 0.653^{■ ■} 1.210 (NS)

TT vs TA + AA P= =0.511; OR= =0.808; 95% CI: 0.427■ ■ 1.527 (NS)

TT + AA vs TA P= =0.641; OR= =0.960; 95% CI: 0.809^{■ ■} 1.139 (NS)

TT + TA vs AA P= =0.526; OR= = 1.056; 95% CI: 0.893^{■ ■} 1.248 (NS)

(S) significant ; (NS) non significant

EXAMPLE 4

Among the markers in LD with rs4897554 or rs9399005, the polymorphisms rs 12192108, rsl2663962, rsl931000, rs2095252, rs7747551, rs9388949 and rs9493157 are associated with Alzheimer disease in a univariate analysis.

The inventors have genotyped SNPs in LD with rs4897554 or rs9399005 by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the AD group and on the control subjects are described in Table 8.

Table 8. Genotype distributions of CTGF SNPs that are in linkage disequilibrium either with rs9399005 or rs4897554 in the various study groups.

Polymorphism Genotype Alzheimer MCI DLB FTP Control rslll54654 TT 237(62.4%) 256 (60.1%) 52(57.8%) , (b,_~ 55.6%)

i.b/?o. 208 (

)

TC 123 (32.4%) 148 (34.7%) 35 (38.9%) ,,,„ 145 (38.8%)

(33.3%1³)

CC 20 (5.3%) 22 (5.2%) 3 (3.3%) 3 (3.0%) 21 (5.6%) rsl2192108 TT 239 (63.1%) 253 (60.2%) 52(57.8%) , 5.8%)

(0,_~3.3%?. 208 (5 )

TC 120(31.7%) 145 (34.5%) 35 (38.9%) ,,, ^ 143 (38.3%)

(33. i /o)

CC 20(5.3%) 22(5.2%) 3(3.3%) 3(3.1%) 22(5.9%) rsl2192391 AA 284(75.1%) 321 (78.5%) 61(70.1%) ,„ 64(70.8%)

(cD.o%. 2

)

η 29

AG 86(22.8%) 74(18.1%) 26(29.9%)

( ,_~ (26.8%) 3U._~ί%. 100

)

GG 8(2.1%) 14(3.4%) 0(0.0%) 4(4.2%) 9(2.4%) rsl2194675 AA 291 (76.6%) 337 (79.1%) 63 (70.0%) ,,_n 265 (71.4%)

(69.7%)

27

AG 80(21.1%) 75(17.6%) 27(30.0%) ,„„ΐ[ 97(26.1%)

(27.3%) GG 9 (2.4%) 14 (3.3%) 0 (0.0%) 3 (3.0%) 9 (2.4%)

65

rs 12200211 AA 284 (74.9%) 331 (77.9%) 63 (70.0%) 261 (70.0%)

(65.7%)

30

AG 86 (22.7%) 80 (18.8%) 27 (30.0%) 101 (27.1 %)

(30.3%)

GG 9 (2.4%) 14 (3.3%) 0 (0.0%) 4 (4.0%) 11 (2.9%)

66

rsl2214123 GG 283 (74.9%) 332 (78.1 %) 62 (69.7%) 261 (70.2%)

(66.7%)

29

AG 86 (22.8%) 79 (18.6%) 27 (30.3%) 101 (27.2%)

(29.3%)

AA 9 (2.4%) 14 (3.3%) 0 (0.0%) 4 (4.0%) 10 (2.7%)

73

rs 12526196 TT 305 (80.3%) 321 (75.5%) 71 (78.9%) 302 (81.2%)

(73.7%)

25

TC 73 (19.2%) 102 (24.0%) 17 (18.9%) 69 (18.5%)

(25.3%)

cc 2 (0.5%) 2 (0.5%) 2 (2.2%) 1 (1.0%) 1 (0.3%)

63

rs 12663962 GG 242 (63.9%) 261 (61.3%) 53 (58.9%) 210 (56.0%)

(63.6%)

34

TG 119 (31.4%) 145 (34.0%) 34 (37.8%) 143 (38.1 %)

(34.3%)

TT 18 (4.7%) 20 (4.7%) 3 (3.3%) 2 (2.0%) 22 (5.9%)

66

rsl3211178 GG 284 (74.7%) 331 (78.3%) 63 (70.0%) 262 (70.4%)

(66.7%)

29

AG (23.2%) 78 (18.4%) 27 (30.0%) 101 (27.2%)

(29.3%)

AA 8 (2.1%) 14 (3.3%) 0 (0.0%) 4 (4.0%) 9 (2.4%)

63

rs 1930994 AA 237 (62.4%) 256 (60.1 %) 52 (57.8%) 205 (55.3%)

(63.6%)

33

AT 123 (32.4%) 148 (34.7%) 35 (38.9%) 145 (39.1 %)

(33.3%)

TT 20 (5.3%) 22 (5.2%) 3 (3.3%) 3 (3.0%) 21 (5.7%)

28

rs 1931000 99 (26.1 %) 133 (31.2%) 30 (33.3%) 148 (39.6%)

(28.3%)

47

198 (52.1 %) 207 (48.6%) 41 (45.6%) 161 (43.0%)

(47.5%)

24

83 (21.8%) 86 (20.2%) 19 (21.1 %) 65 (17.4%)

(24.2%) cc 62 rs 1931004 249 (68.2%) 290 (69.5%) 60 (69.8%) 228 (65.3%)

(64.6%)

34

TC 106(29.0%) 117(28.1%) 23 (26.7%) 107 (30.7%)

(35.4%)

TT 10 (2.7%) 10 (2.4%) 3 (3.5%) 0 (0.0%) 14 (4.0%)

30

rs2095252 102(26.8%) 137 (32.2%) 32(36.0%) 149 (39.8%)

(30.3%)

47

198 (52.1%) 205 (48.2%) 39 (43.8%) 163 (43.6%)

(47.5%)

22

80(21.1%) 83(19.5%) 18(20.2%) 62 (16.6%)

(22.2%)

79

rs2151532 TT 310(81.6%) 351 (82.8%) 78 (87.6%) 301 (81.4%)

(79.8%)

20

TC 67 (17.6%) 69 (16.3%) 11 (12.4%) 67(18.1%)

(20.2%)

cc 3 (0.8%) 4 (0.9%) 0 (0.0%) 0 (0.0%) 2 (0.5%) rs6917644 AA 291 (77.2%) 325 (76.3%) 66 (74.2%) 287 (78.0%)

AG 77 (20.4%) 97 (22.8%) 22(24.7%) ₃ 74(20.1%) GG 9 (2.4%) 4 (0.9%) 1 (1.1%) 3(3.0%) 7(1.9%)

60

rs7747551 AA 235 (61.8%) 250(59.0%) 52(57.8%) 202 (54.4%)

(60.6%)

36

TA 124(32.6%) 150(35.4%) 32(35.6%) 145 (39.1%)

(36.4%)

TT 21 (5.5%) 24 (5.7%) 6 (6.7%) 3 (3.0%) 24 (6.5%)

66

rs7748518 GG 249 (65.5%) 288 (67.6%) 59 (65.6%) 233 (62.6%)

(66.7%)

33

AG 120(31.6%) 126 (29.6%) 31 (34.4%) 122 (32.8%)

(33.3%)

AA 11 (2.9%) 12 (2.8%) 0 (0.0%) 0 (0.0%) 17 (4.6%)

64

rs7768619 TT 245 (65.3%) 284 (68.9%) 59 (65.6%) 236 (64.1%)

(65.3%)

34

TC 123 (32.8%) 128(31.1%) 31 (34.4%) 123 (33.4%)

(34.7%)

CC 7 (1.9%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 9 (2.4%)

51

rs928501 CC 218(57.5%) 220(51.8%) 39 (43.3%) 199 (53.8%)

(52.0%)

37

AC 137 (36.1%) 176(41.4%) 41 (45.6%) 133 (35.9%)

(37.8%)

AA 24 (6.3%) 29 (6.8%) 10(11.1%) 10 38 (10.3%) (10.2%) rs9388949 TT 117(30.9%) 124(29.3%) 19(21.1%) 101(27.0%)

(28.6%)

189(49.9%) 210(49.6%) 53 (58.9%) „^ 173 (46.3%)

(4-b. Vo)

73(19.3%) 89(21.0%) 18(20.0%) ^ 100(26.7%)

4. J 7o) rs9402376 TT 235 (62.3%) 252(59.7%) 53 (58.9%) 208 (55.5%)

(bi.b /o)

TC 122(32.4%) 147 (34.8%) 32(35.6%) ,„ . I⁵. 144(38.4%)

(JJAVO)

CC 20(5.3%) 23(5.5%) 5(5.6%) 3(3.0%) 23(6.1%) rs9493157 CC 232(61.1%) 247 (58.0%) 52(57.8%) ,„ , 215(57.2%)

(bv.b /o)

TC 131(34.5%) 149 (35.0%) 32(35.6%) ,,,„1³ 122(32.4%)

(ίί.ί/ο)

TT 17(4.5%) 30(7.0%) 6(6.7%) 6(6.1%) 34(9.0%)

In AD patients, 239 (63.1%) carry the rsl2192108T/T genotype, whereas into the controls group 208 (55.8%) carry this genotype. This difference is significant (p=0.042; OR=1.164; 95% CI: 1.006-1.347).

In AD patients, 242 (63.9%) carry the rsl2663962G/G genotype, whereas into the controls group 210 (56.0%) carry this genotype. This difference is significant (p=0.028; OR=1.178; 95% CI: 1.018-1.364).

In AD patients, 99 (26.1%) carry the rsl931000G/G genotype, whereas into the controls group 148 (39.6%) carry this genotype. This difference is significant (p=8.310^"5; OR=0.733; 95% CI: 0.629-0.856).

In AD patients, 102 (26.8%) carry the rs2095252G/G genotype, whereas into the controls group 149 (39.8%) carry this genotype. This difference is significant (p=1.610^"4; OR=0.744; 95% CI: 0.638 - 0.868).

In AD patients, 235 (61.8%) carry the rs7747551A/A genotype, whereas into the controls group 202 (54.4%) carry this genotype. This difference is significant (p=0.040; OR=1.164; 95% CI: 1.007- 1.347).

In AD patients, 73 (19.3%) carry the rs9388949C/C genotype, whereas into the controls group 100 (26.7%) carry this genotype. This difference is significant (p=0.015; OR=0.808; 95% CI: 0.681 -0.960). In AD patients, 17 (4.5%) carry the rs9493157T/T genotype, whereas into the controls group 34 (9.0%) carry this genotype. This difference is significant (p=0.012; OR=0.681; 95% CI: 0.505 - 0.920). All the statistical analyses are described into Table 9. Table 9. Association between markers, in LD with rs4897554 and rs9399005, in the CTGF locus with AD disease in univariate analysis.

