Recent studies suggest progranulin (GRN) is a neurotrophic factor. Loss-of-function mutations in the progranulin gene (GRN) cause frontotemporal lobar degeneration (FTLD), a progressive neurodegenerative disease affecting ∼10% of early-onset dementia patients. Using an enzyme-linked immunosorbent assay, we previously showed that GRN is detectable in human plasma and can be used to predict GRN mutation status. This study also showed a wide range in plasma GRN levels in non-GRN mutation carriers, including controls. We have now performed a genome-wide association study of 313,504 single-nucleotide polymorphisms (SNPs) in 533 control samples and identified on chromosome 1p13.3 two SNPs with genome-wide significant association with plasma GRN levels (top SNP rs646776; p = 1.7 × 10⁻³⁰). The association of rs646776 with plasma GRN levels was replicated in two independent series of 508 controls (p = 1.9 × 10⁻¹⁹) and 197 FTLD patients (p = 6.4 × 10⁻¹²). Overall, each copy of the minor C allele decreased GRN levels by ∼15%. SNP rs646776 is located near sortilin (SORT1), and the minor C allele of rs646776 was previously associated with increased SORT1 mRNA levels. Supporting these findings, overexpression of SORT1 in cultured HeLa cells dramatically reduced GRN levels in the conditioned media, whereas knockdown of SORT1 increased extracellular GRN levels. In summary, we identified significant association of a locus on chromosome 1p13.3 with plasma GRN levels through an unbiased genome-wide screening approach and implicated SORT1 as an important regulator of GRN levels. This finding opens avenues for future research into GRN biology and the pathophysiology of neurodegenerative diseases.

Many patients with dementia with Lewy bodies (DLB) have overlapping Alzheimer's disease (AD)-related pathology, which may contribute to white matter (WM) diffusivity alterations on diffusion tensor imaging (DTI). Consecutive patients with DLB (n = 30), age- and sex-matched AD patients (n = 30), and cognitively normal controls (n = 60) were recruited. All subjects underwent DTI, 18F 2-fluoro-deoxy-d-glucose, and (11)C Pittsburgh compound B positron emission tomography scans. DLB patients had reduced fractional anisotropy (FA) in the parietooccipital WM but not elsewhere compared with cognitively normal controls, and elevated FA in parahippocampal WM compared with AD patients, which persisted after controlling for β-amyloid load in DLB. The pattern of WM FA alterations on DTI was consistent with the more diffuse posterior parietal and occipital glucose hypometabolism of 2-fluoro-deoxy-d-glucose positron emission tomography in the cortex. DLB is characterized by a loss of parietooccipital WM integrity, independent of concomitant AD-related β-amyloid load. Cortical glucose hypometabolism accompanies WM FA alterations with a concordant pattern of gray and WM involvement in the parietooccipital lobes in DLB.

Excessive daytime sleepiness is a commonly reported problem in dementia with Lewy bodies (DLB). We examined the relationship between nighttime sleep continuity and the propensity to fall asleep during the day in clinically probable DLB compared to Alzheimer's disease (AD) dementia.

The objective of our study was to evaluate whether cognitively normal (CN) elderly participants showing elevated cortical beta-amyloid (Aβ) deposition have a consistent neuroanatomical signature of brain atrophy that may characterize preclinical Alzheimer's disease (AD). 115 CN participants who were Aβ-positive (CN +) by amyloid PET imaging; 115 CN participants who were Aβ-negative (CN -); and 88 Aβ-positive mild cognitive impairment or AD participants (MCI/AD +) were identified. Cortical thickness (FreeSurfer) and gray matter volume (SPM5) were measured for 28 regions-of-interest (ROIs) across the brain and compared across groups. ROIs that best discriminated CN - from CN + differed for FreeSurfer cortical thickness and SPM5 gray matter volume. Group-wise discrimination was poor with a high degree of uncertainty in terms of the rank ordering of ROIs. In contrast, both techniques showed strong and consistent findings comparing MCI/AD + to both CN - and CN + groups, with entorhinal cortex, middle and inferior temporal lobe, inferior parietal lobe, and hippocampus providing the best discrimination for both techniques. Concordance across techniques was higher for the CN - and CN + versus MCI/AD + comparisons, compared to the CN - versus CN + comparison. The weak and inconsistent nature of the findings across technique in this study cast doubt on the existence of a reliable neuroanatomical signature of preclinical AD in elderly PiB-positive CN participants.

