The Serum Metabolic Profiling On Different Differentiation Of HCC Patients

Objective To explore the characteristic ions in serum about different differentiation of hepatocellular carcinoma (HCC), and to establish models that could distinguish the stages. And then analyze these substances involved in the metabolic pathways. Aimed at analyzing the relationship between the discrepant metabolites and the degree of differentiation in hepatocellular carcinoma patients. So it could provide help for clinical judgment about the pathologic stage of HCC, and enhance the medical quality, and improve the prognosis.Methods UPLC-MS platform was used to analyze serum samples of69HCC patients in the experimental group, among them there were29cases of highly differentiated,26cases of moderately differentiated,14cases of lowly differentiated, and the control group included20health volunteers and20patients with HBV-induced cirrhosis; The data was pretreated and normalized by the software of Mzmine2.0; through OPLS-DA (orthogonal partial least squares discriminant analysis) of SIMCA-P+12.0.1.0, a disease distinguishing model was established; and then, the VIP (variable importance in projection) values, variable confidence intervals in the VIP plot and coefficient plot were conducted to screen the metabolic ions; and independent sample-nonparametric test was used to make the final screening, and the irons would be identified; at last,ROC curve was used to evaluate the effectiveness which they could diagnose the different pathological stages of HCC.Results We established serum metabonomic OPLS-DA model to diagnose different differentiation of HCC based on UPLC-MS, and then we identified14metabolic ions in total associated with the development of disease. Among them, LysoPE(18:2), PC(18:1), LysoPC(16:1), LysoPC(20:4), CE(18:1), PA(P-16:0e), Cer(d18:0/14:0) level was lower in the moderately and lowly differentiated group than the group of highly differentiated, and each of them was lower compared with the group of health and cirrhosis, and the difference was statistically significant(P<0.05);LysoPC(22:6), Prostaglandin El, Indoleacrylic acid, SM(d18:1/14:0), Octadecanamide, and1,2,4-Nonadecanetriol level was lower in the lowly differentiated group than the group of highly and moderately differentiated, and each of them was lower compared with the group of health and cirrhosis, and the difference was statistically significant(P<0.05).14-Methylhexadecanoic acid level in the moderately differentiated group was lower than the highly, lowly differentiated group and the healthy group (P<0.05), however, it had no statistical difference with the cirrhosis group (P>0.05).And the ROC curve indicated that these characteristic irons would play high efficiency in the diagnosis of the degree of HCC. The area under the curve (AUC) was between0.7to0.9, and the AUC of (Cer (d18:0/14:0)) had reached to0.977, that was to say, it had higher diagnostic accuracy.Conclusions We successed in screening metabolic ions associated with the differentiation of HCC and at the same time they had high diagnostic efficiency based on the UPLC-MS platform, combined with multivariate statistical analysis. To explore the trends of the endogenous metabolites in the pathological process of hepatocellular carcinoma, and it would help the clinical diagnosis and treatment.And also, these ions were considered to be potential diagnostic biomarkers and benefit in further clinical study as novel drug targets. It would be a powerful weapon in the area clinical research.