Background: Metabolic dysfunction Associated Fatty Liver Disease (MAFLD) is increasingly recognized as a major health burden in developed countries. It can eventually progress to HCC and up to 25% of MAFLD-HCC arise in the absence of severe liver fibrosis, posing a challenge for early detection and treatment (De A et al., 2020, J clin Exp Hepatol). We previously reported the existence of 2 phenotypes of MAFLD-HCC by metabolomics analysis according to fibrosis level (F0F1 vs. F3F4) (Buchard et al., 2021, Metabolites, Buchard et al., 2021, AASLD Hepatology). The aim of our current study is to explore lipid pathways and identify potential biomarkers related to MAFLD-HCC. Methods: Fifty-six pairs (F0F1=28, F3F4=28) of human MAFLD-HCC (TT) and non-tumor tissues (NTT) and five healthy tissues were collected from CRB. Foie. A non-targeted metabolomics strategy was applied using LC-MS. Based on the results of LC-MS, qRT-PCR regarding sphingomyelin synthase 2 (SGMS2), sphingomyelin phosphodiesterase 1 (SMPD1), choline Kinase alpha (CHKA) and membrane-bound O-acyltransferase 7 (MBOAT7) was performed. Results: Firstly, LC-MS analysis shown that the comparison between the two groups of MAFLD-TT and MAFLD-NTT revealed the presence of two different lipids profiles according to the fibrosis severity (F0F1 vs. F3F4). Most of sphingolipids including ceramides (Cer) and sphingomyelins (SM), and glycerophospholipids, including phosphatidylcholine (PC), phosphatidylethanolamine (PE) and phosphatidylinositol (PI) were increased in MAFLD-HCC-F0F1 while they decreased in MAFLD-HCC-F3F4 (Fig. 1A). Secondly, the results of qRT-PCR indicated that the RNA expression of SGMS2, SMPD1 remain unchanged in MAFLD-TT compared with NTT, regardless of fibrosis level. In contrast, the RNA expression of CHKA and MBOAT7 were exclusively up-regulated in MAFLD-TT-F0F1 compared to NTT-F0F1 using healthy tissues as control (Fig. 1B). These results were in accordance with our metabolomics data that have shown that PC content were highly accumulated in MAFLD-TT-F0F1. Conclusion: Metabolomics and transcriptomic analysis allow us to discriminate MAFLD-HCC according to fibrosis severity. Our findings identified that CHKA and MBOAT7 could be proposed as biomarkers to MAFLD-HCC patients without or with low level of fibrosis. In addition, the different expression of CHKA support the idea that choline could be used as a more efficient tracer of PET-scan in MAFLD-HCC patients.