Identification of a Tumor Immunological Phenotype-Related Gene Signature for Predicting Prognosis, Immunotherapy Efficacy, and Drug Candidates in Hepatocellular Carcinoma

Abstract

Hepatocellular carcinoma (HCC) is the predominant subtype of primary liver cancer and represents a highly heterogeneous disease, making it hard to predict the prognosis and therapy efficacy. Here, we established a novel tumor immunological phenotype-related gene index (TIPRGPI) consisting of 11 genes by Univariate Cox regression and the least absolute shrinkage and selection operator (LASSO) algorithm to predict HCC prognosis and immunotherapy response. TIPRGPI was validated in multiple datasets and exhibited outstanding performance in predicting the overall survival of HCC. Multivariate analysis verified it as an independent predictor and a TIPRGPI-integrated nomogram was constructed to provide a quantitative tool for clinical practice. Distinct mutation profiles, hallmark pathways, and infiltration of immune cells in tumor microenvironment were shown between the TIPRGPI high and low-risk groups. Notably, significant differences in tumor immunogenicity and tumor immune dysfunction and exclusion (TIDE) were observed between the two risk groups, suggesting a better response to immune checkpoint blockade (ICB) therapy of the low-risk group. Besides, six potential drugs binding to the core target of the TIPRGPI signature were predicted via molecular docking. Taken together, our study shows that the proposed TIPRGPI was a reliable signature to predict the risk classification, immunotherapy response, and drugs candidate with potential application in the clinical decision and treatment of HCC. The novel “TIP genes”-guided strategy for predicting the survival and immunotherapy efficacy, we reported here, might be also applied to more cancers other than HCC.

Hallmark-guided subtypes of hepatocellular carcinoma for the identification of immune-related gene classifiers in the prediction of prognosis, treatment efficacy, and drug candidates

“…Gene expression data of HCC patients were downloaded from The Cancer Genome Atlas (TCGA) and International Cancer Genome Consortium (ICGC) databases. Those patients with incomplete overall survival (OS) information and with an OS time of <30 days were excluded as reported before ( 20 , 21 ). The clinicopathological features and other types of survival outcomes including progression-free survival (PFS), disease-free survival (DFS), and disease-specific survival (DSS) were also collected.…”

“…The patient population and the clinicopathological characteristics are summarized in Supplementary Table 1 . The normalized RNA sequencing profiles were retrieved and preprocessed as previously reported ( 20 , 23 , 24 ). For the analysis of somatic mutation information, we gathered the available mutation annotation format (MAF) file from the TCGA data portal ( ) using the “maftools” package ( 25 ).…”

“…The “pheatmap” package was utilized to show the IRDEGs’ expression patterns between the two HCC subtypes. As previously described, Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis for these IRDEGs were conducted via the “clusterProfiler” package with an adjusted p < 0.05 ( 20 , 23 , 32 ).…”

“…The concordance index was further used to assess its performance. Moreover, decision curve analysis (DCA) was used to explore the potential clinical benefit of HGSIS as described ( 20 ). Additionally, Kaplan–Meier analysis with a log-rank test was applied to the output of the nomogram-based classifier for the whole TCGA cohort to further compare the differences in OS, DFS, PFS, and DSS between different risk classes.…”

“…The process of molecular docking was completed with Glide of Schrodinger as previously reported ( 20 ). Specifically, the crystal structure of the key target of HGSIS was derived from the RCSB PDB database ( ), followed by the recognition of its active site using the DeepSite tool ( 53 ) from the PlayMolecule platform ( ).…”