Abstract
Hepatocellular carcinoma (HCC) has been recognized as the third leading cause of cancer-related deaths worldwide. There is increasing evidence that the abnormal expression of autophagy-related genes plays an important role in the occurrence and development of HCC. Therefore, the study of autophagy-related genes can further elucidate the genetic drivers of cancer and provide valuable therapeutic targets for clinical treatment. In this study, we used 232 autophagy-related genes extracted from the Human Autophagy Database (HADb) and Molecular Signatures Database (MSigDB) to construct 1884 autophagy-related gene pairs. On this basis, we developed a prognostic model based on autophagy-related gene pairs using least absolute shrinkage and selection operator (LASSO) Cox regression to evaluate the prognosis of patients after liver cancer resection. We then used 845 liver cancer samples from three different databases to test the reliability of the risk signature through survival analysis, receiver operating characteristic (ROC) curve analysis, univariate and multivariate analysis. To further explore the underlying biological mechanisms, we conducted an enrichment analysis of autophagy-related genes. Finally, we combined the signature with independent prognostic factors to construct a nomogram. Based on the autophagy-related gene pair (ARGP) signature, we can divide patients into high- or low-risk groups. Survival analysis and ROC curve analysis verified the validity of the signature (AUC: 0.786—0.828). Multivariate Cox regression showed that the risk score can be used as an independent predictor of the clinical outcomes of liver cancer patients. Notably, this model has a more accurate predictive effect than most prognostic models for hepatocellular carcinoma. Moreover, our model is a powerful supplement to the HCC staging indicator, and a nomogram comprising both indicators can provide a better prognostic effect. Based on pairs of multiple autophagy-related genes, we proposed a prognostic model for predicting the overall survival rate of HCC patients after surgery, which is a promising prognostic indicator. This study confirms the importance of autophagy in the occurrence and development of HCC, and also provides potential biomarkers for targeted treatments.
Highlights
Hepatocellular carcinoma, the predominant primary tumor of the liver, has been recognized as the third leading cause of cancerrelated death worldwide (Forner et al, 2018)
autophagy-related gene pair (ARGP) were constructed using a total of 269 autophagy-related genes that are represented in all three data sets
We used univariate Cox regression analysis to screen 117 prognostic ARGPs that were significantly associated with overall survival (p < 0.001), and established a prognostic gene model of ARGP using Lasso penalty score Cox regression in the The Cancer Genome Atlas (TCGA) dataset
Summary
Hepatocellular carcinoma, the predominant primary tumor of the liver, has been recognized as the third leading cause of cancerrelated death worldwide (Forner et al, 2018). Many patients are diagnosed when the cancer has already metastasized and a series of severe complications have occurred, indicating that the liver cancer has reached an advanced stage (Cabibbo et al, 2010). In site of recent advances in surgical resection or liver transplantation, the 5-year survival rate of HCC patients remains relatively low (Bosetti et al, 2014; Singal and El-Serag, 2015). There is increasing evidence that abnormal expression of autophagy-related genes plays a pathogenic role in the development of multiple human diseases, including cancer (Mizushima, 2018). As autophagy plays a key role in hepatocellular carcinoma, prognostic signatures based on autophagy-related genes can help us explore the genetic control mechanism of hepatocellular carcinoma and provide valuable therapeutic targets (Lin et al, 2018). Few studies have used autophagy-related genes to construct prognostic signatures for HCC
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
