WO2013179061A1 - Tomm40 as marker for parkinson's disease - Google Patents

Abstract

A method for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features, the method comprising the step of assessing the subject's genotype for TOMM40, optionally in intron 6 (IVS6) of the TOMM40 gene, optionally at position rs10524523. The method may comprise the step of determining whether the subject's genotype for TOMM40, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally determining whether there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally whether the subject's genotype at rs10524523 is SA/L or whether the difference in the length of the poly-T stretch is more than or equal to 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs. If the TOMM40 genotype, optionally in intron 6, optionally at rs 10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is SA/L or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, then the subject is considered to be at higher risk of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or at higher risk of disease progression, optionally at higher risk of developing Parkinson's Disease Dementia (PDD), optionally Dementia with Lewy Bodies (DLB), optionally of developing DLB within a one or two year period of onset of motor symptoms of Parkinson's Disease or motor Parkinsonisian features.

Description

TO 40 AS MARKER FOR PARKINSON'S DISEASE

The present invention relates to assessment and treatment relating to Parkinson's Disease and motor Parkinsonisian features.

Old age dementias are chronic and relentlessly progressive neurodegenerative disorders which are set to become the world's largest socioeconomic healthcare problem, as life expectancy continues to increase. The number of dementia patients world-wide was estimated at 24 million in 2001 and is predicted to increase to 42 million in 2020 and may reach 80 million sufferers by 2040 (Ferri et al, 2005, Comas-Herrera et al, 2005) Parkinson's disease (PD) is the second most common progressive neurodegenerative disease of ageing after Alzheimer's disease (AD), albeit considerably less prevalent. The key neuropathological substrate of PD and the two PD- associated forms of dementia [Dementia with Lewy bodies (DLB) and Parkinson's disease dementia (PDD)] are a selective neuronal loss of dopaminergic neurons in the substantia nigra and widespread intraneuronal inclusions of aggregated a-synuclein, known as Lewy bodies (LB) and Lewy neuritis (McKeith et al, 1996, Gibb and Lees, 1988, Spillantini et al, 997). DLB and PDD share common neuropsychological features of "subcortical" impairment with prominent attentional and planning (dysexecutive) deficits associated with hallucinations, fluctuating confusion and behavioral abnormalities (McKeith et al, 2005, Emre et al, 2007) The timing of the onset of dementia in relation to the onset of motor manifestations is the only criterion differentiating DLB from PDD. The former occurs prior to or within a year (or possibly two years) of the onset of motor symptoms, whilst PDD develops much later in the course of the disease (Emre et al, 2007, Lippa et al, 2007), affecting approximately 75% of PD patients, with a mean time from motor disease onset of 10-12 years (Aarsland and Kurtz, 2010). . DLB is the second most common age related neurodegenerative dementia³, with a reported prevalence of more than 20% of all dementia cases (Mc Keith et al, 1996, Zaccai et al, 2005) . PDD accounts for 3-4% of all dementia cases (Aarsland et al, 2001, De lau et al, 2005) .

In the vast majority of cases, PD, as well as DLB and PDD, are non familial and thought to be multifactorial. Studies on the genetic factors predisposing to DLB and PDD have implicated the microtubule-associated protein tau (MAPT) H1 haplotype (Seto-Salvia et al, 2011 , Williams-Gray et al, 2009, Goris et al, 2007) and ΑΡΟΕε4 (Lamb et al, 1998, , Huang et al, 2006, Pankratz et al, 2006, Papapetropoulos et al, 2007, Kobayashi et al, 2011 , Meeus et al, 2012) but these findings were not replicated by others (Papapetropoulos et al, 2007, Jasinska-Myga, 2007, Ezquerra et al, 2008, Williams-Gray et al, 2009). APOE maps to chromosome 19 and is in linkage disequilibrium with the Translocase of Outer Mitochondrial Membrane 40 homolog (TOMM40) gene. Recently, a poly- T repeat polymorphism (rs10524523) has been identified within the intervening sequence of intron 6 (IVS6) and, through a combined deep sequencing and phylogenetic approach, was shown to be strongly associated with the age of onset of dementia in Alzheimer's disease (Lutz et al, 2010, Roses et al, 2010). The very long (VL) IVS6 poly T repeat was associated with increased risk and lower age of onset, independent of the risk imparted by the APOE ε4 allele. A co-dominant trait with the age of onset determined by the combined effects of the TOMM40 and APOE genes has been proposed as a possible susceptibility mechanism, although the close proximity of TOMM40 and APOE may obscure disease risk associated with other variants within this region of linkage disequilibrium (Roses et al, 2010) See also WO 2010/019550, US 2011/0166185 and US 2011/0189165, incorporated herein by reference. Further publications on genetic factors and Alzheimer's Disease include Cruchaga et al (2011) Arch Neurol 68(8), 1013-1019 "Association and expression analyses with single- nucleotide polymorphisms in TOMM40 in Alzheimer disease."; Maruszak et al (2012) J Alzheimers Dis 28(2), 309-322 "TOMM40 rs10524523 polymorphism's role in late-onset Alzheimer's disease and in longevity"; Johnson et al (2011) Alzheimers Dement 7(4), 456-465 The effect of TOMM40 poly-T length on gray matter volume and cognition in middle-aged persons with APOE ε3/ε3 genotype; Bekris LM, Lutz F and Yu CE: J Hum Genet. 2012 Jan;57(1):18-25. Functional analysis of APOE locus genetic variation implicates regional enhancers in the regulation of both TOMM40 and APOE; and Linnertz et al (2012) PLos One 7(2), e30994 "Characterisation of the Poly-T variant in the TOMM40 gene in diverse populations."

The listing or discussion of an apparently prior-published document in this specification should not necessarily be taken as an acknowledgement that the document is part of the state of the art or is common general knowledge.

Any document referred to herein is hereby incorporated by reference in its entirety.

A first aspect of the invention provides a method for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features, the method comprising the step of assessing the subject's genotype for TOMM40, optionally in intron 6 (IVS6) of the TOM 40 gene, optionally at position rs10524523. The subject may, for example, be a subject considered to have at least early motor Parkinsonisian features, for example early signs of motor symptoms such as tremor, rigidity and bradykinesia.

The aiding in categorising or determining prognosis may be aiding in determining whether the subject has a relatively high or relatively low likelihood of developing or worsening dementia associated with Parkinson's Disease or motor Parkinsonisian features. As noted above, two PD- associated forms of dementia are generally termed Dementia with Lewy bodies (DLB) and Parkinson's disease dementia (PDD). The timing of the onset of dementia in relation to the onset of motor manifestations is the only criterion differentiating DLB from PDD. The former occurs prior to or within one or two years of the onset of motor symptoms, whilst PDD develops much later in the course of the disease^7,8, affecting approximately 75% of PD patients, with a mean time from motor disease onset of 10-12 years (Aarsland and Kurtz, 2010). Dementia occurring two or more years after onset of motor symptoms may be defined as PDD, as discussed in Example 1. The method of the invention may, for example, be useful for aiding in determining whether the subject has a relatively high or relatively low likelihood of developing Dementia with Lewy bodies (DLB) ie dementia occurring within a one or two year period of onset of motor symptoms of Parkinson's Disease or motor Parkinsonisian features.

The terms "Parkinson's Disease" and "motor Parkinsonisian features" will be well known to those skilled in the art. A clinical diagnosis of Parkinson's Disease may be made as well know to those skilled in the art. As defined by the Queen Square brain bank clinical diagnostic criteria (and for example, as set out in Example 1), such a diagnosis may be made when the subject has bradykinesia and at least one of the following: muscular rigidity, rest tremor, postural instability with asymmetrical onset and good response to levodopa in the initial phases of the disease as validating evidence (Gibb and Lees, 1989). See, for example, Nelson, P.T., Kryscio, R.J., Jicha, G.A. & Abner, E.L. Relative preservation of MMSE scores in autopsy-proven dementia with Lewy bodies. Neurology 2009; 73: 1127-1133. A diagnosis of DLB or PDD may be made as well known to those skilled in the art, for example if, in addition to a diagnosis of Parkinson's Disease or motor Parkinsonisian features, the presence of dementia is identified, for example as set out in Example 1 , "Clinical assessment" section. For example, neuropsychological features of subcortical impairment may be present, with prominent attentional and planning (dysexecutive) deficits associated with hallucinations, fluctuating confusion and behavioural abnormalities (Emre et al, 2007), as noted above.

The method for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD) may be applied to any subject and may, for example, be applied to a subject who may have other recognised risk factors for Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD) , for example a subject who is known to have a genotype (other than at TOMM40) linked with Parkinson's Disease, at least in a particular population, for example an LRRK2 genotype linked with Parkinson's Disease, for example LRRK2 G2019S, as known to those skilled in the art (Hulihan et al, Lancet Neurology, 2008). In some populations, the LRRK2 G2019S genotype has a known penetrance of ~80% by age of 80 i.e. 80% of those with the mutation have PD by age of 80 (Hulihan et al, Lancet Neurology, 2008). Other risk factors may include one or two or three or more of head trauma, male gender, age (for example over 60), diabetes, one or more of a number of genes for familial PD (a total of around 16 have been identified), hypertension, rural living, exposure to pesticides or heavy metals. The method for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features may typically be applied with a subject who has been diagnosed with PD/ motor Parkinsonisian features but not DLB or PDD. However, it may also be useful with a subject who has been diagnosed with PD/ motor Parkinsonisian features and presents with early symptoms of hallucinations, cognitive decline or dysexecutive syndrome suggesting possibly an early stage of DLB or PDD The subject typically is a human subject.

A clinician may wish to take into account the subject's genotype at TO M40, as discussed herein, alongside other parameters as noted above in arriving at a diagnosis of PD/ motor Parkinsonisian features or DLB or PDD. Accordingly, assessing a subject's genotype at TO M40, as discussed herein, may be useful in aiding in a diagnosis of PD/ motor Parkinsonisian features or DLB or PDD.

A subject may be considered otherwise at risk of Parkinson's Disease or motor Parkinsonisian features from a combination of known risk factors, such as one or two or three or more of: a number of genes for familial PD (a total of 16, for example an LRRK2 genotype linked with Parkinson's Disease, as discussed above and well known to those skilled in the art); age; male gender; diabetes; hypertension; rural living; exposure to pesticides, heavy metals etc; and head trauma. (Wirdefeldt et al, 2011).

The subject's genotype may be determined at the time of considering a diagnosis of PD/motor Parkinsonisian features, or risk thereof; or it may be (or have been) determined separately, for example as part of a wider genetic characterisation of the subject. The subject's characterisation may then be stored; and accessed when considering a diagnosis of PD/ motor Parkinsonisian features, or risk thereof, for the subject. The genotyping may be done on any suitable tissue from the subject, for example on a blood sample from the subject, as will be well known to those skilled in the art. The genotyping may be done by any suitable technique for determining genotype, as will be well known to those skilled in the art and as discussed further below.

The term TOMM40 will be well known to those skilled in the art. TOMM40 sequences are well known and a human genomic sequence may be found at, for example,

LOCUS NG_007084 10612 bp DNA linear PRI 19-FEB-2012

DEFINITION Homo sapiens apolipoprotein E (APOE), RefSeqGene on chromosome 19.

ACCESSION NG_007084

VERSION NG_007084.2 Gl:163954918

KEYWORDS RefSeqGene.

SOURCE Homo sapiens (human)

Nucleotides 1 to 2908 relates to TOMM40

The location of intron 6 (I S6) of the TOMM40 gene and position rs10524523 are also well known to those skilled in the art and are discussed in, for example, Lutz et al (2010) Alzheimers Dement 6(2), 125-131 (see for example Figure 3: diagram of exons 6(E6) to 10 (E10); and Linnertz et al (2012) PLoS ONE, 7(2), e30994; and in Example 1. Position rs10524523 is considered to be the site of a deletion/insertion polymorphism, particularly a poly-T deletion/insertion polymorphism in which the number of T residues varies between alleles. Figure 1 shows a distribution of poly-T repeat lengths at this position. See also Linnertz et al (2012), supra.

