Paired plasma lipidomics and proteomics analysis in the conversion from mild cognitive impairment to Alzheimer's disease.

BACKGROUND: Alzheimer's disease (AD) is a neurodegenerative condition for which there is currently no available medication that can stop its progression. Previous studies suggest that mild cognitive impairment (MCI) is a phase that precedes the disease. Therefore, a better understanding of the molecular mechanisms behind MCI conversion to AD is needed. METHOD: Here, we propose a machine learning-based approach to detect the key metabolites and proteins involved in MCI progression to AD using data from the European Medical Information Framework for Alzheimer's Disease Multimodal Biomarker Discovery Study. Proteins and metabolites were evaluated separately in multiclass models (controls, MCI and AD) and together in MCI conversion models (MCI stable vs converter). Only features selected as relevant by 3/4 algorithms proposed were kept for downstream analysis. RESULTS: Multiclass models of metabolites highlighted nine features further validated in an independent cohort (0.726 mean balanced accuracy). Among these features, one metabolite, oleamide, was selected by all the algorithms. Further in-vitro experiments in rodents showed that disease-associated microglia excreted oleamide in vesicles. Multiclass models of proteins stood out with nine features, validated in an independent cohort (0.720 mean balanced accuracy). However, none of the proteins was selected by all the algorithms. Besides, to distinguish between MCI stable and converters, 14 key features were selected (0.872 AUC), including tTau, alpha-synuclein (SNCA), junctophilin-3 (JPH3), properdin (CFP) and peptidase inhibitor 15 (PI15) among others. CONCLUSIONS: This omics integration approach highlighted a set of molecules associated with MCI conversion important in neuronal and glia inflammation pathways.

Plasma proteomic profiles of UK Biobank participants with multiple sclerosis.

OBJECTIVE: We aimed to describe plasma protein biomarkers of multiple sclerosis risk and to explore protein biomarkers of disease severity using radiological outcome measures. METHODS: Multiple sclerosis cases and controls were identified in UK Biobank, a longitudinal cohort study of ~500,000 British adults. Plasma proteins were assayed in ~50,000 UK Biobank participants using the Olink proximity extension assay. We performed case-control association testing to examine the association between 2911 proteins and multiple sclerosis, using linear models adjusted for confounding covariates. Associations with radiological lesion burden and brain volume were determined in a subset of the cohort with available magnetic resonance imaging, using normalized T2-hyperintensity volume or whole brain volume as the outcome measure. RESULTS: In total, 407 prevalent multiple sclerosis cases and 39,979 healthy controls were included. We discovered 72 proteins associated with multiple sclerosis at a Bonferroni-adjusted p value of 0.05, including established markers such as neurofilament light chain and glial fibrillary acidic protein. We observed a decrease in plasma Granzyme A, a marker of T cell and NK cell degranulation, which was specific to multiple sclerosis. Higher levels of plasma proteins involved in coagulation were associated with lower T2 lesion burden and preserved brain volume. INTERPRETATION: We report the largest plasma proteomic screen of multiple sclerosis, replicating important known associations and suggesting novel markers, such as the reduction in granzyme A. While these findings require external validation, they demonstrate the power of biobank-scale datasets for discovering new biomarkers for multiple sclerosis.

Artificial intelligence for biomarker discovery in Alzheimer's disease and dementia.

With the increase in large multimodal cohorts and high-throughput technologies, the potential for discovering novel biomarkers is no longer limited by data set size. Artificial intelligence (AI) and machine learning approaches have been developed to detect novel biomarkers and interactions in complex data sets. We discuss exemplar uses and evaluate current applications and limitations of AI to discover novel biomarkers. Remaining challenges include a lack of diversity in the data sets available, the sheer complexity of investigating interactions, the invasiveness and cost of some biomarkers, and poor reporting in some studies. Overcoming these challenges will involve collecting data from underrepresented populations, developing more powerful AI approaches, validating the use of noninvasive biomarkers, and adhering to reporting guidelines. By harnessing rich multimodal data through AI approaches and international collaborative innovation, we are well positioned to identify clinically useful biomarkers that are accurate, generalizable, unbiased, and acceptable in clinical practice. HIGHLIGHTS: Artificial intelligence and machine learning approaches may accelerate dementia biomarker discovery. Remaining challenges include data set suitability due to size and bias in cohort selection. Multimodal data, diverse data sets, improved machine learning approaches, real-world validation, and interdisciplinary collaboration are required.

