Eleven susceptibility loci for late-onset Alzheimer’s disease (LOAD) were identified by previous studies; however, a large portion of the genetic risk for this disease remains unexplained. We conducted a large, two-stage meta-analysis of genome-wide association studies (GWAS) in individuals of European ancestry. In stage 1, we used genotyped and imputed data (7,055,881 SNPs) to perform meta-analysis on 4 previously published GWAS data sets consisting of 17,008 Alzheimer’s disease cases and 37,154 controls. In stage 2,11,632 SNPs were genotyped and tested for association in an independent set of 8,572 Alzheimer’s disease cases and 11,312 controls. In addition to the APOE locus (encoding apolipoprotein E), 19 loci reached genome-wide significance (P < 5 × 10−8) in the combined stage 1 and stage 2 analysis, of which 11 are newly associated with Alzheimer’s disease.

We undertook a two-stage genome-wide association study of Alzheimer's disease involving over 16,000 individuals. In stage 1 (3,941 cases and 7,848 controls), we replicated the established association with the APOE locus (most significant SNP: rs2075650, p= 1.8×10−157) and observed genome-wide significant association with SNPs at two novel loci: rs11136000 in the CLU or APOJ gene (p= 1.4×10−9) and rs3851179, a SNP 5′ to the PICALM gene (p= 1.9×10−8). Both novel associations were supported in stage 2 (2,023 cases and 2,340 controls), producing compelling evidence for association with AD in the combined dataset (rs11136000: p= 8.5×10−10, odds ratio= 0.86; rs3851179: p= 1.3×10−9, odds ratio= 0.86). We also observed more variants associated at p< 1×10−5 than expected by chance (p=7.5×10−6), including polymorphisms at the BIN1, DAB1 and CR1 loci.

A blood-based protein biomarker, or set of protein biomarkers, that could predict onset and progression of Alzheimer's disease (AD) would have great utility; potentially clinically, but also for clinical trials and especially in the selection of subjects for preventative trials. We reviewed a comprehensive list of 21 published discovery or panel-based (> 100 proteins) blood proteomics studies of AD, which had identified a total of 163 candidate biomarkers. Few putative blood-based protein biomarkers replicate in independent studies but we found that some proteins do appear in multiple studies; for example, four candidate biomarkers are found to associate with AD-related phenotypes in five independent research cohorts in these 21 studies: α-1-antitrypsin, α-2-macroglobulin, apolipoprotein E, and complement C3. Using SomaLogic's SOMAscan proteomics technology, we were able to conduct a large-scale replication study for 94 of the 163 candidate biomarkers from these 21 published studies in plasma samples from 677 subjects from the AddNeuroMed (ANM) and the Alzheimer's Research UK/Maudsley BRC Dementia Case Registry at King's Health Partners (ARUK/DCR) research cohorts. Nine of the 94 previously reported candidates were found to associate with AD-related phenotypes (False Discovery Rate (FDR) q-value < 0.1). These proteins show sufficient replication to be considered for further investigation as a biomarker set. Overall, we show that there are some signs of a replicable signal in the range of proteins identified in previous studies and we are able to further replicate some of these. This suggests that AD pathology does affect the blood proteome with some consistency.

Biomarkers for Alzheimer's disease (AD) based on non-invasive methods are highly desirable for diagnosis, disease progression, and monitoring therapeutics. We aimed to study the use of hippocampal volume, entorhinal cortex (ERC) thickness, and whole brain volume (WBV) as predictors of cognitive change in patients with AD. 120 AD subjects, 106 mild cognitive impairment (MCI), and 99 non demented controls (NDC) from the multi-center pan-European AddNeuroMed study underwent MRI scanning at baseline and clinical evaluations at quarterly follow-up up to 1 year. The rate of cognitive decline was estimated using cognitive outcomes, Mini-Mental State Examination (MMSE) and Alzheimer disease assessment scale-cognitive (ADAS-cog) by fitting a random intercept and slope model. AD subjects had smaller ERC thickness and hippocampal and WBV volumes compared to MCI and NDC subjects. Within the AD group, ERC> WBV was significantly associated with baseline cognition (MMSE, ADAS-cog) and disease severity (Clinical Dementia Rating). Baseline ERC thickness was associated with both longitudinal MMSE and ADAS-cog score changes and WBV with ADAS-cog decline. These data indicate that AD subjects with thinner ERC had lower baseline cognitive scores, higher disease severity, and predicted greater subsequent cognitive decline at one year follow up. ERC is a region known to be affected early in the disease. Therefore, the rate of atrophy in this structure is expected to be higher since neurodegeneration begins earlier. Focusing on structural analyses that predict decline can identify those individuals at greatest risk for future cognitive loss. This may have potential for increasing the efficacy of early intervention.