The NCBI snp database shows rs 10524523 as

tactggcatgagccattgcatctggc[-/ 1 I I I I I I I I I I I I I I I I I I jttttttttttttttttgagatgggg

Lutz et al (2010) supra, Linnertz and Example 1 (see, for example, Figure 1) classifies the poly-T repeat lengths as:

Short (S): <19

Long (L): 9 to 29

Very long (VL): >30

The method of the invention may comprise the step of determining whether the subject's genotype for TOMM40, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally determining whether there is a difference in the length of a poly-T stretch at rs10524523 between the patient's alleles, optionally whether the subject's genotype at rs10524523 is S VL or whether the difference in the length of the poly-T stretch (at rs10524523) is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs.

Without being bound by theory, the inventors consider that TOMM40 heterozygosity, for example in intron 6, for example at rs10524523, for example genotype S/VL at this position or a difference of roughly around 15 base pairs between the poly-T stretches at this position, may affect the way in which the TOMM40 gene is expressed (for example transcribed, processed and/or translated), which may also affect the way in which the complex containing the TO M40 polypeptide is assembled The skilled person will be well aware of techniques and reagents suitable for and useful in assessing genotype for TOMM40, optionally in intron 6, optionally at rs 0524523, for example for assessing whether the genotype is heterozygous or determining whether there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally whether the subject's genotype at rs10524523 is S/VL or whether the difference in the length of the poly-T stretch (at rs10524523) is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs. Any suitable technique may be used. For example, the information can be extracted from a genomic sequence of the subject. Typically, polymerase chain reaction (PCR) methods may be used, for example PCR restriction fragment length polymorphism (PCR- RFLP) methods. For example a PCR based technique may be used, for example as described in Example 1 or in Linnertz et al (2012). Examples of suitable techniques are described in Example 1 , in Lutz et al (2010) supra or in Linnertz et al (2012), supra, for example. Examples of suitable primer sequences for analysis at position rs10524523 are given in Example 1 and in Linnertz et al (2012), for example

5' -TGCTGACCTCAAGCTGTCCTC-3' and

5'-GAGGCTGAGAAGGGAGGATT-3';

or Outer: 5'-CTGGGCTCAAATGAACC-3' and 5'-CAAATGTGATTTTATAGGGCCA-3' Inner: 5'-GAGATGGGGTCTCACT-3' and 5'-ACAGGGAAAGAAAACAAGCCTG-3'; or Outer: 5'-TGGCCTCCCAAACTG-3' and 5'-GGCGTGGTGGCAC-3'

Inner: 5'-GTTAGATGAAG I I I I I AAA I T I I I I GTAG-3' and 5'- GAGAAGGGAGGATTGCT-3'

Length of poly-T stretches may be determined by, for example, sequencing techniques or by length determination, for example using capillary electrophoresis. Techniques other than or in addition to PCR may also be used, for example ligase based detection techniques such as the Ligase Detection Reaction (LDR), as well known to those skilled in the art. Other examples of suitable techniques include QB replicase and NASBA (nucleic acid sequence based amplification), also called 3SR, for example as described in Compton (1991) Nature 350, 91-92 and AIDS (1993), Vol 7 (Suppl 2), S108 or SDA (strand displacement amplification), for example as described in Walker et al (1992) Nucl. Acids Res. 20, 1691-1696. The polymerase chain reaction is particularly preferred because of its simplicity.

Typically, the genotype for TO 40 is determined on a sample obtained from the subject. Any sample that contains TO M40 nucleic acid, typically genomic DNA, in sufficient quantity to be assessed by the chosen method may be used. For example, a blood, urine, hair, skin or other epithelial cell sample may be used, as well known to those skilled in the art. The techniques for assessing TOMM40 genotype, optionally in intron 6, optionally at rs 0524523, may make use of one or more nucleic acids capable of hybridising to the TOMM40 nucleic acid, typically genomic DNA. The hybridising nucleic acid(s) which is (are) used in the methods of the invention may further comprise a detectable label, as will be well known to those skilled in the art.

By "detectable label" is included any convenient radioactive label such as ³²P, ³³P or ³⁵S which can readily be incorporated into a nucleic acid molecule using well known methods; any convenient fluorescent or chemi!uminescent label which can readily be incorporated into a nucleic acid is also included. In addition the term "detectable label" also includes a moiety which can be detected by virtue of binding to another moiety (such as biotin which can be detected by binding to streptavidin); and a moiety, such as an enzyme, which can be detected by virtue of its ability to convert a colourless compound into a coloured compound, or vice versa (for example, alkaline phosphatase can convert colourless o-nitrophenylphosphate into coloured o-nitrophenol). In one embodiment, the nucleic acid probe may occupy a certain position in a fixed assay and whether the nucleic acid hybridises to the said TOMM40 nucleic acid can be determined by reference to the position of hybridisation in the fixed assay. The detectable label may also be a fiuorophore-quencher pair as described in Tyagi & Kramer (1996) Nature Biotechnology 14, 303-308.

Other types of labels and tags are disclosed above. The nucleic acid may be branched nucleic acid (see Urdea ef at (1991) Nuc!. Acids Symposium Series 24, 197-200).

The method of the invention need not include a step of assessing the APOE genotype of the subject. Whilst the present inventors have identified that the APOE genotype ε4/ε4 may be linked to a marginally significant earlier onset of dementia (see table 5 of Example 1) it is not considered that consideration of APOE genotype alongside TOM 40 genotype provides useful further information on the likelihood of developing PDD or DLB, or developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features. Thus, in an embodiment, the method does not include the step of assessing the APOE genotype of the subject.

The subject may be considered to be at higher risk of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); disease progression, for example at higher risk of developing Dementia with Lewy Bodies (DLB) or Parkinson's Disease Dementia (PDD), for example of developing DLB (typically within a one or two year period of onset of motor symptoms of Parkinson's Disease or motor Parkinsonisian feature,) if the subject's TOMM40 genotype, for example in intron 6, for example at rs10524523, is heterozygous, for example when there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, for example when the subject's genotype at rs10524523 is S VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, for example more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs.

It is considered, for example, that a subject in the very early stages of PD (for example within one year or less of the onset of motor symptoms) found to have the S/VL rs10524523 genotype would have around a 25% risk of developing DLB, whereas a subject found not to carry this genotype would have around a 95% chance of not developing DLB.

It will be appreciated that determining the TOM 40 genotype, as set out above, in the sample may in itself allow categorising or determining prognosis in a subject with PD/motor Parkinsonisian features (or otherwise at risk thereof), or selection of a therapeutic strategy for a subject with PD/ motor Parkinsonisian features (or at risk thereof); or more typically it may be used by the clinician as an aid in categorising or determining prognosis or selection of a therapeutic strategy.

For example, the clinician may take into account the number and severity of the various PD/ motor Parkinsonisian feature symptoms (or risk factors) and/or DLB or PDD symptoms known to those skilled in the art, for example hallucinations or mild cognitive symptoms, as mentioned herein. It will be appreciated that the clinician will wish to take in to account these or other factors, as well as consider the TOMM40 genotype, as set out above, before categorising or determining prognosis or selection of a therapeutic strategy.

The method of the invention may further comprise the step of selecting a treatment regime making use of the information on the subject's TOMM40 genotype. Determination of the subject's TOMM40 genotype, as set out above, will be useful to the clinician in determining how to manage (or potentially prevent or delay) the PD/ motor Parkinsonisian features in the subject. For example, since the heterozygous genotype, as indicated above, is considered to be associated with a higher risk of developing dementia, the clinician may use the information concerning the TO M40 genotype to facilitate decision making regarding treatment of the subject. Thus, if the TOMM40 genotype is indicative of a lower probability of developing dementia unnecessary treatments may be avoided. If the TO 40 genotype is indicative of a higher probability of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD), for example of developing or worsening dementia, therapy directed to disease modification and/or to dementia may be the preferred treatment, particularly if motor Parkinsonisian features or early signs of dementia are present. Even if it is not appropriate to alter the type of therapy carried out, determining whether the TO M40 genotype is indicative of a higher probability of developing or worsening dementia may help the clinician to decide or advise on appropriate counselling or advice to the subject and/or family. More aggressive treatment, treatment directed to disease modification and/or to dementia and/or more frequent monitoring may be chosen if a subject has a TOMM40 genotype indicative of a higher probability of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD), for example of developing or worsening dementia, particularly if motor Parkinsonisian features or early signs of dementia are present.

Thus, for example, if the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs, then the selected treatment regime may comprise treating the subject with, for example, one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non- motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination, treatments for insulin resistance or anti-inflammatory compounds, for example as referred to in lclat Aviles- Olmos et al, Brain 2012 (Feb 17) Parkinson's disease, insulin resistance and novel agents of neuroprotection or Tansey and Goldberg, Neurobiol.Dis 2010 Mar; 37(3): 510- 8. Epub 2009 Nov 10 Neuroinflammation in Parkinson's disease: its role in neuronal death and implications for therapeutic intervention. Compounds mentioned at http://alzheimers.orq.uk site/scripts/documents info.php?documentlD=106 (for example acetylcholine esterase inhibitors Aricept (donepezil hydrochloride), Exelon (rivastigmine), Reminyl (galantamine); memantine, trade name Ebixa) may also be potentially useful for such subjects.

If the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is not heterozygous, optionally wherein there is no or less than 15, 14, 13, 12, 11 , 10, 9, 8, 7, 6, 5, 4, 3 or 2 base pair difference in the length of a poly-T stretch at rs 10524523 between the subject's alleles or wherein the subject's genotype at rs10524523 is not S VL, then the selected treatment regime may not require any additional components (for example over compounds that may be required more generally for treating PD/motor Parkinsonisian features).

The method may further comprise the step of performing tests to exclude other causes of dementia, such as carrying out a blood test for vitamin 12, as will be well known to those skilled in the art, or for other treatable metabolic causes. , Magnetic Resonance Imaging and metabolic imaging may also be performed.

A further aspect of the invention provides one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non- motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop- ), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds referred to in lclat Aviles-Olmos et al, Brain 2012 or Tansey and Goldberg, Neurobiol.Dis 2010 or compounds mentioned at http://alzheimers.orq.uk site/scripts/documents info.php?documentlD=106 for use in treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype), wherein the subject is a subject for whom it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs 0524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is SA/L or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs.

A further aspect of the invention provides the use of one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non-motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B- AO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds referred to in lclat Aviles-Olmos et al, Brain 2012 or Tansey and Goldberg, Neurobiol.Dis 2010 or compounds mentioned at http://alzheimers.org.uk/site/scripts/documents info.php?documentlD=106 in the manufacture of a medicament for treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype),, wherein the subject is a subject for whom it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is SA L or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs.

A further aspect of the invention provides a method for treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype), the method comprising administering one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non-motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds referred to in lclat Aviles-Olmos et al, Brain 2012 or Tansey and Goldberg, Neurobiol.Dis 2010 or compounds mentioned at http://alzheimers.orq.uk/site/scripts/documents info.php?documentlD=106 to the subject, wherein the subject is a subject for whom it has been determined that the TO M40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs.

The compound may alternatively (or in addition) be a modulator, for example inhibitor or activator of TOMM40, for example an antibody (including antibody fragment, as well known to those skilled in the art) or an RNAi molecule, directed at TOMM40 polypeptide or gene, as appropriate. The compound may alternatively (or in addition) be a mitochondrial modulator or an anti-oxidant, such as Idebenone, a potent antioxidant and inhibitor of lipid peroxidation, interacting with the mitochondrial electron transport chain and facilitating mitochondrial electron flux in by-passing complex I (Haefeli et al., 2011 ) or EPI-743 (Sadun et al, Arch Neurol. 2012 Mar;69(3):331-8). The compound may be a PPARy modulator, for example PPARy inhibitor or activator, for example piaglitazone, as discussed in, for example, Aviles-Olmos et al (2012), supra.

The subject may be administered an additional anti Parkinson's Disease or motor Parkinsonisian features treatment, for example selected from L Dopa with a peripheral dopa decarboxylase inhibitor, dopamine agonists, elective type B monoamine oxidase inhibitors, Amantadine, tolcapone and other agents.

A further aspect of the invention provides a kit of parts useful for assessing a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features, comprising (1) an agent which is specifically capable of use in determining a subject's genotype at TOM 40, optionally intron 6, optionally rs 0524523.

