A novel combination of serum microRNAs for the detection of early gastric cancer.

BACKGROUND

The aim of this study was to identify serum miRNAs that discriminate early gastric cancer (EGC) samples from non-cancer controls using a large cohort.

METHODS

This retrospective case-control study included 1417 serum samples from patients with EGC (seen at the National Cancer Center Hospital in Tokyo between 2008 and 2012) and 1417 age- and gender-matched non-cancer controls. The samples were randomly assigned to discovery and validation sets and the miRNA expression profiles of whole serum samples were comprehensively evaluated using a highly sensitive DNA chip (3D-Gene) designed to detect 2565 miRNA sequences. Diagnostic models were constructed using the levels of several miRNAs in the discovery set, and the diagnostic performance of the model was evaluated in the validation set.

RESULTS

The discovery set consisted of 708 samples from EGC patients and 709 samples from non-cancer controls, and the validation set consisted of 709 samples from EGC patients and 708 samples from non-cancer controls. The diagnostic EGC index was constructed using four miRNAs (miR-4257, miR-6785-5p, miR-187-5p, and miR-5739). In the discovery set, a receiver operating characteristic curve analysis of the EGC index revealed that the area under the curve (AUC) was 0.996 with a sensitivity of 0.983 and a specificity of 0.977. In the validation set, the AUC for the EGC index was 0.998 with a sensitivity of 0.996 and a specificity of 0.953.

CONCLUSIONS

A novel combination of four serum miRNAs could be a useful non-invasive diagnostic biomarker to detect EGC with high accuracy. A multicenter prospective study is ongoing to confirm the present observations.