Circulating microRNAs as biomarkers for early detection of diffuse-type gastric cancer using a mouse model.

BACKGROUND: Diffuse-type gastric cancer (DGC) exhibits rapid disease progression and a poor prognosis. There are no effective serum biomarkers for early detection of DGC. We have established an E-cadherin/p53 double conditional knockout (DCKO) mouse line that recapitulates human DGC morphologically and molecularly. In this study we tried to identify circulating microRNAs (miRNAs) as non-invasive biomarkers for DGC diagnosis using DCKO mice. METHODS: We performed miRNA microarray and quantitative reverse transcription-PCR analyses of tissue and serum samples from DCKO mice with DGC and age-matched littermate controls. RESULTS: Comparative analyses showed that mouse and human primary gastric cancers have similar miRNA expression patterns. Next, we selected some candidate miRNAs highly expressed in sera and cancer tissues of DCKO mice for further evaluation. TaqMan quantitative RT-PCR analyses indicated that four of them, miR-103, miR-107, miR-194 and miR-210, were significantly upregulated in sera of both early and advanced-stage DGC-bearing mice compared with in corresponding controls. Receiver-operating characteristic curve analyses demonstrated that these four miRNAs can discriminate DGC-positive cases from normal ones with high sensitivity and specificity. CONCLUSION: These observations suggest that this mouse model of DGC is useful for identifying serum biomarkers, and we found circulating miRNAs that can accurately detect DGC at an early stage.