Plasma circRNA microarray profiling identifies novel circRNA biomarkers for the diagnosis of ovarian cancer

Abstract

Abstract Background Circular RNA (circRNA), a class of RNA with a covalent closed circular structure that widely existed in serum and plasma, has been considered an ideal liquid biopsy marker in many diseases. In this study, we employed microarray and qRT-PCR to evaluate the potential circulating circRNAs with diagnostic efficacy in ovarian cancer. Methods We used microarray to explore the circRNA expression profile in ovarian cancer patients’ plasma and quantitative real-time (qRT)-PCR approach to assessing the candidate circRNA’s expression. Then the receiver operating characteristic (ROC) curve was employed to analyze the diagnostic values of candidate circRNAs. The diagnostic model circCOMBO was a combination of hsa_circ_0003972 and hsa_circ_0007288 built by binary logistic regression. Then bioinformatic tools were used to predict their potential mechanisms. Results Hsa_circ_0003972 and hsa_circ_0007288 were downregulated in ovarian cancer patients’ plasma, tissues, and cell lines, comparing with the controls. Hsa_circ_0003972 and hsa_circ_0007288 exhibited diagnostic values with the Area Under Curve (AUC) of 0.724 and 0.790, respectively. circCOMBO showed a better diagnostic utility (AUC: 0.781), while the combination of circCOMBO and carbohydrate antigen 125 (CA125) showed the highest diagnostic value (AUC: 0.923). Furthermore, the higher expression level of hsa_circ_0007288 in both plasma and ovarian cancer tissues was associated with lower lymph node metastasis potential in ovarian cancer. Conclusions Our results revealed that hsa_circ_0003972 and hsa_circ_0007288 may serve as novel circulating biomarkers for ovarian cancer diagnosis.