Pathway analysis of metabolic genes and serum metabolites in HCC reveal aberrations associated with fatty acid and lipid metabolism

Objective: Hepatocellular carcinoma (HCC) is among thedeadliest cancers in Thailand. The Thailand Initiative inGenomics and Expression Research for Liver Cancer (TIGERLC)cohort was established to discover molecular markersthat can be used for early detection and diagnosis and tounderstand possible molecular mechanisms underlying thedisease. Previous work on the discovery cohort of 199 cancerpatients found 491 metabolic genes that were correlated with40 tumor-specific tissue metabolites and 75 serum metabolites,which were also associated with patient outcome. In this study,we analyzed those metabolic genes at the pathway level todetermine possible underlying molecular mechanisms of HCCin this cohort. Methods: We performed pathway analysis of 491 metabolicgenes from HCC samples in the TIGER-LC cohort. Theexpression of metabolic genes was derived from transcriptomeprofiles using the Affymetrix Human Transcriptome Array 2.0.We used Gene-Set Enrichment Analysis (GSEA) and IngenuityPathway Analysis (IPA) tools. The two-class comparison wasused in both analyses, where the classes were the matchedtumor and non-tumor samples from the cohort. Results: Biological pathways with statistically significant resultsfrom GSEA were associated with fatty acid metabolic processes,lipid metabolic processes, and small molecule metabolicprocesses. Analysis by IPA revealed similar trends, with the topmolecular and cellular functions identified as lipid metabolism,molecular transport, and small molecule biochemistry. Furtherscrutiny in IPA resulted in several interesting transcriptionfactors that might explain the connection between previousresults and the molecular functions identified in this study. Conclusions: Pathway analysis of metabolic genes associatedwith patient subtypes in previous studies revealed pathwaysassociated with fatty acid and lipid metabolism that mightexplain in part the changes in serum metabolites of HCCpatients in the TIGER-LC cohort. These pathways andassociated transcription factors in the networks might beimportant in HCC early detection and intervention in the Thaipopulation. DOI: 10.20892/j.issn.2095-3941.2018.S101