Serum exosomal miR-122 as a potential diagnostic and prognostic biomarker of colorectal cancer with liver metastasis

Abstract

Background: Liver is the most common site for metastatic spread of CRC at the time of diagnosis which leads to high mortality. This study aimed to identify novel circulating exosomal miRNAs as biomarkers of colorectal cancer (CRC) with liver metastasis (LM). Materials and methods: Candidate miRNAs were selected through integrated analysis of Gene Expression Omnibus (GEO) database as well as clinical samples. Exosomes isolated from serum and cultured media were identified by using transmission electron microscopy (TEM) and western blot. The expression levels and diagnostic value of candidate miRNAs were further tested and validated through qRT-PCR and receiver operating characteristic curve (ROC) analysis. The association of candidate miRNA expressions with patients' prognosis was analyzed with logistic regression and Cox proportional hazards regression models. Results: After integrated analysis of three GEO datasets and clinical samples, miR-122 was discovered to be remarkably overexpressed in tissues of CRC patients. Then we revealed that elevated serum miR-122 was tumor-derived by being packaged into exosomes. The expressions of serum exosomal miR-122 were significantly upregulated in CRC patients, especially in those with LM. Serum exosomal miR-122 expressions could differentiate CRC patients with LM from healthy controls and patients without LM with area under the ROC curve (AUC) of 0.89 and 0.81. Uni- and multivariate logistic regression showed that serum exosomal miR-122 was an independent prognostic indicator of CRC patients. Conclusions: Serum exosomal miR-122 was a novel potential diagnostic and prognostic biomarker in CRC patients with LM.

Keywords: colorectal cancer; diagnosis; exosomes; miRNA; prognosis.; serum.

Figure 1

MiR-122 expressions were remarkably upregulated in CRC with LM. (A) Hotmap of representative miRNAs significantly deregulated in CRC with LM based on GEO datasets. (B) The expression value of miR-122 in three GEO datasets. (C-D) The expression levels of miR-122 were significantly upregulated in CRC tissues and cells.

Figure 2

Elevated serum miR-122 was tumor-derived. (A) Exosomes were identified using TEM and western blot. (B) The expressions of miR-122 in serum did not differ from that in pure exosomes isolated from equivolumetric serum. (C) The expressions of serum exosomal miR-122 were significantly downregulated after tumor being resected. (D) The expressions of exosomal miR-122 in serum of CRC patients were positively correlated with that in CRC tissues. (E) Exosomal miR-122 expressions in the culture media from both cell lines increased with time and with increasing numbers of cells. (F) The expressions of exosomal miR-122 in the culture media were significantly downregulated after being treated with GW4869.

Figure 3

High expressions of serum exosomal miR-122 in CRC patients confirmed in a small set of subjects. (A) The relative expression levels of exosomal miR-122 were significantly upregulated in the serums of CRC patients with liver metastasis. (B) The diagnostic utility of serum exosomal miR-122 to differentiate CRC patients with LM from healthy subjects. (C) The diagnostic utility of serum exosomal miR-122 to differentiate CRC patients with LM from CRC patients without LM.

Figure 4

The diagnostic value of circulating exosomal miR-122 validated in an independent cohort. (A) The expression levels of serum exosomal miR-122 were significantly upregulated in CRC patients with/without LM. (B) The diagnostic utility of serum exosomal miR-122 to differentiate CRC patients with LM from healthy subjects. (C) The diagnostic utility of serum exosomal miR-122 to differentiate CRC patients with LM from CRC patients without LM. (D-E) CRC patients, especially those with LM, who have high serum exosomal miR-122 expressions may suffer from poor OS.