Thus the kit may contain PCR primers suitable for use in determining TOMM40 genotype, for example at position rs 0524523. Examples of such primers include forward primer 5' -TGCTGACCTCAAGCTGTCCTC-3' and reverse primer 5'- GAGGCTGAGAAGGGAGGATT-3';

or Outer: 5'-CTGGGCTCAAATGAACC-3' and 5 -CAAATGTG ATTTTATAG G G CCA-3 ' Inner: 5 -GAGATGGGGTCTCACT-3' and 5'-ACAGGGAAAGAAAACAAGCCTG-3'; or Outer: 5'-TGGCCTCCCAAACTG-3' and 5'-GGCGTGGTGGCAC-3'

Inner: 5'-GTTAGATGAAGTTTTTAAATTTTTTGTAG-3' and 5'-

GAG AAG GG AG G ATTGCT-3' . These "outer" and "inner" primers may be used on their own to amplify the area of interest; or they may be used together in a "nested primer" approach, as well know to those skilled in the art, and as described in Example 1.

Such primers may be used for PCR amplification, typically followed by sequencing or separation of alleles by gel electrophoresis, for example capillary gel electrophoresis. Suitable sequence methods, including suitable internal controls and verification procedures, for example involving two independent PCR amplifications for each sample, are described in Example 1. The kits may usefully further comprise a component for testing for a further PD/motor Parkinsonisian features related parameter, for example genes associated with familial PD (as noted above), or MAPT, gene or APOE.

The kits usefully may contain controls and detection material, for example an internal control sequence, such as the sequence corresponding to a T8 poly-T sequence at position rs10524523, as described in Example 1 , "Addition of an Internal Standard" section.

A further aspect of the invention provides an agent which is specifically capable of use in determining a subject's genotype at TOM 40, optionally intron 6, optionally rs10524523, for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features. The subject may be a subject considered to have at least early motor Parkinsonisian features, as discussed above. A further aspect of the invention provides the use of an agent which is specifically capable of use in determining a subject's genotype at TO M40, optionally intron 6, optionally rs10524523, in the manufacture of a medicament for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features. The subject may be a subject considered to have at least early motor Parkinsonisian features, as discussed above. The agent may typically be a nucleic acid which selectively hybridises to TOM 40 nucleic acid (typically genomic DNA), optionally in or flanking intron 6, optionally in or flanking rs 10524523 (for example within 100 bases either side of intron 6, or rs10524523). Thus, for example, the agent may be one or more of the PCR primers identified above or in Example 1.

The term "selectively hybridising" will be well known to those skilled in the art, and includes the meaning that the nucleic acid has sufficient nucleotide sequence similarity with the said human nucleic acid that it can hybridise under moderately or highly stringent conditions. As is well known in the art, the stringency of nucleic acid hybridization depends on factors such as length of nucleic acid over which hybridisation occurs, degree of identity of the hybridizing sequences and on factors such as temperature, ionic strength and CG or AT content of the sequence. Thus, any nucleic acid which is capable of selectively hybridising as said is useful in the practice of the invention.

TOMM40 sequences are well known, as indicated above and in Example 1.

Nucleic acids which can selectively hybridise to the said human nucleic acid include nucleic acids which have >95% sequence identity, preferably those with >98%, more preferably those with >99% sequence identity, over at least a portion of the nucleic acid with the said human nucleic acid. As is well known, human genes usually contain introns such that, for example, a mRNA or cDNA derived from a gene would not match perfectly along its entire length with the said human genomic DNA but would nevertheless be a nucleic acid capable of selectively hybridising to the said human DNA, other than intronic sequences. Nucleic acids which span the intron-exon boundaries of the said TOMM40 gene may not be able to selectively hybridise to the said TOMM40 mRNA or cDNA but may be useful in the present invention for analysing intronic sequences. Typical moderately or highly stringent hybridisation conditions which lead to selective hybridisation are known in the art, for example those described in Molecular Cloning, a laboratory manual, 2nd edition, Sambrook et al (eds), Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA, incorporated herein by reference.

An example of a typical hybridisation solution when a nucleic acid is immobilised on a nylon membrane and the probe nucleic acid is ~ 500 bases or base pairs is:

6 x SSC (saline Na⁺ citrate)

0.5% Na⁺ dodecyl sulphate (SDS)

100 :g/ml denatured, fragmented salmon sperm DNA

The hybridisation is performed at 68 °C. The nylon membrane, with the nucleic acid immobilised, may be washed at 68°C in 1 x SSC or, for high stringency, 0.1 x SSC.

20 x SSC may be prepared in the following way. Dissolve 175.3 g of NaCI and 88.2 g of Na⁺ citrate in 800 ml of H₂0. Adjust the pH to 7.0 with a few drops of a 10 N solution of NaOH. Adjust the volume to 1 litre with H₂0. Dispense into aliquots. Sterilize by autoclaving.

An example of a typical hybridisation solution when a nucleic acid is immobilised on a nylon membrane and the probe is an oligonucleotide of between 15 and 50 bases is:

3.0 M trimethylammonium chloride (TMACI)

0.01 M Na⁺ phosphate (pH 6.8)

1 mm EDTA (pH 7.6)

0.5% SDS

100 :g/ml denatured, fragmented salmon sperm DNA

0.1 % nonfat dried milk

The optimal temperature for hybridization is usually chosen to be 5 °C below the T, for the given chain length. T, is the irreversible melting temperature of the hybrid formed between the probe and its target sequence. Jacobs et al (1988) Nucl. Acids Res. 16, 4637 discusses the determination of TjS. The recommended hybridization temperature for 17-mers in 3 M TMACI is 48-50 °C; for 19-mers, it is 55-57 °C; and for 20-mers, it is 58-66 °C.

By "nucleic acid which selectively hybridises" is also included nucleic acids which will amplify DNA from the TOMM40 nucleic acid, typically genomic DNA, by any of the well known amplification systems such as those mentioned herein, in particular the polymerase chain reaction (PCR).

Although the nucleic acid which is useful in the methods of the invention may be RNA or DNA, DNA is preferred. Although the nucleic acid which is useful in the methods of the invention may be double-stranded or single-stranded, single-stranded nucleic acid is preferred under some circumstances such as in nucleic acid amplification reactions.

The nucleic acid which is useful in the methods of the invention may be any suitable size. However, for certain diagnostic, probing or amplifying purposes, it is preferred if the nucleic acid has fewer than 10 000, more preferably fewer than 000, more preferably still from 10 to 100, and in further preference from 15 to 30 base pairs (if the nucleic acid is double-stranded) or bases (if the nucleic acid is single stranded). As is described more fully below, single-stranded DNA primers, suitable for use in a polymerase chain reaction, are particularly preferred.

The nucleic acid for use in the methods of the invention is a nucleic acid capable of hybridising to the TOMM40 genomic DNA. Fragments of the said TOMM40 gene and are also preferred nucleic acids for use in the methods of the invention.

It is particularly preferred if the nucleic acid for use in the methods of the invention is an oligonucleotide primer which can be used to amplify a portion of the said TOMM40 nucleic acid, particularly TOMM40 genomic DNA. Preferred nucleic acids for use in the invention are those that selectively hybridise to the TO M40 genomic DNA and do not hybridise to other related genomic DNA. Such selectively hybridising nucleic acids can be readily obtained, for example, by reference to whether or not they hybridise (or are predicted to hybridise, using well known calculations) to the said TOMM40 genomic DNA and not to other genomic sequences. Conveniently, in some embodiments, the nucleic acid capable of hybridising to the TOMM40 genomic DNA and which is used in the methods of the invention further comprises a detectable label, as discussed above. A further aspect of the invention provides a method for selecting a test Parkinson's Disease or motor Parkinsonisian features or otherwise at risk thereof subject in whom to assess a therapeutic strategy or treatment regime for the treatment of Parkinson's Disease or motor Parkinsonisian features, which may include a therapeutic strategy or treatment regime intended for treating (for example preventing, delaying or reducing) dementia either in the early stages, i.e. within a year (DLB) or late stages (PDD) of the disease, the method comprising the step of assessing the test subject's genotype for TOMM40, optionally in intron 6 (IVS6) of the TOMM40 gene, optionally at position rs10524523. The method may comprise the step of determining whether the test subject's genotype for TOMM40, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally determining whether there is a difference in the length of a poly- T stretch at rs10524523 between the subject's alleles, optionally whether the subject's genotype at rs10524523 is S/VL or whether the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs. As above, the subject typically is a human subject. The subject may be a subject considered to have at least early motor Parkinsonisian features, as discussed above

The test subject may be selected if the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs.

It will be appreciated that the therapeutic strategy or treatment regime may be assessed both in subjects selected as indicated above; and in subjects selected as not meeting that criterion; and the results obtained in the two subject groups compared.

A further aspect of the invention provides a method for assessing a therapeutic strategy or treatment regime for the treatment of Parkinson's Disease or motor Parkinsonisian features (which may include a therapeutic strategy or treatment regime intended for treating (for example preventing, delaying or reducing) PDD or DLB), or at risk thereof (other than through determination of TOMM40 genotype), the method comprising the step of evaluating the results of the therapeutic strategy or treatment regime from a subject or subjects who have been selected according to the method of the preceding aspect of the invention.

A further aspect of the invention provides an in vitro method for assessing the suitability of a test compound for the treatment of a subject with or at risk of Parkinson's Disease, motor Parkinsonisian features, PDD or DLB, the method comprising the step of determining the effect of the test compound on a cell for which it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the cell's alleles, optionally wherein the cell's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19, 20 or more base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 or more base pairs.

An assay may also be in an animal model of Parkinson's Disease, for example a transgenic, neurotoxin or virus-based animal model of PD, as known to those skilled in the art.

Thus, a further aspect of the invention provides a method for assessing the suitability of a test compound for the treatment of a subject with or at risk of Parkinson's Disease, motor Parkinsonisian features, PDD or DLB, the method comprising the step of determining the effect of the test compound on a cell in a non-human animal for which it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the animal's alleles, optionally wherein the animal's genotype at rs10524523 is S VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 9 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, for example more than or equal to 15 base pairs. A further aspect of the invention provides a screening method for aiding in identifying a compound likely to be useful in treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof, the method comprising the step of determining the effect of a test compound on TO M40 nucleic acid, protein or activity level; and selecting a compound that modulates, for example reduces or increases, said level. This screening method may be performed on a test compound together with (either before or after) the method of the preceding aspect of the invention.

The effect of the test compound may be determined in vitro.

The effect of the test compound may be determined in vivo in a non-human test animal.

The invention will now be described in more detail by reference to the following, non- limiting, Figures and Examples.

Fig 1. A: Distribution of the TOMM40 rs10524523 allele length, and their clustering in three groups, following Lutz et al. 2010¹⁶. B: Kaplan Meyer survival curve according to TOMM40 rs10524523 genotype; the legend presents the number of individuals observed for each genotype, and the significance level of each genotype according to the model presented in Table 1 (ns = not significant, ^**, p-value < 0.001). Fig 2. A: Distribution of TOMM40 rs10524523 absolute allele length differences over the three clinical phenotypes. The dashed line represents a difference of 15, the minimum difference observed in the S VL genotype, and not observed in any other genotype. The bold black line in each plot represents the median of the distribution of the allele length difference. B: Distribution of the S/VL haplotype over the three clinical phenotypes. DLB S/VL - other = 11 - 4, PDD S VL - other = 21 - 35 and Parkinson's disease with no dementia (PDnD) S/VL - other = 12 - 37.

Figure 3: analysis of how Positive Predictive Value (PPV) and Negative Predictive Value (NPV) change across the allele lengths difference (ALD) spectrum. In the plot the calculated values are represented by the solid red line, and the black dashed lines represent the 95% confidence intervals, calculated through resampling. The vertical blue dashed line indicates the ALD value of 15, which was chosen as the threshold value for the confusion matrix in table 2.

