Altered expression of microRNA (miRNA) is strongly implicated in cancer, and recent studies have shown that the silencing of some miRNAs is associated with CpG island hypermethylation. To identify epigenetically silenced miRNAs in gastric cancer (GC), we screened for miRNAs induced by treatment with 5-aza-2'-deoxycytidine and 4-phenylbutyrate. We found that miR-34b and miR-34c are epigenetically silenced in GC and that their downregulation is associated with hypermethylation of the neighboring CpG island. Methylation of the miR-34b/c CpG island was frequently observed in GC cell lines (13/13, 100%) but not in normal gastric mucosa from Helicobacter pylori-negative healthy individuals. Transfection of a precursor of miR-34b and miR-34c into GC cells induced growth suppression and dramatically changed the gene expression profile. Methylation of miR-34b/c was found in a majority of primary GC specimens (83/118, 70%). Notably, analysis of non-cancerous gastric mucosae from GC patients (n = 109) and healthy individuals (n = 85) revealed that methylation levels are higher in gastric mucosae from patients with multiple GC than in mucosae from patients with single GC (27.3 versus 20.8%; P < 0.001) or mucosae from H. pylori-positive healthy individuals (27.3 versus 20.7%; P < 0.001). These results suggest that miR-34b and miR-34c are novel tumor suppressors frequently silenced by DNA methylation in GC, that methylation of miR-34b/c is involved in an epigenetic field defect and that the methylation might be a predictive marker of GC risk.

Our results suggest the Type II-O pit pattern is a useful hallmark of the premalignant stage of CRCs with MSI and CIMP, which could serve to improve the efficacy of colonoscopic surveillance.

No abstract

BackgroundMetachronous gastric cancer (GC) can develop after endoscopic resection of GC and cannot be predicted based on clinical signature. Aberrant DNA methylation in noncancerous gastric mucosa is strongly implicated in gastric carcinogenesis and could be a useful biomarker of GC risk. We evaluated the clinical utility of DNA methylation as a biomarker of metachronous GC risk.MethodWe carried out scheduled follow-up endoscopy in 129 patients after curative endoscopic resection of GC. Biopsy specimens were collected from noncancerous mucosa in the gastric antrum and body, after which quantitative methylation analysis of miR-34b/c, SFRP1, SFRP2, SFRP5, DKK2 and DKK3 was carried out using bisulfite pyrosequencing. The utility of the methylation for predicting the risk of metachronous GC development was assessed using Kaplan–Meier and Cox proportional hazards model analyses.ResultsDuring the follow-up period, 17 patients (13 %) developed metachronous GCs. The cumulative incidence of metachronous GC was significantly higher among patients with elevated miR-34b/c, SFRP2 and DKK2 methylation in their gastric body. MiR-34b/c showed the strongest association with the risk of metachronous GC, and the cumulative incidence of metachronous GC was much higher in the high-miR-34b/c-methylation group than the low-methylation group. Multivariate analysis adjusted for age, sex, H. pylori status and pathological findings showed miR-34b/c methylation in gastric body to be an independent predictor of metachronous GC risk.ConclusionOur results suggest that methylation of miR-34b/c in the mucosa of the noncancerous gastric body may be a useful biomarker for predicting the risk of metachronous GC.Electronic supplementary materialThe online version of this article (doi:10.1007/s00535-013-0861-7) contains supplementary material, which is available to authorized users.

Aberrant DNA methylation could potentially serve as a biomarker for colorectal neoplasms. In this study, we assessed the feasibility of using DNA methylation detected in bowel lavage fluid (BLF) for colorectal cancer screening. A total of 508 BLF specimens were collected from patients with colorectal cancer (n ¼ 56), advanced adenoma (n ¼ 53), minor polyp (n ¼ 209), and healthy individuals (n ¼ 190) undergoing colonoscopy. Methylation of 15 genes (miR-1-1, miR-9-1, miR-9-3, miR-34b/c, miR-124-1, miR-124-2, miR-124-3, miR-137, SFRP1, SFRP2, APC, DKK2, WIF1, LOC386758, and ZNF582) was then analyzed in MethyLight assays, after which receiver operating characteristic (ROC) curves were analyzed to assess the diagnostic performance of BLF methylation. Through analyzing BLF specimens in a training set (n ¼ 345), we selected the three genes showing the greatest sensitivity for colorectal cancer detection (miR-124-3, 71.8%; LOC386758, 79.5%; and SFRP1, 74.4%). A scoring system based on the methylation of those three genes (M-score) achieved 82% sensitivity and 79% specificity, and the area under the ROC curve (AUC) was 0.834. The strong performance of this system was then validated in an independent test set (n ¼ 153; AUC ¼ 0.808). No significant correlation was found between M-score and the clinicopathologic features of the colorectal cancers. Our results demonstrate that DNA methylation in BLF specimens may be a useful biomarker for the detection of colorectal cancer. Cancer Prev Res; 7(10);