Investigating associations between blood metabolites, later life brain imaging measures, and genetic risk for Alzheimer's disease.

BACKGROUND: Identifying blood-based signatures of brain health and preclinical pathology may offer insights into early disease mechanisms and highlight avenues for intervention. Here, we systematically profiled associations between blood metabolites and whole-brain volume, hippocampal volume, and amyloid-ß status among participants of Insight 46-the neuroscience sub-study of the National Survey of Health and Development (NSHD). We additionally explored whether key metabolites were associated with polygenic risk for Alzheimer's disease (AD). METHODS: Following quality control, levels of 1019 metabolites-detected with liquid chromatography-mass spectrometry-were available for 1740 participants at age 60-64. Metabolite data were subsequently clustered into modules of co-expressed metabolites using weighted coexpression network analysis. Accompanying MRI and amyloid-PET imaging data were present for 437 participants (age 69-71). Regression analyses tested relationships between metabolite measures-modules and hub metabolites-and imaging outcomes. Hub metabolites were defined as metabolites that were highly connected within significant (pFDR < 0.05) modules or were identified as a hub in a previous analysis on cognitive function in the same cohort. Regression models included adjustments for age, sex, APOE genotype, lipid medication use, childhood cognitive ability, and social factors. Finally, associations were tested between AD polygenic risk scores (PRS), including and excluding the APOE region, and metabolites and modules that significantly associated (pFDR < 0.05) with an imaging outcome (N = 1638). RESULTS: In the fully adjusted model, three lipid modules were associated with a brain volume measure (pFDR < 0.05): one enriched in sphingolipids (hippocampal volume: ß = 0.14, 95% CI = [0.055,0.23]), one in several fatty acid pathways (whole-brain volume: ß = - 0.072, 95%CI = [- 0.12, - 0.026]), and another in diacylglycerols and phosphatidylethanolamines (whole-brain volume: ß = - 0.066, 95% CI = [- 0.11, - 0.020]). Twenty-two hub metabolites were associated (pFDR < 0.05) with an imaging outcome (whole-brain volume: 22; hippocampal volume: 4). Some nominal associations were reported for amyloid-ß, and with an AD PRS in our genetic analysis, but none survived multiple testing correction. CONCLUSIONS: Our findings highlight key metabolites, with functions in membrane integrity and cell signalling, that associated with structural brain measures in later life. Future research should focus on replicating this work and interrogating causality.

Multiomics profiling of human plasma and cerebrospinal fluid reveals ATN-derived networks and highlights causal links in Alzheimer's disease.

INTRODUCTION: This study employed an integrative system and causal inference approach to explore molecular signatures in blood and CSF, the amyloid/tau/neurodegeneration [AT(N)] framework, mild cognitive impairment (MCI) conversion to Alzheimer's disease (AD), and genetic risk for AD. METHODS: Using the European Medical Information Framework (EMIF)-AD cohort, we measured 696 proteins in cerebrospinal fluid (n = 371), 4001 proteins in plasma (n = 972), 611 metabolites in plasma (n = 696), and genotyped whole-blood (7,778,465 autosomal single nucleotide epolymorphisms, n = 936). We investigated associations: molecular modules to AT(N), module hubs with AD Polygenic Risk scores and APOE4 genotypes, molecular hubs to MCI conversion and probed for causality with AD using Mendelian randomization (MR). RESULTS: AT(N) framework associated with protein and lipid hubs. In plasma, Proprotein Convertase Subtilisin/Kexin Type 7 showed evidence for causal associations with AD. AD was causally associated with Reticulocalbin 2 and sphingomyelins, an association driven by the APOE isoform. DISCUSSION: This study reveals multi-omics networks associated with AT(N) and causal AD molecular candidates.

Genetic risk for attention-deficit/hyperactivity disorder predicts cognitive decline and development of Alzheimer's disease pathophysiology in cognitively unimpaired older adults.

