Abstract
BACKGROUND: Hepatocellular carcinoma (HCC) being a complex disease, commonly exhibits multifaceted presentations, rendering its treatment challenging and necessitating specific approaches. The tumor immune microenvironment is crucial in cancer treatment, and cholesterol metabolism is a key component that helps cells grow and produce vital metabolites. However, the reprogramming of cholesterol metabolism in the tumor microenvironment (TME) can promote HCC development, and cancer classifiers relating to cholesterol metabolism are currently limited. Despite significant progress, further research is needed to improve early detection, liver function, and treatment options to improve patient outcomes. METHODS: To evaluate the expression abundance of tumor immune microenvironment (TIME) and cholesterol metabolism in 8 types of liver cancer cells, we comprehensively evaluated the immune cell composition, extracellular matrix alterations, and activity of relevant signaling pathways in the TIME through nine liver cancer patients, stromal scoring, immune scoring, tumor purity scoring, immune infiltration analysis, and pathway enrichment. Subsequently, we utilized machine learning techniques to construct prognostic models for both cholesterol metabolism and the tumor immune microenvironment, further exploring the tumor mutation burden, immune infiltration levels, and drug sensitivity in different subtypes of HCC patients. RESULTS: Our study constructed three cancer screening models to identify HCC patients with high cholesterol metabolism and low TIME, who have a poorer prognosis. On the contrary, patients with low cholesterol metabolism and high TIME often have better prognosis. Furthermore, we identified chemical compounds, such as BPD-00008900, ML323, Doramapimod, and AZD2014, which display better chemotherapy results for high-risk patients in specific sub-groups.