Univariate Analysis : Alzheimer vs Control

rsl l l54654¹ TT vs TC vs CC p=0.103; OR= 1.221; 95% CI: 0.960-1.551 (NS)

TT vs TC + CC p=0.060; OR=1.150; 95% CI: 0.994-1.330 (NS)

TT + CC vs TC p=0.067; OR=0.869; 95% CI: 0.749-1.010 (NS)

TT + TC vs CC p=0.831; OR=0.966; 95% CI: 0.705-1.324 (NS)

Γ8ΐ2192108' TT vs TC vs CC P^: =0.071; OR= = 1.247; 95% CI: 0.981-1.584 (NS)

TT vs TC + CC P^: =0.042; OR= = 1.164; 95% CI: 1.006-1.347 (S)

TT + CC vs TC P^: =0.055; OR= =0.863; 95% CI: 0.743-1.003 (NS)

TT + TC vs CC P^: =0.711; OR= =0.943; 95% CI: 0.690-1.287 (NS) rsl2192391 ^u AA vs AG vs GG P^: =0.205; OR= = 1.203; 95% CI: 0.904-1.601 (NS)

AA vs AG + GG P^: =0.179; OR= = 1.117; 95% CI: 0.950-1.312 (NS)

AA + GG vs AG P^: =0.198; OR= =0.897; 95% CI: 0.759-1.059 (NS)

AA + AG vs GG P^: =0.785; OR= =0.935; 95% CI: 0.578-1.514 (NS) rsl2194675^u AA vs AG vs GG P^: =0.155; OR= = 1.231; 95% CI: 0.924-1.640 (NS)

AA vs AG + GG P^: =0.108; OR= =1.144; 95% CI: 0.971-1.347 (NS)

AA + GG vs AG P^: =0.101; OR= =0.868; 95% CI: 0.733-1.028 (NS)

AA + AG vs GG P^: =0.959; OR= =0.988; 95% CI: 0.619-1.577 (NS) rsl2200211 ^u AA vs AG vs GG P^: =0.140; OR= = 1.234; 95% CI: 0.933-1.633 (NS)

AA vs AG + GG P^: =0.128; OR= = 1.133; 95% CI: 0.965-1.330 (NS)

AA + GG vs AG P^: =0.165; OR= =0.889; 95% CI: 0.753-1.049 (NS)

AA + AG vs GG P^: =0.625; OR= =0.895; 95% CI: 0.573-1.398 (NS) rsl2214123^u AA vs AG vs GG P^: =0.179; OR= =0.825; 95% CI: 0.622-1.093 (NS)

AA vs AG + GG P^: =0.789; OR= =0.940; 95% CI: 0.596-1.483 (NS)

AA + GG vs AG P^: =0.164; OR= =0.889; 95% CI: 0.753-1.049 (NS)

AA + AG vs GG P^: =0.149; OR= = 1.126; 95% CI: 0.958-1.322 (NS) rsl2526196^u TT vs TC vs CC P^: =0.692; OR= =0.932; 95% CI: 0.656-1.323 (NS)

TT vs TC + CC P^: =0.749; OR= =0.971; 95% CI: 0.810-1.164 (NS)

TT + CC vs TC P^: =0.817; OR= = 1.022; 95% CI: 0.851-1.227 (NS)

TT + TC vs CC P^: =0.583; OR= = 1.401; 95% CI: 0.421-4.663 (NS)

Γ812663962¹ TT vs TG vs GG P^: =0.039; OR= =0.775; 95% CI: 0.609-0.987 (S)

TT vs TG + GG P^: =0.494; OR= =0.894; 95% CI: 0.650-1.232 (NS)

TT + GG vs TG P^: =0.052; OR= =0.862; 95% CI: 0.741-1.002 (NS)

TT + TG vs GG P^: =0.028; OR= = 1.178; 95% CI: 1.018-1.364 (S) rsl3211178^u AA vs AG vs GG P^: =0.209; OR= =0.833; 95% CI: 0.627-1.108 (NS)

AA vs AG + GG P^: =0.772; OR= =0.931; 95% CI: 0.575-1.508 (NS)

AA + GG vs AG P^: =0.207; OR= =0.899; 95% CI: 0.762-1.061 (NS)

AA + AG vs GG P^: =0.186; OR= = 1.114; 95% CI: 0.949-1.309 (NS) Γ81930994¹ TT vs TA vs AA p=0.086; OR=0.810; 95% CI 0.637-1.030 (NS) TT vs TA + AA p=0.811; OR=0.962; 95% CI 0.702-1.318 (NS) TT + AA vs TA p=0.055; OR=0.864; 95% CI 0.744-1.003 (NS) TT + TA vs AA p=0.048; OR=1.158; 95% CI 1.001-1.340(NS) rsl931000² AA vs AG vs GG p=0.001; OR=1.432; 95% CI 1.168-1.755 (NS)

AA vs AG + GG p=0.124; OR=1.153; 95% CI 0.962-1.381 (NS) AA + GG vs AG p=0.013; OR=1.200; 95% CI 1.039-1.385 (NS) AA + AG vs GG p=8.3 10°; OR=0.733; 95% CI: 0.629-0.856 (NS) rs 1931004 1,2 TT vs TC vs CC p=0.310; OR=0.870; 95% CI: 0.665 - 1.138 (NS)

TT vs TC + CC p=0.349; OR=0.821; 95% CI: 0.543 - 1.240 (NS) TT + CC vs TC p=0.637; OR=0.962; 95% CI: 0.820 - 1.129 (NS) TT + TC vs CC p=0.413; OR=1.067; 95% CI: 0.913 - 1.247 (NS) rs2095252^z AA vs AG vs GG p=0.001; OR= 1.423 95% CI: 1.159 1.747 (S)

AA vs AG + GG p=0.117; OR= 1.158 95% CI: 0.964 1.392 (NS) AA + GG vs AG p=0.019; OR= 1.187 95% CI: 1.028 1.370 (S) AA + AG vs GG p=1.6 10^{^}; OR=0.744; 95% CI: 0.638 - 0.868 (S) rs2151532^z TT vs TC vs CC p=0.994; OR= :0.999: 95% CI: 0.705 1.416 (NS)

TT vs TC + CC p=0.936; OR= : 1.008: 95% CI: 0.838 1.211 (NS) TT + CC vs TC p=0.865; OR= :0.984: 95% CI: 0.816 1.186 (NS) TT + TC vs CC p=0.677; OR= :1.210: 95% CI: 0.493 2.969 (NS) rs6917644 1,2 AA vs AG vs GG p=0.713; OR=0.945; 95% CI: 0.700 1.277 (NS)

AA vs AG + GG p=0.793; OR=0.977; 95% CI: 0.823 1.161 (NS) AA + GG vs AG p=0.915; OR=1.010; 95% CI: 0.845 1.207 (NS) AA + AG vs GG p=0.649; OR=1.123; 95% CI: 0.682 1.850 (NS)

Γ87747551¹ TT vs TA vs AA p=0.061; OR=0.797; 95% CI: 0.629 1.010 (NS)

TT vs TA + AA p=0.587; OR=0.920; 95% CI: 0.680 1.244 (NS) TT + AA vs TA p=0.066; OR=0.869; 95% CI: 0.748 1.009 (NS) TT + TA vs AA p=0.040; OR=1.164; 95% CI: 1.007 1.347 (S)

Γ87748518¹ AA vs AG vs GG p=0.264; OR=0.864; 95% CI: 0.669 1.116 (NS)

AA vs AG + GG p=0.229; OR=0.789; 95% CI: 0.536 1.161 (NS) AA + GG vs AG p=0.721; OR=0.973; 95% CI: 0.835 1.133 (NS) AA + AG vs GG p=0.409; OR=1.065; 95% CI: 0.917 1.236 (NS) rs7768619^z TT vs TC vs CC p=0.645; OR= 1.066 95% CI: 0.811 1.401 (NS)

TT vs TC + CC p=0.732; OR= 1.027 95% CI: 0.883 1.193 (NS) TT + CC vs TC p=0.857; OR=0.986 95% CI: 0.846 1.149 (NS) TT + TC vs CC p=0.588; OR=0.871 95% CI: 0.529 1.435 (NS) rs928501¹ AA vs AC vs CC p=0.104; OR= :0.831: 95% CI: 0.664 1.039 (NS)

AA vs AC + CC p=0.053; OR= :0.769: 95% CI: 0.589 1.003 (NS) AA + CC vs AC p=0.954; OR= : 1.004: 95% CI: 0.865 1.166 (NS) AA + AC vs CC p=0.304; OR= : 1.079: 95% CI: 0.934 1.246 (NS) rs9388949 1,2 TT vs TC vs CC p=0.031; OR=1.247; 95% CI: 1.021 1.523 (NS)

TT vs TC + CC p=0.243; OR=1.099; 95% CI: 0.938 1.286 (NS) TT + CC vs TC p=0.321; OR=1.075; 95% CI: 0.932 1.240 (NS) TT + TC vs CC p=0.015; OR=0.808; 95% CI: 0.681 0.960 (S) rs9402376¹ TT vs TC vs CC p=0.081; OR=1.237; 95% CI: 0.974 1.570 (NS)

TT vs TC + CC p=0.056; OR=1.153; 95% CI: 0.996 1.333 (NS) TT + CC vs TC P= =0.084; OR 76; 95% CI: 0.754 - 1.018 (NS)

TT + TC vs CC P= =0.625; OR 26; 95% CI: 0.680 - 1.261 (NS)

Γ89493157¹ TTTT vvss TTCC vvss CCCC pP== =00..008866;; OORR=0.816; 95% CI: 0.648 - 1.029 (NS)

TT vs TC + CC P= =0.012; OR .81 ; 95% CI: 0.505 - 0.920 (S)

TT + CC vs TC P= =0.645; OR 36; 95% CI: 0.891 - 1.206 (NS)

TT + TC vs CC P= =0.387; OR '66; 95% CI: 0.922 - 1.234 (NS)

(S) significant ; (NS) non significant

barkers in LD with rs4897554

²Markers in LD with rs9399005 EXAMPLE 5

A multivariate analysis, performed on all the markers associated in univariate analysis, reveals that three main SNPs are associated with an increased risk of developing Alzheimer disease. The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis (polymorphism were analyzed one by one). Similarly to the univariate analysis, the genotypes were considered as binary variables. In this analysis, we included 10 polymorphisms: rs9399005, rs4897554, rsl931001, rsl2192108, rsl2663962, rsl931000, rs2095252, rs7747551, rs9388949 and rs9493157. In this analysis, five polymorphisms were still significantly associated to AD (rs4897554 (p=2.4 10^"4; OR=1.811; 95% CI: 1.319-2.485), rs9399005 (p=0.033; OR=0.694; 95% CI: 0.496-0.971), rsl931001 (p=0.006; OR=0.791; 95% CI: 0.670-0.934), rsl931000 (p=0.005; OR=0.732; 95% CI: 0.590-0.908), rs9388949 (p=0.047; OR=1.315; 95% CI: 1.004-1.724)) whereas the other covariates were excluded from the analysis. See Table 10.

Table 10. Multivariate analysis between SNPs in the CTGF locus with Alzheimer disease. Multivariate Analysis : Alzheimer vs Control

rs4897554 AA + AG vs GG P= =2.4 10^"4 OR=1.811; 95% CI: 1.319-2.485 rs9399005 TT vs TC + CC P= =0.033; OR=0.694; 95% CI: 0.496-0.971 rsl931001 AA+ CC vs AC P= =0.006; OR=0.791; 95% CI: 0.670-0.934 rsl931000 AA +AG vs GG P= =0.005; OR=0.732; 95% CI: 0.590-0.908 rs9388949 TT + TC vs CC P= =0.047; OR=1.315; 95% CI: 1.004-1.724 rsl2192108 TT vs TC + CC excluded

rsl2663962 TT + TG vs GG excluded

rs2095252 AA + AG vs GG excluded rs7747551 TT + TA vs AA excluded

rs9493157 TT vs TC + CC excluded

EXAMPLE 6

Association between SNPs rs9399005, rs4897554, rs9493150, 12527379 and rs6918698 in the CTGF locus with MCI disease in univariate analysis.

The inventors have investigated whether patients with MCI are also affected by CTGF allelic variants. The TAG polymorphisms were also genotyped on 426 subjects with Mild cognitive impairment disease. The genotype distribution is indicated in Table 3.

In the MCI subjects group, 406 (96.0%) subjects carry either the rs9399005 C/C or rs9399005C/T genotypes whereas 327 (88.4%) of the controls carry the same genotypes. This difference is significant (p=l. l 10^"4; OR=0.564; 95% CI: 0.422-0.754).

In the MCI subjects group, 346 (81.4%) carry the genotypes rs4897554A/A or rs4897554A/G versus 248 (95%) for the control. This difference is significant (p=2.2 10^"6; OR=2.087; 95% CI: 1.539-2.830)

Similarly, 36 (8.6%) MCI patients carry the rs9493150C/C genotype whereas up to 50 (13.5%) of the controls carry this genotype. This difference is significant (p=0.030; OR=0.778; 95% CI: 0.620-0.976).

Similarly, in MCI patients, 141 (33.1%) carry the rsl2527379G/G genotype, 204 (47.9%) carry the rsl2527379A/G genotype and 81 (19.0%) carry the rsl2527379A/A genotype whereas into the controls group 151 (40.5%) carry the rsl2527379G/G genotype, 162 (43.4%) carry the rsl2527379A/G genotype and 60 (16.1%) carry the rsl2527379A/A genotype. This difference is significant (p=0.041; OR=1.227; 95% CI: 1.008-1.494).

Similarly, 328 (79.4%) MCI patients carry the rs6918698G/G or rs6918698G/C genotypes whereas up to 265 (71.5%) of the controls carry these two genotypes. This difference is significant (p=0.010; OR=0.805; 95% CI: 0.683 - 0.948). See Table 11.

Table 11. Association between TAG SNPs in the CTGF locus with MCI disease in univariate analysis.

