Discerning the molecular mediators between cardiometabolic risk factors and Alzheimer's disease - PROJECT SUMMARY Cardiometabolic traits have long been associated with Alzheimer’s disease (AD) risk, but there is a funda- mental gap in our understanding of the molecular pathways that link these conditions. This knowledge gap hin- ders efforts to develop new biomarkers and therapeutics targeting these pathways. The long-term goal of this application is to improve the diagnosis, prognosis, and treatment of AD by identifying molecular markers for AD and its risk factors. The objective is to identify markers of cardiometabolic pathways that contribute to AD in the largest available cohorts. The hypothesis is that plasma proteins and metabolites mediate the effects of cardi- ometabolic pathways on AD risk. The rationale is that identifying these molecular mediators will be a key step in enabling the development of prognostic panels and new AD treatments. This proposal leverages large-scale omics data from the UK Biobank and Million Veteran Program (MVP) to tackle the following specific aims: 1) identify plasma protein mediators of cardiometabolic traits and AD; 2) identify plasma metabolite mediators of these pathways; and 3) identify genomic evidence for a causal effect of cardiometabolic traits on AD. Aim 1, will use plasma proteomic data from the UK Biobank for a mediation analysis to identify protein mediators of cardi- ometabolic-AD risk, with replication in MVP and two leading AD cohorts. Genomic data will be used in a Mende- lian Randomization (MR)-based analysis to test the causal directions of these protein effects. Aim 2 will analyze plasma metabolite data from the UK Biobank using a similar mediation and MR analysis (replication in MVP and an AD cohort), with an added multiomic integration analysis incorporating proteomics data. Aim 3 will use gen- otype data in an MR analysis to estimate the causal effect of each cardiometabolic trait on AD, with replication across multiple populations in MVP and with external genomic summary statistics. The innovation is the breadth of molecular data analyzed (thousands of markers across three omic data types), depth of the sample (48,500 proteomic, 260,208 metabolomic, and 439,947 genomic samples in UK Biobank), and translatability (multiple populations). The significance of this proposal lies in its ability to identify the molecular mediators of cardiomet- abolic-AD risk across various populations, providing robust preliminary evidence for the follow-up analysis of these molecules as potential prognostic biomarkers. This K01 proposal provides training in the physiology of cardiometabolic traits, biobank genotyping/phenotyping, and imaging, cognitive, and brain measures of AD, which is critical for Dr. Daniel Panyard’s transition to research independence. His mentoring team, led by Dr. Themistocles Assimes, includes leading experts in AD (Roussos, Cruchaga, Hohman; advisors), cardiometa- bolic disease and biobank-scale omic data analysis (Tsao, co-mentor), and multiomics statistical methodology (Tang, advisor). This robust research training plan will position Dr. Panyard as a leading scientist, prepared to secure R-level funding and launch an independent career as a molecular epidemiologist in AD and aging.
