Deep learning, a state-of-the-art machine learning approach, has shown outstanding performance over traditional machine learning in identifying intricate structures in complex high-dimensional data, especially in the domain of computer vision. The application of deep learning to early detection and automated classification of Alzheimer’s disease (AD) has recently gained considerable attention, as rapid progress in neuroimaging techniques has generated large-scale multimodal neuroimaging data. A systematic review of publications using deep learning approaches and neuroimaging data for diagnostic classification of AD was performed. A PubMed and Google Scholar search was used to identify deep learning papers on AD published between January 2013 and July 2018. These papers were reviewed, evaluated, and classified by algorithm and neuroimaging type, and the findings were summarized. Of 16 studies meeting full inclusion criteria, 4 used a combination of deep learning and traditional machine learning approaches, and 12 used only deep learning approaches. The combination of traditional machine learning for classification and stacked auto-encoder (SAE) for feature selection produced accuracies of up to 98.8% for AD classification and 83.7% for prediction of conversion from mild cognitive impairment (MCI), a prodromal stage of AD, to AD. Deep learning approaches, such as convolutional neural network (CNN) or recurrent neural network (RNN), that use neuroimaging data without preprocessing for feature selection have yielded accuracies of up to 96.0% for AD classification and 84.2% for MCI conversion prediction. The best classification performance was obtained when multimodal neuroimaging and fluid biomarkers were combined. Deep learning approaches continue to improve in performance and appear to hold promise for diagnostic classification of AD using multimodal neuroimaging data. AD research that uses deep learning is still evolving, improving performance by incorporating additional hybrid data types, such as –omics data, increasing transparency with explainable approaches that add knowledge of specific disease-related features and mechanisms.

The metabolic basis of Alzheimer disease (AD) is poorly understood, and the relationships between systemic abnormalities in metabolism and AD pathogenesis are unclear. Understanding how global perturbations in metabolism are related to severity of AD neuropathology and the eventual expression of AD symptoms in at-risk individuals is critical to developing effective disease-modifying treatments. In this study, we undertook parallel metabolomics analyses in both the brain and blood to identify systemic correlates of neuropathology and their associations with prodromal and preclinical measures of AD progression. Quantitative and targeted metabolomics (Biocrates AbsoluteIDQ [identification and quantification] p180) assays were performed on brain tissue samples from the autopsy cohort of the Baltimore Longitudinal Study of Aging (BLSA) (N = 44, mean age = 81.33, % female = 36.36) from AD (N = 15), control (CN; N = 14), and “asymptomatic Alzheimer’s disease” (ASYMAD, i.e., individuals with significant AD pathology but no cognitive impairment during life; N = 15) participants. Using machine-learning methods, we identified a panel of 26 metabolites from two main classes—sphingolipids and glycerophospholipids—that discriminated AD and CN samples with accuracy, sensitivity, and specificity of 83.33%, 86.67%, and 80%, respectively. We then assayed these 26 metabolites in serum samples from two well-characterized longitudinal cohorts representing prodromal (Alzheimer’s Disease Neuroimaging Initiative [ADNI], N = 767, mean age = 75.19, % female = 42.63) and preclinical (BLSA) (N = 207, mean age = 78.68, % female = 42.63) AD, in which we tested their associations with magnetic resonance imaging (MRI) measures of AD-related brain atrophy, cerebrospinal fluid (CSF) biomarkers of AD pathology, risk of conversion to incident AD, and trajectories of cognitive performance. We developed an integrated blood and brain endophenotype score that summarized the relative importance of each metabolite to severity of AD pathology and disease progression (Endophenotype Association Score in Early Alzheimer’s Disease [EASE-AD]). Finally, we mapped the main metabolite classes emerging from our analyses to key biological pathways implicated in AD pathogenesis. We found that distinct sphingolipid species including sphingomyelin (SM) with acyl residue sums C16:0, C18:1, and C16:1 (SM C16:0, SM C18:1, SM C16:1) and hydroxysphingomyelin with acyl residue sum C14:1 (SM (OH) C14:1) were consistently associated with severity of AD pathology at autopsy and AD progression across prodromal and preclinical stages. Higher log-transformed blood concentrations of all four sphingolipids in cognitively normal individuals were significantly associated with increased risk of future conversion to incident AD: SM C16:0 (hazard ratio [HR] = 4.430, 95% confidence interval [CI] = 1.703–11.520, p = 0.002), SM C16:1 (HR = 3.455, 95% CI = 1.516–7.873, p = 0.003), SM (OH) C14:1 (HR = 3.539, 95% CI = 1.373–9.122, p = 0.009), and SM C18:1 (HR = 2.255, 95% CI = 1.047–4.855, p = 0.038). The sphingolipid species identified map to several biologically relevant pathways implicated in AD, including tau phosphorylation, amyloid-β (Aβ) metabolism, calcium homeostasis, acetylcholine biosynthesis, and apoptosis. Our study has limitations: the relatively small number of brain tissue samples may have limited our power to detect significant associations, control for heterogeneity between groups, and replicate our findings in independent, autopsy-derived brain samples. We present a novel framework to identify biologically relevant brain and blood metabolites associated with disease pathology and progression during the prodromal and preclinical stages of AD. Our results show that perturbations in sphingolipid metabolism are consistently associated with endophenotypes across preclinical and prodromal AD, as well as with AD pathology at autopsy. Sphingolipids may be biologically relevant biomarkers for the early detection of AD, and correcting perturbations in sphingolipid metabolism may be a plausible and novel therapeutic strategy in AD.

