Is endothelial dysfunction induced by aromatase inhibitors reversible after treatment?

Introduction: Breast cancer accounts for about one-third of all new female cancer diagnoses and remains the second leading cause of cancer-related mortality among women. Aromatase inhibitors (AIs) are a standard therapy for postmenopausal women with hormone receptor-positive breast cancer, improving disease-free survival compared to tamoxifen. However, prolonged AI use is linked to increased cardiovascular (CV) risk, including hypertension, dyslipidemia, and endothelial dysfunction, likely due to estrogen depletion. We previously demonstrated early impairment in endothelial function during AI therapy. This study assessed whether AI-induced endothelial dysfunction is reversible after AI discontinuation. Methods: Patients were recruited before or within one month of AI initiation (Pre/Early AI), at multiple time points during AI therapy, and after AI discontinuation (Post-AI) from two prospective studies: Aromatase Inhibitors and Vascular Health (AIVH) and Vascular Assessment in Breast Cancer Survivors Taking Aromatase Inhibitors (VABC). Patients with hypertension, hyperlipidemia, diabetes, or tobacco use were excluded. Vascular assessments, including the EndoPAT ratio for endothelial function and large and small artery elasticity indices for arterial stiffness, were conducted using the non-invasive EndoPAT 2000 and the HDI/PulseWave CR-2000 CV Profiling System. Lower EndoPAT ratios predict increased risk for future CV events and adverse outcomes. Plasma levels of estradiol, lipid profiles, interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-alpha) were also measured. To evaluate the longitudinal changes in endothelial function during AI therapy, a model was fit using generalized estimating equations to assess how the EndoPAT ratio changed over time on AI. Results: EndoPAT ratio, was significantly impaired during AI therapy compared to the Pre/Early AI (Pre/Early AI mean: 1.92; AI mean: 1.03; p < 0.0001). The EndoPAT ratio declined as early as 6 months on AI (model-based mean at 6 months: 1.19, p -value = 0.0097) and showed a progressive decline with increasing duration of AI use (e.g., model-based mean at 5 years on AI: 0.92; p = 0.0003). Importantly, after AI discontinuation, the EndoPAT ratio was only partially and not significantly restored (mean: 1.12) despite the full restoration of estradiol levels and the long post-treatment follow-up period (mean: 3.93 years; range: 1.53-5.34). No significant differences were observed in large and small artery elasticity, blood pressure, or lipid profiles between groups. Circulating IL-6 and TNF-alpha significantly decreased following AI discontinuation compared to during AI ( p < 0.05). Conclusion: AI therapy is associated with significant and progressive endothelial dysfunction, which does not fully recover after treatment cessation, highlighting the importance of CV monitoring in breast cancer patients receiving long-term AI therapy.