Breast cancer is a heterogeneous disease, which is the most common malignancy in women. The incidence and mortality rates of breast cancer indicate that it is the leading cause of cancer-related with deaths. circRNAs operate as part of competing endogenous RNAs (ceRNAs) mechanisms, which play critical roles in the different biological processes of breast cancer such as proliferation, migration, and apoptosis. The goal of the present study is to identify the potential predictive biomarker for breast cancer diagnosis in the circRNA network by in vitro and in silico analyzes. 40 miRNAs were obtained from the miRWalk database and their combinatorial target genes (potential ceRNAs) were identified with ComiR. We stated that the cancer-specific circRNA genes in MCF-7 cells using the cancer-specific circRNA (CSDC) database, and obtained the ones showing potential ceRNA activity in our previous analysis among them. Identified genes with remarkable expression differences between BCa and normal breast tissue were determined by the GEPIA database. Moreover, the Spearman correlation test in the GEPIA database was used for the statistical analysis of the relationship between DCAF7 and SOGA1, SOGA1 and AVL 9, DCAF7 and AVL 9 gene pairs. And also, DCAF7, SOGA1, and AVL9 gene expression levels were detected in MCF-7 and MCF-10A cells by RT-qPCR method. DCAF7, SOGA1, and AVL9 gene were significantly more expressed to BCa tissue and MCF-7 cells than normal breast tissue and MCF-10 A cells. And also, DCAF7 and SOGA1, SOGA1 and AVL9, DCAF7 and AVL9 genes pairs were found to be significantly correlated with BCa. These genes may be considered as potential predictive biomarkers to discriminate BCa patients from healthy persons. Our preliminary results can supply a new perspective for in vitro and vivo studies in the future.