A spliceosome-associated gene signature aids in predicting prognosis and tumor microenvironment of hepatocellular carcinoma

Abstract

Splicing alterations have been shown to be key tumorigenesis drivers. In this study, we identified a novel spliceosome-related genes (SRGs) signature to predict the overall survival (OS) of patients with hepatocellular carcinoma (HCC). A total of 25 SRGs were identified from the GSE14520 dataset (training set). Univariate and least absolute shrinkage and selection operator (LASSO) regression analyses were utilized to construct the signature using genes with predictive significance. We then constructed a risk model using six SRGs (BUB3, IGF2BP3, RBM3, ILF3, ZC3H13, and CCT3). The reliability and predictive power of the gene signature were validated in two validation sets (TCGA and GSE76427 dataset). Patients in training and validation sets were divided into high and low-risk groups based on the gene signature. Patients in high-risk groups exhibited a poorer OS than in low-risk groups both in the training set and two validation sets. Next, risk score, BCLC staging, TNM staging, and multinodular were combined in a nomogram for OS prediction, and the decision curve analysis (DCA) curve exhibited the excellent prediction performance of the nomogram. The functional enrichment analyses demonstrated high-risk score patients were closely related to multiple oncology characteristics and invasive-related pathways, such as Cell cycle, DNA replication, and Spliceosome. Different compositions of the tumor microenvironment and immunocyte infiltration ratio might contribute to the prognostic difference between high and low-risk score groups. In conclusion, a spliceosome-related six-gene signature exhibited good performance for predicting the OS of patients with HCC, which may aid in clinical decision-making for individual treatment.

Integrated multi-omics analysis and machine learning based on O_linked_glycosylation genes refine molecular subtypes and prognosis for hepatocellular carcinoma

“…Accompanied by the vigorous development of high-throughput sequencing technology in recent years, numerous prognostic signatures have been constructed grounded on LASSO and stepwise COX algorithms with the aim to stratify HCC patients and guide their prognosis (41,42). After a system retrieval of the relevant literature published over the past three years, we nally included 60 unique signatures to compare with our CMLS.…”