Example 1 :

Heterozygosity of TOMM40 poly-T repeat length is strongly predictive of risk of early-onset dementia in Parkinson patients Two forms of dementia are often associated with Parkinson's disease (PD), known as "Dementia with Lewy bodies" (DLB) and "Parkinson's disease dementia" (PDD). The prevalence of these progressively debilitating neurodegenerative diseases of old age is increasing in parallel with the exponential increase of life expectancy. There is, still currently, no available treatment that is considered to prevent or modify the disease course. We present results from the analysis of genes MAPT, APOE and TOMM40 in a sample of neuropathologicaliy confirmed PD cases with and without dementia from the Parkinson's UK Tissue Bank. This is the largest genetic study with autopsy confirmed diagnosis of PD, providing an unbiased population sample of dementia distribution in PD patients. We observed that the SA L genotype of intron 6 (IVS6) of the Translocase Outer Mitochondrial Membrane homologue 40 (TOMM40) gene was strongly associated with dementia, especially with the DLB dementia form, which typically occurs within a year of onset of motor symptoms. We also demonstrate that the TOMM40 IVS6 SA L genotype has a strong predictive value as a biomarker for early dementia in PD.

We tested for association between known APOE, TOMM40 and MAPT alleles and dementia in 120 selected cases from the Parkinson's UK Tissue Bank. Selection was based on neuropathologicaliy confirmed diagnosis of PD and the availability of reliable clinical evidence of presence or absence of significant cognitive decline and of features such as visual hallucinations and fluctuating confusion⁸. The neuropathological and demographic characteristics of the subjects are given in Table 3.

We characterized the H1/H2 MAPT haplotypes based on four SNPs (rs242557, rs3785883, rs247 738 and rs9468), and differentiated APOE ε2/ε3/ε4 genotypes based on two SNPs (rs429358 and rs7412). We found 16 allelic lengths of the multivariate fragment length of the TOMM40 IVS6 poly- T (rs10524523) and applied the previously reported length classification in three groups; short (S), long (L) and very long (VL)²⁵. Histograms of fragment lengths (fig 1A) strongly suggest that three distinct distributions with specific modes are present and the subdivision of fragment lengths into three groups reflects this observation. Whether analyzed as raw poly- T allelic lengths or in the three categorical groups, the IVS6 poly-T repeats are in Hardy-Weinberg equilibrium (p- value of 1 for the raw data after 10,000,000 replicates, and p-value of 0.68 after 10000 replicates). The results from a Cox proportional hazards model (Table 4A) showed that the S/VL TOMM40 rs10524523 genotype was strongly associated with dementia, with an odds ratio (OR) of 3.83 (p value 0.002). We did not observe any other association with dementia of any other TOMM40 genotype, nor for any genotype of either APOE or MAPT. Using a recently proposed methodology based on Schoenfeld residuals, we estimated that polymorphism in TOMM40 repeat length contributes 11% in the observed variation for dementia occurrence in our population sample ²⁷. Because the S/VL genotype is the most extreme heterozygote in terms of allele length differences, we repeated the same Cox analysis substituting the rs10524523 genotype with the allelic length difference (ALD) and mean pair length (Table 4B). The results showed that the only significant effect on dementia was that of ALD, with a multiplicative increased risk of 1.07 for each point of increase in the difference of the two alleles. Furthermore, the DLB phenotype was strongly over-represented in individuals carrying the S/VL genotype (Fisher exact test p-value 0.003) (Fig 2A).

In view of the strong association between the TOMM40 ISV6 Poly- T S/VL genotype and DLB, we then tested for the predictive value of this genotype with respect to DLB. Table 5 summarizes the confusion matrix obtained testing for presence of DLB and TOMM40 S VL, showing a sensitivity of 73%, with a corresponding specificity of 69%. Our sample yielded a prevalence of DLB of 0.125. Using this value, the positive and negative predictive values of a TOMM40 ISV6 poly-T S/VL genotype are 25% and 95%, respectively, suggesting that a patient in the very early stages of PD found to harbor this genotype would carry a risk of 25% of developing DLB, whereas a patient found not to carry this genotype would have a 95% chance of not developing early dementia.

We did not observe any significant association between either MAPT or APOE with dementia. There was a positive signal of the APOE ε4/ε4 genotype only when APOE was analyzed alone (supplemental information, Table 9). This effect, however, disappeared in all other models where the three genes were entered together and the only significant signal was of the S/VL genotype of TOMM40 IVS6.

Our analysis reveals a striking effect of the TOM 40 ISV6 poly-T (rs10524523) polymorphism on the latency of the onset of dementia from the onset of motor symptoms, S/VL being over represented in DLB patients. It has recently been demonstrated that the IVS6 poly-T is a putative regulatory region of the TOMM40 promoter, conferring an enhancer effect and acting independently of the closely linked APOE gene²⁷. These results confirm the "non-APOE effect" role of this and (potentially other) TOMM40 variants on dementia risk. They also suggest that non-APOE ε4 cis- regulatory variants may contribute to a promoter-enhancer haplotype structure that influences the regulation of both the TOMM40 and APOE genes. Thus, the genetic association found in multiple recent studies between TO M40 variants and Alzheimer's Disease^25,26 may be due to a dementia- modifier effect by TOMM40 that is haplotype and cell type specific.

Because selection of cases was based on neuropathologica! confirmation of PD independently of presence or absence of dementia, ours is an unbiased population sample of the distribution of both dementia and the TOMM40 IVS6 poly-T alleles in the PD population; hence our attempt to test rs10524523 as a predictive marker for DLB, using a confusion matrix. Based on our findings, rs10524523 may have clinical utility in predicting risk estimates of early dementia in the very early stages of Parkinsonism.

In conclusion, here we present the first evidence that TOM 40 is both strongly associated with and predictive of DLB in a population sample of PD patients. Our findings represent an important novel insight with potential implications both in the clinical setting suggesting that the TOMM40 IVS6 polyT may be a powerful predictive biomarker of early dementia in PD and, by pointing towards a possible pathogenic role of perturbations of mitochondrial function and dynamics, for drug discovery of novel disease modifying therapeutics.

Methods Summary

The complete methods are provided in the Supplementary Methods and Data, which include clinical, neuropathological and genotypic data, data quality controls, and detailed statistical methods. In brief, from an original sample of 268 cases from the Parkinson's UK Tissue Bank, complete genotyping data were obtained in 218 cases. For each sample the clinical and neuropathological records were reviewed and 120 cases were selected based on neuropathological diagnosis of PD and reliable clinical information of presence or absence of dementia, with the characteristic features of visual hallucinations and fluctuating confusion. A Cox proportional hazard model was used to analyze the effects of genotype, adjusting for age of onset of motor disease, vascular risk score and stratifying the model for gender. McKeith scores using semi quantitative grading of LB density in five brain regions according to the DLB Consortium criteria³, together with the TOMM40 genotype, age of onset of motor symptoms, and age of onset of dementia were used in dementia patients only to ascertain whether DLB and PDD, show a distinctive clustering pattern. Finally, the positive and negative predictive values of the TOMM40 S/VL genotype were calculated from a confusion matrix.

Table 1 : A: Cox proportional hazards model stratified by gender, including TOMM40 rs10524523 (baseline risk S/S), APOE (baseline risk ε3/ε3), MAPT (baseline risk H2/H2), age of onset and Vascular Risk Score as predictors. B: Cox model including TOMM40 rs10524523 allele length difference, TOMM40 rs10524523 mean pair length, APOE (baseline risk ε3/ε3), MAPT (baseline risk H2/H2), age of onset and Vascular Risk Score as predictors

Model A: model stratified by Sex

OR Lower 0.95 CI Upper .095 CI P-value

UL 1.79 0.21 15.50 0.598

L/VL 2.51 0.71 8.94 0.155

S/L 1.48 0.43 5.09 0.531

S/VL 3.83 1.66 8.87 0.002

VL/VL 1.45 0.54 3.87 0.461 ε2/ε3 0.44 0.18 1.07 0.071 ε2/ε4 2.20 0.35 13.91 0.403 ε3/ε4 0.94 0.40 2.17 0.876 ε4/ε4 2.51 0.64 9.82 0.186

Η1 Η1 1.33 0.30 5.95 0.707

Η1 Η2 1.16 0.25 5.28 0.852

Age onset PD 0.98 0.79 1.22 0.883

Vascular Risk Score 0.97 0.93 1.01 0.131

Model B: model stratified by Sex

Allele length

difference 1.07 1.03 1.11 0.0003 ***

Mean Pair Length 1.02 0.97 1.07 0.4044 ε2/ε3 0.46 0.19 1.12 0.0871 ε2/ε4 1.46 0.29 7.24 0.6446 ε3/ε4 0.98 0.53 1.80 0.9412 ε4/ε4 2.74 0.99 7.53 0.0512

Η1 Η1 1.33 0.30 5.89 0.7106

Η1 Η2 1.20 0.26 5.49 0.8108

Age onset PD 1.00 0.80 1.23 0.9679

Vascular Risk Score 0.97 0.93 1.01 0.1137

Table 2: Confusion matrix for TOMM40 allele length difference, with the resulting sensitivity, specificity, positive predictive value and negative predictive value for a 15 bases allele length difference threshold (see supplementary material, picture 1).

References

1 Ferri, CP. et al. Global prevalence of dementia: a Delphi consensus study. Lancet 2005 Dec 17; 366 (9503): 2112-2117

2 Comas-Herrera, A., Wittenberg, R., Pickard, I., Knapp, M. & MRC-CFAS. Cognitive impairment in older people: its implications for future demand for services and costs. PSSRU Discussion Paper 2005; 1728.

3 McKeith, I.G. et al. Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy Bodies (DLB): report of the consortium on DLB international workshop. Neurology 1996; 47: 1113-1 124.

4 Gibb, W.R.G. & Lees, A.J. The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson's disease. J Neurol Neurosurg Psychiatry 1988; 51 : 745-752.

5 Spillantini, M.G. et al. Alpha-synuclein in Lewy bodies. Nature 1997; 388: 839-840. 6 McKeith, I.G. et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology 2005; 65:1863-1872.

7 Emre, M. et al. Clinical diagnostic criteria for dementia associated with Parkinson's disease. Mov Disord 2007; 22: 1689-1707.

8 Lippa CF, et al. DLB and PDD boundary issues: diagnosis, treatment, molecular pathology, and biomarkers. Neurology. 2007 Mar 13;68(11):812-9.

9 Aarsland, D. & Kurz, M.W. The epidemiology of dementia associated with Parkinson's disease. Brain Path. 2010; 20(30): 633 - 639.

10 Zaccai, J., McCracken, C. & Brayne, C. A systematic review of prevalence and incidence studies of dementia with Lewy bodies. Age Ageing 2005; 34: 561-566. 11 De Lau, L.M.L., Schipper, C.M.A., Hofman, A., Koudstaal, P.J. & Breteler, M.M.B. Prognosis of Parkinson's disease: risk of dementia and mortality: the Rotterdam study. Arch Neurol 2005; 62: 265-1269. 12 Aarsland, D. et al. Risk of dementia in Parkinson's disease: a community-based, prospective study. Neurology 2001 ;56:730-736.

13 Seto-Salvia, N., et al. Dementia risk in Parkinson disease: disentangling the role of MAPT haplotypes. Arch Neurol 2011 ; 68 (3): 359-364.

14 Goris, A. et al. Tau and alpha-synuclein in susceptibility to, and dementia in, Parkinson's disease. Ann Neurol 2007; 62: 145-53.

15 Williams-Gray, C.H. et al. The distinct cognitive syndromes of Parkinson's disease: 5 year follow-up of the Campaign cohort. Brain 2009; 132: 2958-2969. (B)

16 Papapetropoulos, S. et al. Phenotypic associations of tau and ApoE in Parkinson's disease. Neurosci Lett. 2007 Mar 6;414(2):141-4. Epub 2007 Jan 3.

17 Huang, X., Chen, P., Kaufer, D.I., Troster, A.I. & Poole, C. Apolipoprotein E and dementia in Parkinson disease: a meta-analysis. Arch Neurol 2006; 63: 189-193.

18 Pankratz, N. et al. Presence of an APOE4 allele results in significantly earlier onset of Parkinson's disease and a higher risk with dementia. Mov Dis 2006; 21 : 45-49.

19 Lamb, H. et al. Apolipoprotein E and alpha-1 antichymotrypsin polymorphism genotyping in Alzheimer's disease and in dementia with Lewy bodies. Distinctions between diseases. Neurology 1998; 50: 388-391.

