Alzheimer's disease (AD) is highly heritable and recent studies have identified over 20 diseaseassociated genomic loci. Yet these only explain a small proportion of the genetic variance, indicating that undiscovered loci remain. Here, we performed the largest genome-wide association study of clinically diagnosed AD and AD-by-proxy (71,880 cases, 383,378 controls). AD-by-proxy, based on parental diagnoses, showed strong genetic correlation with AD (rg=0.81). Meta-analysis identified 29 risk loci, implicating 215 potential causative genes. Associated genes are strongly expressed in immune-related tissues and cell types (spleen, liver and microglia). Gene-set analyses indicate 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 the genetic factors that contribute to AD risk and add novel insights into the neurobiology of AD.

BackgroundAlzheimer’s disease (AD) neuropathology is associated with neuroinflammation, but there are few useful biomarkers. Mutant variants of triggering receptor expressed on myeloid cells 2 (TREM2) have recently been linked to late-onset AD and other neurodegenerative disorders. TREM2, a microglial receptor, is involved in innate immunity. A cleaved fragment, soluble TREM2 (sTREM2), is present in the cerebrospinal fluid (CSF).MethodsWe developed and used a novel enzyme-linked immunosorbent assay to investigate the potential value of CSF sTREM2 as an AD biomarker in two independent cohorts: an AD/mild cognitive impairment (MCI)/control cohort (n = 100) and an AD/control cohort (n = 50).ResultsWe found no significant difference in sTREM2 levels between groups of controls and patients with AD or MCI. However, among all controls there was a positive correlation between sTREM2 and age (Spearman rho = 0.50; p < 0.001; n = 75). In the AD/MCI/control cohort, CSF sTREM2 correlated positively with total Tau (T-tau) (Spearman rho 0.57; p < 0.001; n = 50), phosphorylated Tau (P-tau) (Spearman rho 0.63; p < 0.001; n = 50) and amyloid-β1–42 (Aβ42) (Spearman rho 0.35; p = 0.01; n = 50) in control subjects. Among controls with a CSF Aβ42 above a cut-off value (700 pg/ml) in this cohort, the positive correlation between sTREM2 and Aβ42 was stronger (Spearman rho = 0.44; p = 0.002; n = 46).ConclusionssTREM2 in CSF correlates with aging in controls, and with the neurodegenerative markers CSF T-tau/P-tau among controls who are negative for AD CSF core biomarkers Aβ42, T-tau or P-tau.Electronic supplementary materialThe online version of this article (doi:10.1186/s13195-016-0182-1) contains supplementary material, which is available to authorized users.

IMPORTANCE Cerebral amyloid-β aggregation is an early event in Alzheimer disease (AD). Understanding the association between amyloid aggregation and cognitive manifestation in persons without dementia is important for a better understanding of the course of AD and for the design of prevention trials. OBJECTIVE To investigate whether amyloid-β aggregation is associated with cognitive functioning in persons without dementia. DESIGN, SETTING, AND PARTICIPANTS This cross-sectional study included 2908 participants with normal cognition and 4133 with mild cognitive impairment (MCI) from 53 studies in the multicenter Amyloid Biomarker Study. Normal cognition was defined as having no cognitive concerns for which medical help was sought and scores within the normal range on cognitive tests. Mild cognitive impairment was diagnosed according to published criteria. Study inclusion began in 2013 and is ongoing. Data analysis was performed in January 2017. MAIN OUTCOMES AND MEASURES Global cognitive performance as assessed by the Mini-Mental State Examination (MMSE) and episodic memory performance as assessed by a verbal word learning test. Amyloid aggregation was measured with positron emission tomography or cerebrospinal fluid biomarkers and dichotomized as negative (normal) or positive (abnormal) according to study-specific cutoffs. Generalized estimating equations were used to examine the association between amyloid aggregation and low cognitive scores (MMSE score ≤27 or memory z score≤−1.28) and to assess whether this association was moderated by age, sex, educational level, or apolipoprotein E genotype. RESULTS Among 2908 persons with normal cognition (mean [SD] age, 67.4 [12.8] years), amyloid positivity was associated with low memory scores after age 70 years (mean difference in amyloid positive vs negative, 4% [95% CI, 0%–7%] at 72 years and 21% [95% CI, 10%–33%] at 90 years) but was not associated with low MMSE scores (mean difference, 3% [95% CI, −1% to 6%], P = .16). Among 4133 patients with MCI (mean [SD] age, 70.2 [8.5] years), amyloid positivity was associated with low memory (mean difference, 16% [95% CI, 12%–20%], P < .001) and low MMSE (mean difference, 14% [95% CI, 12%–17%], P < .001) scores, and this association decreased with age. Low cognitive scores had limited utility for screening of amyloid positivity in persons with normal cognition and those with MCI. In persons with normal cognition, the age-related increase in low memory score paralleled the age-related increase in amyloid positivity with an intervening period of 10 to 15 years. CONCLUSIONS AND RELEVANCE Although low memory scores are an early marker of amyloid positivity, their value as a screening measure for early AD among persons without dementia is limited.

