Exosomal Long Non-Coding RNA CEBPA-AS1 Inhibits Tumor Apoptosis and Functions as a Non-Invasive Biomarker for Diagnosis of Gastric Cancer

Traditional non-invasive diagnostic markers for gastric cancer (GC) exhibit insufficient sensitivity and specificity. Circulating exosomes are clinically useful non-invasive biomarkers for tumor diagnosis. In addition to their potential role in cancer biology, circulating long non-coding RNAs (lncRNAs) are a new class of promising cancer biomarkers. In the present study, we aimed to identify lncRNAs in circulating exosomes with potential as biomarkers for GC detection.

CEBPA-AS1-containing exosomes secreted from GC cells could promote cell proliferation, inhibit apoptosis, and induce GC progression, indicating that exosomal CEBPA-AS1 is involved in cell-to-cell communication in GC carcinogenesis. Exosomal CEBPA-AS1 is a promising new biomarker for clinical diagnosis of GC.

We compared the expression of CEBPA-AS1 between GC cells and gastric epithelial cells. The biological function of exosomal CEBPA-AS1 was determined by cell phenotype experiments and rescue assays. We also compared the expression of CEBPA-AS1 in cancerous tissue from GC patients and corresponding adjacent normal tissues, as well as the expression of CEBPA-AS1 in plasma exosomes of GC patients and healthy controls. Diagnostic accuracy was assessed by the receiver operating characteristic (ROC) curve and area under the curve (AUC).

CEBPA-AS1 was highly expressed in both GC cells and in exosomes secreted by GC cells. In addition, CEBPA-AS1-containing exosomes secreted by GC cells could promote cell proliferation and inhibit apoptosis, thereby inducing the malignant behavior of GC cells. The level of CEBPA-AS1 was also significantly increased in tissues and plasma exosomes of GC patients. Stability tests showed that most plasma CEBPA-AS1 was encased in exosomes, thus avoiding degradation by RNases. We evaluated the diagnostic accuracy of exosome-derived CEBPA-AS1. The AUC value of CEBPA-AS1 in discriminating GC patients from healthy controls was 0.824, which was higher than the diagnostic accuracy of other traditional tumor biomarkers.