Introduction Metabolite, lipid, and lipoprotein lipid profiling can provide novel insights into mechanisms underlying incident dementia and Alzheimer’s disease. Methods We studied eight prospective cohorts with 22,623 participants profiled by nuclear magnetic resonance or mass spectrometry metabolomics. Four cohorts were used for discovery with replication undertaken in the other four to avoid false positives. For metabolites that survived replication, combined association results are presented. Results Over 246,698 person-years, 995 and 745 cases of incident dementia and Alzheimer’s disease were detected, respectively. Three branched-chain amino acids (isoleucine, leucine, and valine), creatinine and two very low density lipoprotein (VLDL)-specific lipoprotein lipid subclasses were associated with lower dementia risk. One high density lipoprotein (HDL; the concentration of cholesterol esters relative to total lipids in large HDL) and one VLDL (total cholesterol to total lipids ratio in very large VLDL) lipoprotein lipid subclass was associated with increased dementia risk. Branched-chain amino acids were also associated with decreased Alzheimer’s disease risk and the concentration of cholesterol esters relative to total lipids in large HDL with increased Alzheimer’s disease risk. Discussion Further studies can clarify whether these molecules play a causal role in dementia pathogenesis or are merely markers of early pathology.

Circulating metabolites were consistently associated with cognition, dementia, and lifestyle factors, opening new avenues for prevention of cognitive decline and dementia.

Data on associations of apolipoproteins A-I and B (apo A-I, apo B) and HDL cholesterol (HDL-C) with dementia and Alzheimer's disease (AD) are conflicting. Our aim was to examine, whether apo B, apoA-I, their ratio, or HDL-C are significant, independent predictors of incident dementia and AD in the general population free of dementia at baseline. We analyzed the results from two Finnish prospective population-based cohort studies in a total of 13,275 subjects aged 25 to 74 years with mainly Caucasian ethnicity. The followup time for both cohorts was 10 years. We used Cox proportional hazards regression to evaluate hazard ratios (HR) for incident dementia (including AD) (n = 220) and for AD (n = 154). Cumulative incidence function (CIF) analysis was also performed to adjust the results for competing risks of death. Adjusted for multiple dementia and AD risk factors, log-transformed apo A-I, log HDL-C, log apo B, and log apo B/A-I ratio were not associated with incident dementia or AD. HDL-C was inversely associated with AD risk when adjusted for competing risks but no other statistically significant associations were observed in the CIF analyses. Apo A-I, HDL-C, apo B, or apo B/A-I ratio were not associated with future dementia or AD. HDL-C was inversely

Objective: Investigation of the clinical potential of extensive phenotype data and machine learning (ML) in the prediction of mortality in acute coronary syndrome (ACS). Methods: The value of ML and extensive clinical data was analyzed in a retrospective registry study of 9066 consecutive ACS patients (January 2007 to October 2017). Main outcome was sixmonth mortality. Prediction models were developed using two ML methods, logistic regression and extreme gradient boosting (xgboost). The models were fitted in training set of patients treated in 2007-2014 and 2017 (81%, n ¼ 7344) and validated in a separate validation set of patients treated in 2015-2016 with full GRACE score data available for comparison of model accuracy (19%, n ¼ 1722). Results: Overall, six-month mortality was 7.3% (n ¼ 660). Several variables were found to be significantly associated with six-month mortality by both ML methods. The xgboost scored the best performance: AUC 0.890 (0.864-0.916). The AUC values for logistic regression and GRACE score were 0.867(0.837-0.897) and 0.822 (0.785-0.859), respectively. The AUC value of xgboost was better when compared to logistic regression (p ¼ .012) and GRACE score (p < .00001). Conclusions: The use of extensive phenotype data and novel machine learning improves prediction of mortality in ACS over traditional GRACE score. KEY MESSAGES The collection of extensive cardiovascular phenotype data from electronic health records as well as from data recorded by physicians can be used highly effectively in prediction of mortality after acute coronary syndrome. Supervised machine learning methods such as logistic regression and extreme gradient boosting using extensive phenotype data significantly outperform conventional risk assessment by the current golden standard GRACE score. Integration of electronic health records and the use of supervised machine learning methods can be easily applied in a single centre level to model the risk of mortality.