Gastric cancer-associated long non-coding RNA profiling and noninvasive biomarker screening based on a high-risk population cohort

Abstract

Background: Effective noninvasive biomarkers of gastric cancer (GC) are critical for early detection and improvement of prognosis. We performed genome-wide long non-coding RNA (lncRNA) microarray analysis to identify and validate novel GC biomarkers depending on a high-risk population cohort.

Methods: LncRNA profiles were described using the Human LncRNA Microarray between GC and control plasma samples. The differential candidate lncRNAs were validated in two stages by quantitative reverse transcription polymerase chain reaction (qRT-PCR). We further evaluated the joint effect between the GC-associated lncRNA and Helicobacter pylori (H. pylori) infection on the risk of cardia and non-cardia GC, respectively.

Results: Different lncRNA expression profiles were identified between GC and control plasma with a total of 1206 differential lncRNAs including 470 upregulated and 736 downregulated in GC compared with the control group. The eight significantly upregulated lncRNAs (RP11-521D12.1, AC011995.3, RP11-5P4.3, RP11-244 K5.6, RP11-422 J15.1, CTD-2306 M5.1, CTC-428G20.2, and AC009133.20) in GC cases both in the present study and a similar microarray screening study by our collaborative team were selected for a two-stage validation. After the large sample size validation, the subjects with higher expression of RP11-244 K5.6 showed a significantly increased risk of GC with an adjusted odds ratio (OR) as 2.68 and 95% confidence interval (CI) as 1.15-6.24. Joint effects between RP11-244 K5.6 expression and H. pylori infection on the risk of GC were evaluated with no statistical significance.

Conclusions: Our study found different lncRNA expression profiles between GC and control plasma and preliminarily identified RP11-244 K5.6 as a potential noninvasive biomarker for GC screening.

Keywords: LncRNAs; biomarkers; gastric cancer; microarray; plasma.

FIGURE 1

Flow process of study design.

FIGURE 2

LncRNA expression profiles comparison between GC and control groups by microarray analysis. (A) Hierarchical clustering visualized the differential expression of lncRNAs among subjects. Each row represents the expression level of a differential lncRNA. Each column represents a plasma sample from each subject (test—gastric cancer plasma sample; ctrl—control plasma sample). (B) The volcano plot showed the differential expression of lncRNAs. Red dots represent upregulated lncRNAs and green dots represent downregulated lncRNAs in gastric cancer plasma (Fold change≥2 and p ≤ 0.05).

FIGURE 3

Small sample size validation of the 8 candidate lncRNAs. A paired t‐test was conducted to compare the 8 candidate lncRNAs between GC and control plasma samples. The differential lncRNAs were selected according to the criteria of fold change (FC) ≥ 2.0 and p‐value ≤0.05.