Biological and clinical implications of hsa_circ_0086720 in gastric cancer and its clinical application

Abstract

Background: Circular RNAs (circRNAs) are thought to be vital participants in carcinogenesis and have the characteristics of being stable, specific, and well conserved. However, their clinical significance and application value in gastric cancer (GC) are still poorly understood. Hsa_circ_0086720 was found to be a dysregulated circRNA in GC by microarray screening and was further explored for its clinical significance and application.

Methods: Hsa_circ_0086720 was detected in GC cell lines, tissues, and plasma, and the clinicopathological correlations were investigated. The existence, stability, origin, and change in the plasma hsa_circ_0086720 level were verified in early GC patients. Moreover, receiver operating characteristic and Kaplan-Meier survival curves were constructed to analyze the diagnostic and prognostic values, and bioinformatics analysis was used to identify the potential functions. Finally, risk factors and nomogram predicting were established.

Results: Hsa_circ_0086720 was found to be downregulated in gastric carcinogenesis, and tissue hsa_circ_0086720 was negatively associated with perineural invasion, Borrmann type, disease-free survival, and overall survival. Hsa_circ_0086720 was stable in circulating plasma and was actively secreted by cells in gastric carcinogenesis. As a biomarker for early GC screening, plasma hsa_circ_0086720 had good sensitivity and specificity, and its stability met the clinical application requirements. Bioinformatics analysis suggested that dysregulated hsa_circ_0086720 has important functions in gastric carcinogenesis. Univariate Cox regression analysis identified factors associated with overall survival time and disease-free survival time. The nomograms showed good accuracy of predicting survival time.

Conclusion: Hsa_circ_0086720 is a novel biomarker for screening early GC and predicting the prognosis of advanced-stage patients.

Keywords: CircRNA; biomarker; gastric cancer; hsa_circ_0086720; prognosis.

FIGURE 1

Hsa_circ_0086720 expression is downregulated during gastric carcinogenesis. (A) Hsa_circ_0086720 expression levels in four gastric cancer cell lines (AGS, BGC‐823, HGC‐27, and MGC‐803) and a normal gastric epithelial cell line (GES‐1) were determined by qRT‐PCR. (B) The expression levels of hsa_circ_0086720 in cancer tissues (n = 96) and adjacent normal tissues (n = 96). (C) Compared to expression in adjacent normal tissues, Hsa_circ_0086720 expression was significantly downregulated in 82.3% (79/96) of gastric cancer tissues. (D) Hsa_circ_0086720 expression was significantly downregulated in early‐stage EGC (n = 24). Data are shown as the means ± SDs of two independent experiments. A higher ΔC _t value means a lower hsa_circ_0086720 level. Asterisks represent significant differences (*p < 0.05, **p < 0.01, ***p < 0.001)

FIGURE 2

Kaplan–Meier survival plot. Patients in the low hsa_circ_0086720 expression group had longer overall survival (OS) and disease‐free survival (DFS) times than those in the high expression group; this trend was not only confined to advanced clinical stages (stages III and IV; B, D) but also occurred in all clinical stages (A, C)

FIGURE 3

Clinical diagnostic value of hsa_circ_0086720 as a biomarker. (A) DNA sequencing results of plasma hsa_circ_0086720. The qRT‐PCR products of hsa_circ_0086720 were sequenced. Primer amplification sequences contain the back‐splice junction site. (B) Hsa_circ_0086720 levels in the plasma of patients in the EGC stage (n = 40) were increased significantly compared to those in healthy controls (n = 42). (C, D) EGC patients were finally pathologically diagnosed with intramucosal adenocarcinoma designated as locally cancerous. (E) ROC curve of plasma hsa_circ_0086720 as a biomarker for EGC screening. The ROC curve was up to 0.771 (95% CI, 0.663–0.879; p < 0.001). (F) ROC curve of hsa_circ_0086720 in differentiating gastric cancer tissues from controls. The AUC of hsa_circ_0086720 in cancer tissues was only up to 0.670 (95% CI, 0.594–0.746; p < 0.001). Data are two independent experiments. A higher ΔC _t value means a lower hsa_circ_0086720 level. Asterisks represent significant differences (***p < 0.001)

FIGURE 4

The origin and stability of hsa_circ_0086720 in plasma. (A‐C) Gastric cell culture experiments. The normal human gastric mucosa epithelial cell line GES‐1 and gastric cancer cell lines BGC‐823 and MGC‐803 were cultured in serum‐free medium. qRT‐PCR was used to detect the hsa_circ_0086720 levels in medium after 0, 8, 24, and 48 h of incubation. *p < 0.05, **p < 0.01. (D) Results of freeze‐thaw experiments (p > 0.05). Four blood samples were randomly selected and then equally divided into four parts. After 0, 2, 4, and 8 freeze‐thaw cycles, the hsa_circ_0086720 levels were detected. (E) Results of the incubation experiments (p > 0.05). Four blood samples were randomly selected and then equally divided into four parts and placed at 4 °C. After 0, 2, 4, and 8 h of incubation, the hsa_circ_0086720 levels were detected

FIGURE 5

Prediction of hsa_circ_0086720 function. (A) Prediction of hsa_circ_0086720/miRNA interactions. (B) A network map showing hsa_circ_0086720, four miRNAs, and their downstream targets is presented. (C) Venn diagram revealing the number of common downstream targets of four miRNAs. (D) The hsa_circ_0086720‐miRNA axes related to GO analysis. (E) The hsa_circ_0086720‐miRNA axes related to KEGG pathway analysis

FIGURE 6

Nomogram for risk assessment of survival time. (A, B) The nomograms showed good accuracy in predicting OS (C‐index, 0.822) and DFS (C‐index, 0.772). (C, D) Calibration curves reflected good discriminative ability of prognosis