Exosomal miR-1290 is a potential biomarker of high-grade serous ovarian carcinoma and can discriminate patients from those with malignancies of other histological types

Abstract

Background: microRNAs (miRNAs) stably exist in circulating blood encapsulated in extracellular vesicles such as exosomes; therefore, serum miRNAs have the potential to serve as novel cancer biomarkers. New diagnostic markers to detect high grade serous ovarian cancer (HGSOC) are urgently needed. The aim of this study was to identify miRNAs specific to HGSOC and analyze whether serum miRNA can discriminate HGSOC patients from healthy controls or patients with ovarian malignancies of other histological types.

Methods: Exosomes from ovarian cancer cell lines were collected and exosomal miRNAs extracted. miRNA microarray analysis revealed several elevated miRNAs specific to HGSOC. Among these, we focused on miR-1290. Sera from 70 ovarian cancer patients and 13 healthy controls were gathered and its expression levels detected by quantitative real-time polymerase chain reaction.

Results: In HGSOC patients, serum miR-1290 was significantly overexpressed compared to in healthy controls (3.52 fold; P = 0.03), unlike in patients with ovarian cancers of other histological types. The relative expression of miR-1290 was higher in advanced stages of HGSOC than in early stages (4.23 vs. 1.58; P = 0.23). Its expression significantly decreased after operation (5.87 to 1.17; P < 0.01), indicating that this miRNA reflects tumor burden. A receiver operating characteristic curve analysis showed that at the cut-off of 1.20, the sensitivity and specificity were 63% and 85% respectively for discriminating patients with HGSOC (area under the curve [AUC] = 0.71) from healthy controls, and at the cut-off of 1.55, the sensitivity and specificity were 47% and 85% respectively for discriminating patients with HGSOC (AUC = 0.76) from those with malignancies of other histological types.

Conclusions: Serum miR-1290 is significantly elevated in patients with HGSOC and can be used to discriminate these patients from those with malignancies of other histological types; it is a new potential diagnostic biomarker for HGSOC.

Keywords: Biomarker; HGSOC; Ovarian cancer; miR-1290; microRNA.

Fig. 1

The collection of exosomes from EOC cell lines and the analyses of exosomal miRNAs. (a) Electron microscopy. Exosomes were immunogold-labeled with anti-CD63 antibody and transmission electron micrographs of isolated exosomes secreted from 2 HGSOC cell lines (HeyA8, upper panel;TYK-nu, middle panel) are shown. Immortalized ovarian surface epithelium (IOSE) cells (bottom panel) were used as a non-malignant control. Scale bar, 100 nm. (b) Venn diagrams showing the relative miRNAs expressed in HeyA8 and TYK-nu vs. IOSEs by exosomal miRNA microarray. Circles with solid lines represent counts of exosomal miRNAs with expression levels increased by > 3.0-fold. Circles with dotted lines represent counts of miRNAs with expression levels increased by > 4.0-fold. (c) Summary of exosomal miRNA microarray. Exosomal miRNAs that were upregulated by more than 4.0-fold in both HeyA8 and TYK-nu vs. IOSEs are listed. (d) qRT-PCR of miR-1290. The relative expression of miR-1290 in HGSOC-derived exosomes is shown. IOSE-derived exosomes were used as a control. Data are the mean ± standard deviation (SD) of 3 experiments. **, P < 0.01

Fig. 2

Serum miR-1290 expression is significantly elevated in HGSOC patients. (a) Relative miR-1290 expression levels in sera from healthy controls (n = 13) and EOC patients (n = 70) are shown. The average miR-1290 expression level in healthy controls was normalized to 1.0. The box-and-whisker plots indicate the median and interquartile range. (b) CA125 values in sera from healthy controls (n = 13) and EOC patients (n = 70) are shown. The box-and-whisker plots indicate the median and interquartile range. (c) ROC curves for the identification of patients with EOC (n = 70) vs. healthy controls (n = 13) based on the expression of CA125 (gray line), miR-1290 (black line), and the combination of both (dotted line). The AUC values are shown on the graphs. (d) ROC curves for the identification of patients with ovarian cancer of each histological subtype (serous: n = 30, clear-cell: n = 18, endometrioid: n = 12, mucinous: n = 10) vs. healthy controls (n = 13). (e) Relative miR-1290 expression levels in sera from healthy controls (n = 13) and HGSOC patients (n = 30) are shown. **, P < 0.01; *, P < 0.05; n.s., not significant

Fig. 3

Serum miR-1290 expression reflects HGSOC tumor burden. (a) Relative serum miR-1290 expression levels between HGSOC patients at early (I + II) (n = 8) and advanced (III + IV) stage (n = 22) are shown. The average miR-1290 expression level in healthy controls (n = 13) was normalized to 1.0. The box-and-whisker plots indicate the median and interquartile range. (b) Relative miR-1290 expression levels in matched serum samples from HGSOC patients before and after primary debulking surgery (PDS). Individual changes in the serum expression of miR-1290 in HGSOC patients before and after PDS are shown. The average miR-1290 expression level in healthy controls (n = 13) was normalized to 1.0. **, P < 0.01; n.s., not significant

Fig. 4

Diagnostic performance of miR-1290 for the discrimination of HGSOC from non-HGSOC. (a) Relative miR-1290 expression levels in sera from HGSOC patients (n = 30) and patients with EOC other than HGSOC (n = 40). The average miR-1290 expression level in healthy controls (n = 13) was normalized to 1.0. The box-and-whisker plots indicate the median and interquartile range. (b) CA125 expression levels in sera from HGSOC patients (n = 30) and patients with EOC other than HGSOC (n = 40). The box-and-whisker plots indicate the median and interquartile range. (c) ROC curves for the identification of HGSOC patients from patients with EOC other than HGSOC (n = 40) based on the expression of CA125 (gray line), miR-1290 (black line), and the combination of both (dotted line). **, P < 0.01; n.s., not significant