Abstract. While APOE ε4 is the major genetic risk factor for Alzheimer's disease (AD), amyloid dysmetabolism is an initial or early event predicting clinical disease and is an important focus for secondary intervention trials. To improve identification of cases with increased AD risk, we evaluated recruitment procedures using pathological CSF concentrations of A␤ 42 (pA␤) and APOE ε4 as risk markers in a multi-center study in Norway. In total, 490 subjects aged 40-80 y were included after response to advertisements and media coverage or memory clinics referrals. Controls (n = 164) were classified as normal controls without first-degree relatives with dementia (NC), normal controls with first-degree relatives with dementia (NCFD), or controls scoring below norms on cognitive screening. Patients (n = 301) were classified as subjective cognitive decline or * Correspondence to: Lene Pålhaugen, P.B. 1000, N-1478 Lørenskog, Norway. Tel.: +47 95832775; E-mail: lene. palhaugen@gmail.com. 98 T. Fladby et al. / Detecting At-Risk AD Casesmild cognitive impairment. Subjects underwent a clinical and cognitive examination and MRI according to standardized protocols. Core biomarkers in CSF from 411 and APOE genotype from 445 subjects were obtained. Cases (both self-referrals (n = 180) and memory clinics referrals (n = 87)) had increased fractions of pA␤ and APOE ε4 frequency compared to NC. Also, NCFD had higher APOE ε4 frequencies without increased fraction of pA␤ compared to NC, and cases recruited from memory clinics had higher fractions of pA␤ and APOE ε4 frequency than self-referred. This study shows that memory clinic referrals are pA␤ enriched, whereas self-referred and NCFD cases more frequently are pA␤ negative but at risk (APOE ε4 positive), suitable for primary intervention.

We performed a genome-wide association study of total hip replacements, based on variants identified through whole-genome sequencing, which included 4,657 Icelandic patients and 207,514 population controls. We discovered two rare signals that strongly associate with osteoarthritis total hip replacement: a missense variant, c.1141G>C (p.Asp369His), in the COMP gene (allelic frequency = 0.026%, P = 4.0 × 10, odds ratio (OR) = 16.7) and a frameshift mutation, rs532464664 (p.Val330Glyfs*106), in the CHADL gene that associates through a recessive mode of inheritance (homozygote frequency = 0.15%, P = 4.5 × 10, OR = 7.71). On average, c.1141G>C heterozygotes and individuals homozygous for rs532464664 had their hip replacement operation 13.5 years and 4.9 years earlier than others (P = 0.0020 and P = 0.0026), respectively. We show that the full-length CHADL transcript is expressed in cartilage. Furthermore, the premature stop codon introduced by the CHADL frameshift mutation results in nonsense-mediated decay of the mutant transcripts.

White matter hyperintensities (WMHs) are associated with vascular risk and Alzheimer’s disease. In this study, we examined relations between WMH load and distribution, amyloid pathology and vascular risk in 339 controls and cases with either subjective (SCD) or mild cognitive impairment (MCI). Regional deep (DWMH) and periventricular (PWMH) WMH loads were determined using an automated algorithm. We stratified on Aβ1-42 pathology (Aβ+/−) and analyzed group differences, as well as associations with Framingham Risk Score for cardiovascular disease (FRS-CVD) and age. Occipital PWMH ( p = 0.001) and occipital DWMH ( p = 0.003) loads were increased in SCD-Aβ+ compared with Aβ− controls. In MCI-Aβ+ compared with Aβ− controls, there were differences in global WMH ( p = 0.003), as well as occipital DWMH ( p = 0.001) and temporal DWMH ( p = 0.002) loads. FRS-CVD was associated with frontal PWMHs ( p = 0.003) and frontal DWMHs ( p = 0.005), after adjusting for age. There were associations between global and all regional WMH loads and age. In summary, posterior WMH loads were increased in SCD-Aβ+ and MCI-Aβ+ cases, whereas frontal WMHs were associated with vascular risk. The differences in WMH topography support the use of regional WMH load as an early-stage marker of etiology.

The APOE ε4 prevalence in AD was higher than that in previous studies, which did not require presence of Aβ pathology. Furthermore, our results highlight disease heterogeneity related to age and geographical location.

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.