AUTHOR=Fang Qiongxuan , Chen Hongsong 

TITLE=Development of a Novel Autophagy-Related Prognostic Signature and Nomogram for Hepatocellular Carcinoma

JOURNAL=Frontiers in Oncology

VOLUME=Volume 10 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2020.591356

DOI=10.3389/fonc.2020.591356

ISSN=2234-943X

ABSTRACT=Background: Hepatocellular carcinoma (HCC) is the seventh most common malignancy and the second most common cause of cancer-related deaths. Autophagy plays a crucial role in the progression of HCC development.
Methods: Univariate and Lasso Cox regression analyses were performed to identify the optimal gene model for overall survival (OS) prediction. Patients in The Cancer Genome Atlas (TCGA), GSE14520, and GSE54236 datasets were divided into high-risk and low-risk groups according to the established ATG models. Univariate and multivariate Cox regression analyses were used to identify risk factors of OS for nomogram construction. Calibration and receiver operating characteristic (ROC) curves were used to evaluate the model performance. Real-time PCR was used in HepG2 and Huh7 cell lines to validate gene expression in the presence and absence of an autophagy inhibitor. 
Results: OS was significantly shorter in the high-risk group than in the low-risk group. GSEA enrichment analysis showed that the low-risk group was significantly associated with autophagy- and immune-related pathways. ULK2, PPP3CC, and NAFTC1 might play vital roles in preventing HCC progression. Furthermore, tumor environment analysis by ESTIMATION showed that the low-risk group was related to high immune and stromal scores. Based on the EPIC prediction, CD8+ T and B cell fractions were significantly higher in the low-risk group than in the high-risk group in TCGA and GSE54236 datasets. Finally, three variables were selected based on univariate and multivariate analyses for the development of the nomogram. The calibration plots showed good agreement between the nomogram prediction and actual observations. In HepG2 and Huh7 cells, the genes constituting the gene model were overexpressed when autophagy was inhibited.
Conclusions: We determined the role of autophagy-related genes (ATGs) in the progression of HCC and constructed a novel nomogram for predicting the OS in HCC patients through a combined analysis of TCGA and gene expression omnibus (GEO) databases.