20 Meeus, B. et al. DLB and PDD: a role for mutations in dementia and Parkinson's disease genes? Neurobiol. Aging 2012; 33(3):629.e5-629.e18

21 Kobayashi, S. et al. Apolipoprotein E4 frequencies in a Japanese population with Alzheimer's disease and dementia with Lewy bodies. PLoS One. 2011 Apr

28;6(4):e18569.

22 Jasinska-Myga, B. et al. Apolipoprotein E gene polymorphism, total plasma cholesterol level, and Parkinson disease dementia. Arch Neurol. 2007 Feb;64(2):261-5.

23 Ezquerra. M. et al. Lack of association of APOE and tau polymorphisms with dementia in Parkinson's disease. Neurosci Lett 2008; 448 (1): 20-23.

24 Williams-Gray, C.H. et al. Apoprotein E genotype as a risk factor for susceptibility to and dementia in Parkinson's disease. J Neurol 2009; 256: 493-498. (A)

25 Lutz, M.W., Crenshaw, D.G., Saunders, A.M. & Roses, A.D. Genetic variation at a single locus and age of onset for Alzheimer's disease. Alzheimers Dement. 2010; 6(2): 125-31.

26 Roses, A.D. et al. A TOMM40 variable-length polymorphism predicts the age of late- onset Alzheimer's disease. Pharmacogenomics J. 2010; 10(5):375-84. 27 Bekris, L.M., Lutz, F. & Yu, C.E. Functional analysis of APOE locus genetic variation implicates regional enhancers in the regulation of both TOMM40 and APOE. J. Hum. Genetics 2011 ; 57(10): 1-8.

5 Supplementary materials.

Table 3: Demographic, clinical and neuropathological data of Parkinson's disease patients classified according to presence/absence of dementia and dementia type. DLB = dementia with Lewy bodies; PDD = Parkinson's disease with Dementia; PDnD = 10 Parkinson's disease with no dementia.

DLB (n = 15) PDD (n = 56) PDnD (n = 49)

Females 3 (20%) 17 (30%) 15 (31%)

Age at onset of motor symptoms

70.93 ± 7.12 62.66 ± 8.93 66.47 ± 10.45 (years)

Age at death (years) 75.47 ± 6.48 77.62 ± 6.59 78.53 ± 8.19

Disease duration (years) 4.53 ± 1.96 14.96 ± 6.26 12.06 ± 6.72

Age at onset of dementia (years) 71.33 ± 6.91 74.55 ± 6.74 -

Time to dementia (years) 0.40 ± 0.51 11.89 ± 5.63 -

Dementia duration (years) 4.13 ± 1.85 3.07 ± 2.28 -

Hallucinations reported 14 (93%) 44 (80%) 9 (18%)

Mean Dysautonomia score 1.80 ± 1.01 2.57 ± 1.01 1.87 ± 0.87

RBD reported 10 (67%) 16 (29%) 5 (10%)

Cognitive Fluctuations 5 (33%) 12 (21%) 0 (0%)

aSN Braak Stage 6 13 53 40

aSN Braak Stage 5 1 3 6

aSN Braak Stage 4 0 0 1

aSN Braak Stage 3 0 0 2

Tau Braak Stage C 2 1 1

Tau Braak Stage B 3 6 4

Tau Braak Stage A 10 49 44

McKeith SFC 2.40 ± 1.18 1.86 ± 1.03 1.10 ± 0.71

McKeith AC 2.79 ± 0.58 2.60 ± 0.84 2.00 ± 0.89

McKeith TC 2.54 ± 0.78 1.61 ± 1.08 0.75 ± 0.63

McKeith PC 1.91 ± 1.14 1.06 ± 0.91 0.44 ± 0.60

McKeith Entorhinal 2.43 ± 0.85 2.36 ± 1.08 1.52 ± 1.11 Αβ Striatal (pos/neg) 5/1 10/9 6/15

Λβ Frontal (pos/neg) 14/1 39/17 26/23

Αβ Occipital (pos/neg) 8/2 16/8 13/4

Αβ Hippo-entorhinal (pos/neg) 13/2 38/17 24/25

aSN: Alpha-synuclein; AC: Anterior Cingulate; PC: Parietal Cortex; RBD: REM-sleep Behaviour Disorder; SFC: Superior Frontal Cortex; TC: Temporal Cortex.

There was significant difference between DLB and PDD patients in the age at onset of 5 motor symptoms (p = 0.003). There was no significant difference in age at death between the three groups, while the overall disease duration was significantly shorter in DLB compared to PDnD and PDD (p = 5.2 e-7) and, also, between PDD and PDnD (p = 0.02). The dysautonomia score was not significantly different between patients with and without dementia (p = 0.1 ), but it was significantly different between DLB and PDD (p = 10 0.007). DLB and PDD did not differ in the age at onset of dementia (p = 0.11) and for dementia duration (p = 0.1) but were significantly different for time to dementia from motor onset, as expected (p = 3.59 e-11).

The fisher test for sex distribution across the three clinical classes produced a non significant p-value (0.75), while fisher test for reported hallucinations, RBD and cognitive

15 fluctuations were all strongly significant (p = 2.6 e-11 for hallucinations, p = 8.2 e-5 for RBD and 5.3 e-5 for cognitive fluctuations), indicating a different distribution of these symptoms across the three groups. There was a difference between DLB and PDD in reported RBD (p = 0.01), but not for hallucinations or cognitive fluctuations.

The fisher test for the distribution of the aSN Braak staging showed a non significant p-

20 value of 0.33, indicating that the three clinical groups did not vary for a-synuclein topographical distribution, at time of autopsy. Similarly, the fisher test for the distribution of the BNE tau Braak staging gave a non significant p-value of 0.12, again indicating that the three clinical groups did not differ for topographical distribution of tau deposition, at time of autopsy.

25 For presence or absence of Αβ, and correction for multiple testing, we did not observe any significant difference for striatal, occipital cortex and hippocampal/ entorhinal cortex (p-values, respectively of 0.05 - striatal, 0.77 - occipital, 0.01 hippo-entorhinal) across the three clinical groups, while the p-value for frontal cortex was significantly different between the groups (p-value = 0.009), indicating that the proportion of positives is much

30 greater in patients with dementia, more so in DLB patients. After correcting for multiple testing, all cKeith scores were significantly higher in patients with dementia vs PDnD patients (all comparisons' p-values « 0.001), but only McKeith scores for temporal and parietal cortex differed when comparing DLB vs PDD (McKeith TC, p-value 0.001 ; McKeith PC, p-value 0.003).

5

Table 4: Age of onset of dementia

Beta SE t-value p-value

L/L 1.15 5.04 0.23 0.82

IJVL -0.47 3.02 -0.16 0.88

S/L 2.02 3.07 0.66 0.51

S/VL -2.39 2.07 -1.15 0.25

VL VL 0.91 2.33 0.39 0.70

ε2/ε3 2.22 2.10 1.06 0.29

ε2/ε4 -8.47 3.46 -2.45 0.02

ε3/ε4 -3.22 2.02 -1.60 0.12

ε4/ε4 -2.27 3.39 -0.67 0.51

Η1 Η1 -0.14 3.68 -0.04 0.97

Η1 Η2 -0.54 3.80 -0.14 0.89

Age Onset Motor 0.48 0.07 7.31 9.81 e-10

Vascular Risk Score 0.46 0.48 0.95 0.34

Sex (Male) -3.66 1.38 -2.65 0.01

10

Table 5: latency of dementia

Beta SE t-value p-value

L/L 1.15 5.04 0.23 0.82

UVL -0.47 3.02 -0.16 0.88

S/L 2.02 3.07 0.66 0.51

S/VL -2.39 2.07 -1.15 0.25

VL/VL 0.91 2.33 0.39 0.70

ε2/ε3 2.22 2.10 1.06 0.29

ε2/ε4 -8.47 3.46 -2.45 0.02

ε3/ε4 -3.22 2.02 -1.60 0.12

ε4/ε4 -2.27 3.39 -0.67 0.51 H1 H1 -0.14 3.68 -0.04 0.97

H1 H2 -0.54 3.80 -0.14 0.89

Age Onset Motor -0.52 0.07 -7.96 _***

7.94e-1 1

Vascular Risk Score 0.46 0.48 0.95 0.34

Sex (male) -3.66 1.38 -2.65 0.01 _*

The most significant factor in both Age of Onset of Dementia and Latency of Dementia was the age of onset of motor symptoms, with both the APOE genotype ε4/ε4 and Male Sex conferring a marginally significant earlier onset of dementia

5 Table 6: Cox model stratified by sex, including TOMM40 rs10524523 (baseline risk S/S), APOE (baseline risk ε3/ε3), MAPT (baseline risk H2/H2), Age of Onset and Vascular Risk Score as predictors. - PDnD and PDD only, excluding DLB

OR Lower 0.95 CI Upper .095 CI P-value

L/L 1.38 0.14 13.32 0.78

UVL 2.03 0.45 9.12 0.35

S/L 1.61 0.41 6.30 0.49

S VL 2.99 1.18 7.58 0.02

VL/VL 1.80 0.63 5.15 0.27

ε2/ε3 0.64 0.26 1.59 0.33

ε2/ε4 0.00 0.00 Inf 1.00

ε3/ε4 0.91 0.33 2.52 0.86

ε4/ε4 3.93 0.92 16.75 0.06

Η1 Η1 1.13 0.25 5.22 0.87

Η1 Η2 0.95 0.20 4.51 0.95

Age onset PD 0.92 0.88 0.97 0.00

Vascular Risk Score 1.05 0.82 1.35 0.69

Table 7. Cox model including TOMM40 '523 allele length difference, TOMM40 10 rs 0524523 mean pair length, APOE (baseline risk ε3/ε3), MAPT (baseline risk H2/H2), Age of Onset and Vascular Risk Score as predictors. - PDnD and PDD only, excluding DLB

OR Lower 0.95 CI Upper .095 CI P-value

Allele length difference 1.05 1.01 ΪΤ 0 0.01 10 ^*

Mean Pair Length 1.03 0.97 1.08 0.3197

ε2/ε3 0.64 0.26 1.60 0.3418 ε2/ε4 0.00 0.00 Inf 0.9961

ε3/ε4 0.93 0.46 1.86 0.8278

ε4/ε4 3.71 1.33 10.34 0.0124 ^*

Η1 Η1 1.09 0.24 4.94 0.9135

Η1 Η2 0.90 0.19 4.26 0.8936

Age onset PD 0.92 0.88 0.96 0.0003 ^***

Vascular Risk Score 1.04 0.82 1.33 0.7337

When PDD and PDnD are analysed excluding DLB, the effect of TOMM40 S/VL and TOMM40 ALD was still present. In both cases, there is also a strong effect of age of onset of extrapyramidal motor symptoms. In the analysis using the ALD score, there was 5 a significant effect of the APOE ε4/ε4 as an enhancing risk factor.

Cox model for each gene alone

Table 8: Cox model for TOMM40 '523 allele (baseline risk S/S), Age of Onset and0 Vascular Risk Score as predictors. Model stratified by sex.

OR Lower 0.95 CI Upper .095 CI P-value

L/L 5.41 1.06 27.54 0.042

LVL 2.58 0.89 7.50 0.082

S/L 2.00 0.71 5.64 0.191

S/VL 3.68 1.61 8.40 0.002

VLA L 1.44 0.54 3.81 0.464

Age onset PD 0.98 0.94 1.02 0.342

Vascular Risk Score 0.96 0.78 1.18 0.676

Table 9: Cox model for APOE (baseline risk ε3/ε3), Age of Onset and Vascular Risk5 Score as predictors. Model stratified by sex.

OR Lower 0.95 CI Upper .095 CI P-value ε2/ε3 0.60 0.25 1.42 0.24

ε2/ε4 1.79 0.37 8.67 0.47

ε3/ε4 1.07 0.59 1.92 0.83 ε4/ε4 2.93 1.10 7.81 0.03

Age onset PD 0.98 0.94 1.02 0.23

Vascular Risk Score 1.00 0.83 1.21 0.97

Table 10: Cox model for MAPT (baseline risk H2/H2), age of onset and Vascular Risk Score as predictors. Model stratified by sex.

