Non-invasive Detection of Hepatocellular Carcinoma Serum Metabolic Profile Through Surface-enhanced Raman Spectroscopy

The present study aims to identify distinctive Raman spectrum metabolic peaks to predict hepatocellular carcinoma (HCC). We performed a label-free, non-invasive surface-enhanced Raman spectroscopy (SERS) test on 230 serum samples including 47 HCC, 60 normal controls (NC), 68 breast cancer (BC) and 55 lung cancer (LC) by mixing Au@AgNRs with serum directly. Based on the observed SERS spectra, discriminative metabolites including tryptophan, phenylalanine, and etc. were found in HCC, when compared with BC, LC, and NC (P<0.05 in all). Common metabolites-proline, valine, adenine and thymine were found in HCC, BC and LC with compared to NC group (P<0.05). Importantly, Raman spectra of HCC serum biomarker AFP were firstly detected to analyze the HCC prominent peak. Orthogonal partial least squares discriminant analysis was adopted to assess the diagnostic accuracy; area under curve value of HCC is 0.991. This study provides new insights into the HCC metabolites detection through Raman spectroscopy.