Abstract
BACKGROUND: An important stage in controlling gene expression is RNA alternative splicing (AS), and aberrant AS can trigger the development and spread of malignancies, including hepatocellular carcinoma (HCC). A crucial component of AS is cleavage and polyadenylation-specific factor 4 (CPSF4), a component of the CPSF complex, but it is unclear how CPSF4-related AS molecules describe immune cell infiltration in the total tumor microenvironment (TME). METHODS: Using RNA-sequencing data and clinical data from TCGA-LIHC from the Cancer Genome Atlas (TCGA) database, the AS genes with differential expression were found. The univariate Cox analysis, KM analysis, and Spearman analysis were used to identify the AS genes related to prognosis. Screening of key AS genes that are highly correlated with CPSF4. Key genes were screened using Cox regression analysis and stepwise regression analysis, and prognosis prediction models and the topography of TME cell infiltration were thoroughly analyzed. RESULTS: A model consisting of seven AS genes (STMN1, CLSPN, MDK, RNFT2, PRR11, RNF157, GHR) was constructed that was aimed to predict prognostic condition. The outcomes of the HCC samples in the high-risk group were considerably worse than those in the lower risk group (p < 0.0001), and different risk patient groups were formed. According to the calibration curves and the area under the ROC curve (AUC) values for survival at 1, 2, and 3 years, the clinical nomogram performs well in predicting survival in HCC patients. These values were 0.76, 0.70, and 0.69, respectively. Moreover, prognostic signature was markedly related to immune infiltration and immune checkpoint genes expression. CONCLUSION: By shedding light on the function of CPSF4 and the seven AS genes in the formation and progression of HCC, this research analysis contributes to the development of more useful prognostic, diagnostic, and possibly therapeutic biomarkers.