ASCVD risk assessment in T2DM facilitated by novel computational immunology programs

Project Summary Type 2 diabetes mellitus (T2DM) is a primary cause of the marked increase of ASCVD, accounting for over 30% of symptomatic cases. Accordingly, the ASCVD risk in T2DM patients is 2-4 fold more than in the non-T2DM population. The concomitant increase of T2DM and ASCVD in the US prompted the urgency for accurate ASCVD prediction followed by an effective mitigation strategy. However, recent clinical trials and extensive cohort studies revealed underachievement by controlling conventional lipid/hypertension/glycemic factors, warranting new directions additional to the existing mitigation paradigm. Furthermore, multiple studies comparing existing ASCVD risk prediction models, including those derived from general or T2DM populations, achieved only marginal or underperformance in multiple independent cohorts, endorsing the significance of this significant unmet biomedical challenge. Our proposed study aims to fill this critical scientific and clinical gap in increased atherosclerosis cardiovascular disease risk (ASCVD) in T2DM. Supported by our recent discoveries and development of multiple immune cell annotation tools, we propose to test the overarching hypothesis that the foaming process in circulating monocytes can be directly altered by T2DM towards pathogenic foaming, thus directly contributing to the elevated risk of ASCVD. Application of our newly developed cell function annotation tools uncovered distinct macrophage-derived foam cell development programs, followed by ASCVD causal gene signature identification and proof-of-concept ASCVD predictive modeling by incorporating a selected gene list using the original machine learning program. Our new work further revealed that super-networks governed by three upstream regulators are uniquely altered by T2DM and associated with future ASCVD incidence. Hence, in two aims: (AIM 1) we will determine the impact of suppressed super-networks governed by identified master regulators (SNW1, NCOR2, CITED2) on pathogenic foaming program using established foam cell development system with quantitative methods and single cell transcriptomics; (AIM 2) will assess the efficacy of T2DM- ASCVD signatures on predictive modeling using existing and innovative feature selection and feature extraction tools. Completing this project will provide crucial information to significantly advance our understanding of ASCVD risk in the T2DM population. Our innovative computational programs, derived from novel and in-depth mechanistic investigation, will offer translatable ASCVD prediction tools and molecular targets for drug development to achieve personalized intervention for T2DM patients to mitigate ASCVD risks.