WITHDRAWN: Comparison of Leucocyte Telomere Length, Atherosclerotic Cardiovascular disease risk and the output of a Deep learning Cardiac BioAge risk prediction tool: A UK Biobank Cohort study

Abstract

Significance. That a retinal image based Deep Learning (DL) Cardiac BioAge Model may be a useful novel tool that can personalise an individuals risk of Atherosclerotic cardiovascular disease (ASCVD) events. Purpose. To determine whether the results issued by our DL Cardiac BioAge model are consistent with the known trends of cardiovascular disease (CVD) risk and the biomarker Leucocyte Telomere Length, in a cohort of individuals from the UK Biobank. Methods. Individuals were divided by sex, ranked by Z adjusted log T/S Leucocyte Telomere length (LTL) and then grouped into deciles. The retinal images were then presented to the DL model and individuals Cardiac BioAges determined. Individuals within each LTL decile was then ranked by Cardiac BioAge, and the mean of the CVD risk biomarkers in the top and bottom quartiles compared. The relationship between an individuals Cardiac BioAge, the CVD biomarkers and LTL were determined using traditional correlation statistics. Results. The DL Cardiac BioAge model was able to accurately stratify individuals by the traditional CVD risk biomarkers, and for both males and females those issued with a Cardiac BioAge in the top quartile of their chronological peer group had a significantly higher mean SBP, HbA1C and 10-year Pooled Cohort Equation ASCVD scores compared to those individuals in the bottom quartile. Cardiac BioAge was associated with LTL shortening for both males and females. (Males: -0.220, P <0.001; Females: -0.174, P <0.001) Conclusion In this small cohort study increasing CVD risk; as assessed by both traditional biomarkers, ASCVD risk scoring and a DL Cardiac BioAge CVD risk model, was inversely related to LTL. At a population level our data supports the growing body of evidence that suggests that LTL shortening is a surrogate marker for increasing CVD risk and that this risk can be captured by our novel DL Cardiac BioAge model.