Our objective was to examine associations between glucose metabolism, as measured by (18)F-fluorodeoxyglucose positron emission tomography (FDG PET), and age and to evaluate the impact of carriage of an apolipoprotein E (APOE) ε4 allele on glucose metabolism and on the associations between glucose metabolism and age. We studied 806 cognitively normal (CN) and 70 amyloid-imaging-positive cognitively impaired participants (35 with mild cognitive impairment and 35 with Alzheimer's disease [AD] dementia) from the Mayo Clinic Study of Aging, Mayo Alzheimer's Disease Research Center and an ancillary study who had undergone structural MRI, FDG PET, and (11)C-Pittsburgh compound B (PiB) PET. Using partial volume corrected and uncorrected FDG PET glucose uptake ratios, we evaluated associations of regional FDG ratios with age and carriage of an APOE ε4 allele in CN participants between the ages of 30 and 95 years, and compared those findings with the cognitively impaired participants. In region-of-interest (ROI) analyses, we found modest but statistically significant declines in FDG ratio in most cortical and subcortical regions as a function of age. We also found a main effect of APOE ε4 genotype on FDG ratio, with greater uptake in ε4 noncarriers compared with carriers but only in the posterior cingulate and/or precuneus, lateral parietal, and AD-signature meta-ROI. The latter consisted of voxels from posterior cingulate and/or precuneus, lateral parietal, and inferior temporal. In age- and sex-matched CN participants the magnitude of the difference in partial volume corrected FDG ratio in the AD-signature meta-ROI for APOE ε4 carriers compared with noncarriers was about 4 times smaller than the magnitude of the difference between age- and sex-matched elderly APOE ε4 carrier CN compared with AD dementia participants. In an analysis in participants older than 70 years (31.3% of whom had elevated PiB), there was no interaction between PiB status and APOE ε4 genotype with respect to glucose metabolism. Glucose metabolism declines with age in many brain regions. Carriage of an APOE ε4 allele was associated with reductions in FDG ratio in the posterior cingulate and/or precuneus, lateral parietal, and AD-signature ROIs, and there was no interaction between age and APOE ε4 status. The posterior cingulate and/or precuneus and lateral parietal regions have a unique vulnerability to reductions in glucose metabolic rate as a function both of age and carriage of an APOE ε4 allele.

To test the hypotheses predicted in a hypothetical model of Alzheimer disease (AD) biomarkers that rates of β-amyloid (Aβ) accumulation on PET imaging are not related to hippocampal neurodegeneration whereas rates of neurodegenerative brain atrophy depend on the presence of both amyloid and neurodegeneration in a population-based sample.

We aimed to identify new candidate genes potentially involved in early-onset Alzheimer's disease (EOAD). Exome sequencing was conducted on 45 EOAD patients with either a family history of Alzheimer's disease (AD, <65 years) or an extremely early age at the onset (≤55 years) followed by multiple variant filtering according to different modes of inheritance. We identified 29 candidate genes potentially involved in EOAD, of which the gene TYROBP, previously implicated in AD, was selected for genetic and functional follow-up. Using 3 patient cohorts, we observed rare coding TYROBP variants in 9 out of 1110 EOAD patients, whereas no such variants were detected in 1826 controls (p = 0.0001), suggesting that at least some rare TYROBP variants might contribute to EOAD risk. Overexpression of the p.D50_L51ins14 TYROBP mutant led to a profound reduction of TREM2 expression, a well-established risk factor for AD. This is the first study supporting a role for genetic variation in TYROBP in EOAD, with in vitro support for a functional effect of the p.D50_L51ins14 TYROBP mutation on TREM2 expression.