Background: While Alzheimer disease (AD) and vascular dementia (VaD) may be accelerated by hypercholesterolemia, the mechanisms underlying this association are unclear. We tested whether dysregulation of cholesterol catabolism, through its conversion to primary bile acids (BAs), was associated with dementia pathogenesis.Methods and findings: We used a 3-step study design to examine the role of the primary BAs, cholic acid (CA), and chenodeoxycholic acid (CDCA) as well as their principal biosynthetic precursor, 7α-hydroxycholesterol (7α-OHC), in dementia. In Step 1, we tested whether serum markers of cholesterol catabolism were associated with brain amyloid accumulation, white matter lesions (WMLs), and brain atrophy. In Step 2, we tested whether exposure to bile acid sequestrants (BAS) was associated with risk of dementia. In Step 3, we examined plausible mechanisms underlying these findings by testing whether brain levels of primary BAs and gene expression of their principal receptors are altered in AD. Step 1: We assayed serum concentrations CA, CDCA, and 7α-OHC and used linear regression and mixed effects models to test their associations with brain amyloid accumulation (N= 141), WMLs, and brain atrophy (N= 134) in the Baltimore Longitudinal Study of Aging (BLSA). The BLSA is an ongoing, community-based cohort study that began in 1958. Participants in the BLSA neuroimaging sample were approximately 46% male with a mean age of 76 years; longitudinal analyses included an average of 2.5 follow-up magnetic resonance imaging (MRI) visits. We used the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (N= 1,666) to validate longitudinal neuroimaging results in BLSA. ADNI is an ongoing, community-based cohort study that began in 2003. Participants were approximately 55% male with a mean age of 74 years; longitudinal analyses included an average of 5.2 follow-up MRI visits. Lower serum concentrations of 7α-OHC, CA, and CDCA were associated with higher brain amyloid deposition (p= 0.041), faster WML accumulation (p= 0.050), and faster brain atrophy mainly (false discovery rate [FDR]p= <0.001–0.013) in males in BLSA. In ADNI, we found a modest sex-specific effect indicating that lower serum concentrations of CA and CDCA were associated with faster brain atrophy (FDRp= 0.049) in males. Step 2: In the Clinical Practice Research Datalink (CPRD) dataset, covering >4 million registrants from general practice clinics in the United Kingdom, we tested whether patients using BAS (BAS users; 3,208 with ≥2 prescriptions), which reduce circulating BAs and increase cholesterol catabolism, had altered dementia risk compared to those on non-statin lipid-modifying therapies (LMT users; 23,483 with ≥2 prescriptions). Patients in the study (BAS/LMT) were approximately 34%/38% male and with a mean age of 65/68 years; follow-up time was 4.7/5.7 years. We found that BAS use was not significantly associated with risk of all-cause dementia (hazard ratio (HR) = 1.03, 95% confidence interval (CI) = 0.72–1.46,p= 0.88) or its subtypes. We found a significant difference between the risk of VaD in males compared to females (p= 0.040) and a significant dose–response relationship between BAS use and risk of VaD (p-trend = 0.045) in males. Step 3: We assayed brain tissue concentrations of CA and CDCA comparing AD and control (CON) samples in the BLSA autopsy cohort (N= 29). Participants in the BLSA autopsy cohort (AD/CON) were approximately 50%/77% male with a mean age of 87/82 years. We analyzed single-cell RNA sequencing (scRNA-Seq) data to compare brain BA receptor gene expression between AD and CON samples from the Religious Orders Study and Memory and Aging Project (ROSMAP) cohort (N= 46). ROSMAP is an ongoing, community-based cohort study that began in 1994. Participants (AD/CON) were approximately 56%/36% male with a mean age of 85/85 years. In BLSA, we found that CA and CDCA were detectable in postmortem brain tissue samples and were marginally higher in AD samples compared to CON. In ROSMAP, we found sex-specific differences in altered neuronal gene expression of BA receptors in AD. Study limitations include the small sample sizes in the BLSA cohort and likely inaccuracies in the clinical diagnosis of dementia subtypes in primary care settings. Conclusions: We combined targeted metabolomics in serum and amyloid positron emission tomography (PET) and MRI of the brain with pharmacoepidemiologic analysis to implicate dysregulation of cholesterol catabolism in dementia pathogenesis. We observed that lower serum BA concentration mainly in males is associated with neuroimaging markers of dementia, and pharmacological lowering of BA levels may be associated with higher risk of VaD in males. We hypothesize that dysregulation of BA signaling pathways in the brain may represent a plausible biologic mechanism underlying these results. Together, our observations suggest a novel mechanism relating abnormalities in cholesterol catabolism to risk of dementia.