Background and aims: Metabolomics serves as an important tool in distinguishing changes in metabolic pathways and the diagnosis of human disease. Hepatocellular carcinoma (HCC) is a malignance present of heterogeneous metabolic disorder and lack of effective biomarker for surveillance and diagnosis. In this study, we searched for potential metabolite biomarkers of HCC using tissue and serum metabolomics approach.
Methods: A total of 30 pairs of matched liver tissue samples from HCC patients and 90 serum samples (30 HCC patients, 30 liver cirrhosis patients, and 30 healthy individuals) were assessed. Metabolomics was performed through ultra performance liquid chromatography-mass spectrometry in conjunction with multivariate and univariate statistical analyses.
Results: A total of six differential metabolites including chenodeoxycholic acid (CDCA), glycocholic acid (GCA), LPC20:5, LPE18:0, succinyladenosine and uridine were present in HCC tissue and serum samples. CDCA, LPC20:5, succinyladenosine and uridine were used to construct a diagnostic model based on logistic regression. The four-metabolite panel discriminated HCC from liver cirrhosis with an AUC score of 0.938, sensitivity of 93.3% and specificity of 86.7%. For all HCC and cirrhosis patients, the diagnostic accuracy increased to 96.7% and 90.0%, respectively.
Conclusion: The combination of CDCA, LPC20:5, succinyladenosine and uridine can be used as a biomarker panel to improve HCC sensitivity and specificity. This panel significantly benefits HCC diagnostics and reveals new insight into HCC pathogenesis.
Keywords: Diagnostic ability; Hepatocellular carcinoma; Non-targeted metabolomics; Tissue and serum.
Copyright © 2018 Elsevier B.V. All rights reserved.