Univariate Analysis : MCI vs Control

rsl0872387 AA vs AC vs CC p=0.162; OR=1.180; 95% CI: 0.936-1.489 (NS)

AA vs AC + CC p=0.137; OR=1.113; 95% CI: 0.966-1.282 (NS) AA + CC vs AC p=0.192; OR=0.908; 95% CI: 0.786-1.050 (NS)

AA + AC vs CC p=0.643; OR=0.931; 95% CI: 0.687-1.261 (NS) rs9483364 AA vs AG vs GG p=0.462; OR=0.924; 95% CI: 0.750-1.140 (NS)

AA vs AG + GG p=0.247; OR=0.873; 95% CI: 0.695-1.098 (NS)

AA + GG vs AG p=0.657; OR=1.033; 95% CI: 0.895-1.191 (NS)

AA + AG vs GG p=0.785; OR=1.020; 95% CI: 0.887-1.173 (NS) rs 12527379 AA vs AG vs GG p=0.041; OR=1.227; 95% CI: 1.008-1.494 (S)

AA vs AG + GG p=0.279; OR=1.107; 95% CI: 0.921-1.330 (NS)

AA + GG vs AG p=0.207; OR=1.094; 95% CI: 0.951-1.258 (NS)

AA + AG vs GG p=0.031; OR=0.853; 95% CI: 0.738-0.985 (NS) rs 1931002 AA vs AG vs GG p=0.625; OR=0.917; 95% CI: 0.648-1.298 (NS)

AA vs AG + GG p=0.518; OR=1.324; 95% CI: 0.565-3.103 (NS)

AA + GG vs AG p=0.401; OR=0.922; 95% CI: 0.763-1.114 (NS)

AA + AG vs GG p=0.501; OR=1.066; 95% CI: 0.885-1.283(NS) rs4897554 AA vs AG vs GG p=0.987; OR=1.002; 95% CI: 0.787-1.275 (NS)

AA vs AG + GG p=1.7 10^"4; OR=1.404; 95% CI: 1.176-1.677 (S)

AA + GG vs AG p=6.7 10^-12; OR=0.557; 95% CI: 0.472-0.659 (S)

AA + AG vs GG p=2.2 10^-6; OR=2.087; 95% CI: 1.539-2.830 (S) rs6918698 CC vs CG vs GG p=0.071; OR= :0.834: 95% CI: 0.684 - 1.015 (NS)

CC vs CG + GG p=0.010; OR= :0.805: 95% CI: 0.683 - 0.948 (S)

CC + GG vs CG p=0.059; OR= =1.145; 95% CI: 0.995 - 1.318 (NS)

CC + CG vs GG p=0.694; OR= = 1.032: 95% CI: 0.881 - 1.210 (NS) rs9399005 TT vs TC vs CC p==0.021; OR==0.769; 95% CI: 0.616-0.961 (NS)

TT vs TC + CC p==l. l 10^-4; OR==0.564; 95% CI: 0.422-0.754 (S)

TT + CC vs TC p==0.170; OR==1.105; 95% CI: 0.958-1.263 (NS)

TT + TC vs CC p==0.434; OR==1.057; 95% CI: 0.920-1.216 (NS) rs9493150 CC vs CG vs GG p==0.277; OR==0.891; 95% CI: 0.724-1.097 (NS)

CC vs CG + GG p==0.030; OR==0.778; 95% CI: 0.620-0.976 (S)

CC + GG vs CG p==0.198; OR==1.098; 95% CI: 0.952-1.266 (NS)

CC + CG vs GG p==0.922; OR==1.007; 95% CI: 0.876-1.158 (NS)

(S) significant ; (NS) non significant

EXAMPLE 7

Association between SNPs rs9399005 and rs4897554 in the CTGF locus with MCI disease in multivariate analysis.

The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis. In this analysis, the inventors have included the rs9399005, rs4897554, rs9493150, 12527379 and rs6918698 polymorphisms as covariates. In this analysis rs9399005 and rs4897554 are still significantly associated to MCI disease whereas the other polymorphisms were excluded by the software (for rs9399005 polymorphism: p=0.046; OR=0.741; 95% CI: 0.552-0.995 and for rs4897554 polymorphism: p=5.4 10^"5; OR=1.901; 95% CI: 1.392-2.597). See Table 12. Table 12. Association between TAG SNPs in the CTGF locus with MCI disease in multivariate analysis.

Multivariate Analysis : MCI vs Control

rs4897554 AA + AG vs GG p=5.4 10^"s; OR=1.901; 95% CI: 1.392-2.597 rs9399005 TT vs TC + CC p=0.046; OR=0.741; 95% CI: 0.552-0.995 rs9493150 CC vs CG + GG excluded

rsl2527379 AA vs AG vs GG excluded

rs6918698 CC vs CG + GG excluded EXAMPLE 8

Among the additional markers in the CTGF locus, rs 1931001 is associated with MCI disease in a univariate analysis.

The inventors have genotyped additional SNPs by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the MCI group and on the control subjects are described in Table 6. Among these SNPs, only one marker detects significant difference between cases and controls.

In MCI patients, 195 (45.9%) carry the rsl931001C/C genotype, whereas into the controls group 144 (38.7%) carry this genotype. This difference is significant (p=0.041; OR=1.159; 95% CI: 1.006-1.335). See Table 13.

Table 13. Association between additional markers in the CTGF locus with MCI disease in univariate analysis.

Univariate Analysis : MCI vs Control

rsl029122 TT vs TC vs CC p=0.932; OR=0.992; 95% CI 0.817-1.203 (NS)

TT vs TC + CC p=0.899; OR= 1.010; 95% CI 0.871-1.170 (NS)

TT + CC vs TC p=0.716; OR=0.974; 95% CI 0.847-1.120 (NS)

TT + TC vs CC p=0.763; OR=1.027; 95% CI 0.864-1.221(NS)

rsl 1154653 TT vs TG vs GG p=0.428; OR=1.199; 95% CI 0.766-1.876(NS)

TT vs TG + GG p=0.763; OR=1.148; 95% CI 0.468-2.816 (NS)

TT + GG vs TG p=0.464; OR= 1.100; 95% CI 0.853-1.419 (NS)

TT + TG vs GG p=0.428; OR=0.905; 95% CI 0.707-1.158(NS) rsl 0293697 TT vs TC vs CC p=0.067; OR=0.803; 95% CI 0.636-1.015 (NS)

TT vs TC + CC p=0.250; OR=0.921; 95% CI 0.799- 1.060(NS)

TT + CC vs TC p=0.942; OR= 1.005; 95% CI 0.870-1.161 (NS)

TT + TC vs CC p=0.022; OR= 1.472; 95% CI 1.057-2.051 (NS) rs 1931001 AA vs AC vs CC p=0.050: OR=0.804; 95% CI: 0.647-1.000 (NS)

AA vs AC + CC p=0.394: OR=0.902; 95% CI: 0.712-1.143 (NS)

AA + CC vs AC p=0.128: OR=0.897; 95% CI: 0.781-1.032 (NS)

AA + AC vs CC p=0.041; OR=1.159; 95% CI: 1.006-1.335 (S)

rs9321315 TT vs TA vs AA p=0.459; OR=0.894; 95% CI: 0.666 - 1.202 (NS)

TT vs TA + AA p=0.962; OR=1.013; 95% CI: 0.585 - 1.756 (NS)

TT + AA vs TA p=0.384; OR=0.928; 95% CI: 0.785 - 1.098 (NS)

TT + TA vs AA p=0.403; OR=1.072; 95% CI: 0.910 - 1.263 (NS)

(S) significant ; (NS) non significant

EXAMPLE 9

Among the markers in LD with rs4897554 or rs9399005, the polymorphisms rsl2192391, 5 rsl2194675, rsl2200211, rsl2214123, rsl3211178, rsl931000, rs2095252 and rs7768619 are associated with MCI disease in a univariate analysis.

The inventors have genotyped SNPs in LD with rs4897554 or rs9399005 by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the MCI group and on the control subjects are described in Table 8. 10 In MCI patients, 321 (78.5%) carry the rsl2192391A/A genotype, whereas 264 (70.8%) within the control group carry this genotype. This difference is significant (p=0.013; OR=1.227; 95% CI: 1.043-1.444).

In MCI patients, 291 (76.6%) carry the rsl2194675A/A genotype, whereas into the controls group 265 (71.4%) carry this genotype. This difference is significant (p=0.012; OR=1.231; 15 95% CI: 1.046-1.447).

In MCI patients, 331 (77.9%) carry the rs 12200211 A/A genotype, whereas into the controls group 261 (70.2%) carry this genotype. This difference is significant (p=0.011; OR=1.229; 95% CI: 1.048-1.442).

In MCI patients, 332 (78.1%) carry the rsl2214123G/G genotype, whereas into the controls 0 group 261 (70.0%) carry this genotype. This difference is significant (p=0.010; OR=1.232;

95% CI: 1.050-1.446).

In MCI patients, 331 (78.3%) carry the rsl3211178G/G genotype, whereas into the controls group 262 (70.4%) carry this genotype. This difference is significant (p=0.012; OR=1.229; 95% CI: 1.047-1.443).

5 In MCI patients, 133 (31.2%) carry the rsl931000G/G genotype, whereas into the controls group 148 (39.6%) carry this genotype. This difference is significant (p=0.014; OR=0.833; 95% CI: 0.720-0.963). In MCI patients, 137 (32.2%) carry the rs2095252G/G genotype, whereas into the controls group 149 (39.8%) carry this genotype. This difference is significant (p=0.025; OR=0.848; 95% CI: 0.733 - 0.980).

In MCI patients, 284 (68.9%) carry the rs7768619T/T genotype, 128 (31.1%) carry the rs7768619T/C genotype, none carries the rs7768619C/Cgenotype whereas into the controls group 236 (64.1%) carry the rs7768619T/T genotype, 123 (33.4%) carry the rs7768619T/C genotype and 9 (2.4%) carry the rs7768619C/C genotype. This difference is significant (p=0.043; OR=1.338; 95% CI: 1.009 - 1.775). All the statistical analyses are described into Table 14.

Table 14. Association between markers, in LD with rs4897554 and rs9399005, in the CTGF locus with MCI disease in univariate analy sis.

Univariate Analysis : MCI vs Control

rsl l l54654¹ TT vs TC vs CC P= =0.245; OR= = 1.148; 95% CI: 0.910-1.449 (NS)

TT vs TC + CC P= =0.200; OR= = 1.096; 95% CI: 0.952-1.262 (NS)

TT + CC vs TC P= =0.238; OR= =0.917; 95% CI: 0.794-1.059 (NS)

TT + TC vs CC P= =0.778; OR= =0.957; 95% CI: 0.704-1.301 (NS)

rsl2192108¹ TT vs TC vs CC P= =0.230; OR= = 1.153; 95% CI: 0.914-1.455 (NS)

TT vs TC + CC P= =0.203; OR= = 1.096; 95% CI: 0.952-1.263 (NS)

TT + CC vs TC P= =0.265; OR= =0.921; 95% CI: 0.797-1.065 (NS)

TT + TC vs CC P= =0.685; OR= =0.939; 95% CI: 0.693-1.273 (NS)

rsl2192391 ^u AA vs AG vs GG P= =0.068; OR= = 1.293; 95% CI: 0.981-1.704 (NS)

AA vs AG + GG P= =0.013; OR= = 1.227; 95% CI: 1.043-1.444 (S)

AA + GG vs AG P= =0.004; OR= =0.777; 95% CI: 0.655-0.921 (S)

AA + AG vs GG P= =0.406; OR= = 1.197; 95% CI: 0.783-1.831 (NS)

rsl2194675^u AA vs AG vs GG P= =0.059; OR= = 1.305; 95% CI: 0.990-1.719 (NS)

AA vs AG + GG P= =0.012; OR= = 1.231; 95% CI: 1.046-1.447 (S)

AA + GG vs AG P= =0.004; OR= =0.777; 95% CI: 0.655-0.921 (S)

AA + AG vs GG P= =0.471; OR= = 1.169; 95% CI: 0.765-1.788 (NS)

rsl2200211 ^u AA vs AG vs GG P= =0.040; OR= = 1.326; 95% CI: 1.012-1.736 (S)

AA vs AG + GG P= =0.011; OR= = 1.229; 95% CI: 1.048-1.442 (S)

