Hepatocellular carcinoma (HCC) is a common high-mortality cancer, mainly due to diagnostic difficulties during its early clinical stages. In this study, we aimed to identify genes that are important for HCC diagnosis and treatment, and we investigated the underlying mechanism of prognostic differences. Differentially expressed genes (DEGs) were identified by using the limma package, and receiver operating characteristic curve analysis was performed to identify diagnostic markers for HCC. Bioinformatics and clinical specimens were used to assess epithelial cell transforming 2 (ECT2) in terms of expression, prognostic value, pathways, and immune correlations. In vitro experiments were used to investigate the underlying mechanism and function of ECT2, and the results were confirmed through in vivo experiments. The integrated analysis revealed 53 upregulated DEGs, and one candidate biomarker for diagnosis (ECT2) was detected. High expression of ECT2 was found to be an independent prognostic risk factor for HCC. ECT2 expression showed a strong correlation with tumor-associated macrophages. We found that ECT2 overexpression increased the migration and proliferation of HCC cells. It also promoted the expression of PLK1, which subsequently interacted with PTEN and interfered with its nuclear translocation, ultimately enhancing aerobic glycolysis and promoting M2 macrophage polarization. M2 macrophages suppress the functions of NK cells and T cells, and this was confirmed in the in vivo experiments. Overall, ECT2 may promote the polarization of M2 macrophages by enhancing aerobic glycolysis and suppressing the functions of immune cells. ECT2 could serve as a candidate diagnostic and prognostic biomarker for HCC.