Abstract
BACKGROUND: MicroRNAs (miRNAs) are recognized as potential biomarkers for detection of breast cancer. Many experiments have been done to explore the aberrant expression of circulating miRNAs in breast cancer patients. However, an overwhelming number of miRNAs are identified and some expression characteristics are inconsistent between studies. OBJECTIVE: To distinguish the candidates for breast cancer detection from spurious miRNAs. METHODS: One solution is to take the intersections between studies. As repeated efforts can improve reliability and reduce error, valuable candidate miRNAs in this study are defined as those validated and consistently reported by multiple studies. Data validated by real-time PCR (RT-PCR) were collected, and the most frequently reported miRNAs with consistent regulation were identified by using a vote-counting strategy based on the number of relevant studies, total sample size and fold change. RESULTS: The top four miRNAs (miR-21, -155, -222 and -10b) are consistently regulated in comparisons of the pre-operative patients and the control group. And blood sample type was found to affect the regulation characteristics of miRNAs. CONCLUSION: MiR-21, -155, -222 and -10b are reliable candidate biomarkers for detection of breast cancer.