AA + GG vs AG P= =0.006; OR= =0.790; 95% CI: 0.669-0.934 (S)

AA + AG vs GG P= =0.780; OR= = 1.059; 95% CI: 0.709-1.581 (NS)

rsl2214123^u AA vs AG vs GG P= =0.045; OR= =0.757; 95% CI: 0.577-0.994 (S)

AA vs AG + GG P= =0.618; OR= = 1.110; 95% CI: 0.736-1.676 (NS)

AA + GG vs AG P= =0.004; OR= =0.783; 95% CI: 0.662-0.925 (S)

AA + AG vs GG P= =0.010; OR= = 1.232; 95% CI: 1.050-1.446 (S) rsl2526196^u TT vs TC vs CC P= =0.053; OR= =0.720; 95% CI: 0.516-1.004 (NS)

TT vs TC + CC P= =0.054; OR= =0.846; 95% CI: 0.713-1.003 (NS)

TT + CC vs TC P= =0.062; OR= = 1.178; 95% CI: 0.992-1.398 (NS)

TT + TC vs CC P= =0.647; OR= = 1.324; 95% CI: 0.398-4.407 (NS) Γ812663962¹ TT vs TG vs GG p=0.127; OR=0.834; 95% CI 0.660-1.053 (NS) TT vs TG + GG p=0.459; OR=0.889; 95% CI 0.651 -1.213 (NS) TT + GG vs TG p=0.228; OR=0.915; 95% CI 0.792-1.057 (NS) TT + TG vs GG p=0.131; OR=1.115; 95% CI 0.968-1.284 (NS) rsl3211178¹ AA vs AG vs GG p=0.057; OR=0.767; 95% CI 0.583-1.008 (NS)

AA vs AG + GG p=0.457; OR=1.175; 95% CI 0.768-1.797 (NS) AA + GG vs AG p=0.004; OR=0.779; 95% CI 0.659-0.921 (S) AA + AG vs GG p=0.012; OR=1.229; 95% CI 1.047-1.443 (S) rsl 930994¹ TT vs TA vs AA P= =0.209; OR= =0.861 95% CI: 0.682- 1.087 (NS)

TT vs TA + AA P= =0.757; OR= =0.953 95% CI: 0.701- 1.296 (NS)

TT + AA vs TA P= =0.205; OR= =0.911 95% CI: 0.789- 1.052 (NS)

TT + TA vs AA P= =0.168; OR= = 1.104 95% CI: 0.959- 1.271(NS) rsl931000² AA vs AG vs GG P= =0.028; OR= = 1.244 95% CI: 1.023- 1.512 (S)

AA vs AG + GG P= =0.312; OR= =1.097 95% CI: 0.917- 1.311 (NS)

AA + GG vs AG P= =0.117; OR= = 1.118 95% CI: 0.973- 1.286 (NS)

AA + AG vs GG P= =0.014; OR= =0.833 95% CI: 0.720- 0.963 (S) rsl931004^u TT vs TC vs CC P= =0.138; OR= =0.819 95% CI: 0.630 - 1.066 (NS)

TT vs TC + CC P= =0.207; OR= =0.767 95% CI: 0.508 - 1.158 (NS)

TT + CC vs TC P= =0.431; OR= =0.939 95% CI: 0.803 - 1.098 (NS)

TT + TC vs CC P= =0.215; OR= = 1.101 95% CI: 0.946 - 1.281 (NS) rs2095252² AA vs AG vs GG P= =0.037; OR= = 1.232 95% CI: 1.013 - 1.499 (S)

AA vs AG + GG P= =0.280; OR= = 1.105 95% CI: 0.922 - 1.325 (NS)

AA + GG vs AG P= =0.188; OR= = 1.098 95% CI: 0.955 - 1.263 (NS)

AA + AG vs GG P= =0.025; OR= =0.848 95% CI: 0.733 - 0.980 (S) rs2151532² TT vs TC vs CC P= =0.723; OR= = 1.064 95% CI: 0.756 - 1.496 (NS)

TT vs TC + CC P= =0.600; OR= = 1.050 95% CI: 0.875 - 1.259 (NS)

TT + CC vs TC P= =0.494; OR= =0.938 95% CI: 0.779 - 1.128 (NS)

TT + TC vs CC P= =0.519; OR= = 1.324 95% CI: 0.565 - 3.102 (NS) rs6917644^u AA vs AG vs GG P= =0.822; OR= =0.966 95% CI: 0.713 - 1.308 (NS)

AA vs AG + GG P= =0.570; OR= =0.953 95% CI: 0.807 - 1.125 (NS)

AA + GG vs AG P= =0.363; OR= = 1.082 95% CI: 0.913 - 1.283 (NS)

AA + AG vs GG P= =0.257; OR= =0.699 95% CI: 0.377 - 1.297 (NS)

Γ87747551¹ TT vs TA vs AA P= =0.219; OR= =0.866 95% CI: 0.689 - 1.089 (NS)

TT vs TA + AA P= =0.633; OR= =0.931 95% CI: 0.696 - 1.247 (NS)

TT + AA vs TA P= =0.281; OR= =0.924 95% CI: 0.800 - 1.067 (NS)

TT + TA vs AA P= =0.200; OR= = 1.096 95% CI: 0.952 - 1.262 (NS)

Γ87748518¹ AA vs AG vs GG P= =0.089; OR= =0.805 95% CI: 0.626 - 1.034 (NS)

AA vs AG + GG P= =0.191; OR= =0.778 95% CI: 0.534 - 1.133 (NS)

AA + GG vs AG P= =0.327; OR= =0.928 95% CI: 0.798 - 1.078 (NS)

AA + AG vs GG P= =0.141; OR= = 1.116 95% CI: 0.964 - 1.291 (NS) rs7768619² TT vs TC vs CC P= =0.043; OR= = 1.338 95% CI: 1.009 - 1.775 (S)

TT vs TC + CC P= =0.156; OR= = 1.114 95% CI: 0.960 - 1.293 (NS)

TT + CC vs TC P= =0.482; OR= =0.947 95% CI: 0.815 - 1.101 (NS)

TT + TC vs CC P= =0.999; OR= =0.000 95% CI: 0.000 - (NS)

Γ8928501¹ AA vs AC vs CC P= =0.755; OR= =0.966 95% CI: 0.778 - 1.199 (NS) AA vs AC + CC p=0.083; OR=0.800: 95% CI: 0.621■ 1.030 (NS)

AA + CC vs AC p=0.115; OR=1.122: 95% CI: 0.972■ 1.295 (NS)

AA + AC vs CC p=0.570; OR=0.960: 95% CI: 0.835■ 1.104 (NS) rs9388949 1,2 TT vs TC vs CC p=0.117; OR=1.168; 95% CI: 0.962 1.418 (NS)

TT vs TC + CC p=0.470; OR=1.059; 95% CI: 0.907 1.236 (NS)

TT + CC vs TC p=0.339; OR=1.070; 95% CI: 0.931 1.230 (NS)

TT + TC vs CC p=0.060; OR=0.854; 95% CI: 0.725 1.006 (NS)

Γ89402376¹ TT vs TC vs CC p=0.250: OR= 1.145 95% CI: 0.909 1.441 (NS)

TT vs TC + CC p=0.226: OR= 1.091 95% CI: 0.948 1.256 (NS)

TT + CC vs TC p=0.297: OR= 0.926 95% CI: 0.802 1.070 (NS)

TT + TC vs CC p=0.680: OR= 0.939 95% CI: 0.697 1.265 (NS)

Γ89493157¹ TT vs TC vs CC p=0.636: OR= 0.949 95% CI: 0.764 1.179 (NS)

TT vs TC + CC p=0.273: OR= 0.867 95% CI: 0.671 1.119 (NS)

TT + CC vs TC p=0.534: OR= 1.048 95% CI: 0.904 1.214 (NS)

TT + TC vs CC p=0.993: OR= 1.001 95% CI: 0.869 1.152 (NS)

(S) significant ; (NS) non significant

barkers in LD with rs4897554

EXAMPLE 10

A multivariate analysis, performed on all the markers associated in univariate analysis, reveals that four main SNPs are associated with an increased risk of developing MCI.

The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis (polymorphism were analyzed one by one). Similarly to the univariate analysis, the genotypes were considered as binary variables. In this analysis, we included 11 polymorphisms: rs9399005, rs4897554, rsl931001, rsl2192391, rsl2194675, rsl2200211, rsl2214123, rsl3211178, rsl931000, rs2095252 and rs7768619. In this analysis, four polymorphisms were still significantly associated to MCI (rs4897554 (p=1.3 10^"4; OR=1.841; 95% CI: 1.347-2.512), rs9399005 (p=2.7 10^"4; OR=0.493; 95% CI: 0.337-0.722), rsl931001 (p=0.042; OR=1.191; 95% CI: 1.006-1.409), rsl3211178 (p=0.021; OR=1.248; 95% CI: 1.034-1.505) whereas the other covariates were excluded from the analysis. See Table 15.

Table 15. Multivariate analysis between SNPs in the CTGF locus with MCI disease.

Multivariate Analysis : MCI vs Control

rs4897554 AA + AG vs GG p=1.3 10^"4; OR=1.841; 95% CI: 1.347-2.512 rs9399005 TT vs TC + CC p=2.7 10^"4; OR=0.493; 95% CI: 0.337-0.722 rsl931001 AA+ CC vs AC p=0.042; OR=l.191; 95% CI: 1.006-1.409 rsl3211178 AA +AG vs GG p=0.021; OR=1.248; 95% CI: 1.034-1.505 rsl2192391 AA vs AG + GG excluded

rsl2200211 AA vs AG + GG excluded

rsl2214123 AA + AG vs GG excluded

rsl931000 AA + AG vs GG excluded

rsl2194675 AA vs AG + GG excluded

rs2095252 AA + AG vs GG excluded

rs7768619 TT + TC vs CC excluded

EXAMPLE 11

Association between SNPs rs4897554 in the CTGF locus with DLB disease.

The TAG polymorphisms were also genotyped on 90 subjects with DLB disease. The genotype distribution is indicated in Table 3.

in DLB patients, 33 (36.7%) carry the rs4897554A/A genotype, 44 (48.9%) carry the rs4897554A/G genotype and 13 (14.4%) carry the rs4897554G/G genotype whereas into the controls group 57 (21.8%) carry the rs4897554A/A genotype, 191 (73.2%) carry the rs4897554A/G genotype and 13 (5.0%) carry the rs4897554G/G. This difference is significant. AA vs AG + GG (p=0.006; OR=1.439; 95% CI: 1.110-1.867), AA + GG vs AG (p=3.4 10^"5; OR=0.592; 95% CI: 0.462-0.759), AA + AG vs GG (p=0.005; OR=1.795; 95% CI: 1.197-2.691). See Table 16.

Table 16. Association between TAG SNPs in the CTGF locus with DLB disease in univariate analysis.