Clinical and neuropathological characteristics associated with G4C2 repeat expansions in chromosome 9 open reading frame 72 (C9ORF72), the most common genetic cause of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia, are highly variable. To gain insight on the molecular basis for the heterogeneity among C9ORF72 mutation carriers, we evaluated associations between features of disease and levels of two abundantly expressed "c9RAN proteins" produced by repeat-associated non-ATG (RAN) translation of the expanded repeat. For these studies, we took a departure from traditional immunohistochemical approaches and instead employed immunoassays to quantitatively measure poly(GP) and poly(GA) levels in cerebellum, frontal cortex, motor cortex, and/or hippocampus from 55 C9ORF72 mutation carriers [12 patients with ALS, 24 with frontotemporal lobar degeneration (FTLD) and 19 with FTLD with motor neuron disease (FTLD-MND)]. We additionally investigated associations between levels of poly(GP) or poly(GA) and cognitive impairment in 15 C9ORF72 ALS patients for whom neuropsychological data were available. Among the neuroanatomical regions investigated, poly(GP) levels were highest in the cerebellum. In this same region, associations between poly(GP) and both neuropathological and clinical features were detected. Specifically, cerebellar poly(GP) levels were significantly lower in patients with ALS compared to patients with FTLD or FTLD-MND. Furthermore, cerebellar poly(GP) associated with cognitive score in our cohort of 15 patients. In the cerebellum, poly(GA) levels similarly trended lower in the ALS subgroup compared to FTLD or FTLD-MND subgroups, but no association between cerebellar poly(GA) and cognitive score was detected. Both cerebellar poly(GP) and poly(GA) associated with C9ORF72 variant 3 mRNA expression, but not variant 1 expression, repeat size, disease onset, or survival after onset. Overall, these data indicate that cerebellar abnormalities, as evidenced by poly(GP) accumulation, associate with neuropathological and clinical phenotypes, in particular cognitive impairment, of C9ORF72 mutation carriers.

Intronic hexanucleotide (G4C2) repeat expansions in C9orf72 are genetically associated with frontotemporal lobar degeneration (FTLD) and amyotrophic lateral sclerosis (ALS). The repeat RNA accumulates within RNA foci but is also translated into disease characterizing dipeptide repeat proteins (DPR). Repeat-dependent toxicity may affect nuclear import. hnRNPA3 is a heterogeneous nuclear ribonucleoprotein, which specifically binds to the G4C2 repeat RNA We now report that a reduction of nuclear hnRNPA3 leads to an increase of the repeat RNA as well as DPR production and deposition in primary neurons and a novel tissue culture model that reproduces features of the C9orf72 pathology. In fibroblasts derived from patients carrying extended C9orf72 repeats, nuclear RNA foci accumulated upon reduction of hnRNPA3. Neurons in the hippocampus of C9orf72 patients are frequently devoid of hnRNPA3. Reduced nuclear hnRNPA3 in the hippocampus of patients with extended C9orf72 repeats correlates with increased DPR deposition. Thus, reduced hnRNPA3 expression in C9orf72 cases leads to increased levels of the repeat RNA as well as enhanced production and deposition of DPR proteins and RNA foci.

Progranulin (PGRN), a widely secreted growth factor, is involved in multiple biological functions, and mutations located within the PGRN gene (GRN) are a major cause of frontotemporal lobar degeneration with TDP-43-positive inclusions (FLTD-TDP). In light of recent reports suggesting PGRN functions as a protective neurotrophic factor and that sortilin (SORT1) is a neuronal receptor for PGRN, we used a Sort1-deficient (Sort1-/-) murine primary hippocampal neuron model to investigate whether PGRN's neurotrophic effects are dependent on SORT1. We sought to elucidate this relationship to determine what role SORT1, as a regulator of PGRN levels, plays in modulating PGRN's neurotrophic effects.

We and others have recently reported an association between amyotrophic lateral sclerosis (ALS) and single nucleotide polymorphisms on chromosome 9p21 in several populations. Here we show that the associated haplotype is the same in all populations and that several families previously shown to have genetic linkage to this region also share this haplotype. The most parsimonious explanation of these data are that there is a single founder for this form of disease.

