Differential diagnostics of gastric cancer and precancerous changes of the gastric mucosa using analysis of expression of six microRNAS.

Abstract

The lack of specific symptoms for the early detection of gastric cancer leads to the fact that it is often diagnosed at a late stage, when the prognosis is unfavorable. The analysis of molecular markers in addition to standard diagnostic procedures is a promising approach for improving the preoperative diagnosis of both gastric cancer and precancerous changes in the mucosa. Therefore, the aim of our study was to analyze the diagnostic significance of using miRNA expression to diagnosis gastric cancer and precancerous conditions (dysplasia) in histological material. In this work, 122 samples of archival histological material in the form of paraffin blocks were used: 34 samples of gastric adenocarcinoma, 54 samples of gastric ulcers with dysplasia and 34 samples of normal gastric mucosa obtained from patients after bariatric surgery. The expression level of miRNA-145-5p, -150-5p, -20a-5p, -21-5p, -31-5p, -34a-5p, -375 was determined using real-time RT-PCR. Samples were stratified into different groups using the C-RT decision tree algorithm. All miRNAs, except miRNA-20a, were included in the decision tree, which allows stratification of samples for normal mucosa, dysplasia, and gastric cancer. Normal mucosa can be distinguished from gastric cancer only by miRNA-34a, -21, -375. Diagnostic characteristics for the detection of dysplasia: specificity - 97%, sensitivity - 87%; for the detection of gastric cancer: specificity - 91%, sensitivity - 93%. The sufficiently high values of the diagnostic characteristics for detecting dysplasia of the gastric mucosa and gastric cancer obtained in our study indicate the possibility of using expression data of a small amount of miRNAs for the effective separation of samples with tumor and precancerous changes in the stomach tissue.