Attention-deficit/hyperactivity disorder (ADHD) persists in older age and is postulated as a risk factor for cognitive impairment and Alzheimer's Disease (AD). However, these findings rely primarily on electronic health records and can present biased estimates of disease prevalence. An obstacle to investigating age-related cognitive decline in ADHD is the absence of large-scale studies following patients with ADHD into older age. Alternatively, this study aimed to determine whether genetic liability for ADHD, as measured by a well-validated ADHD polygenic risk score (ADHD-PRS), is associated with cognitive decline and the development of AD pathophysiology in cognitively unimpaired (CU) older adults. We calculated a weighted ADHD-PRS in 212 CU individuals without a clinical diagnosis of ADHD (55-90 years). These individuals had baseline amyloid-ß (Aß) positron emission tomography, longitudinal cerebrospinal fluid (CSF) phosphorylated tau at threonine 181 (p-tau181), magnetic resonance imaging, and cognitive assessments for up to 6 years. Linear mixed-effects models were used to test the association of ADHD-PRS with cognition and AD biomarkers. Higher ADHD-PRS was associated with greater cognitive decline over 6 years. The combined effect between high ADHD-PRS and brain Aß deposition on cognitive deterioration was more significant than each individually. Additionally, higher ADHD-PRS was associated with increased CSF p-tau181 levels and frontoparietal atrophy in CU Aß-positive individuals. Our results suggest that genetic liability for ADHD is associated with cognitive deterioration and the development of AD pathophysiology. Findings were mostly observed in Aß-positive individuals, suggesting that the genetic liability for ADHD increases susceptibility to the harmful effects of Aß pathology.

Disentangling Independent and Mediated Causal Relationships Between Blood Metabolites, Cognitive Factors, and Alzheimer's Disease.

Background: Education and cognition demonstrate consistent inverse associations with Alzheimer's disease (AD). The biological underpinnings, however, remain unclear. Blood metabolites reflect the end point of biological processes and are accessible and malleable. Identifying metabolites with etiological relevance to AD and disentangling how these relate to cognitive factors along the AD causal pathway could, therefore, offer unique insights into underlying causal mechanisms. Methods: Using data from the largest metabolomics genome-wide association study (N ≈ 24,925) and three independent AD cohorts (N = 4725), cross-trait polygenic scores were generated and meta-analyzed. Metabolites genetically associated with AD were taken forward for causal analyses. Bidirectional two-sample Mendelian randomization interrogated univariable causal relationships between 1) metabolites and AD; 2) education and cognition; 3) metabolites, education, and cognition; and 4) education, cognition, and AD. Mediating relationships were computed using multivariable Mendelian randomization. Results: Thirty-four metabolites were genetically associated with AD at p < .05. Of these, glutamine and free cholesterol in extra-large high-density lipoproteins demonstrated a protective causal effect (glutamine: 95% confidence interval [CI], 0.70 to 0.92; free cholesterol in extra-large high-density lipoproteins: 95% CI, 0.75 to 0.92). An AD-protective effect was also observed for education (95% CI, 0.61 to 0.85) and cognition (95% CI, 0.60 to 0.89), with bidirectional mediation evident. Cognition as a mediator of the education-AD relationship was stronger than vice versa, however. No evidence of mediation via any metabolite was found. Conclusions: Glutamine and free cholesterol in extra-large high-density lipoproteins show protective causal effects on AD. Education and cognition also demonstrate protection, though education's effect is almost entirely mediated by cognition. These insights provide key pieces of the AD causal puzzle, important for informing future multimodal work and progressing toward effective intervention strategies.

Metabolic correlates of late midlife cognitive outcomes: findings from the 1946 British Birth Cohort.

Investigating associations between metabolites and late midlife cognitive function could reveal potential markers and mechanisms relevant to early dementia. Here, we systematically explored the metabolic correlates of cognitive outcomes measured across the seventh decade of life, while untangling influencing life course factors. Using levels of 1019 metabolites profiled by liquid chromatography-mass spectrometry (age 60-64), we evaluated relationships between metabolites and cognitive outcomes in the British 1946 Birth Cohort (N = 1740). We additionally conducted pathway and network analyses to allow for greater insight into potential mechanisms, and sequentially adjusted for life course factors across four models, including sex and blood collection (Model 1), Model 1 + body mass index and lipid medication (Model 2), Model 2 + social factors and childhood cognition (Model 3) and Model 3 + lifestyle influences (Model 4). After adjusting for multiple tests, 155 metabolites, 10 pathways and 5 network modules were associated with cognitive outcomes. Of the 155, 35 metabolites were highly connected in their network module (termed 'hub' metabolites), presenting as promising marker candidates. Notably, we report relationships between a module comprised of acylcarnitines and processing speed which remained robust to life course adjustment, revealing palmitoylcarnitine (C16) as a hub (Model 4: ß = -0.10, 95% confidence interval = -0.15 to -0.052, P = 5.99 × 10-5). Most associations were sensitive to adjustment for social factors and childhood cognition; in the final model, four metabolites remained after multiple testing correction, and 80 at P < 0.05. Two modules demonstrated associations that were partly or largely attenuated by life course factors: one enriched in modified nucleosides and amino acids (overall attenuation = 39.2-55.5%), and another in vitamin A and C metabolites (overall attenuation = 68.6-92.6%). Our other findings, including a module enriched in sphingolipid pathways, were entirely explained by life course factors, particularly childhood cognition and education. Using a large birth cohort study with information across the life course, we highlighted potential metabolic mechanisms associated with cognitive function in late midlife, suggesting marker candidates and life course relationships for further study.