Frontotemporal lobar degeneration is heterogeneous; cases with tau- and synuclein-negative, ubiquitin-positive neuronal inclusions are the most common, and some have mutations in the gene for progranulin (PGRN). The purpose of this study was to determine whether there were distinctive clinical and neuropathologic features of frontotemporal lobar degeneration with ubiquitin-positive inclusions with PGRN mutations. A retrospective review of medical records and semiquantitative neuropathologic analysis was performed on 18 PGRN(+) and 24 PGRN(-) cases. Clinically, PGRN(+) cases had more frequent language impairment and parkinsonism. Pathologically, PGRN(+) cases had smaller brains, more marked global atrophy, and more frontal atrophy. There was no difference in the frequency of hippocampal sclerosis. The pathology of PGRN(+) cases was relatively homogeneous, whereas PGRN(-) cases were more heterogenous. PGRN(+) cases had greater density of cortical ubiquitin-immunoreactive lesions, especially dystrophic neurites in layer II. Intranuclear inclusions were present in all PGRN(+) and 42% of PGRN(-) cases. The results suggest that frontotemporal lobar degeneration with ubiquitin-positive inclusions due to PGRN mutations has several characteristic features, including ubiquitin-immunoreactive neuritic pathology in superficial cortical layers and neuronal intranuclear inclusions. On the other hand, there is no histopathologic feature or combination of features that is pathognomonic. Neuronal intranuclear inclusions are virtually always present, but they can be detected in PGRN(-) cases.

Visually preserved metabolism in posterior cingulate cortex relative to hypometabolism in precuneus and cuneus, the cingulate island sign, is a feature of dementia with Lewy bodies (DLB) on FDG-PET. Lower cingulate island sign ratio (posterior cingulate cortex/cuneus+precuneus; FDG-CISr) values have been associated with a higher Braak neurofibrillary tangle stage in autopsied DLB. Using voxel-wise analysis, we assessed the patterns of regional cortical perfusion and metabolism, and using an atlas-based approach, we measured perfusion cingulate island sign ratio on arterial spin labeling MRI (ASL-CISr), and its associations with FDG-CISr, uptake on tau-PET and clinical severity in DLB. Our study sample (n = 114) included clinically probable DLB patients (n = 19), age-matched patients with probable Alzheimer's disease dementia (AD; n = 19) and matched controls (n = 76) who underwent MRI with 3-dimensional pseudo-continuous arterial spin labeling, 18F-FDG-PET and 18F-AV-1451 tau PET. Patterns of cortical perfusion and metabolism were derived from quantitative maps using Statistical Parametric Mapping. DLB patients showed hypoperfusion on ASL-MRI in precuneus, cuneus and posterior parieto-occipital cortices, compared to controls, and relatively spared posterior cingulate gyrus, similar to pattern of hypometabolism on FDG-PET. DLB patients had higher ASL-CISr and FDG-CISr than AD patients (p <0.001). ASL-CISr correlated with FDG-CISr in DLB patients (r = 0.67; p =0.002). Accuracy of distinguishing DLB from AD patients was 0.80 for ASL-CISr and 0.91 for FDG-CISr. Lower ASL-CISr was moderately associated with a higher composite medial temporal AV-1451 uptake (r = -0.50; p =0.03) in DLB. Lower perfusion in precuneus and cuneus was associated with worse global clinical scores. In summary, the pattern of cortical hypoperfusion on ASL-MRI is similar to hypometabolism on FDG-PET, and respective cingulate island sign ratios correlate with each other in DLB. Non-invasive and radiotracer-free ASL-MRI may be further developed as a tool for the screening and diagnostic evaluation of DLB patients in a variety of clinical settings where FDG-PET is not accessible.

Late-life measures of white matter (WM) microstructural integrity may predict cognitive status, cognitive decline, and incident mild cognitive impairment (MCI) or dementia. We considered participants of the Atherosclerosis Risk in Communities study who underwent cognitive assessment and neuroimaging in 2011-2013 and were followed through 2016-2017 (n = 1775 for analyses of prevalent MCI and dementia, baseline cognitive performance, and longitudinal cognitive change and n = 889 for analyses of incident MCI, dementia, or death). Cross-sectionally, both overall WM fractional anisotropy and overall WM mean diffusivity were strongly associated with baseline cognitive performance and risk of prevalent MCI or dementia. Longitudinally, greater overall WM mean diffusivity was associated with accelerated cognitive decline, as well as incident MCI, incident dementia, and mortality, but WM fractional anisotropy was not robustly associated with cognitive change or incident cognitive impairment. Both cross-sectional and longitudinal associations were attenuated after additionally adjusting for likely downstream pathologic changes. Increased WM mean diffusivity may provide an early indication of dementia pathogenesis.

