مقالات پذیرفته شده در نهمین کنگره بین المللی زیست پزشکی
Extracellular Vesicle Derived MicroRNAs as Non-Invasive Biomarkers for Distinguishing Benign from Malignant Breast Lesions Categorized as Breast Imaging Reporting and Data System Category Four
Extracellular Vesicle Derived MicroRNAs as Non-Invasive Biomarkers for Distinguishing Benign from Malignant Breast Lesions Categorized as Breast Imaging Reporting and Data System Category Four
Zahra Nasrollahi,1,*
1. Kermanshah University of Medical Sciences, Faculty of Medicine
Introduction: Early detection of breast cancer is crucial for improving patient outcomes. Conventional imaging modalities, such as
mammography and ultrasound, often yield indeterminate findings, particularly in Breast Imaging Reporting and Data
System category 4 (BI-RADS 4) lesions carry a risk of malignancy ranging from 2% to 95%. Consequently,
invasive biopsies are frequently performed, leading to unnecessary patient anxiety and healthcare costs. Extracellular
vesicles (EVs), which are nanoscale particles released by cells into the circulation, carry microRNAs (miRNAs) that
reflect the molecular state of their parent cells. EV-derived miRNAs are stable in plasma and present a promising
non-invasive approach for cancer diagnostics. This study investigates the utility of EV-derived miRNAs, particularly
from HER2+ and CD24+ EV subpopulations, as biomarkers to distinguish benign from malignant BI-RADS 4 breast
lesions.
Methods: A prospective cohort of 113 women with BI-RADS 4 breast lesions (86 benign, 27 malignant) was analyzed. Plasma
samples were collected before biopsy. HER2+ and CD24+ EVs were isolated using the Track-Etched Magnetic
Nanopore (TENPO) platform. EV RNA was extracted, and global miRNA profiling was performed using nextgeneration sequencing. Differentially expressed miRNAs were identified, and a diagnostic panel was constructed
using LASSO regression. Validation was performed with quantitative PCR.
For comparison, additional studies analyzing tumor-derived EVs and circulating miRNAs in early-stage breast cancer
were included to confirm diagnostic performance.
Results: HER2+ EVs: 19 miRNAs were differentially expressed; miR-340-5p showed the highest discriminative power
with AUC = 0.87 (95% CI: 0.80–0.94).
CD24+ EVs: 11 miRNAs were differentially expressed; miR-223-3p and miR-126-3p achieved AUC = 0.75.
A four-miRNA panel (miR-340-5p, miR-598-3p, miR-15b-5p, miR-126-3p) reached diagnostic accuracy = 0.88,
validated by qPCR.
Complementary studies using tumor-derived EVs (miR-9, miR-16, miR-21, miR-429) reported 96.8%
sensitivity and 80% specificity, while circulating miRNA panels analyzed with machine learning achieved 92.5%
accuracy, 95% specificity, and 88% sensitivity.
These results indicate that EV-derived miRNAs reliably reflect the malignant status of BI-RADS 4 lesions and can
serve as minimally invasive diagnostic biomarkers for this condition.
Conclusion: EV-derived miRNAs, particularly from HER2+ and CD24+ subpopulations, demonstrate high diagnostic accuracy in
distinguishing benign from malignant BI-RADS 4 breast lesions. Implementation of a four-miRNA panel could
significantly reduce unnecessary biopsies. Further validation in multicenter cohorts is warranted to confirm clinical
applicability. Integrating EV-based liquid biopsy with imaging may enhance early breast cancer diagnosis and patient
management.
Keywords: Extracellular vesicles; EV-derived microRNA; Liquid biopsy; Breast cancer diagnosis; miR-340-5p