Univariate Analysis : DLB vs Control

rsl0872387 AA vs AC vs CC P= =0.376; OR= 1.196; 95% CI 0.805-1.776 (NS)

AA vs AC + CC P= =0.534; OR= 1.077; 95% CI 0.853-1.360 (NS)

AA + CC vs AC P= =0.856; OR= 0.978; 95% CI 0.772-1.240 (NS)

AA + AC vs CC P= =0.336; OR= 0.740; 95% CI 0.400-1.367 (NS)

rs9483364 AA vs AG vs GG P= =0.446; OR= 0.875; 95% CI 0.622-1.233 (NS)

AA vs AG + GG P= =0.901; OR= 1.023; 95% CI 0.718-1.456 (NS)

AA + GG vs AG P= =0.213; OR= 0.856; 95% CI 0.670-1.093 (NS)

AA + AG vs GG P= =0.259; OR= 1.143; 95% CI 0.906-1.441 (NS) rs 12527379 AA vs AG vs GG P= =0.294; OR= 1.186; 95% CI 0.863-1.631 (NS)

AA vs AG + GG P= =0.374; OR= 1.142; 95% CI 0.852-1.531 (NS)

AA + GG vs AG P= =0.862; OR= 1.021; 95% CI 0.810-1.287 (NS)

AA + AG vs GG P= =0.391; OR= 0.901; 95% CI 0.709-1.144 (NS) rs 1931002 AA vs AG vs GG p=0.184; OR=0.635; 95% CI 0.325-1.241(NS) AA vs AG + GG p=0.999; OR=0.000; 95% CI 0.000- (NS) AA + GG vs AG p=0.249; OR=0.817; 95% CI 0.580-1.152(NS)

AA + AG vs GG p=0.207; OR=1.247; 95% CI 0.885-1.755 (NS) rs4897554 AA vs AG vs GG p=0.421; OR=1.198; 95% CI: 0.772-1.859 (NS)

AA vs AG + GG p=0.006; OR=1.439; 95% CI: 1.110-1.867 (S)

AA + GG vs AG p=3.4 10^-5; OR=0.592; 95% CI: 0.462-0.759 (S)

AA + AG vs GG p=0.005; OR=1.795; 95% CI: 1.197-2.691 (S)

rs6918698 CC vs CG vs GG p=0.927; OR= 1.015; 95% CI 0.735-1.403 (NS)

CC vs CG + GG p=0.439; OR=0.899; 95% CI 0.686-1.178 (NS)

CC + GG vs CG p=0.129; OR=1.200; 95% CI 0.948-1.519 (NS)

CC + CG vs GG p=0.341; OR=0.871; 95% CI 0.655-1.127 (NS) rs9399005 TT vs TC vs CC p=0.265; OR=0.818; 95% CI 0.574-1.165 (NS)

TT vs TC + CC p=0.178; OR=0.737; 95% CI 0.473-1.148 (NS)

TT + CC vs TC p=0.851; OR=1.023; 95% CI 0.808-1.294 (NS)

TT + TC vs CC p=0.510; OR= 1.081; 95% CI 0.858-1.361 (NS) rs9493150 CC vs CG vs GG p=0.960; OR=1.009; 95% CI: 0.721 - 1.411 (NS)

CC vs CG + GG p=0.269; OR=0.801; 95% CI: 0.541-1.187 (NS)

CC + GG vs CG p=0.116; OR=1.207; 95% CI: 0.955 - 1.524 (NS)

CC + CG vs GG p=0.418; OR=0.908; 95% CI: 0.718-1.147 (NS)

(S) significant ; (NS) non significant

EXAMPLE 12

Among the additional markers in the CTGF locus, rs 1029122 is associated with DLB disease. The inventors have genotyped additional SNPs by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the DLB group and on the control subjects are described in Table 6. Among these SNPs, only one marker detects significant difference between cases and controls.

In DLB patients, 36 (40.4%) carry the rsl029122T/T or rsl029122C/C genotypes, whereas into the controls group 197 (53.0%) carry these genotypes. This difference is significant (p=0.035; OR=1.287; 95% CI: 1.018-1.628). See Table 17.

Table 17. Association between additional markers in the CTGF locus with DLB disease in univariate analysis.

Univariate Analysis : DLB vs Control

rsl029122 TT vs TC vs CC P= =0.970; OR= =1.006; 95% CI: 0.723-1.402 (NS)

TT vs TC + CC P= =0.269; OR= =0.865; 95% CI: 0.668-1.119 (NS)

TT + CC vs TC P= =0.035; OR= =1.287; 95% CI: 1.018-1.628 (S)

TT + TC vs CC P= =0.166; OR= =0.792; 95% CI: 0.570-1.102 (NS) rsl 1154653 TT vs TG vs GG p=0.481; OR=1.291; 95% CI 0.635-2.625 (NS)

TT vs TG + GG p=0.999; OR=0.000; 95% CI 0.000- (NS)

TT + GG vs TG p=0.270; OR=1.239; 95% CI 0.847-1.814 (NS)

TT + TG vs GG p=0.358; OR=0.838; 95% CI 0.574-1.222 (NS) rsl 0293697 TT vs TC vs CC p=0.165; OR=0.760; 95% CI 0.515-1.120 (NS)

TT vs TC + CC p=0.282; OR=0.881; 95% CI 0.699-1. HO(NS)

TT + CC vs TC p=0.589; OR=1.067; 95% CI 0.843-1.351 (NS)

TT + TC vs CC p=0.183; OR= 1.403; 95% CI 0.852-2.308 (NS) rsl931001 AA vs AC vs CC p=0.502; OR=0.884; 95% CI 0.617-1.267 (NS)

AA vs AC + CC p=0.862; OR=1.033; 95% CI 0.715-1.493 (NS)

AA + CC vs AC p=0.279; OR=0.880; 95% CI 0.698-1.109 (NS)

AA + AC vs CC p=0.319; OR=1.125; 95% CI 0.892-1.420 (NS) rs9321315 TT vs TA vs AA p=0.869 OR=0.960; 95% CI 0.587-1.567 (NS)

TT vs TA + AA p=0.732 OR=0.830; 95% CI 0.286-2.408 (NS)

TT + AA vs TA p=0.989 OR= 1.002; 95% CI 0.763-1.315 (NS)

TT + TA vs AA p=0.934 OR=1.011; 95% CI 0.774-1.322 (NS)

(S) significant ; (NS) non significant

EXAMPLE 13

Among the markers in LD with rs4897554 or rs9399005, polymorphism rs9388949 is associated with DLB disease.

The inventors have genotyped SNPs in LD with rs4897554 or rs9399005 by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the DLB group and on the control subjects are described in Table 8. In DLB patients, 37 (41.1%) carry the rs9388949T/T or rs9388949C/C genotypes, whereas into the controls group 201 (53.7%) carry these genotypes. This difference is significant (p=0.032; OR=1.290; 95% CI: 1.022-1.629). See Table 18.

Table 18. Association between markers, in LD with rs4897554 and rs9399005, in the CTGF locus with DLB disease in univariate analysis.

Univariate Analysis : DLB vs Control

rsl 1154654 TT vs TC vs CC p=0.525; OR=1.137; 95% CI 0.766-1.687 (NS)

TT vs TC + CC p=0.711; OR=1.045 ; 95% CI 0.828-1.319 (NS)

TT + CC vs TC p=0.983; OR=1.003; 95% CI 0.792-1.269 (NS)

TT + TC vs CC p=0.386; OR=0.761; 95% CI 0.411-1.410 (NS) rsl2192108¹ TT vs TC vs CC p=0.515; OR=1.139; 95% CI 0.769-1.687 (NS)

TT vs TC + CC p=0.730; OR= 1.042; 95% CI 0.826-1.315 (NS)

TT + CC vs TC p=0.923; OR=1.012; 95% CI 0.799-1.281 (NS)

TT + TC vs CC p=0.341; OR=0.742; 95% CI 0.401-1.371 (NS) rsl2192391 ^u AA vs AG vs GG p=0.771; OR=1.072; 95% CI: 0.671-1.715 (NS) AA vs AG + GG p=0.903; OR=0.984; 95% CI: 0.763-1.270 (NS) AA + GG vs AG p=0.563; OR=1.079; 95% CI: 0.835-1.394 (NS) AA + AG vs GG p=0.999; OR=0.000; 95% CI: 0.000- (NS) rs 12194675¹ AA vs AG vs GG P^: =0 866; OR=1.041; 95% CI 0.655-1.652 (NS)

AA vs AG + GG P^: =0 788; OR=0.966; 95% CI 0.751-1.243 (NS)

AA + GG vs AG P^: =0 460; OR= 1.100; 95% CI 0.854-1.418 (NS)

AA + AG vs GG P^: =0 999; OR=0.000; 95% CI 0.000- (NS) rsl2200211¹ AA vs AG vs GG P^: =0 624; OR=1.121; 95% CI 0.710-1.770 (NS)

AA vs AG + GG P^: =0 996; OR=1.001; 95% CI 0.778-1.286 (NS)

AA + GG vs AG P^: =0 578; OR=1.074; 95% CI 0.834-1.383 (NS)

AA + AG vs GG P^: =0 999; OR=0.000; 95% CI 0.000- (NS) rsl2214123' AA vs AG vs GG p=0.717; OR=0.918; 95% CI 0.580-1.455 (NS)

AA vs AG + GG p=0.999; OR=0.000; 95% CI 0.000- (NS)

AA + GG vs AG P= =0 547; OR=1.081; 95% CI: 0.839 1 392 (NS)

AA + AG vs GG P= =0 927; OR=0.988; 95% CI: 0.768 1 271 (NS) rs 12526196¹ TT vs TC vs CC P= =0 385; OR=0.791; 95% CI: 0.466 1 342 (NS)

TT vs TC + CC P= =0 621; OR=0.931; 95% CI: 0.700 1 237 (NS)

TT + CC vs TC P= =0 941; OR=1.011; 95% CI: 0.753 1 358 (NS)

TT + TC vs CC P= =0 083; OR=2.904; 95% CI: 0.870 9 697 (NS)

Γ812663962¹ TT vs TG vs GG P= =0 440; OR=0.856; 95% CI: 0.577 1 270 (NS)

TT vs TG + GG P= =0 345; OR=0.744; 95% CI: 0.402 1 375 (NS)

TT + GG vs TG P= =0 950; OR=0.992; 95% CI: 0.783 1 258 (NS)

TT + TG vs GG P= =0 620; OR=1.061; 95% CI: 0.840 1 340 (NS) rsl3211178¹ AA vs AG vs GG p=0.737; OR=0.924; 95% CI 0.581-1.468 (NS)

AA vs AG + GG p=0.999; OR=0.000; 95% CI 0.000- (NS) AA + GG vs AG p=0.588; OR=1.072; 95% CI 0.833-1.381 (NS) AA + AG vs GG p=0.936; OR=0.990; 95% CI 0.770-1.273 (NS) rsl930994¹ TT vs TA vs AA p=0.488; OR=0.870; 95% CI 0.586-1.291 (NS)

TT vs TA + AA p=0.378; OR=0.758; 95% CI 0.409-1.404 (NS) TT + AA vs TA p=0.973; OR=0.996; 95% CI 0.786-1.261 (NS) TT + TA vs AA p=0.666; OR=1.053; 95% CI 0.834-1.329 (NS) rsl931000^z AA vs AG vs GG p=0.241; OR=1.207; 95% CI 0.881-1.655 (NS)

AA vs AG + GG p=0.410; OR=1.128; 95% CI 0.847-1.502 (NS) AA + GG vs AG p=0.667; OR=1.052; 95% CI 0.835-1.326 (NS) AA + AG vs GG p=0.275; OR=0.874; 95% CI 0.686-1.113 (NS) rs 1931004 1,2 TT vs TC vs CC p=0.462; OR=0.849; 95% CI 0.548-1.314 (NS)

TT vs TC + CC p=0.823; OR=0.930; 95% CI 0.493-1.755 (NS) TT + CC vs TC p=0.478; OR=0.909; 95% CI 0.698-1.184 (NS) TT + TC vs CC p=0.436; OR=1.107; 95% CI 0.857-1.428 (NS) rs2095252² AA vs AG vs GG p=0.375; OR=1.156; 95% CI 0.840-1.590 (NS)

AA vs AG + GG p=0.414; OR=1.130; 95% CI 0.843-1.513 (NS) AA + GG vs AG p=0.968; OR=1.005; 95% CI 0.796-1.268 (NS) AA + AG vs GG p=0.500; OR=0.921; 95% CI 0.724- 1.170 (NS) rs2151532² TT vs TC vs CC p=0.146; OR=1.644; 95% CI: 0.841-3.211 (NS)

TT vs TC + CC p=0.163; OR=1.275; 95% CI: 0.906-1.794 (NS) TT + CC vs TC p=0.198; OR=0.799; 95% CI: 0.567-1.125 (NS) TT + TC vs CC p=0.999; OR=0.000; 95% CI: 0.000- (NS) rs6917644 1,2 AA vs AG vs GG p=0.582; OR: :0.874; 95% CI 0.542-1.411 (NS)

AA vs AG + GG p=0.440; OR: :0.900; 95% CI 0.689-1.176 (NS)

AA + GG vs AG p=0.339; OR: :1.142; 95% CI 0.870-1.500 (NS)

AA + AG vs GG p=0.619; OR: :0.766; 95% CI 0.267-2.197 (NS)

Γ87747551¹ TT vs TA vs AA p=0.665; OR: :0.920; 95% CI 0.630-1.343 (NS)

TT vs TA + AA p=0.946; OR: =1.016; 95% CI 0.640-1.614 (NS)

TT + AA vs TA p=0.537; OR: :0.927; 95% CI 0.730-1.178 (NS)

TT + TA vs AA p=0.569; OR: :1.070; 95% CI 0.848-1.350 (NS)