Tau deposition is a key biological feature of Alzheimer's disease that is closely related to cognitive impairment. However, it remains poorly understood why certain individuals may be more susceptible to tau deposition while others are more resistant. The recent availability of in vivo assessment of tau burden through positron emission tomography provides an opportunity to test the hypothesis that common genetic variants may influence tau deposition. We performed a genome-wide association study of tau-positron emission tomography on a sample of 754 individuals over age 50 (mean age 72.4 years, 54.6% men, 87.6% cognitively unimpaired) from the population-based Mayo Clinic Study of Aging. Linear regression was performed to test nucleotide polymorphism associations with AV-1451 (18F-flortaucipir) tau-positron emission tomography burden in an Alzheimer's-signature composite region of interest, using an additive genetic model and covarying for age, sex and genetic principal components. Genome-wide significant associations with higher tau were identified for rs76752255 (P = 9.91 × 10-9, β = 0.20) in the tau phosphorylation regulatory gene PPP2R2B (protein phosphatase 2 regulatory subunit B) and for rs117402302 (P  = 4.00 × 10-8, β = 0.19) near IGF2BP3 (insulin-like growth factor 2 mRNA-binding protein 3). The PPP2R2B association remained genome-wide significant after additionally covarying for global amyloid burden and cerebrovascular disease risk, while the IGF2BP3 association was partially attenuated after accounting for amyloid load. In addition to these discoveries, three single nucleotide polymorphisms within MAPT (microtubule-associated protein tau) displayed nominal associations with tau-positron emission tomography burden, and the association of the APOE (apolipoprotein E) ɛ4 allele with tau-positron emission tomography was marginally nonsignificant (P  = 0.06, β = 0.07). No associations with tau-positron emission tomography burden were identified for other single nucleotide polymorphisms associated with Alzheimer's disease clinical diagnosis in prior large case-control studies. Our findings nominate PPP2R2B and IGF2BP3 as novel potential influences on tau pathology which warrant further functional characterization. Our data are also supportive of previous literature on the associations of MAPT genetic variation with tau, and more broadly supports the inference that tau accumulation may have a genetic architecture distinct from known Alzheimer's susceptibility genes, which may have implications for improved risk stratification and therapeutic targeting.

β-Amyloid (Aβ) pathology is common in patients with probable dementia with Lewy bodies (DLB). However, the pathologic basis and the differential diagnostic performance of Aβ PET are not established in DLB. Our objective was to investigate the pathologic correlates of 11C-Pittsburgh compound B(PiB) uptake on PET in cases with antemortem diagnosis of probable DLB or Lewy body disease (LBD) at autopsy.

Recombinant adeno-associated virus (rAAV) vectors have emerged as the safe vehicles of choice for long-term gene transfer in mammalian nervous system. Recombinant adeno-associated virus-mediated localized gene transfer in adult nervous system following direct inoculation, that is, intracerebral or intrathecal, is well documented. However, recombinant adeno-associated virus delivery in defined neuronal populations in adult animals using less-invasive methods as well as avoiding ectopic gene expression following systemic inoculation remain challenging. Harnessing the capability of some recombinant adeno-associated virus serotypes for retrograde transduction may potentially address such limitations (Note: The term retrograde transduction in this manuscript refers to the uptake of injected recombinant adeno-associated virus particles at nerve terminals, retrograde transport, and subsequent transduction of nerve cell soma). In some studies, recombinant adeno-associated virus serotypes 2/6, 2/8, and 2/9 have been shown to exhibit transduction of connected neuroanatomical tracts in adult animals following lower limb intramuscular recombinant adeno-associated virus delivery in a pattern suggestive of retrograde transduction. However, an extensive side-by-side comparison of these serotypes following intramuscular delivery regarding tissue viral load, and the effect of promoter on transgene expression, has not been performed. Hence, we delivered recombinant adeno-associated virus serotypes 2/6, 2/8, or 2/9 encoding enhanced green fluorescent protein (eGFP), under the control of either cytomegalovirus (CMV) or human synapsin (hSyn) promoter, via a single unilateral hindlimb intramuscular injection in the bicep femoris of adult C57BL/6J mice. Four weeks post injection, we quantified viral load and transgene (enhanced green fluorescent protein) expression in muscle and related nervous tissues. Our data show that the select recombinant adeno-associated virus serotypes transduce sciatic nerve and groups of neurons in the dorsal root ganglia on the injected side, indicating that the intramuscular recombinant adeno-associated virus delivery is useful for achieving gene transfer in local neuroanatomical tracts. We also observed sparse recombinant adeno-associated virus viral delivery or eGFP transduction in lumbar spinal cord and a noticeable lack thereof in brain. Therefore, further improvements in recombinant adeno-associated virus design are warranted to achieve efficient widespread retrograde transduction following intramuscular and possibly other peripheral routes of delivery.