Circulating Metabolome and White Matter Hyperintensities in Women and Men.

BACKGROUND: White matter hyperintensities (WMH), identified on T2-weighted magnetic resonance images of the human brain as areas of enhanced brightness, are a major risk factor of stroke, dementia, and death. There are no large-scale studies testing associations between WMH and circulating metabolites. METHODS: We studied up to 9290 individuals (50.7% female, average age 61 years) from 15 populations of 8 community-based cohorts. WMH volume was quantified from T2-weighted or fluid-attenuated inversion recovery images or as hypointensities on T1-weighted images. Circulating metabolomic measures were assessed with mass spectrometry and nuclear magnetic resonance spectroscopy. Associations between WMH and metabolomic measures were tested by fitting linear regression models in the pooled sample and in sex-stratified and statin treatment-stratified subsamples. Our basic models were adjusted for age, sex, age×sex, and technical covariates, and our fully adjusted models were also adjusted for statin treatment, hypertension, type 2 diabetes, smoking, body mass index, and estimated glomerular filtration rate. Population-specific results were meta-analyzed using the fixed-effect inverse variance-weighted method. Associations with false discovery rate (FDR)-adjusted P values (PFDR)<0.05 were considered significant. RESULTS: In the meta-analysis of results from the basic models, we identified 30 metabolomic measures associated with WMH (PFDR<0.05), 7 of which remained significant in the fully adjusted models. The most significant association was with higher level of hydroxyphenylpyruvate in men (PFDR.full.adj=1.40×10-7) and in both the pooled sample (PFDR.full.adj=1.66×10-4) and statin-untreated (PFDR.full.adj=1.65×10-6) subsample. In men, hydroxyphenylpyruvate explained 3% to 14% of variance in WMH. In men and the pooled sample, WMH were also associated with lower levels of lysophosphatidylcholines and hydroxysphingomyelins and a larger diameter of low-density lipoprotein particles, likely arising from higher triglyceride to total lipids and lower cholesteryl ester to total lipids ratios within these particles. In women, the only significant association was with higher level of glucuronate (PFDR=0.047). CONCLUSIONS: Circulating metabolomic measures, including multiple lipid measures (eg, lysophosphatidylcholines, hydroxysphingomyelins, low-density lipoprotein size and composition) and nonlipid metabolites (eg, hydroxyphenylpyruvate, glucuronate), associate with WMH in a general population of middle-aged and older adults. Some metabolomic measures show marked sex specificities and explain a sizable proportion of WMH variance.

Sex-Specific Metabolic Pathways Were Associated with Alzheimer's Disease (AD) Endophenotypes in the European Medical Information Framework for AD Multimodal Biomarker Discovery Cohort.

BACKGROUND: physiological differences between males and females could contribute to the development of Alzheimer's Disease (AD). Here, we examined metabolic pathways that may lead to precision medicine initiatives. METHODS: We explored whether sex modifies the association of 540 plasma metabolites with AD endophenotypes including diagnosis, cerebrospinal fluid (CSF) biomarkers, brain imaging, and cognition using regression analyses for 695 participants (377 females), followed by sex-specific pathway overrepresentation analyses, APOE ε4 stratification and assessment of metabolites' discriminatory performance in AD. RESULTS: In females with AD, vanillylmandelate (tyrosine pathway) was increased and tryptophan betaine (tryptophan pathway) was decreased. The inclusion of these two metabolites (area under curve (AUC) = 0.83, standard error (SE) = 0.029) to a baseline model (covariates + CSF biomarkers, AUC = 0.92, SE = 0.019) resulted in a significantly higher AUC of 0.96 (SE = 0.012). Kynurenate was decreased in males with AD (AUC = 0.679, SE = 0.046). CONCLUSIONS: metabolic sex-specific differences were reported, covering neurotransmission and inflammation pathways with AD endophenotypes. Two metabolites, in pathways related to dopamine and serotonin, were associated to females, paving the way to personalised treatment.