Γ87748518¹ AA vs AG vs GG p=0.257; OR: :0.779; 95% CI 0.5056-1.199 (NS)

AA vs AG + GG p=0.998; OR: :0.000; 95% CI 0.000- (NS) AA + GG vs AG p=0.766; OR: =1.038; 95% CI 0.814-1.323 (NS)

AA + AG vs GG p=0.606; OR: =1.066; 95% CI 0.837-1.357 (NS) rs7768619^z TT vs TC vs CC P= =0.530; OR= =1.156; 95% CI 0.736-1.813 (NS)

TT vs TC + CC P= =0.800; OR= =1.032; 95% CI 0.810-1.314 (NS)

TT + CC vs TC P= =0.854; OR= =1.023; 95% CI 0.802-1.304 (NS)

TT + TC vs CC P= =0.999; OR= =0.000; 95% CI 0.000- (NS) rs928501¹ AA vs AC vs CC P= =0.154; OR= =1.275; 95% CI 0.913-1.780 (NS)

AA vs AC + CC P= =0.815; OR= =1.045; 95% CI 0.723-1.512 (NS)

AA + CC vs AC P= =0.093; OR= =1.221; 95% CI 0.967- 1.541(NS)

AA + AC vs CC P= =0.076; OR= =0.811; 95% CI 0.643-1.022 (NS) rs9388949 1,2 TT vs TC vs CC P= =0.920; OR= =1.017; 95% CI 0.737-1.402 (NS)

TT vs TC + CC P= =0.253; OR= =0.850; 95% CI 0.644-1.123 (NS)

TT + CC vs TC P= =0.032; OR= = 1.290; 95% CI 1.022-1.629 (S)

TT + TC vs CC P= =0.189; OR= =0.828; 95% CI 0.624-1.098 (NS)

Γ89402376¹ TT vs TC vs CC P= =0.576; OR= =1.116; 95% CI 0.760-1.640 (NS)

TT vs TC + CC P= =0.557; OR= =1.072; 95% CI 0.849-1.354 (NS)

TT + CC vs TC P= =0.617; OR= =0.941; 95% CI 0.740-1.195 (NS)

TT + TC vs CC P= =0.836; OR= =0.949; 95% CI 0.577-1.561 (NS)

Γ89493157¹ TT vs TC vs CC P= =0.761; OR= =0.946; 95% CI 0.662-1.353 (NS)

TT vs TC + CC P= =0.452; OR= =0.841; 95% CI 0.536-1.320 (NS)

TT + CC vs TC P= =0.630; OR= =1.061; 95% CI 0.834-1.351 (NS)

TT + TC vs CC P= =0.976; OR= =0.996; 95% CI 0.789-1.258 (NS)

(S) significant ; (NS) non significant

barkers in LD with rs4897554

'Markers in LD with rs9399005 EXAMPLE 14

A multivariate analysis, performed on all the markers associated in univariate analysis, reveals that two main SNPs are associated to an increase risk to develop DLB disease.

The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis (polymorphism were analyzed one by one). Similarly to the univariate analysis, the genotypes were considered as binary variables. In this analysis, we included 3 polymorphisms: rs4897554, rsl029122, and rs9388949. In this analysis, two polymorphisms were still significantly associated to DLB disease rs4897554 (p=0.003; OR=1.872; 95% CI: 1.240-2.825), rsl029122 (p=0.057; OR=1.274; 95% CI: 0.992-1.637) whereas the other covariates were excluded from the analysis. See Table 19.

Table 19. Multivariate analysis between SNPs in the CTGF locus with DLB disease.

Multivariate Analysis : DLB vs Control

rs4897554 AA + AG vs GG p=0.003; OR=1.872; 95% CI: 1.240-2.825 rsl029122 TT + CC vs TC p=0.057; OR=1.274; 95% CI: 0.992-1.637 rs9388949 TT + CC vs TC excluded

EXAMPLE 15

Association between SNPs rs9399005, rs4897554 in the CTGF locus with FTD disease in an univariate analysis.

The TAG polymorphisms were also genotyped on 90 subjects with FTD disease. The genotype distribution is indicated in Table 3.

In the FTD subjects group, 96 (97.0%) subjects carry either the rs9399005 C/C or rs9399005C/T genotypes whereas 327 (88.4%) of the controls carry the same genotypes. This difference is significant (p=0.018; OR=0.487; 95% CI: 0.269-0.885).

In FTD patients, 36 (36.7%) carry the rs4897554A/A genotype, 45 (45.9%) carry the rs4897554A/G genotype and 17 (17.3%) carry the rs4897554G/G genotype whereas into the controls group 57 (21.8%) carry the rs4897554A/A genotype, 191 (73.2%) carry the rs4897554A/G genotype and 13 (5.0%) carry the rs4897554G/G. This difference is significant. AA vs AG + GG (p=0.005; OR=1.442; 95% CI: 1.120-1.856), AA + GG vs AG (p=2.1 10^"6; OR=0.558; 95% CI: 0.438-0.710), AA + AG vs GG (p=3.7 10^"4; OR=2.001; 95% CI: 1.365-2.932) See Table 20.

Table 20. Association between TAG SNPs in the CTGF locus with FTD disease in univariate analysis.

Univariate Analysis : FTD vs Control

rsl0872387 AA vs AC vs CC P= =0.096; OR= =1.396; 95% CI: 0.942-2.071 (NS)

AA vs AC + CC P= =0.136; OR= =1.190; 95% CI: 0.947-1.497 (NS)

AA + CC vs AC P= =0.318; OR= =0.888; 95% CI: 0.703-1.121 (NS) AA + AC vs CC p=0.262; OR=0.704; 95% CI: 0.381-1.301 (NS) rs9483364 AA vs AG vs GG p=0.978; OR= :0.995; 95% CI 0.718-1.380 (NS)

AA vs AG + GG p=0.648; OR= =0.919; 95% CI 0.639-1.321 (NS)

AA + GG vs AG p=0.579; OR: =1.066; 95% CI 0.851-1.335 (NS)

AA + AG vs GG p=0.800; OR: =0.972; 95% CI 0.778-1.214 (NS) rs 12527379 AA vs AG vs GG p=0.045; OR: = 1.369; 95% CI 1.007-1.860 (NS)

AA vs AG + GG p=0.154; OR: =1.221; 95% CI 0.928-1.606 (NS)

AA + GG vs AG p=0.472; OR: =1.085; 95% CI 0.869-1.355 (NS)

AA + AG vs GG p=0.065; OR: =0.800; 95% CI 0.630-1.014 (NS) rs 1931002 AA vs AG vs GG p=0.680; OR: =1.121; 95% CI 0.653-1.924 (NS)

AA vs AG + GG p=0.999; OR: =0.000; 95% CI 0.000- (NS)

AA + GG vs AG p=0.496; OR: =1.102; 95% CI 0.833-1.458 (NS)

AA + AG vs GG p=0.582; OR: =0.925; 95% CI 0.699-1.222 (NS) rs4897554 AA vs AG vs GG p==0.702; OR==1.085; 95% CI 0.715-1.644 (NS)

AA vs AG + GG p==0.005; OR==1.442; 95% CI 1.120-1.856 (S)

AA + GG vs AG p==2.1 10^-6; OR==0.558; 95% CI 0.438-0.710 (S)

AA + AG vs GG p==3.7 10^"4; OR==2.001; 95% CI 1.365-2.932 (S) rs6918698 CC vs CG vs GG P= =0.116; OR= =0.779; 95% CI 0.571-1.064 (NS)

CC vs CG + GG P= =0.105; OR= =0.796; 95% CI 0.604-1.049 (NS)

CC + GG vs CG P= =0.557; OR= =1.070; 95% CI 0.853-1.342 (NS)

CC + CG vs GG P= =0.332; OR= =1.131; 95% CI 0.882-1.448 (NS) rs9399005 TT vs TC vs CC P= =0.027; OR= =0.666; 95% CI 0.464-0.954 (NS)

TT vs TC + CC P= =0.018; OR= =0.487; 95% CI 0.269-0.885 (S)

TT + CC vs TC P= =0.931; OR= =1.010; 95% CI 0.805-1.267 (NS)

TT + TC vs CC P= =0.152; OR= =1.178; 95% CI 0.941-1.473 (NS) rs9493150 CC vs CG vs GG P= =0.806; OR= =0.960; 95% CI 0.694-1.328 (NS)

CC vs CG + GG P= =0.426; OR= =0.864; 95% CI 0.603-1.238 (NS)

CC + GG vs CG P= =0.457; OR= =1.090; 95% CI 0.868-1.369 (NS)

CC + CG vs GG P= =0.847; OR= =0.978; 95% CI 0.781-1.225 (NS)

(S) significant ; (NS) non significant

EXAMPLE 16

Association between SNPs rs9399005, rs4897554 in the CTGF locus with FTD disease in a multivariate analysis.

The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis. In this analysis, the inventors have included the rs9399005 and rs4897554, polymorphisms as covariates. In this analysis rs9399005 and rs4897554 were still significantly associated to FTD disease (for rs9399005 polymorphism: p=0.043; OR=0.534; 95% CI: 0.291-0.981 and for rs4897554 polymorphism: p=0.001; OR=1.899; 95% CI: 1.294-2.786). See Table 21. Table 21. Association between TAG SNPs in the CTGF locus with FTD disease in multivariate analysis.

Multivariate Analysis : FTD vs Control

rs4897554 AA + AG vs GG p=0.001; OR=1.899; 95% CI: 1.294-2.786 rs9399005 TT vs TC + CC p=0.043; OR=0.534; 95% CI: 0.291-0.981

5

EXAMPLE 17

Among the additional markers in the CTGF locus, rsl 1154653 and rs9321315 are associated with FTD disease.

The inventors have genotyped additional SNPs by the Illumina's GoldenGate Assay 10 according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the MCI group and on the control subjects are described in Table 6. Among these SNPs, only one marker detects significant difference between cases and controls.

In FTD patients, 84 (84.8%) carry the rsl l l54653G/G genotype, whereas into the controls group 342 (91.9%) carry this genotype. This difference is significant (p=0.036; OR=0.701; 15 95% CI: 0.503-0.977).

In FTD patients, 87 (87.9%) carry the rs9321315A/A genotype, whereas into the controls group 281 (75.1%) carry this genotype. This difference is significant (p=0.008; OR=1.549; 95% CI: 1.121-2.141). See Table 22.

Table 22. Association between additional markers in the CTGF locus with FTD disease in 0 univariate analysis.

Univariate Analysis : FTD vs Control

rsl029122 TT vs TC vs CC p=0.381; OR=1.151; 95% CI 0.840-1.578 (NS)

TT vs TC + CC p=0.668; OR=1.052; 95% CI 0.834-1.328 (NS)

TT + CC vs TC p=0.664; OR=1.050; 95% CI 0.841-1.311 (NS)

TT + TC vs CC p=0.286; OR=0.848; 95% CI 0.626-1.148 (NS) rsl 1154653 TT vs TG vs GG p=0.069; OR=1.786; 95% CI 0.955-3.339 (NS)

TT vs TG + GG p=0.999; OR=0.000; 95% CI 0.000- (NS)

TT + GG vs TG p=0.022; OR= 1.481; 95% CI 1.059-2.072 (S)

TT + TG vs GG p=0.036; OR=0.701; 95% CI 0.503-0.977 (S) rsl 0293697 TT vs TC vs CC p=0.470; OR= 1.162; 95% CI 0.773-1.747 (NS)

TT vs TC + CC p=0.581; OR=1.066; 95% CI 0.849-1.340(NS)

TT + CC vs TC p=0.773; OR=0.966; 95% CI 0.767-1.218 (NS)

TT + TC vs CC p=0.461; OR=0.754; 95% CI 0.355-1.600 (NS) rsl931001 AA vs AC vs CC p=0.127; OR=0.759; 95% CI 0.533-1.081 (NS) AA vs AC + CC P= =0.684; OR= =0.924; 95% CI: 0.632-1.352 (NS)

AA + CC vs AC P= =0.139; OR= =0.845; 95% CI: 0.675-1.056 (NS)

AA + AC vs CC P= =0.080; OR= =1.221; 95% CI: 0.977-1.526 (NS) rs9321315 TT vs TA vs AA P= =0.012; OR= =0.461; 95% CI: 0.252-0.842 (S)

TT vs TA + AA P= =0.666; OR= =0.791; 95% CI: 0.273-2.293 (NS)

TT + AA vs TA P= =0.010; OR= =0.642; 95% CI: 0.459-0.898 (S)

TT + TA vs AA P= =0.008; OR= =1.549; 95% CI: 1.121-2.141 (S)

(S) significant ; (NS) non significant

EXAMPLE 18

Among the markers in LD with rs4897554 or rs9399005, polymorphism rsl931000 is associated with FTD disease.

