Leveraging Circulating Extracellular Vesicle-Encapsulated MicroRNAs to Construct a Longitudinal Predictive Model for Kidney and Cardiovascular Events

Background: Chronic kidney disease (CKD) causes premature death through a harmful inter-organ crosstalk involving the kidneys, heart, and metabolic risk. This interplay can be conceptualized as the cardiovascular-kidney-metabolic (CKM) syndrome. However, a method for longitudinal prediction of kidney and cardiovascular events has yet to be established. Methods: To develop a predictive model that can be implemented in clinical practice using microRNA (miRNA) within circulating small extracellular vesicles (cEVs), we performed a transcriptomic analysis of miRNAs within cEVs in a derivation cohort of 36 CKD patients without kidney replacement therapy (KRT). After developing a Lasso-Cox prediction model using miRNAs, the risk score equation was optimized using a validation cohort of 234 individuals. The primary outcome was a kidney outcome of a 30% reduction in creatinine-based estimated glomerular filtration rate (eGFR) and KRT initiation. Results: The cEVs within the cluster enriched for CKD-G5 were depleted of 23 miRNAs, which target signaling pathways such as cellular senescence and cell cycle, as identified through KEGG pathway analysis. The Lasso-Cox analysis constructed the equation including the expression levels of hsa-let-7d-5p, hsa-miR-24-3p, and hsa-miR-126-3p (HR 30.7, 95% CI 3.4-275 for high-risk score versus low-risk score). The Lasso-Cox analysis in the validation cohort optimized the equation (M3V2 equation), including these three miRNAs, serum cystatin C, and urinary protein levels. The M3V2 high-risk group was at greater risk for the kidney outcome (HR 7.8, 95% CI 4.75-13.0 versus low-risk group) and also the composite of all-cause deaths, initiating KRT, and admissions owing to major adverse cardiovascular and cerebrovascular events (HR 10.3, 95% CI 4.86-22.0), respectively. These predictive abilities were independent of underlying cardiovascular disease and origins of CKD and were comparable to, yet exceeded those of CKM stages. Conclusion: This study identified the miRNA signature encapsulated in cEVs, which can contribute to the development of novel predictive models and provide insights into the pathogenesis of CKD and the CKM syndrome. Funding: Private Foundation Support