A marker of Alzheimer's disease (AD) that can accurately diagnose disease at the earliest stage would significantly support efforts to develop treatments for early intervention. We have sought to determine the sensitivity and specificity of peripheral blood gene expression as a diagnostic marker of AD using data generated on HT-12v3 BeadChips. We first developed an AD diagnostic classifier in a training cohort of 78 AD and 78 control blood samples and then tested its performance in a validation group of 26 AD and 26 control and 118 mild cognitive impairment (MCI) subjects who were likely to have an AD-endpoint. A 48 gene classifier achieved an accuracy of 75% in the AD and control validation group. Comparisons were made with a classifier developed using structural MRI measures, where both measures were available in the same individuals. In AD and control subjects, the gene expression classifier achieved an accuracy of 70% compared to 85% using MRI. Bootstrapping validation produced expression and MRI classifiers with mean accuracies of 76% and 82%, respectively, demonstrating better concordance between these two classifiers than achieved in a single validation population. We conclude there is potential for blood expression to be a marker for AD. The classifier also predicts a large number of people with MCI, who are likely to develop AD, are more AD-like than normal with 76% of subjects classified as AD rather than control. Many of these people do not have overt brain atrophy, which is known to emerge around the time of AD diagnosis, suggesting the expression classifier may detect AD earlier in the prodromal phase. However, we accept these results could also represent a marker of diseases sharing common etiology.

Introduction Machine learning (ML) may harbor the potential to capture the metabolic complexity in Alzheimer Disease (AD). Here we set out to test the performance of metabolites in blood to categorize AD when compared to CSF biomarkers. Methods This study analyzed samples from 242 cognitively normal (CN) people and 115 with AD‐type dementia utilizing plasma metabolites (n = 883). Deep Learning (DL), Extreme Gradient Boosting (XGBoost) and Random Forest (RF) were used to differentiate AD from CN. These models were internally validated using Nested Cross Validation (NCV). Results On the test data, DL produced the AUC of 0.85 (0.80–0.89), XGBoost produced 0.88 (0.86–0.89) and RF produced 0.85 (0.83–0.87). By comparison, CSF measures of amyloid, p‐tau and t‐tau (together with age and gender) produced with XGBoost the AUC values of 0.78, 0.83 and 0.87, respectively. Discussion This study showed that plasma metabolites have the potential to match the AUC of well‐established AD CSF biomarkers in a relatively small cohort. Further studies in independent cohorts are needed to validate whether this specific panel of blood metabolites can separate AD from controls, and how specific it is for AD as compared with other neurodegenerative disorders.

A growing number of epigenome-wide association studies have demonstrated a role for DNA methylation in the brain in Alzheimer's disease. With the aim of exploring peripheral biomarker potential, we have examined DNA methylation patterns in whole blood collected from 284 individuals in the AddNeuroMed study, which included 89 nondemented controls, 86 patients with Alzheimer's disease, and 109 individuals with mild cognitive impairment, including 38 individuals who progressed to Alzheimer's disease within 1 year. We identified significant differentially methylated regions, including 12 adjacent hypermethylated probes in the HOXB6 gene in Alzheimer's disease, which we validated using pyrosequencing. Using weighted gene correlation network analysis, we identified comethylated modules of genes that were associated with key variables such as APOE genotype and diagnosis. In summary, this study represents the first large-scale epigenome-wide association study of Alzheimer's disease and mild cognitive impairment using blood. We highlight the differences in various loci and pathways in early disease, suggesting that these patterns relate to cognitive decline at an early stage.

Late onset Alzheimer’s disease (AD) is the most common form of dementia with more than 35 million people affected worldwide, and no curative treatment available. AD is highly heritable and recent genome-wide meta-analyses have identified over 20 genomic loci associated with AD, yet only explaining a small proportion of the genetic variance indicating that undiscovered loci exist. Here, we performed the largest genome-wide association study of clinically diagnosed AD and AD-by-proxy (71,880 AD cases, 383,378 controls). AD-by-proxy status is based on parental AD diagnosis, and showed strong genetic correlation with AD (rg=0.81). Genetic meta analysis identified 29 risk loci, of which 9 are novel, and implicating 215 potential causative genes. Independent replication further supports these novel loci in AD. Associated genes are strongly expressed in immune-related tissues and cell types (spleen, liver and microglia). Furthermore, gene-set analyses indicate the genetic contribution of biological mechanisms involved in lipid-related processes and degradation of amyloid precursor proteins. We show strong genetic correlations with multiple health-related outcomes, and Mendelian randomisation results suggest a protective effect of cognitive ability on AD risk. These results are a step forward in identifying more of the genetic factors that contribute to AD risk and add novel insights into the neurobiology of AD to guide new drug development.