The inventors have genotyped SNPs in LD with rs4897554 or rs9399005 by the Illumina's GoldenGate Assay according to the manufacturer's instructions. The genotype frequencies of each polymorphism on the FTD group and on the control subjects are described in Table 8. In FTD patients, 28 (28.3%) carry the rsl931000G/G genotype, whereas into the controls group 148 (39.6%) carry this genotype. This difference is significant (p=0.040; OR=0.776; 95% CI: 0.609-0.988). See Table 23.

Table 23. Association between markers, in LD with rs4897554 and rs9399005, in the CTGF locus with FTD disease in univariate analysis.

Univariate Analysis : FTD vs Control

rsl l l54654¹ TT vs TC vs CC p=0.115; OR= 1.374; 95% CI: 0.926-2.040 (NS)

TT vs TC + CC p=0.152; OR= 1.182; 95% CI: 0.940-1486 (NS)

TT + CC vs TC p=0.321; OR=0.889; 95% CI: 0.704-1.122 (NS)

TT + TC vs CC p=0.305; OR=0.725; 95% CI: 0.392-1.341 (NS) rsl2192108¹ TT vs TC vs CC p=0.128; OR=1.357; 95% CI: 0.916-2.012 (NS)

TT vs TC + CC p=0.182; OR=1.169; 95% CI: 0.929-1.470 (NS)

TT + CC vs TC p=0.396; OR=0.904; 95% CI: 0.715-1.142 (NS)

TT + TC vs CC p=0.274; OR=0.710; 95% CI: 0.384-1.311 (NS) rsl2192391 ^u AA vs AG vs GG p=0.252; OR=0.787; 95% CI: 0.523-1.186 (NS)

AA vs AG + GG p=0.328; OR=0.888; 95% CI: 0.700-1.127 (NS)

AA + GG vs AG p=0.506; OR=1.087; 95% CI: 0.850-1.390 (NS)

AA + AG vs GG p=0.357; OR=1.326; 95% CI: 0.728-2.416 (NS) rsl2194675^u AA vs AG vs GG p=0.689; OR=0.918; 95% CI: 0.601-1.400 (NS)

AA vs AG + GG p=0.736; OR=0.959; 95% CI: 0.753-1.222 (NS)

AA + GG vs AG p=0.821; OR=1.029; 95% CI: 0.802-1.321 (NS)

AA + AG vs GG p=0.735; OR=1.121; 95% CI: 0.578-2.176 (NS) rsl2200211 ^u AA vs AG vs GG p=0.374; OR=0.834; 95% CI: 0.559-1.244 (NS)

AA vs AG + GG p=0.409; OR=0.906; 95% CI: 0.716-1.146 (NS)

AA + GG vs AG p=0.524; OR=1.082; 95% CI: 0.849-1.380 (NS)

AA + AG vs GG p=0.584; OR=1.177; 95% CI: 0.657-2.109 (NS) rsl2214123' AA vs AG vs GG p=0.420; OR: 1.181; 95% CI 0.788-1.771 (NS) AA vs AG + GG p=0.484; OR: 1.235; 95% CI 0.684-1.229 (NS) AA + GG vs AG p=0.672; OR: 1.054; 95% CI 0.825-1.347 (NS) AA + AG vs GG p=0.503; OR: 0.922; 95% CI 0.728-1.168 (NS) rs 12526196¹ TT vs TC vs CC p=0.084; OR: 0.645; 95% CI 0.392-1.061 (NS)

TT vs TC + CC p=0.104; OR: 0.807; 95% CI 0.623-1.045 (NS) TT + CC vs TC p=0.140; OR: 1.218; 95% CI 0.938-1.582 (NS) TT + TC vs CC p=0.348; OR: 1.946; 95% CI 0.484-7.815 (NS)

Γ812663962¹ TT vs TG vs GG p=0.087; OR: 0.707; 95% CI 0.475-1.052 (NS)

TT vs TG + GG p=0.139; OR: 0.575; 95% CI 0.277-1.196 (NS) TT + GG vs TG p=0.488; OR: 0.921; 95% CI 0.730-1.162 (NS) TT + TG vs GG p=0.172; OR: 1.173; 95% CI 0.933-1.474 (NS) rsl3211178¹ AA vs AG vs GG p=0.366; OR: 1.207; 95% CI 0.803-1.815(NS)

AA vs AG + GG p=0.387; OR: 1.303; 95% CI 0.715-2.374 (NS) AA + GG vs AG p=0.672; OR: 1.054; 95% CI 0.825-1.347 (NS) AA + AG vs GG p=0.470; OR: 0.916; 95% CI 0.723-1.161 (NS) rsl930994¹ TT vs TA vs AA p=0.102: OR=0.719: 95% CI 0.484-1.068 (NS)

TT vs TA + AA p=0.299: OR=0.722: 95% CI 0.390-1.335 (NS) TT + AA vs TA p=0.295: OR=0.883: 95% CI 0.699-1.115 (NS) TT + TA vs AA p=0.136: OR= 1.190: 95% CI 0.947-1.497 (NS) rsl931000² AA vs AG vs GG p=0.028: OR=1.407: 95% CI 1.038-1.906 (S)

AA vs AG + GG p=0.122: OR=1.233: 95% CI 0.945-1.609 (NS) AA + GG vs AG p=0.430: OR=1.094: 95% CI 0.876-1.366 (NS) AA + AG vs GG p=0.040: OR=0.776: 95% CI 0.609-0.988 (S) rs 1931004 1,2 TT vs TC vs CC p=0.604: OR=0.894: 95% CI 0.587-1.364 (NS)

TT vs TC + CC p=0.999: OR=0.000: 95% CI 0.000- (NS) TT + CC vs TC p=0.375: OR=1.114: 95% CI 0.878-1.413 (NS) TT + TC vs CC p=0.892: OR=0.984: 95% CI 0.777-1.246 (NS) rs2095252² AA vs AG vs GG p=0.062: OR=1.338: 95% CI 0.985-1.817 (NS)

AA vs AG + GG p=0.193: OR=1.199: 95% CI 0.912-1.576 (NS) AA + GG vs AG p=0.489: OR= 1.082: 95% CI 0.866-1.351 (NS) AA + AG vs GG p=0.083: OR=0.810: 95% CI 0.639-1.028 (NS) rs2151532^z TT vs TC vs CC p=0.826: OR=0.941: 95% CI 0.549-1.614 (NS)

TT vs TC + CC p=0.726: OR=0.952: 95% CI 0.721 -1.257(NS) TT + CC vs TC p=0.634: OR=1.070: 95% CI 0.810-1.414 (NS) TT + TC vs CC p=0.999: OR=0.000: 95% CI 0.000- (NS) rs6917644 1,2 AA vs AG vs GG p=0.088: OR=0.689: 95% CI 0.449-1.057 (NS)

AA vs AG + GG p=0.087: OR=0.806: 95% CI 0.629-1.032 (NS) AA + GG vs AG p=0.126: OR=1.221: 95% CI 0.946-1.576 (NS) AA + AG vs GG p=0.495: OR=1.269: 95% CI 0.640-2.520 (NS)

Γ87747551¹ TT vs TA vs AA p=0.161: OR=0.760: 95% CI 0.518-1.116 (NS)

TT vs TA + AA p=0.202: OR=0.672: 95% CI 0.365-1.238 (NS) TT + AA vs TA p=0.621: OR=0.944: 95% CI 0.750-1.188 (NS) TT + TA vs AA p=0.274: OR=1.135: 95% CI 0.905-1.422 (NS) rs7748518¹ AA vs AG vs GG p=0.175; OR=0.748; 95% CI: 0.492-1.138 (NS) AA vs AG + GG p=0.998; OR=0.000; 95% CI: 0.000- (NS) AA + GG vs AG p=0.919; OR=1.012; 95% CI: 0.800- 1.281(NS)

AA + AG vs GG p=0.459; OR=1.092; 95% CI: 0.865-1.380 (NS) rs7768619² TT vs TC vs CC p=0.542; OR=1.145; 95% CI 0.741-1.749 (NS)

TT vs TC + CC p=0.829; OR=1.026; 95% CI 0.812-1.296 (NS)

TT + CC vs TC p=0.813; OR=1.029; 95% CI 0.814-1.301 (NS)

TT + TC vs CC p=0.999; OR=0.000; 95% CI 0.000- (NS)

Γ8928501¹ AA vs AC vs CC p=0.826; OR=1.038 95% CI 0.746-1.443 (NS)

AA vs AC + CC p=0.985; OR=0.996 95% CI 0.690-1.439 (NS)

AA + CC vs AC p=0.741; OR=1.040 95% CI 0.826-1.309 (NS)

AA + AC vs CC p=0.758; OR=0.966 95% CI 0.773- 1.207 (NS) rs9388949 1,2 TT vs TC vs CC p=0.646; OR: :1.074; 95% CI 0.792-1.455 (NS)

TT vs TC + CC p=0.757; OR: :1.040; 95% CI 0.812-1.331 (NS)

TT + CC vs TC p=0.904; OR: :1.014; 95% CI 0.811-1.267 (NS)

TT + TC vs CC p=0.653; OR: :0.943; 95% CI 0.729-1.219 (NS)

Γ89402376¹ TT vs TC vs CC p=0.173; OR: =1.308; 95% CI 0.889-1.925 (NS)

TT vs TC + CC p=0.273; OR: =1.135; 95% CI 0.905-1.424 (NS)

TT + CC vs TC p=0.578; OR: =0.937; 95% CI 0.744-1.180 (NS)

TT + TC vs CC p=0.238; OR: =0.692; 95% CI 0.375-1.275 (NS)

Γ89493157¹ TT vs TC vs CC p=0.433; OR: =0.869; 95% CI 0.611-1.235 (NS)

TT vs TC + CC p=0.329; OR: =0.800; 95% CI 0.510- 1.253(NS)

TT + CC vs TC p=0.933; OR: =1.010; 95% CI 0.798-1.278 (NS)

TT + TC vs CC p=0.634; OR: =1.057; 95% CI 0.842-1.325 (NS)

(S) significant ; (NS) non significant

barkers in LD with rs4897554

'Markers in LD with rs9399005 EXAMPLE 19

A multivariate analysis, performed on all the markers associated in univariate analysis, reveals that three main SNPs are associated to an increase risk to develop FTD disease.

The inventors have performed a multivariate analysis (linear regression, stepwise procedure) to confirm the associations found previously in univariate analysis (polymorphism were analyzed one by one). Similarly to the univariate analysis, the genotypes were considered as binary variables. In this analysis, we included 5 polymorphisms: rs4897554, rs9399005, rsl 1154653 and rs9321315 and rsl931000. In this analysis, three polymorphisms were still significantly associated to FTD disease rs4897554 (p=4.6 10^"4; OR=2.007; 95% CI: 1.359- 2.965), rs9321315 (p=0.013; OR=1.538; 95% CI: 1.096-2.159) rsl l l54653 (p=0.017; OR=0.637; 95% CI: 0.440-0.923) whereas the other covariates were excluded from the analysis. See Table 24.

Table 24. Multivariate analysis between SNPs in the CTGF locus with FTD disease. Multivariate Analysis : FTD vs Control

rs4897554 AA + AG vs GG p=4.6 10^"4; OR=2.007; 95% CI: 1.359-2.965 rs9321315 TT + TA vs AA p=0.013; OR=1.538; 95% CI: 1.096-2.159 rsl 1154653 TT + TG vs GG p=0.017; OR=0.637; 95% CI: 0.440-0.923 rs9399005 TT vs TC + CC excluded

rsl 931000 AA + AG vs GG excluded

CONCLUSION

Significant association was shown between several polymorphisms and dementias.