A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer's disease.

Late-onset Alzheimer's disease is a prevalent age-related polygenic disease that accounts for 50-70% of dementia cases. Currently, only a fraction of the genetic variants underlying Alzheimer's disease have been identified. Here we show that increased sample sizes allowed identification of seven previously unidentified genetic loci contributing to Alzheimer's disease. This study highlights microglia, immune cells and protein catabolism as relevant to late-onset Alzheimer's disease, while identifying and prioritizing previously unidentified genes of potential interest. We anticipate that these results can be included in larger meta-analyses of Alzheimer's disease to identify further genetic variants that contribute to Alzheimer's pathology.

Assessing Genetic Overlap and Causality Between Blood Plasma Proteins and Alzheimer's Disease.

BACKGROUND: Blood plasma proteins have been associated with Alzheimer's disease (AD), but understanding which proteins are on the causal pathway remains challenging. OBJECTIVE: Investigate the genetic overlap between candidate proteins and AD using polygenic risk scores (PRS) and interrogate their causal relationship using bi-directional Mendelian randomization (MR). METHODS: Following a literature review, 31 proteins were selected for PRS analysis. PRS were constructed for prioritized proteins with and without the apolipoprotein E region (APOE+/-PRS) and tested for association with AD status across three cohorts (n = 6,244). An AD PRS was also tested for association with protein levels in one cohort (n = 410). Proteins showing association with AD were taken forward for MR. RESULTS: For APOE ɛ3, apolipoprotein B-100, and C-reactive protein (CRP), protein APOE+ PRS were associated with AD below Bonferroni significance (pBonf, p < 0.00017). No protein APOE- PRS or AD PRS (APOE+/-) passed pBonf. However, vitamin D-binding protein (protein PRS APOE-, p = 0.009) and insulin-like growth factor-binding protein 2 (AD APOE- PRS p = 0.025, protein APOE- PRS p = 0.045) displayed suggestive signals and were selected for MR. In bi-directional MR, none of the five proteins demonstrated a causal association (p < 0.05) in either direction. CONCLUSION: Apolipoproteins and CRP PRS are associated with AD and provide a genetic signal linked to a specific, accessible risk factor. While evidence of causality was limited, this study was conducted in a moderate sample size and provides a framework for larger samples with greater statistical power.

A genome-wide association study of plasma phosphorylated tau181.

Plasma phosphorylated tau at threonine-181 (P-tau181) demonstrates promise as an accessible blood-based biomarker specific to Alzheimer's Disease (AD), with levels recently demonstrating high predictive accuracy for AD-relevant pathology. The genetic underpinnings of P-tau181 levels, however, remain elusive. This study presents the first genome-wide association study of plasma P-tau181 in a total sample of 1153 participants from 2 independent cohorts. No loci, other than those within the APOE genomic region (lead variant = rs429358, beta = 0.32, p =8.44 × 10-25) demonstrated association with P-tau181 at genome-wide significance (p < 5 × 10-08), though rs60872856 on chromosome 2 came close (beta = -0.28, p = 3.23 × 10-07, nearest gene=CYTIP). As the APOE ε4 allele is already a well-established genetic variant associated with AD, this study found no evidence of novel genetic associations relevant to plasma P-tau181, though presents rs60872856 on chromosome 2 as a candidate locus to be further evaluated in future larger size GWAS.

Mendelian randomization identifies blood metabolites previously linked to midlife cognition as causal candidates in Alzheimer's disease.