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline, robust microgliosis, neuroinflammation, and neuronal loss. Genome-wide association studies recently highlighted a prominent role for microglia in late-onset AD (LOAD). Specifically, inositol polyphosphate-5-phosphatase (INPP5D), also known as SHIP1, is selectively expressed in brain microglia and has been reported to be associated with LOAD. Although INPP5D is likely a crucial player in AD pathophysiology, its role in disease onset and progression remains unclear. We performed differential gene expression analysis to investigate INPP5D expression in AD and its association with plaque density and microglial markers using transcriptomic (RNA-Seq) data from the Accelerating Medicines Partnership for Alzheimer's Disease (AMP-AD) cohort. We also performed quantitative real-time PCR, immunoblotting, and immunofluorescence assays to assess INPP5D expression in the 5xFAD amyloid mouse model. Differential gene expression analysis found that INPP5D expression was upregulated in LOAD and positively correlated with amyloid plaque density. In addition, in 5xFAD mice, Inpp5d expression increased as the disease progressed, and selectively in plaque-associated microglia. Increased Inpp5d expression levels in 5xFAD mice were abolished entirely by depleting microglia with the colony-stimulating factor receptor-1 antagonist PLX5622. Our findings show that INPP5D expression increases as AD progresses, predominantly in plaque-associated microglia. Importantly, we provide the first evidence that increased INPP5D expression might be a risk factor in AD, highlighting INPP5D as a potential therapeutic target. Moreover, we have shown that the 5xFAD mouse model is appropriate for studying INPP5D in AD.

Background: Alzheimer’s disease (AD) is characterized by robust microgliosis and phenotypic changes that accompany disease pathogenesis. Accumulating evidence from genetic studies suggests the importance of phospholipase C γ 2 (PLCG2) in late-onset AD (LOAD) pathophysiology. However, the role ofPLCG2in AD is still poorly understood.Methods: Using bulk RNA-Seq (N=1249) data from the Accelerating Medicines Partnership-Alzheimer’s Disease Consortium (AMP-AD), we investigated whetherPLCG2expression increased in the brains of LOAD patients. We also evaluated the relationship betweenPLCG2expression levels, amyloid plaque density, and expression levels of microglia specific markers (AIF1 and TMEM119). Finally, we investigated the longitudinal changes ofPLCG2expression in the 5xFAD mouse model of AD. To further understand the role ofPLCG2in different signaling pathways, differential gene expression and co-expression network analyses were performed using bulk RNA-Seq and microglial single-cell RNA-Seq data. To substantiate the human analyses, we performed differential gene expression analysis on wild-type (WT) and inactivatedPlcg2mice and used immunostaining to determine if the differentially expressed genes/pathways were altered by microglial cell coverage or morphology.Results: We observed significant upregulation ofPLCG2expression in three brain regions of LOAD patients and significant positive correlation ofPLCG2expression with amyloid plaque density. These findings in the human brain were validated in the 5xFAD amyloid mouse model, which showed disease progression-dependent increases inPlcg2expression associated with amyloid pathology. Of note, increasedPlcg2expression levels in 5xFAD mice were abolished by reducing microglia. Furthermore, using bulk RNA-Seq data, we performed differential expression analysis by comparing cognitively normal older adults (CN) with 75th percentile (high) and 25th percentile (low)PLCG2gene expression levels to identify pathways related to inflammation and the inflammatory response. The findings in the human brain were validated by differential expression analyses between WT andplcg2inactivated mice.PLCG2co-expression network analysis of microglial single-cell RNA-Seq data identified pathways related to the inflammatory response including regulation of I-kappaB/NF-kappa B signaling and response to lipopolysaccharide.Conclusions: Our results provide further evidence thatPLCG2plays an important role in AD pathophysiology and may be a potential target for microglia-targeted AD therapies.