OR Lower 0.95 CI Upper .095 CI P-value

H1 H1 1.28 0.28 5.82 0.75

H1 H2 1.08 0.23 5.01 0.92

Age onset PD 0.98 0.94 1.02 0.23

Vascular Risk Score 1.01 0.82 1.23 0.94

Methods

Genotyping, clinical and neuropathological assessment were all conducted blinded to one another. Initially, the first 217 cases were included for whom DNA of sufficient quality was obtained. Of 217 genotyped cases, 93 were excluded due to incomplete clinical (n = 38) or neuropathological (n = 16) data or when neuropathology excluded the diagnosis of Parkinson's disease or Dementia with Lewy bodies, according to established criteria (n = 43)¹'². 120 genotyped cases with a neuropathologically confirmed diagnosis of PD and reliable clinical evidence allowing to differentiate between Parkinson's disease without dementia (PDnD, N=49), Parkinson's disease with dementia (PDD, N=56), or dementia with Lewy bodies (DLB, N=15) were selected for analysis.

Clinical assessment

Clinical summaries were extrapolated for each patient by reviewing medical records, which included neurological, psychiatric, and geriatric examinations, GP printouts and notes, formal cognitive testing and neuroimaging results. Clinical records were assessed by four neurologists (SM, PP, C.R., L.M.) and 20 randomly selected cases were subsequently reviewed by two neurologists with expertise in movement disorders (SM and PP), blinded to one another's evaluation, to confirm diagnosis.

Presence of dementia was ascertained based on DSM IV criteria and was defined as a decline from a preceding normal status in more than one cognitive domain causing a significant impairment in activities of daily living³. Cognitive testing was also used when available: scores equal to or below 24/30 on the standard Mini Mental State Examination or equal to or below 7/10 on the Abbreviated Mental Test (AMT) were considered diagnostic of dementia^4,5. Furthermore, we used recently published guidelines for the clinical diagnosis of dementia in Parkinson's disease⁵. Specifically, we evaluated presence of symptoms such as reported persistent and severe visual hallucinations (i.e. not medication related), behavioral changes (apathy, agitation, paranoid delusions), cognitive fluctuations and REM-behavior disorder (RBD), which are known to be associated with dementia in the context of Lewy body disease (Table 3)^7,8. Since the MMSE may not adequately evaluate core features of dementia with Lewy bodies such as cognitive fluctuations, hallucinations and behavioral abnormalities, an MMSE score above the cut-off of 24 but below 27 did not exclude presence of dementia^6,8'. In addition to the central feature of dementia, all cases diagnosed as having Dementia with Lewy Bodies (DLB) had spontaneous extrapyramidal symptoms and at least one other core feature among those listed in the third report of the DLB consortium, ie persistent hallucinations and/or fluctuating cognition. Cases were classified as PDnD when there was no evidence of dementia within one year preceding death. Cases with clinical data insufficient to determine presence or absence of dementia one year prior to death were excluded.

Clinical diagnosis of Parkinson's disease was based on the presence of at least two symptoms among bradykinesia, rigidity, resting tremor and postural instability, with a good response to levodopa as validating evidence⁹. When dementia appeared concomitantly with parkinsonism or within one year of motor onset, the case was defined as DLB, whilst cases were defined as PDD if dementia occurred two or more years after onset of motor symptoms^{6,7, 0}\

Dysautonomia was scored by adding reported dysautonomic symptoms among the following: constipation; bladder impairment including urinary urgency/frequency, nocturia, and retention; orthostatic hypotension; erectile dysfunction; altered perspiration; hypersalivation; dysphagia. Dysphagia and hypersalivation/drooling were classified as dysautonomia if they occurred in the first half of the disease course and were thus clearly independent from global functional deterioration occurring later in the disease course. When more symptoms pertaining to one system were reported, they would count separately only if clearly different (i.e. urinary urgency and urinary frequency = one point; urinary urgency and urinary retention = 2 points).

As an indirect measure of a possible contribution of vascular impairment to the clinical picture, we scored vascular risk by adding one point for each risk factor among the following, when they were reported in medical records: systemic hypertension, ischaemic heart disease, myocardial infarction, arrythmia (mainly atrial fibrillation), unequivocal family history of vascular disease, hyperlipidaemia, smoker status, neuroimaging reports (CT or MRI) of small vessel disease, clinical or neuroimaging reports of cerebrovascular accidents (stroke or transient ischaemic attacks). Neuropathological assessment

All of the 217 genotyped cases underwent neuropathological analysis by three neuropathologists (FR, SG, I.B). Routine staining with Haematoxylin and Eosin and immunohistochemical analysis with antibodies reactive for alpha-synuclein (clone 42, monoclonal, 1 :300, BD, or clone 2 1 , monoclonal 1:300, Santa Cruz), phosphorylated tau (AT8, monoclonal, 1 :800, NBS bio, or AT8, monoclonal, 1 :800, AB), and Αβ peptide (4G8, monoclonal, 1 :800, Signet, or 4G8, monoclonal, 1:400, Abeam) were carried out on the following regions: brainstem, including medulla, pons, and midbrain - with one cranial and one caudal transaxial section of each; cerebellum with dentate nucleus; basal ganglia, including caudate, putamen and pallidum - with coronal slices of the anterior (nucleus accumbens), middle (anterior commissure) and posterior (mammilary bodies) portions; basal forebrain and limbic areas, including amygdala, insular cortex, nucleus basalis of Meynert, anterior cingulate cortex, hippocampal formation and entorhinal cortex; neocortical regions including superior temporal, post-central (parietal), and superior frontal cortex.

Neuropathological diagnosis of Parkinson's disease was based on presence of Lewy bodies associated with neuronal loss in the substantia nigra, in accordance with accepted criteria¹¹. Diagnosis of dementia with Lewy bodies was based on the same neuropathological findings, in association with a clinical history of dementia occurring within one year of onset of parkinsonism².

Staging of alpha-synuclein pathology was both topographical and semi-quantitative ⁷·¹². Topographical distribution of LB and LD was defined according to recently proposed guidelines. Briefly, presence or absence of alpha-synuclein-immunoreactive LBs or LDs was assessed in nine brain regions, with stages 1 to 6 identifying progression of pathology from lower brainstem to neocortical regions^12,13. Semi-quantitative assessment of a-synuclein-immunoreactive Lewy body-type pathology was carried out in the following cortical regions: frontal (N = 120), temporal (N = 91), parietal (N = 85), entorhinal (N = 115) and anterior cingulate (N = 108) based on the recommendations of the DLB consortium, as follows: 1 = mild (sparse Lewy bodies at x 100 magnification); 2 = moderate (1-3 Lewy bodies at x 100 magnification); 3 = severe (4 or more Lewy bodies at x 100 magnification); 4 = very severe (numerous Lewy bodies and dendrites at x 100 magnification)⁷. Alpha-synuclein aggregates were judged as Lewy bodies when spherical aggregates were clearly associated with a neuronal nucleus and cytoplasm. Thus, extracellular Lewy bodies were not considered.

To stage tau pathology, we used guidelines recently published by the BrainNetEurope (BNE) consortium. Briefly, topographical distribution of phosphorylated tau-positive neuropil threads was divided into three categories - A, B, C - corresponding respectively to stages 1-2, 3-4, and 5-6 of the BNE guidelines, which are based on a widely accepted staging scheme proposed by Braak and colleagues^14,15. Thus, stage A corresponds to tau-positive neuropil threads limited to the trans-entorhinal and/or entorhinal regions, stage B to involvement of the occipito-temporal and/or middle temporal gyri, and stage C to spread of neuropil threads to the occipital peristriate and/or striate areas, with a progression of severity of pathology from A to C.

Presence of Αβ deposition was assessed in the superior frontal cortex and in the hippocampal/entorhinal region in all (N = 120) cases and in the basal ganglia in a subset of 48 cases. Plaque type was assessed by defining presence of cored or diffuse plaques or both. Semi-quantitative assessment of lesion density was also performed, with categories of 0 = absent, 1 = mild, 2 = moderate or 3 = severe Αβ deposition, based on CERAD criteria¹⁶.

Genotyping

For all individuals DNA was extracted from 25 to 50 mg brain tissue (snap frozen or fixed frozen) to produce a sample DNA ranging from 11.3ug/ul to 11.3ug/ul of DNA. Genotyping for APOE and the TOMM40 rs10524523 poly T repeat was carried out by Polymorphic DNA Technologies using Sanger dideoxy DNA sequencing¹⁷ Polymorphic genotyped two SNPs (rs429358, rs7412) used to identify APOE ε2/ε3/ε4 haplotypes, and genotyped the APOE poly T repeat. Genotyping for MAPT and again for APOE was carried out by Newgene, using Sequenom MALDI-TOF technology¹³. Newgene genotyped 6 MAPT SNPs (rs242557, rs1800547, rs3785883, rs 98 997, rs2471738 and rs9468) which allowed to determine MAPT H1/H2 haplotypes, and the same APOE SNPs (rs429358, rs7412) as Polymorphic.

Once the raw SNP data was available, they were used to define the H1/H2 haplotypes in MAPT and the ε2/ε3/ε4 haplotypes in APOE. The TOMM40 IVS6 rs10524523 poly T repeats were classified as 'Short' (poly T repeat shorter than 19 Ts), 'Long' (poly T repeat of 19 to 29 Ts) and 'Very Long' (poly T repeat longer than 30 Ts)¹⁹. For each individual we also calculated the allele length difference and the mean pair length. Assay description of the high-throughput sequencing and analytical validity data for TOMM40 rs10524523 (Polymorphic DNA Technologies, Inc.):

Polymorphic DNA Technologies, Inc. (Alameda CA, polymorphicdna.com) developed two independent assays that use PCR amplification of the region surrounding located at chrl 9:45,403,049-45,403,083 (hg19) in the intron between the 5^th and 6^th coding exons of the gene TOMM40 (mitochondrial import receptor subunit Tom40 homolog). In both cases the PCR amplification step was followed by Sanger sequencing of the PCR templates. The two independent assays are labeled "Assay 1" and "Assay 2". The oligonucleotide primers used in Assay 2 are different from those used in Assay 1 , so that any unknown variation at a primer site for one assay will not interfere with the amplification in the other assay. By using Sanger sequencing, single-base resolution can be achieved using the automation and reliability of the ABI 3730x1 sequencing platform. This platform also provides reference sequence information that confirms the identity of the amplified region and also gives contextual information that can be used to estimate the exact sizes of the homopolymer.

Methods

PCR Primers. In order to achieve a high level of specific amplification a two-step, nested PCR strategy was used to amplify the region flanking '523. Human genomic DNA samples are first PCR-amplified with a pair of "outer" oligonucleotide primers, and those amplification products are then re-amplified with a second pair of "inner" primers.

The primers used in Assay 1 for these amplifications are as follows:

Outer: 5'-CTGGGCTCAAATGAACC-3' and 5'-CAAATGTGATTTTATAGGGCCA-3' Inner: 5'-GAGATGGGGTCTCACT-3' and 5'-ACAGGGAAAGAAAACAAGCCTG-3' The resulting final PCR product maps to the human reference sequence in the range chrl 9:45,402,934-45,403,215. The reference sequence of the amplicon produced in Assay 1 is shown below:

5'-GAGATGGGGTCTCACTATGTCACCTAGGCTTGTCTCCAACTGCTGACCT

CAAGCTGTCCTCTTGCCCCAGCCCTCCAAAGCATTGGGATTACTGGCATGA GCCATTGCATCTGGCT I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I GAGAT GGGGTCTCACCATGTTGCCCGGGCTGGCTTCGAATGCCTAGGCTCAAGCAA TCCTCCCTTCTCAGCCTCCCAAATAGCTGGGATTACAGGCACGTGCCACCAC GCCAGGCTTGTTTTCTTTCCCTGT-3' The primers used in Assay 2 for these amplifications are as follows:

Outer: 5'-TGGCCTCCCAAACTG-3' and 5'-GGCGTGGTGGCAC-3'

Inner: 5'-GTTAGATGAAG I I I I I AAATTTTTTGTAG-3' and 5'-

GAGAAGGGAGGATTGCT-3'

The resulting final PCR product maps to the human reference sequence in the range chrl 9:45,402,906-45,403, 151. The reference sequence of the amplicon produced in Assay 2 is shown below:

5 -GTTAGATGAAG I I I I I AAA I I I I I I GTAGAGATGGGGTCTCACTATGTC

ACCTAGGCTTGTCTCCAACTGCTGACCTCAAGCTGTCCTCTTGCCCCAGC CCTCCAAAGCATTGGGATTACTGGCATGAGCCATTGCATCTGGC I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I GAGATGGGGTCTCACCATGTTGCCC GGGCTGGCTTCGAATGCCTAGGCTCAAGCAATCCTCCCTTCTC-3'

The 3'-ends were chosen to be in locations in which there are no A-residues. PCR Procedures. The first "outer" PCR reaction was performed in a 384-well plate by combining in each well approximately 10 ng of genomic DNA with 1 unit of Klen Taq (Ab Peptides, Inc.), 0.5 pL of 10 X PCR Buffer, 1.0 pL of 5.0 M betaine, 0.175 μΙ_ DMSO, 0.8 pL of a mixture of dNTPs (2.5 mM each), 0.5 pl_ of a mixture of the two outer oligonucleotide primers (2 μΜ each), plus water to bring the entire reaction volume per well to a total of 5.0 μΙ_. Plates were placed in a thermocycler and subjected to 40 cycles of the following conditions: 94° C for 20 s., 55° C for 25 s., and 72° C for 60 s.

