Circulating microRNA profiles in plasma: identification of miR-224 as a novel diagnostic biomarker in hepatocellular carcinoma independent of hepatic function

Abstract

Aims: This study was designed to identify novel microRNAs (miRNAs) in plasma for detecting and monitoring hepatocellular carcinoma (HCC), independent of hepatic function and background liver diseases with different etiologies.

Results: (1) Four oncogenic miRNAs (miR-151, 155, 191 and 224) with high expression in HCC tissues were selected as candidates. (2) Quantitative RT-PCR using plasma samples from 107 HCC patients and 75 healthy volunteers revealed a significantly higher level of plasma miR-224 in HCC patients than in healthy volunteers according to a small-scale analysis (P < 0.0001), two independent large-scale cohort analysis (P < 0.0001, AUC 0.908). (3) miR-224 expression was significantly higher in HCC tissues and HCC cell lines than in normal hepatic tissues and fibroblasts, respectively. (P = 0.0011, 0.0150) (4) Plasma miR-224 reflected tumor dynamics; preoperative plasma levels of miR-224 were significantly reduced in postoperative samples (P = 0.0058), and plasma miR-224 levels were significantly correlated with paired miR-224 levels in HCC tissues (P = 0.0005). (5) Furthermore, plasma miR-224 levels significantly discriminated HCC patients from patients with chronic liver disease (P = 0.0008). A high plasma miR-224 level was significantly correlated with larger tumor size (P = 0.0005) and recurrences (P = 0.0027). The plasma miR-224 level could accurately detect small tumors less than 18 mm preoperatively.

Methods: We performed a systematic review of the NCBI database and selected candidate miRNAs reported as highly expressed in HCC tissue.

Conclusions: Plasma miR-224 may be a sensitive biomarker for screening HCC and monitoring tumor dynamics.

Keywords: circulating microRNA; liquid biopsy; liver cirrhosis; plasma; tumor marker.

Figure 1

A. Selection of the candidate microRNAs from the NCBI database. B. Study design to identify novel plasma miRNA biomarkers for HCC. Using a systematic review of the NCBI database, we selected four candidate miRNAs for cancer detection and monitoring. Candidate miRNAs, previously reported to have an oncogenic role in HCC and previously confirmed in tissues or cell lines but not in body fluids, were selected. Among these, the expression level of plasma miR-224 showed the most significant difference (P < 0.0001). Thus, further analyses of plasma miR-224 were conducted.

Figure 2. Small-scale analyses comparing the plasma levels of four candidate miRNAs between HCC patients and healthy volunteers

Plasma levels of the four candidate miRNAs in 20 HCC patients and 20 healthy volunteers were analyzed using qRT-PCR. The expression level of each miRNA was normalized to that of cel-miR-39 as described in the Materials and Methods. All candidate miRNAs showed significantly higher expression levels in HCC patients than in healthy volunteers. Among these, the expression level of plasma miR-224 showed the most significant difference (P < 0.0001).

Figure 3

A, B. Confirmation of the expression level of miR-224 in HCC tissues and HCC cell lines. miR-224 expression was significantly higher in HCC tissues (P = 0.0011) and HCC cell lines (P = 0.0150) than in normal tissues and fibroblasts, respectively. C. Large-scale analysis of the plasma level of miR-224 in HCC patients using a validation cohort. The expression level of plasma miR-224 was significantly higher in HCC patients than in healthy volunteers (P < 0.0001), and this finding was validated in large-scale analysis (AUC was 0.908). For the large-scale analysis, total RNA extracted from plasma samples obtained from 87 HCC patients and 55 age-matched healthy volunteers were used to analyze the expression level of miR-224 using qRT-PCR. D. The correlation between the plasma and the tissue levels of miR-224 were confirmed in HCC patients using paired samples (P = 0.0005). E. Evaluation of whether the plasma miR-224 levels reflect tumor dynamics. The expression level of miR-224 was significantly reduced in postoperative plasma samples (P = 0.0058).

Figure 4. Evaluation of whether the plasma miR-224 level could satisfactory distinguish early stage HCC patients form healthy volunteers

The expression level of plasma miR-224 was significantly higher in stage I HCC patients than in healthy volunteers (P < 0.0001). The AUC values in the LCSGJ and TNM (AJCC/UICC) staging systems for the plasma miR-224 analysis were 0.888 and 0.899, respectively.

Figure 5. Evaluation of tumor detection based on the plasma miR-224 level, independent of chronic hepatic disease and hepatic function

A. There were no significant differences in the plasma miR-224 levels among patients with HBV, HCV and other disease in the large-scale analysis (P = 0.3087). B, C. The plasma miR-224 levels were significantly higher in the plasma of HCC patients compared with non-HCC outpatients among patients with mild (P < 0.0001) or severe liver dysfunction (P = 0.0008).

Figure 6. Correlation between the plasma level of miR-224 and conventional serum tumor markers and indicators of hepatic functions in HCC patients

No significant correlation was observed between the plasma miR-224 levels and other clinical indicators, such as the tumor makers PIVKA-II (DCP) and AFP, the ICG retention rate (ICG 15R), total bilirubin (T-Bil), albumin (Alb), prothrombin time (PT (%)), aspartate transaminase (AST) and alanine transaminase (ALT).

Figure 7. Post-treatment plasma miR-224 level as a sensitive indicator of residual tumor during HCC treatment

Current tumor markers and imaging modalities, such as Lipiodol^®-CT and dynamic MRI, have limitations in the detection of residual tumor after local therapy for HCC. (A) The plasma miR-224 level was significantly higher in patients with residual tumors than in patients with no remaining cancer cells (P = 0.0318) (B) In these patients, conventional tumor markers and dynamic MRI could not discriminate residual HCC.