Although the diagnosis and therapy approach developed, techniques for the early diagnosis of HCC remain insufficient which results in poor prognosis of patients. The traditional biomarker AFP, however, has been proved with low specificity. Circulating exosomal ncRNAs revealed different profiles reflecting the characteristics of tumour. In this study, we mainly focused on circulating exosomal ncRNAs which might be the fingerprint for HCC, especially for the diagnosis or metastasis prediction. A high throughput lncRNA microarray in exosomes extracted from cell-free plasma was applied. The risk score analysis was employed to screen the potential exosome-derived lncRNAs in two independent sets based on different clinical parameters in 200 paired HCC patients. After a multi-stage validation, we finally revealed three lncRNAs, ENSG00000248932.1, ENST00000440688.1 and ENST00000457302.2, increased in HCC comparing with the both chronic hepatitis (CH) patients and cancer-free controls. ROC curve revealed a higher sensitivity and specificity in predicting the occurrence of HCC from cancer-free controls and CH patients with the area under curve (AUC) of 0.905 and 0.879 by combining AFP. The three lncRNA panel combined with AFP also indicted a fingerprint function in predicting the metastasis of HCC with the AUC of 0.870. In conclusion, ENSG00000248932.1, ENST00000440688.1 and ENST00000457302.2 might be the potential biomarker for the tumorigenesis prediction from CH patients or healthy controls and may also be applied for dynamic monitoring the metastasis of HCC.
Keywords: exosome; fingerprint; lncRNA; plasma; risk score analysis.
© 2019 The Authors. Journal of Cellular and Molecular Medicine published by John Wiley & Sons Ltd and Foundation for Cellular and Molecular Medicine.