The second "inner" PCR reaction was performed in a different 384-well plate by combining in each well 0.8 μΙ_ of the product from the first PCR reaction with 0.25 units of Klen Taq, 0.5 pL of 10 X PCR Buffer, 1.0 pL of 5.0 M betaine, 0.175 μΙ_ DMSO, 0.2 μΙ_ of a mixture of dNTPs (2.5 mM each), 1.0 μΙ_ of a mixture of the two inner oligonucleotide primers (2 μΜ each), plus water to bring the entire reaction volume per well to a total of 5.0 pL. Plates were placed in a thermocycler and subjected to 40 cycles of the following conditions: 94° C for 20 s., 55° C for 25 s., and 72° C for 60 s. These reaction products were then purified using PCR Cleanup Plates (MultiScreen384 PCR, Millipore Corp.), with a final elution of each sample using 16 pL of water.

Addition of an Internal Standard. An internal standard was added to each amplified sample product prior to running of the Sanger sequencing reactions. This standard (PCR product) was then amplified using a plasmid clone of the region as the PCR template. The cloned sequence was the sequence Chr19:45,402,798-45,403,279, but with a synthetic deletion so that the poly-T sequence contains only 8 T-residues (T8). The synthetic allele T8 was used because this is shorter than the smallest natural- occurring allele and so that the T8 sequence can be unambiguously identified as the internal standard rather than a product derived from the original genomic sample. This cloned T8 sequence containing all oligonucleotide priming sites used in both Assay 1 and Assay 2 was amplified with the same primers and under the same conditions in order to create an appropriate internal standard for each assay. To each amplicon from a genomic sample, an aliquot of the appropriate internal standard amplification product was added in the ratio 8:1 (8 parts amplicon from the genomic sample to 1 part amplicon from the plasmid clone.)

Sanger Sequencing Reactions. Using the mixed template described above, fluorescent Sanger sequencing reactions were carried out using the forward "inner" PCR primer as a sequencing primer. Reactions were performed in a new 384-well plate by combining in each well 0.8 μΙ_ of the "spiked" PCR product described above with 0. 25 pL Big Dye Mix (Applied BioSystems, Inc.), 0.215 pL 5X Dilution Buffer (Applied BioSystems, Inc.), 0.25 pL of a 1 μΜ solution of the forward "inner" primer, 0.105 μΙ_ DMSO, 0.6 μΙ_ of 5 M betaine, plus water to bring the entire volume to 3.0 pL. Plates were then placed in a thermocycler and subjected to 30 cycles of the following conditions: 96° C for 0 s., 50° C for 15 s., and 60° C for 120 s. These reaction products were then purified using Sequencing Reaction Cleanup Plates (MultiScreen384 SEQ, Millipore Corp.), with a final elution of each sample with 20 pL of 0.3 mM EDTA solution. Electrophoresis of Sequencing Products. Information about the analysis of the electropherograms can be obtained by Polymorphic DNA technologies. In particular, for the final assessment of the genotypes, in most cases the two sets of genotype calls were identical or very similar. When small differences occured, the two values were averaged to the nearest integer value. If, for example, Assay 1 yielded a genotype of T16 T35 and Assay 2 yielded a genotype of T16/T33, the final consensus genotype reported would be T16/T34. In some others cases, however, only one of the two assays yielded a result, and in this case, the genotype reported was the one from that assay, along with a note saying that result is only from one assay. In rare situations, there was a major discrepancy between the calls of the two assays for a particular sample. This could have resulted either due to an analysis error or when one of the assays failed to amplify one allele. When a major discrepancy occurred, the electropherograms from both assays were re-examined. If the discrepancy was caused by an analysis error, then the errant call is corrected. However, if the error was assay-related, the results are reported "as is" from both assays and such discrepancies were noted in the genotype report.

Final Determination of Genotypes. The above procedures were performed separately on both Assay 1 and Assay 2. Only after all genotype calls have been independently determined are the two sets of genotype calls compared. In most cases, the two sets of genotype calls were identical or very similar. When small differences occurred, the two values were averaged to the nearest integer value. If, for example, Assay 1 yielded a genotype of T16 T35 and Assay 2 yielded a genotype of T16/T33, the final consensus genotype reported would be T16/T34. In some others cases, however, only one of the two assays yielded a result, and in this case, the genotype reported was the one from that assay, along with a note saying that result is only from one assay. In rare situations, there was a major discrepancy between the calls of the two assays for a particular sample. This could have resulted either due to an analysis error or when one of the assays failed to amplify one allele. When a major discrepancy occurred, the electropherograms from both assays were re-examined. If the discrepancy was caused by an analysis error, then the errant call is corrected. However, if the error was assay- related, the results are reported "as is" from both assays and such discrepancies were noted in the genotype report.

Discussion

Method Validation. The method has been validated in three ways. Firstly, the method was applied to cloned DNA with known poly-T sizes and this method confirms the sizes of those poly-T repeats. Secondly, the method was run on genomic DNA samples for which the poly-T genotypes had been independently measured by the sequencing of clones from long-range PCR. In this case, the present method gives results that are the same within the statistical uncertainty of the clones picked. Thirdly, the allele distribution as defined by this method was found to be comparable to that found by cloning methods, again with the caveat that this was identified to result from those cloning methodologies. Statistical methods

Cox Proportional Hazard Model of Genetic effects on Dementia

The TOMM40 IVS6 rs10524523 fragment lengths was classified in three groups, short (S), long (L) and very long(VL)¹⁹. The distribution of fragment length strongly suggests that three ciades are present, and the subdivision of fragment lengths into three groups reflects this observation. Our first analysis used a Cox Proportional Hazard model to model the effects of TOMM40 rs10524523, and the effects of APOE, MAPT, age of onset of motor symptoms and vascular risk score. While the Cox model only takes presence or absence of dementia as a variable, it also automatically adjusts for latency to dementia, making the subdivision between DLB and PDD redundant for this kind of model.

The strong association between the most strongly heterozygous genotype and dementia raised the issue of whether the binning of rs10524523 fragment length could actually dilute some of the effects of rs10524523 on dementia status. We overcame this obstacle by using two natural measures intrinsic to the rs10524523 poly T polymorphism. The first is an absolute value representing the difference between the two allele lengths (ALD) and is directly related to the level of heterozygosity between the two alleles. Closely or perfectly matched alleles would yield a low score, while a high score would reflect highly divergent allele lengths. The second measure that we used was the mean fragment length. This measure is directly related to both homozygosity and to overall length concordance. In this case we would observe a low score for closely matched short alleles, a high score for closely matched long alleles and an intermediate score from divergent length alleles. We run a second cox proportional hazards model substituting TOMM40's genotypes for ALD and mean allele length.

All analyses were computed in R and SAS and gave the same results.

Confusion Matrix

Sensitivity, specificity, positive predictive value and negative predictive value were calculated from the confusion matrix in Table 2, using the R package 'Caret'²⁰ and by double checking the results by doing the calculations by hand^21,22. Table 11 : list of 17 individuals showing either severe A(3deposition in the frontal cortex and/or Braak Tau stage V VI. Highlighted are 4 cases with possible concurrent AD pathology

Bra Bra AB

Age Age Dementia Age Dement B dys ak ak AB Intensity fror

Sex Motor Dementia Latency Death hals ia D vrs aut aSN tau striatal striatal al

M 78 79 1 84 y def y 0 2 6 A NA NA pos

M 62 62 0 69 y def NA 1 2 6 C NA NA os

M 58 59 1 63 y def y 0 4 6 A NA NA pos no

dementi

M 65 NA NA 69 y a NA 3 3 6 C NA NA pos M 72 78 6 78 n def y 3 2 6 A NA NA pos moderat M 66 71 5 72 y def y 0 3 6 A pos e pos F 64 80 16 83 y def NA 0 2 6 A NA NA pos no

dementi

F 70 NA NA 77 n a NA 0 1 6 A pos severe pos no

dementi

M 78 NA NA 89 n a NA 0 1 3 A pos severe pos

M 60 61 1 65 y def NA 0 1 6 B pos severe pos

F 78 78 0 83 y def y 0 2 6 B pos severe pos

75 75 0 78 y def NA 1 6 B os severe pos

53 77 24 80 y def y 0 2 6 A pos mild pos moderat

F 63 81 18 84 y def NA 1 1 6 B pos e pos moderat

M 62 70 8 72 n def NA 2 1 6 C os e pos

M 58 73 15 74 y def NA 4 3 6 A pos severe pos

F 74 74 0 77 y def y 0 1 6 C pos severe pos

Results after removing the 4 possible (Id 2, 11 , 12, 7) concurrent AD cases Table 12: A: Cox model stratified by sex, including TO M40 rs10524523 (baseline risk S/S), APOE (baseline risk ε3/ε3), MAPT (baseline risk H2/H2), age of onset and Vascular Risk Score as predictors. B: Cox model including TOMM40 '523 allele length difference, TOMM40 rs10524523 mean pair length, APOE (baseline risk ε3/ε3), MAPT 5 (baseline risk H2/H2), age of onset and Vascular Risk Score as predictors

Model A: model stratified by Sex

OR Lower 0.95 CI Upper .095 CI P-value

LA 1.68 0.19 15.08 0.644

LA/L 1.76 0.44 7.07 0.427

S/L 1.55 0.44 5.49 0.497

S/VL 3.55 1.52 8.25 0.003 **

VUVL 1.44 0.54 3.85 0.471

ε2/ε3 0.46 0.19 1.12 0.086

ε2/ε4 1.07 0.10 11.81 0.953

ε3/ε4 0.91 0.37 2.21 0.830

ε4/ε4 2.80 0.70 11.21 0.147

Η1Η1 1.18 0.26 5.37 0.828

Η1Η2 1.14 0.25 5.27 0.865

Age onset PD 0.96 0.92 1.00 0.054

Vascular Risk Score 0.99 0.79 1.23 0.916

Model B: model stratified by Sex

Allele length difference 1.07 1.03 1.10 0.0005 ***

Mean Pair Length 1.01 0.96 1.06 0.6304

ε2/ε3 0.47 0.19 1.14 0.0945

ε2/ε4 0.72 0.08 6.13 0.7627

ε3/ε4 0.85 0.45 1.62 0.6243

ε4/ε4 2.97 1.08 8.22 0.0355 * mm 1.21 0.27 5.44 0.8055

Η1 Η2 1.15 0.25 5.31 0.8588

Age onset PD 0.96 0.92 1.00 0.0369 *

Vascular Risk Score 0.99 0.80 1.24 0.9497

Broadly speaking the results do not change: the TOMM40 SA/L genotype and the ALD score are still strongly and positively associated with the onset of dementia. In the model 10 using ADL we also observe a positive increase in OR for the APOE ε4/ε4 genotype. Table 13: Confusion matrix for TO M40 allele length difference, with the resulting sensitivity, specificity, positive predictive value and negative predictive value for a 15 bases allele length difference threshold.

