A photoplethysmography-based aging clock reveals genetic determinants of arterial aging

Abstract

Arterial aging, marked by progressive vascular stiffening, is a contributor to cardiovascular disease. Photoplethysmography (PPG) waveforms offer an easily accessible signal of arterial function, enabling scalable assessment of arterial age at the population level. Here, we present a multimodal deep learning framework integrating raw PPG waveforms with hemodynamic features from UK Biobank participants to construct an arterial aging clock. From this model, we derived ArtAgeGap, an age-independent residual biomarker of arterial aging, which correlated with established measures of vascular stiffening, such as pulse pressure (r=0.60). ArtAgeGap was significantly elevated in individuals with a history of cardiovascular disease, and associated with diabetes (+0.93 years, 95%CI: 0.64-1.21) and hyperlipidemia (+0.49, [0.36-0.61]). In 96,615 participants followed for a median of 13 years, higher ArtAgeGap values predicted incident hypertension, major adverse cardiovascular events, and cardiovascular and all-cause mortality (hazard ratios per +1 year: 1.01-1.06), on top of chronological age. A genome-wide association study in 114,098 individuals identified 60 independent loci associated with ArtAgeGap, including variants previously associated with blood pressure (e.g. NPR3), arterial stiffness (CLCN6), aortic diameter (SLC24A3), and atherosclerosis (HDAC9), as well as 34 novel loci enriched for arterial tissue expression. Integrative analyses with transcriptomic data from human arteries prioritized 28 genes, such as RSG19 and ULK4, whose genetically proxied expression was associated with ArtAgeGap. Single-cell transcriptomic data from arterial tissue revealed strong enrichment of these genes in fibroblasts, implicating fibrotic remodeling mechanisms in arterial aging. Rare variant burden testing further implicated damaging variants in COL21A1, LMNA, TP53BP2, RXRB, and FLOT2, also converging to mechanisms related to extracellular matrix organization and fibrosis regulation. Lastly, Mendelian randomization analysis identified ArtAgeGap as an intermediate biomarker in-between vascular risk factors and outcomes, with central adiposity, higher blood pressure, and type 2 diabetes increasing ArtAgeGap, and higher ArtAgeGap elevating the risks of coronary heart disease and stroke. Together, these findings establish ArtAgeGap as a scalable PPG-based biomarker of arterial aging and provide mechanistic insights into potential therapeutic strategies to mitigate arterial aging.