SNPs rs9399005, rs4897554, rs9493150, rs9483364, rsl0872387, rsl2527379, rs6918698, rsl931001, rsl2192108, rsl2663962, rsl931000, rs2095252, rs7747551, rs9388949 and rs9493157, in the CTGF locus, are associated with an increased risk of developing Alzheimer disease.

rs9399005: susceptibility genotypes CC or TC relative risk: 1.50

rs4897554: susceptibility genotype GG relative risk: 1.99

rs9493150: susceptibility genotypes GG or GC relative risk: 1.34

rs9483364: susceptibility genotypes GG or GA relative risk: 1.59

rs 10872387: susceptibility genotype AA relative risk: 1.17

rs 12527379: susceptibility genotype GG relative risk: 1.35

rs6918698: susceptibility genotypes GG or GC relative risk: 1.28

rs 1931001: susceptibility genotype CC relative risk: 1.25

rs l2192108: susceptibility genotype TT relative risk: 1.16

rs l2663962: susceptibility genotype GG relative risk: 1.18

rsl931000: susceptibility genotypes AA or AG relative risk: 1.36

rs2095252: susceptibility genotypes AA or AG relative risk: 1.34

rs7747551: susceptibility genotype AA relative risk: 1.16

rs9388949: susceptibility genotypes TT or TC relative risk: 1.24

rs9493157: susceptibility genotype CC or TC relative risk: 1.47 These susceptibility genotypes may be combined. For example, subjects carrying the rsl2192108TT genotype have an Alzheimer disease risk multiplied by 1.16. Subjects carrying the rsl2663962GG genotype have an Alzheimer disease risk multiplied by 1.18. Subjects carrying the rsl2192108TT genotype AND the rsl2663962GG genotype have an Alzheimer disease risk of 1.37 (1.16x1.18). These combinations may be made with any susceptibility markers.

The preferred set of markers associated with an increased risk of developing Alzheimer disease includes rs4897554, rs9399005, rsl931001, rsl931000, rs9388949. The relative risks in this set are: 1.81 (rs4897554), 1.44 (rs9399005), 1.26 (rsl931001), 1.37 (rsl931000), 1.32 (rs9388949).

SNPs rs9399005, rs4897554, rs9493150, rsl2527379, rs6918698, rsl931001, rsl2192391, rsl2194675, rsl2200211, rsl2214123, rsl3211178, rsl931000, rs2095252 and rs7768619 in the CTGF locus, are associated with an increased risk of developing with MCI disease.

rs9399005: susceptibility genotypes CC or TC relative risk: 1.77

rs4897554: susceptibility genotype GG relative risk: 2.09

rs9493150: susceptibility genotypes GG or GC relative risk: 1.29

rs 12527379: susceptibility genotype AA or AG relative risk: 1.17

rs6918698: susceptibility genotypes GG or GC relative risk: 1.24

rs 1931001: susceptibility genotype CC relative risk: 1.16

rsl2192391: susceptibility genotype AA relative risk: 1.23

rs 12194675: susceptibility genotype AA relative risk: 1.23

rs 12200211: susceptibility genotype AA relative risk: 1.23

rsl2214123: susceptibility genotype GG relative risk: 1.23

rsl3211178: susceptibility genotype GG relative risk: 1.23

rs 1931000: susceptibility genotypes AA or AG relative risk: 1.20

rs2095252: susceptibility genotypes AA or AG relative risk: 1.18

rs7768619: susceptibility genotype TT relative risk: 1.11 The preferred set of markers associated with an increased risk of developing MCI disease is includes rs4897554, rs9399005, rsl931001, rsl3211178. The relative risks in this set are: 1.84 (rs4897554), 2.01 (rs9399005), 1.19 (rsl931001), 1.25(rsl3211178). SNPs rs4897554, rsl029122, rs9388949 in the CTGF locus are associated with an increased risk of developing DLB disease.

rs4897554: susceptibility genotype GG relative risk: 1.80

rs 1029122: susceptibility genotype CT relative risk: 1.29

rs9388949: susceptibility genotypes TC relative risk: 1.29

The preferred set of markers associated with an increased risk of developing DLB disease is includes rs4897554, rs 1029122. The relative risks in this set are: 1.87 (rs4897554), 1.27 (rs 1029122).

SNPs rs9399005, rs4897554, rsl l l54653, rs9321315 rsl931000 in the CTGF locus are associated with an increased risk of developing FTD disease.

rs9399005: susceptibility genotypes CC or TC relative risk: 2.05

rs4897554: susceptibility genotype GG relative risk: 2.00

rsl 1154653: susceptibility genotype TG relative risk: 1.48

rs9321315: susceptibility genotype AA relative risk: 1.55

rs 1931000: susceptibility genotypes AA or AG relative risk: 1.29 The preferred set of markers associated with an increased risk of developing FTD disease includes rs4897554, rs9321315, rsl l l54653. The relative risks in this set are: 2.01 (rs4897554), 1.54 (rs9321315), 1.57 (rsl l l54653).

Claims

CLAIMS 1. A method of determining whether a subject has or is at risk of developing a dementia, which method comprises detecting the presence of a mutation, deletion or insertion of one or several nucleotides in the CTGF gene locus in a biological sample of said subject.

2. The method of claim 1, wherein said dementia is a dementia with memory and/or cognitive dysfunction.

3. The method of claim 1 or 2, wherein said dementia is a senile dementia, preferably occurring in humans over 65 years old.

4. The method of anyone of the preceding claims, wherein said dementia is selected from the group consisting of mild cognitive impairment, Alzheimer's disease, fronto-temporal dementia and dementia with Lewis bodies.

5. The method of anyone of the preceding claims, wherein said mutation, deletion or insertion of one or several nucleotides is located from within 20 kb, preferably within 15 kb, more preferably within 10 kb and even more preferably within 5 kb upstream the start codon of the CTGF gene to 20 kb, preferably within 15 kb, more preferably within 10 kb and even more preferably within 5 kb downstream the 3'UTR of the CTGF gene.

6. The method of any of the preceding claims, wherein said mutation, deletion or insertion of one or several nucleotides is one or several single nucleotide polymorphism(s) (SNPs).

7. The method of claim 6, wherein said SNP is selected from the SNPs listed in Table 1A or Table IB.

8. The method of claim 7, comprising detecting the presence of a SNP or a combination of SNPs selected from the group consisting of rsl2192108, rs2095252, rsl0872387, rsl2527379, rsl2663962, rsl931000, rsl931001, rs4897554, rs6918698, rs7747551, rs9388949, rs9399005, rs9483364, rs9493150, rs9493157, rsl931001, rsl2192391, rsl2194675, rs 12200211, rsl2214123, rsl3211178, rs7768619, rsl029122, rsl 1154653, and rs9321315., preferably rs4897554, rs9399005, rsl931001, rsl931000, rs9388949, 5 rsl3211178, rsl029122, rs9321315, and/or rsl l l54653.

9. The method of claim 8, comprising detecting the presence of SNP rs9399005 wherein the presence of at least one allele C is indicative of a risk of developing a dementia or of the development of a dementia.

10

10. The method of claim 8, comprising detecting the presence of SNP rs4897554 wherein the presence of two alleles G is indicative of a risk of developing a dementia or of the development of a dementia.

15 11. The method of claim 7, for determining whether a subject has or is at risk of developing Alzheimer's disease, which method comprises detecting the presence of a SNP or combination of SNPs, selected from the group consisting of rsl2192108, rs2095252, rsl0872387, rsl2527379, rsl2663962, rsl931000, rsl931001, rs4897554, rs6918698, rs7747551, rs9388949, rs9399005, rs9483364, rs9493150 and rs9493157, preferably 0 rs7747551, rsl931001, rsl2663962, rsl0872387, and/or rs 12192108.

12. The method of claim 11, wherein the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing,

Alzheimer's disease :

5 genotype CC or TC with respect to rs9399005

genotype GG with respect to rs4897554

genotype GG or GC with respect to rs9493150

genotype GG or GA with respect to rs9483364

genotype AA with respect to rs 10872387

30 genotype GG with respect to rs 12527379

genotype GG or GC with respect to rs6918698

genotype CC with respect to rsl 931001 genotype TT with respect to rs 12192108

genotype GG with respect to rs 12663962

genotype AA or AG with respect to rs 1931000

genotype AA or AG with respect to rs2095252

5 genotype AA with respect to rs7747551

genotype TT or TC with respect to rs9388949; and/or

genotype CC or TC with respect to rs9493157.

13. The method of claim 12, wherein the presence of a combination of risk alleles, in a subject 10 is indicative of a development of, or a risk of developing, Alzheimer's disease, wherein said risk alleles are :

genotype GG with respect to rs4897554,

genotype CC or TC with respect to rs9399005,

genotype CC with respect to rs 1931001,

15 genotype AA or AG with respect to rs 1931000, and

genotype TT or TC with respect to rs9388949.

14. The method of claim 7, for determining whether a subject has or is at risk of developing mild cognitive impairment, which method comprises detecting the presence of a SNP or

20 combination of SNPs, selected from the group consisting of rsl931001, rs2095252, rsl2192391, rsl2194675, rs 1220021 1 , rs12214123, rs12527379, rs1321 1 178, rs1931000, rs4897554, rs69 8698, rs7768619, rs9399005, rs9493150, preferably rs1931001 , rs12192391 , rs12194675, rs1220021 1 , rs12214123, rs1321 1 178, and/or rs7768619.

25 15. The method of claim 14, wherein the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing, mild cognitive impairment:

genotype CC or TC with respect to rs9399005

genotype GG with respect to rs4897554

30 genotype GG or GC with respect to rs9493150

genotype AA or AG with respect to rs 12527379

genotype GG or GC with respect to rs6918698:

genotype CC with respect to rsl931001 genotype AA with respect to rs 12192391

genotype A A with respect to rs 12194675

genotype A A with respect to rs 12200211

genotype GG with respect to rs 12214123

5 genotype GG with respect to rsl3211178

genotype AA or AG with respect to rs 1931000

genotype AA or AG with respect to rs2095252, and/or

genotype TT with respect to rs7768619.

10 16. The method of claim 15, wherein the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, mild cognitive impairment, wherein said risk alleles are

genotype GG with respect to rs4897554

genotype CC or TC with respect to rs9399005

15 genotype CC with respect to rs 1931001, and

genotype GG with respect to rsl3211178.

17. The method of claim 7, for determining whether a subject has or is at risk of developing Dementia with Lewis bodies, which method comprises detecting the presence of a SNP or

20 combination of SNPs, selected from the group consisting of rsl029122, rs4897554, rs9388949, preferably rsl029122.

18. The method of claim 17, wherein the presence of one of the following genotypes, alone or in combinations, in a subject is indicative of a development of, or a risk of developing,

25 dementia with Lewis bodies:

genotype GG with respect to rs4897554

genotype CT with respect to rs 1029122, and/or

genotype TC with respect to rs9388949.

30 19. The method of claim 18, wherein the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, dementia with Lewis bodies, wherein said risk alleles are genotype GG with respect to rs4897554 and

genotype CT with respect to rs 1029122.

20. The method of claim 7, for determining whether a subject has or is at risk of developing 5 Ftonto-Temporal Dementia, which method comprises detecting the presence of a SNP or combination of SNPs, selected from the group consisting of rsl 1154653, rs 1931000 , rs4897554, rs9321315, rs9399005, preferably rsl 1154653 and/or rs9321315.

21. The method of claim 20, wherein the presence of one of the following genotypes, alone or 10 in combinations, in a subject is indicative of a development of, or a risk of developing, fronto- temporal dementia:

genotype CC or TC with respect to rs9399005

genotype GG with respect to rs4897554

genotype TG with respect to rsl 1154653

15 genotype AA with respect to rs9321315, and/or

genotype AA or AG with respect to rsl931000.

22. The method of claim 21, wherein the presence of a combination of risk alleles, in a subject is indicative of a development of, or a risk of developing, fronto-temporal dementia, wherein

20 said risk alleles are

genotype GG with respect to rs4897554

genotype AA with respect to rs9321315, and

genotype TG with respect to rsl 1154653.

25 23. The method of claim 1, which method comprises detecting the presence of a SNP at position rs9399005 or rs4897554 in the CTGF gene locus in a biological sample of said subject, which is indicative of a subject having or being at risk of developing an early form of dementia such as a mild cognitive impairment (MCI).

30 24. The method of claim 1, which method comprises detecting the presence of a SNP at position rs9399005 in the CTGF gene locus in a biological sample of said subject, which is indicative of a subject having or being at risk of developing an advanced form of dementia such as Alzheimer's disease (AD).

25. The method of any one of the preceding claims, wherein the presence of an alteration in the CTGF gene locus is detected by sequencing, selective hybridization and/or selective amplification.