Neurofibrillary tangles are a pathological hallmark of Alzheimer's disease, and their levels correlate with the severity of cognitive dysfunction in humans. However, experimental evidence suggests that soluble tau species cause cognitive deficits and memory impairment. Our recent study suggests that caspase-2 (Casp2)-catalyzed tau cleavage at aspartate 314 mediates synaptic dysfunction and memory impairment in mouse and cellular models of neurodegenerative disorders. Δtau314, the C-terminally-truncated cleavage products, are soluble and present in human brain. In addition, levels of Δtau314 proteins are elevated in the brain of the cognitively impaired individuals compared to the cognitively normal individuals, indicating a possible role for Δtau314 proteins in cognitive deterioration. Here we show that (1) Δtau314 proteins are present in the inferior temporal gyrus of human brains; (2) Δtau314 proteins are generated from all six tau splicing isoforms, (3) levels of both Casp2 and Δtau314 proteins are elevated in cognitively impaired individuals compared to cognitively normal individuals, and (4) levels of Δtau314 proteins show a modest predictive value for dementia. These findings advance our understanding of the characteristics of Δtau314 proteins and their relevance to cognitive dysfunction and shed light on the contribution of Casp2-mediated Δtau314 production to cognitive deterioration.

Parkinson disease (PD) is the second most common neurodegenerative disorder and the leading neurodegenerative cause of motor disability. Pathologic accumulation of aggregated alpha synuclein (AS) protein in brain, and imbalance in the nigrostriatal system due to the loss of dopaminergic neurons in the substantia nigra- pars compacta, are hallmark features in PD. AS aggregation and propagation are considered to trigger neurotoxic mechanisms in PD, including mitochondrial deficits and oxidative stress. The eukaryotic elongation factor-2 kinase (eEF2K) mediates critical regulation of dendritic mRNA translation and is a crucial molecule in diverse forms of synaptic plasticity. Here we show that eEF2K activity, assessed by immuonohistochemical detection of eEF2 phosphorylation on serine residue 56, is increased in postmortem PD midbrain and hippocampus. Induction of aggressive, AS-related motor phenotypes in a transgenic PD M83 mouse model also increased brain eEF2K expression and activity. In cultures of dopaminergic N2A cells, overexpression of wild-type human AS or the A53T mutant increased eEF2K activity. eEF2K inhibition prevented the cytotoxicity associated with AS overexpression in N2A cells by improving mitochondrial function and reduced oxidative stress. Furthermore, genetic deletion of the eEF2K ortholog efk-1 in C. elegans attenuated human A53T AS induced defects in behavioural assays reliant on dopaminergic neuron function. These data suggest a role for eEF2K activity in AS toxicity, and support eEF2K inhibition as a potential target in reducing AS-induced oxidative stress in PD.

Fluorodeoxyglucose positron emission tomography (FDG-PET) imaging based 3D topographic brain glucose metabolism patterns from normal controls (NC) and individuals with dementia of Alzheimer's type (DAT) are used to train a novel multi-scale ensemble classification model. This ensemble model outputs a FDG-PET DAT score (FPDS) between 0 and 1 denoting the probability of a subject to be clinically diagnosed with DAT based on their metabolism profile. A novel 7 group image stratification scheme is devised that groups images not only based on their associated clinical diagnosis but also on past and future trajectories of the clinical diagnoses, yielding a more continuous representation of the different stages of DAT spectrum that mimics a real-world clinical setting. The potential for using FPDS as a DAT biomarker was validated on a large number of FDG-PET images (N=2984) obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database taken across the proposed stratification, and a good classification AUC (area under the curve) of 0.78 was achieved in distinguishing between images belonging to subjects on a DAT trajectory and those images taken from subjects not progressing to a DAT diagnosis. Further, the FPDS biomarker achieved state-of-the-art performance on the mild cognitive impairment (MCI) to DAT conversion prediction task with an AUC of 0.81, 0.80, 0.77 for the 2, 3, 5 years to conversion windows respectively.