Circulating extracellular vesicle profile predicts the response to exercise training in patients with coronary artery disease during cardiac rehabilitation

Background Exercise-based cardiac rehabilitation (CR) is a Class 1A guideline recommendation following percutaneous coronary intervention (PCI). Despite its benefits, responses to aerobic endurance training vary, with patients categorized as responders (R) or non-responders (NR) based on changes in cardiorespiratory fitness. Recent evidence suggests that blood-based biomarkers, such as extracellular vesicles (EVs), could help stratify cardiac patients and personalize CR programs, potentially improving outcomes. Purpose This study explores whether circulating EVs influence CR outcomes in patients with coronary artery disease (CAD) by modulating vascular adaptations to exercise training. Methods A total of 416 CAD patients post-PCI underwent cardiopulmonary exercise testing (CPET) before and after center-based phase II CR. Responders (R) and non-responders (NR) were classified based on improvements in submaximal and maximal exercise capacity — specifically, VO₂ at the first ventilatory threshold (VT1) and VO₂peak. R were defined as those showing increases greater than two times the typical error from the population mean. The groups were matched for confounding factors such as age, disease severity, time since PCI, smoking status, and more. Serum samples were analyzed using nanoparticle tracking analysis (NTA) to measure EV count and size, and flow cytometry to assess inflammatory and coagulation-related surface antigens on EVs. Isolated EVs from both groups were tested in vitro on human aortic endothelial cells (hAECs) to evaluate redox status, angiogenesis, and extracellular matrix secretion. Results Forty male patients (20 R, 20 NR; mean age: 53.3 years) were included. Both groups completed a similar volume of exercise training (measured by METs). NR had a higher number of circulating EVs and elevated surface markers (CD31, CD62P, CD42a) — all known predictors of major adverse cardiovascular events (MACE). In vitro, NR-EVs induced oxidative stress in hAECs, reflected by increased dihydroethidium positivity and levels of MDA, 4-HNE, nitro-tyrosine, and 8OHdG. Additionally, NR-EVs impaired tube formation and promoted collagen production, indicating compromised vascular function. Interestingly, Omega-3 alpha-linolenic acid (ALA), an antioxidative molecule, mitigated the harmful effects of NR-EVs. Conclusions Circulating EVs, particularly those expressing surface markers CD31, CD62P, and CD42a, are linked to CR response in CAD patients. The underlying mechanisms may involve EV-induced oxidative stress and vascular dysfunction in NR. These findings highlight the potential of using EV profiles to personalize CR strategies, aiming to optimize patient outcomes.