Abstract PS4-01-04: Circulating MIRNA Signature Predicts Breast Cancer Metastasis: A Pilot Study for Non-Invasive Risk Stratification

Abstract Background: Breast cancer remains the leading cause of cancer-related mortality in women globally, with metastatic disease accounting for 90% of deaths. Current staging and histopathological features inadequately predict individual recurrence risk. Approximately 20-30% of early-stage patients develop distant metastases despite optimal therapy, while others are overtreated. Circulating microRNAs (MIRNAs) represent a paradigm shift toward liquid biopsy, offering real-time tumor assessment through minimally invasive blood sampling. These stable molecules reflect dynamic tumor behavior and metastatic potential, making them ideal for personalized surveillance strategies. Methods: This prospective proof-of-concept study included 46 breast cancer patients from Sri Shankara Cancer Hospital, India: 37 with metastatic disease (plasma collected at progression, 2024) and 9 disease-free controls (≥3-year follow-up). MIRNA isolation used MIRVANA method with samples processed within 2 hours, stored at -80°C. Expression levels of MIR-21, MIR-30d, and MIR-425 were quantified by TaqMan qRT-PCR, normalized to MIR-16 control. Logistic regression analysis of the transcript levels yielded a probability of predictive accuracy ranging from 0-1. Cut-off of high expression was chosen based on the best sensitivity and specificity by ROC curve analysis. Clinicopathological parameters (CPP) were correlated with MIR expression(Mann-Whitney U test). A combined model of MIR and CPP was built using logistic regression and ROC curve analysis. Results: Patient demographics(in Table) Difference in median age, subtypes and nodal status are interesting to note. MIR-21 showed elevated expression in metastatic cases, particularly hormone receptor-positive tumors. MIR-30d demonstrated significant correlation with grade 3 and node-positive status (p=0.038 for N0 vs N1, p=0.006 for N0 vs N3), with high expression associated with poor progression-free survival (p=0.0057, HR=3.24). MIR-425 exhibited significant differential expression between metastatic and non-metastatic groups with strong correlation to advanced nodal stages (p=0.0018). Univariate analysis with all clinicopathological parametres with DFS revealed LN status as significant predictor (AUC=0.68). Similarly combined MIR score of MIR-425 and MIR-30d achieved 72% predictive accuracy of DFS (AUC=0.716). Interestingly, multivariate analysis of MIR score and LN with DFS, enhanced the AUC to 0.81. Conclusions: This proof-of-concept study suggests circulating MIR-30d and MIR-425 may serve as potential biomarkers for breast cancer recurrence risk stratification. The integrated signature shows promise for enhancing surveillance strategies. Assay also standardized methods for using plasma to measure circulating MIRNA levels. Validation in larger multicenter cohorts is underway for clinical assay development. Citation Format: A. Korlimarla, V. Sasimouli, L. Krishna, S. Shree, H. P.S, B. Sivaraman, S. Appachu, S. Srihari, G. Guhan, P. Advani, S. B.S. Circulating MIRNA Signature Predicts Breast Cancer Metastasis: A Pilot Study for Non-Invasive Risk Stratification abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-01-04.