Tumor-Derived circRNAs as Circulating Biomarkers for Breast Cancer

Early diagnosis is the key to improving the prognosis of breast cancer (BC) patients; however, there are currently no circulating biomarkers that demonstrate good sensitivity and specificity. This study applied circular RNA (circRNA) microarray analysis, screening, and verification in BC plasma samples to identify three tumor-derived differentially expressed circRNAs: hsa_circ_0000091, hsa_circ_0067772, and hsa_circ_0000512. We constructed a diagnostic model using logistic regression analysis in the training set and established an optimal diagnostic model based on the three circRNAs, which showed sensitivity, specificity, and area under the curve (AUC) values of .971, .902, and .974, respectively. We then verified the diagnostic model in the test set which showed satisfactory stability for BC diagnosis. Additionally, the expression of hsa_circ_0000091 in plasma correlated with axillary lymph node (ALN) metastasis, TNM stage, and prognosis of BC patients. Furthermore, hsa_circ_0000091 combined with ultrasound showed predictive ability for ALN metastasis, with an AUC of .808. These findings suggested that the three identified circRNAs can be used as circulating biomarkers for BC diagnosis, with hsa_circ_0000091 potentially representing a prognostic biomarker for BC and novel approach for predicting ALN metastasis.