Clinical Diagnostic and Prognostic Potential of NDRG1 and NDRG2 in Hepatocellular Carcinoma Patients

Abstract

Background: Primary liver cancer is still the most common lethal malignancy. The N-myc downstream-regulated gene family (NDRG1–4) is a group of multifunctional proteins associated with carcinogenesis. However, systematic evaluation of the diagnostic and prognostic values of NDRG1 or NDRG2 expression in liver cancer is poorly investigated. Method: The gene expression matrix of liver hepatocellular carcinoma (LIHC) was comprehensively analyzed by the “limma” and “Dseq2” R packages. The Gene Ontology (GO) and Gene Set Enrichment Analysis (GSEA) were used to identify the biological functional differences. A single-sample GSEA (ssGSEA) was conducted to quantify the extent of immune cell infiltration. Finally, the clinical and prognostic information of LIHC patients was systematically investigated using Kaplan–Meier analysis and logistic and Cox regression analysis. Results: Compared with normal tissues, NDRG1 expression was higher, whereas NDRG2 expression was lower in tumor tissues (P <0.001). The area under the receiver operator characteristic curve (AUROC) of NDRG1 and NDRG2 for LIHC was 0.715 and 0.799, respectively. Kaplan–Meier analysis revealed that NDRG1 and NDRG2 were independent clinical prognostic biomarkers for the overall survival (OS, P = 0.001 and 2.9e−06), progression-free interval (PFI, P = 0.028 and 0.005) and disease-specific survival (DSS, P = 0.027 and P <0.001). The C-indexes and calibration plots of the nomogram suggest that NDRG1 and NDRG2 have an effective predictive performance for OS (C-index: 0.676), DSS (C-index: 0.741) and PFI (C-index: 0.630) of liver cancer patients. The mutation rate of NDRG1 in liver cancer reached up to 14%, and DNA methylation levels of NDRG1 and NDRG2 promoters correlated significantly with clinical prognosis. Conclusions: The mRNA expression and DNA methylation of NDRG superfamily members have the potential for LIHC diagnosis and prognosis via integrative analysis from multiple cohorts.