Plasma microRNAs as potential new biomarkers for early detection of early gastric cancer

Abstract

Background: Early gastric cancer (EGC), compared with advanced gastric cancer (AGC), has a higher 5-year survival rate. However, due to the lack of typical symptoms and the difficulty in diagnosing EGC, no effective biomarkers exist for the detection of EGC, and gastroscopy is the only detection method.

Aim: To provide new biomarkers with high specificity and sensitivity through analyzed the differentially expressed microRNAs (miRNAs) in EGC and AGC and compared them with those in benign gastritis (BG).

Methods: We examined the differentially expressed miRNAs in the plasma of 30 patients with EGC, AGC, and BG by miRNA chip analysis. Then, we analyzed and selected the significantly different miRNAs using bioinformatics. Reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) confirmed the relative transcription level of these miRNAs in another 122 patients, including patients with EGC, AGC, Helicobacter pylori (H. pylori)-negative gastritis (Control-1), and H. pylori-positive atrophic gastritis (Control-2). To establish a diagnostic model for the detection of plasma miRNA in EGC, we chose miRNAs that can be used to determine EGC and AGC from Control-1 and Control-2 and miRNAs in EGC from all other groups.

Results: Among the expression profiles of the miRNA chips in the three groups in the discovery set, of 117 aberrantly expressed miRNAs, 30 confirmed target prediction, whereas 14 were included as potential miRNAs. The RT-qPCR results showed that 14 potential miRNAs expression profiles in the two groups exhibited no differences in terms of H. pylori-negative gastritis (Control-1) and H. pylori-positive atrophic gastritis (Control-2). Hence, these two groups were incorporated into the Control group. A combination of four types of miRNAs, miR-7641, miR-425-5p, miR-1180-3p and miR-122-5p, were used to effectively distinguish the Cancer group (EGC + AGC) from the Control group [area under the curve (AUC) = 0.799, 95% confidence interval (CI): 0.691-0.908, P < 0.001]. Additionally, miR-425-5p, miR-24-3p, miR-1180-3p and miR-122-5p were utilized to distinguish EGC from the Control group (AUC = 0.829, 95%CI: 0.657-1.000, P = 0.001). Moreover, the miR-24-3p expression level in EGC was lower than that in the AGC (AUC = 0.782, 95%CI: 0.571-0.993, P = 0.029), and the miR-4632-5p expression level in EGC was significantly higher than that in AGC (AUC = 0.791, 95%CI: 0.574-1.000, P = 0.024).

Conclusion: The differentially expressed circulatory plasma miR-425-5p, miR-1180-3p, miR-122-5p, miR-24-3p and miR-4632-5p can be regarded as a new potential biomarker panel for the diagnosis of EGC. The prediction and early diagnosis of EGC can be considerably facilitated by combining gastroscopy with the use of these miRNA biomarkers, thereby optimizing the strategy for effective detection of EGC. Nevertheless, larger-scale human experiments are still required to confirm our findings.

Keywords: Biomarker; Early gastric cancer; MicroRNA; Plasma.

Figure 1

Overview of the study design. EGC: Early gastric cancer; AGC: Advanced gastric cancer; BG: Benign gastritis; Control-1: Helicobacter pylori-negative gastritis; Control-2: Helicobacter pylori-positive atrophic gastritis; RT-qPCR: Reverse transcription quantitative real-time polymerase chain reaction.

Figure 2

Expression of the four identified microRNAs between the control (Control-1 + Control-2) and cancer (early gastric cancer + advanced gastric cancer) groups. To assess the diagnostic utility of miR-7641, miR-425-5p, miR-1180-3p and miR-122-5p, a linear combination of the risk score of the four miRNAs weighted by the regression coefficient was used to calculate a risk score factor (RSF) for the four-microRNA panel for each subject. The RSF was calculated as follows: RSF = 1.569 × miR-7641 + 1.312 × miR-425-5p + 1.852 × miR-1180-3p + 1.322 × miR-122-5p. Area under curve = 0.799, 95% confidence interval: 0.691-0.908, P < 0.001. AUC: Area under curve; CI: Confidence interval; miRNA: MicroRNA.

Figure 3

Expression of the four identified microRNAs between the control (Control-1 + Control-2) and early gastric cancer groups. To assess the diagnostic utility of miR-425-5p, miR-24-3p, miR-1180-3p and miR-122-5p, a linear combination of the risk score of the four microRNAs (miRNAs) weighted by the regression coefficient was used to calculate a risk score factor (RSF) for the four-miRNA panel for each subject. The RSF was calculated as follows: RSF = 2.117 × miR-425-5p + 2.234 × miR-24-3p + 2.005 × miR-1180-3p + 1.792 × miR-122-5p. Area under curve = 0.829, 95% confidence interval: 0.657-1.000, P = 0.001. AUC: Area under curve; CI: Confidence interval; miRNA: MicroRNA; EGC: Early gastric cancer.

Figure 4

Expression of the two microRNAs between the advanced and early gastric cancer groups. The expression of miR-24-3p in the early gastric cancer (EGC) group was significantly lower than that in the advanced gastric cancer (AGC) [area under curve (AUC) = 0.782, 95% confidence interval (CI): 0.571-0.993, P = 0.029] group. The expression of miR-4632-5p in the EGC group was significantly higher than that in the AGC (AUC = 0.791, 95%CI: 0.574-1.000, P = 0.024) group. AUC: Area under curve; CI: Confidence interval; miRNA: MicroRNA.