Once more the results obtained in the general analysis are replicated, with a higher sensitivity and a consequent slightly lower positive predictive value but a higher negative predictive value.

References

1 Gibb, W.R.G & Lees, A.J. The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson's disease. J Neurol Neurosurg Psychiatry 1988; 51 : 745-752. 2 McKeith, I.G. et al. Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy Bodies (DLB): report of the consortium on DLB international workshop. Neurology 996; 47: 1 13-1 24.

3 DSM-IV. Diagnostic and statistical manual of mental disorders. Washington: American Psychiatric Association; 1995.

4 Hodkinson, H.M. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing. 1972 Nov; 1(4):233-8.

5 Folstein, M.F., Folstein, S.E. & McHugh, P.R. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189— 98.

6 Emre, M. et al. Clinical diagnostic criteria for dementia associated with Parkinson's disease. Mov Disord 2007; 22: 1689-1707. 7 McKeith, I.G. et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology 2005; 65:1863-1872.

8 Emre, M. Dementia associated with Parkinson's disease. Lancet Neurol 2003; 2: 229- 37.

9 Nelson, P.T., Kryscio, R.J., Jicha, G.A. & Abner, E.L. Relative preservation of MMSE scores in autopsy-proven dementia with Lewy bodies. Neurology 2009; 73: 1127-1 133.

10 Daniel, S.E. & Lees, A.J. Parkinson^'s Disease Society Brain Bank, London: Overview and research. J Neural Transm Suppl 1993;39:165-172.

11 Emre, M. et al. Clinical diagnostic criteria for dementia associated with Parkinson's disease. ov Disord 2007; 22: 1689-1707.

12 Alafuzoff, I. et al. Staging/typing of Lewy body related a-synuclein pathology: a study of the BrainNet Europe Consortium. Acta Neuropathol 2009; 17: 635-652.

13 Braak, H. et al. Staging of brain pathology related to sporadic Parkinson's disease. Neurobiol Aging 2003; 24: 197-21 .

14 Braak, H. & Braak, E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 1991 ;82(4):239-59.

15 Alafuzoff, I. et al. Staging of neurofibrillary pathology in Alzheimer's Disease. A study of the BrainNet Europe Consortium. Brain Pathol 2008; 18:484-496.

16 Mirra, S.S. et al. The Consortium to Establish a Registry for Alzheimer's disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer's disease. Neurology 1991; 41: 479-486.

17 Sanger, F. Sequences, sequences, and sequences. Annu. Rev. Biochem. 1988; 57:1- 28.

18 Edwards, J.R., Ruparel, H. & Ju, J. Mass-spectrometry DNA sequencing. Mutation Research 2005; 573: 3-12

19 Lutz, M.W., Crenshaw, D.G., Saunders, A.M. & Roses, A.D. Genetic variation at a single locus and age of onset for Alzheimer's disease. Alzheimers Dement. 2010;6(2): 125-31.

20 Kuhn. M. Building predictive models in R using the caret package. Journal of Statistical Software 2008; 28(5): 1-26.

21 Altman, D.G. & Bland, J.M. Diagnostic tests 1 : sensitivity and specificity. British Medical Journal 1994; 308: 1552.

22 Altman, D.G. & Bland, J.M. Diagnostic tests 2: predictive values. British Medical Journal 1994; 309: 102.

Example 2: Clinical trial of candidate Parkinson's Disease treatment In order to assess the efficacy of a candidate compound expected to be useful in the treatment of Parkinson's Disease, for example expected to be useful in delaying or diminishing dementia associated with Parkinson's Disease (for example PLB), the candidate compound is assessed in Parkinson's Disease patients (or patients considered to be at risk of Parkinson's Disease) selected on the basis of a relatively high likelihood of rapid development of dementia based on their TO 40 genotype, possibly combined with other factors such as sex (male). Because the selected patients have a relatively high likelihood of rapid development of dementia, the effect (or lack of effect) of the candidate compound should be apparent much earlier in such patients than in unselected PD patients, where the overall likelihood of rapid development of dementia would be much lower. This means that the trial can potentially be conducted more quickly and with fewer patients.

Claims

1. A method for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features, the method comprising the step of assessing the subject's genotype for TO M40, optionally in intron 6 (IVS6) of the TOM 40 gene, optionally at position rs10524523.

2. The method of claim 1 wherein the method comprises the step of determining whether the subject's genotype for TOMM40, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally determining whether there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally whether the subject's genotype at rs10524523 is SA/L or whether the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

3. The method of claim 1 or 2 wherein if the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is SA L or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, then the subject is considered to be at higher risk of developing Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or at higher risk of disease progression, optionally at higher risk of developing Parkinson's Disease Dementia (PDD), optionally Dementia with Lewy Bodies (DLB), optionally of developing DLB within a one or two year period of onset of motor symptoms of Parkinson's Disease or motor Parkinsonian features.

4. The method of any one of claims 1 to 3 further comprising the step of selecting a treatment regime making use of the information on the TOMM40 genotype.

5. The method of any one of claims 1 to 4 wherein if the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs, then the selected treatment regime comprises treating the subject with one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non-motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds for treating insulin resistance or antiinflammatory compounds, for example as referred to in lclat Aviles-Olmos et al, Brain 2012 (Feb 17) Parkinson's disease, insulin resistance and novel agents of neuroprotection, or Tansey and Goldberg, Neurobiol.Dis 2010 Mar;37(3):510-8. Epub 2009 Nov 10 Neuroinflammation in Parkinson's disease: its role in neuronal death and implications for therapeutic intervention; or compounds mentioned at http://alzheimers.orq.uk site/scripts/documents info.php?documentlD=106.

6. The method of any one of claims 1 to 4 wherein if the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is not heterozygous, optionally wherein there is no or less than 15, 14, 13, 12, 11 , 10, 9, 8, 7, 6, 5, 4, 3 or 2 base pair difference in the length of a poly-T stretch at rs10524523 between the subject's alleles or wherein the subject's genotype at rs10524523 is not S/VL, then the selected treatment regime does not comprise any additional components.

7. One or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non-motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds for treating insulin resistance or antiinflammatory compounds, for example as referred to in lclat Aviles-Olmos et al, Brain 2012 (Feb 17) Parkinson's disease, insulin resistance and novel agents of neuroprotection, or Tansey and Goldberg, Neurobiol.Dis 2010 Mar;37(3):510-8. Epub 2009 Nov 10 Neuroinflammation in Parkinson's disease: its role in neuronal death and implications for therapeutic intervention; or compounds mentioned at http://alzheimers.orq.uk/site/scripts/documents info.php?documentlD=106 for use in treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype), wherein the subject is a subject for whom it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

8. Use of one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non-motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti-TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds for treating insulin resistance or anti-inflammatory compounds, for example as referred to in lclat Aviles-Olmos et al, Brain 2012 (Feb 17) Parkinson's disease, insulin resistance and novel agents of neuroprotection, or Tansey and Goldberg, Neurobiol.Dis 2010 Mar;37(3):510-8. Epub 2009 Nov 10 Neuroinflammation in Parkinson's disease: its role in neuronal death and implications for therapeutic intervention; or compounds mentioned at http://alzheimers.orq.uk site/scripts/documents info.php?documentlD=106 in the manufacture of a medicament for treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype), wherein the subject is a subject for whom it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

9. A method for treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype), the method comprising administering one or more agents selected from agents considered to be useful in having a disease modifying effect on the PD and/or its non-motor complications, such as DLB or PDD, or in treating dementia, such as donezepil, Anti- TNF, B-MAO Inhibitors, Selegiline, Cyclosporine A and FK-506, immunofilin ligands such as pentoxifylline and COX-2 Inhibitors, minocycline, Immunoglobulins, NMDA receptor antagonsists, PPAR agonists and modulators, iNOS Inhibitors, Copolymer-1 (Cop-1 ), GLP-1 receptor agonists, Acetylcholinesterase Inhibitors, other antibodies, fusion proteins, therapeutic RNA molecules and combination thereof, compounds for treating insulin resistance or anti-inflammatory compounds, for example as referred to in lclat Aviles-Olmos et al, Brain 2012 (Feb 17) Parkinson's disease, insulin resistance and novel agents of neuroprotection, or Tansey and Goldberg, Neurobiol.Dis 2010 Mar;37(3):510-8. Epub 2009 Nov 10 Neuroinflammation in Parkinson's disease: its role in neuronal death and implications for therapeutic intervention; or compounds mentioned at http://alzheimers.orq.uk/site/scripts/documents info.php?documentlD=106 to the subject, wherein the subject is a subject for whom it has been determined that the TO M40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to

10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

10. The treatment for use, use or method of any one of claims 7 to 9 wherein the subject is administered an additional anti Parkinson's Disease or motor Parkinsonisian features treatment, optionally selected from L Dopa with a peripheral dopa decarboxylase inhibitor, dopamine agonists, elective type B monoamine oxidase inhibitors, Amantadine and tolcapone.

11. An agent which is specifically capable of use in determining the determining a subject genotype at TOMM40, optionally intron 6, optionally rs10524523, for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features..

12. Use of an agent which is specifically capable of use in determining the determining a subject's genotype at TOMM40, optionally intron 6, optionally rs10524523, in the manufacture of a medicament for aiding in determining likelihood of developing or worsening Parkinson's Disease, motor Parkinsonisian features, dementia with Lewy Bodies (DLB) or Parkinson's Dementia (PD); or for categorising or determining prognosis, optionally a relatively high or relatively low likelihood of developing or worsening dementia, for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features; and/or in selecting a therapeutic strategy for a subject with or otherwise at risk of Parkinson's Disease or motor Parkinsonisian features.

13. An agent according to claim 11 or use according to claim 12 wherein the agent is a nucleic acid which selectively hybridises to TOMM40 nucleic acid, optionally in or flanking intron 6, optionally in or flanking rs10524523.

14. A method for selecting a test Parkinson's Disease or motor Parkinsonisian features or at risk thereof subject (other than through determination of TOMM40 genotype) in whom to assess a therapeutic strategy or treatment regime for the treatment of Parkinson's Disease or motor Parkinsonisian features, the method comprising the step of assessing the test subject's genotype for TOMM40, optionally in intron 6 (IVS6) of the TO M40 gene, optionally at position rs10524523.

15. The method of claim 14, comprising the step of determining whether the test subject's genotype for TOMM40, optionally in intron 6, optionally at rs 0524523, is heterozygous, optionally determining whether there is a difference in the length of a poly- T stretch at rs10524523 between the subject's alleles, optionally whether the subject's genotype at rs10524523 is SA L or whether the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

16. The method of claim 15 wherein the test subject is selected if the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the subject's alleles, optionally wherein the subject's genotype at rs10524523 is S/VL or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

17. A method for assessing a therapeutic strategy or treatment regime for the treatment of Parkinson's Disease or motor Parkinsonisian features or at risk thereof (other than through determination of TOMM40 genotype), the method comprising the step of evaluating the results of the therapeutic strategy or treatment regime from a subject or subjects who have been selected according to the method of any one of claims 14 to 16.

18. An in vitro method for assessing the suitability of a test compound for the treatment of a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof, the method comprising the step of determining the effect of the test compound on a cell for which it has been determined that the TOMM40 genotype, optionally in intron 6, optionally at rs10524523, is heterozygous, optionally wherein there is a difference in the length of a poly-T stretch at rs10524523 between the cell's alleles, optionally wherein the cell's genotype at rs 10524523 is SA L or wherein the difference in the length of the poly-T stretch is more than or equal to 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 base pairs, optionally more than or equal to 12, 13, 14, 15, 16, 17 or 18 base pairs.

19. A screening method for identifying a compound likely to be useful in treating a subject with Parkinson's Disease or motor Parkinsonisian features or at risk thereof, the method comprising the step of determining the effect of a test compound on TOMM40 nucleic acid, protein or activity level; and selecting a compound that modulates, for example reduces or increases, said level.

20. The method of claim 19 wherein the effect of the test compound is determined in vitro.

21. The method of claim 19 wherein the effect of the test compound is determined in vivo in a non-human test animal.