There are currently no disease-modifying treatments for Alzheimer's disease (AD), and an understanding of preclinical causal biomarkers to help target disease pathogenesis in the earliest phases remains elusive. Here, we investigated whether 19 metabolites previously associated with midlife cognition-a preclinical predictor of AD-translate to later clinical risk, using Mendelian randomization (MR) to tease out AD-specific causal relationships. Summary statistics from the largest genome-wide association studies (GWASs) for AD and metabolites were used to perform bidirectional univariable MR. Bayesian model averaging (BMA) was additionally performed to address high correlation between metabolites and identify metabolite combinations that may be on the AD causal pathway. Univariable MR indicated four extra-large high-density lipoproteins (XL.HDL) on the causal pathway to AD: free cholesterol (XL.HDL.FC: 95% CI = 0.78 to 0.94), total lipids (XL.HDL.L: 95% CI = 0.80 to 0.97), phospholipids (XL.HDL.PL: 95% CI = 0.81 to 0.97), and concentration of XL.HDL particles (95% CI = 0.79 to 0.96), significant at an adjusted P < 0.009. MR-BMA corroborated XL.HDL.FC to be among the top three causal metabolites, in addition to total cholesterol in XL.HDL (XL.HDL.C) and glycoprotein acetyls (GP). Both XL.HDL.C and GP demonstrated suggestive univariable evidence of causality (P < 0.05), and GP successfully replicated within an independent dataset. This study offers insight into the causal relationship between metabolites demonstrating association with midlife cognition and AD. It highlights GP in addition to several XL.HDLs-particularly XL.HDL.FC-as causal candidates warranting further investigation. As AD pathology is thought to develop decades prior to symptom onset, expanding on these findings could inform risk reduction strategies.

Association between polygenic risk score of Alzheimer's disease and plasma phosphorylated tau in individuals from the Alzheimer's Disease Neuroimaging Initiative.

BACKGROUND: Recent studies suggest that plasma phosphorylated tau181 (p-tau181) is a highly specific biomarker for Alzheimer's disease (AD)-related tau pathology. It has great potential for the diagnostic and prognostic evaluation of AD, since it identifies AD with the same accuracy as tau PET and CSF p-tau181 and predicts the development of AD dementia in cognitively unimpaired (CU) individuals and in those with mild cognitive impairment (MCI). Plasma p-tau181 may also be used as a biomarker in studies exploring disease pathogenesis, such as genetic or environmental risk factors for AD-type tau pathology. The aim of the present study was to investigate the relation between polygenic risk scores (PRSs) for AD and plasma p-tau181. METHODS: Data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) was used to examine the relation between AD PRSs, constructed based on findings in recent genome-wide association studies, and plasma p-tau181, using linear regression models. Analyses were performed in the total sample (n = 818), after stratification on diagnostic status (CU (n = 236), MCI (n = 434), AD dementia (n = 148)), and after stratification on Aß pathology status (Aß positives (n = 322), Aß negatives (n = 409)). RESULTS: Associations between plasma p-tau181 and APOE PRSs (p = 3e-18-7e-15) and non-APOE PRSs (p = 3e-4-0.03) were seen in the total sample. The APOE PRSs were associated with plasma p-tau181 in all diagnostic groups (CU, MCI, and AD dementia), while the non-APOE PRSs were associated only in the MCI group. The APOE PRSs showed similar results in amyloid-ß (Aß)-positive and negative individuals (p = 5e-5-1e-3), while the non-APOE PRSs were associated with plasma p-tau181 in Aß positives only (p = 0.02). CONCLUSIONS: Polygenic risk for AD including APOE was found to associate with plasma p-tau181 independent of diagnostic and Aß pathology status, while polygenic risk for AD beyond APOE was associated with plasma p-tau181 only in MCI and Aß-positive individuals. These results extend the knowledge about the relation between genetic risk for AD and p-tau181, and further support the usefulness of plasma p-tau181 as a biomarker of AD.

Urinary metabolic phenotyping for Alzheimer's disease.

Finding early disease markers using non-invasive and widely available methods is essential to develop a successful therapy for Alzheimer's Disease. Few studies to date have examined urine, the most readily available biofluid. Here we report the largest study to date using comprehensive metabolic phenotyping platforms (NMR spectroscopy and UHPLC-MS) to probe the urinary metabolome in-depth in people with Alzheimer's Disease and Mild Cognitive Impairment. Feature reduction was performed using metabolomic Quantitative Trait Loci, resulting in the list of metabolites associated with the genetic variants. This approach helps accuracy in identification of disease states and provides a route to a plausible mechanistic link to pathological processes. Using these mQTLs we built a Random Forests model, which not only correctly discriminates between people with Alzheimer's Disease and age-matched controls, but also between individuals with Mild Cognitive Impairment who were later diagnosed with Alzheimer's Disease and those who were not. Further annotation of top-ranking metabolic features nominated by the trained model revealed the involvement of cholesterol-derived metabolites and small-molecules that were linked to Alzheimer's pathology in previous studies.