AUTHOR=He Qinglian , Li Ziqi , Yin Jinbao , Li Yuling , Yin Yuting , Lei Xue , Zhu Wei 

TITLE=Prognostic Significance of Autophagy-Relevant Gene Markers in Colorectal Cancer

JOURNAL=Frontiers in Oncology

VOLUME=Volume 11 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.566539

DOI=10.3389/fonc.2021.566539

ISSN=2234-943X

ABSTRACT=Background: Colorectal cancer (CRC) is a common malignant solid tumor with an extremely low survival rate after relapse. Previous investigations had shown that autophagy possesses a crucial function in tumor. However, there is no consensus on the value of autophagy-associated genes in predicting the prognosis of CRC patients. This work screens autophagy-related markers and signaling pathways that may participate in the development of CRC, and establishes a prognostic model of CRC based on autophagy-associated genes. 
Methods: Gene transcripts from TCGA database and autophagy-associated gene data from the GeneCards database were used to obtain expression levels of autophagy-associated genes, followed by Wilcox tests to screen for autophagy-related differentially expressed genes. Then, 11 key autophagy-associated genes were identified through univariate and multivariate Cox proportional hazard regression analysis and used to establish prognostic models. Based on the multivariate Cox analysis, risk scores were calculated and used to classify samples into high-risk and low-risk groups. Kaplan-Meier survival analysis, risk profiling, and independent prognosis analysis were carried out. Receiver operating characteristic analysis was performed to estimate the specificity and sensitivity of the prognostic model. Finally, to identify the relevant signaling pathways, we performed GSEA, GO, and KEGG analysis. 
Results: A total of 301 autophagy-related genes were differentially expressed in CRC. The areas under the 1-year, 3-year, and 5-year receiver operating characteristic curves of the autophagy-based prognostic model for CRC were 0.764, 0.751, and 0.729, respectively. GSEA analysis of the model showed significant enrichment in several tumor relevant pathways and cellular protective biological process. GO analysis showed that the 11 target autophagy genes were chiefly enriched in mRNA processing, RNA splicing, and regulation of the mRNA metabolic process. KEGG analysis showed enrichment mainly in spliceosomes. We constructed 11 prognostic risk assessment models based on autophagy-related genes in CRC. 
Conclusion: We constructed 11 prognostic risk assessment models based on autophagy-associated genes in CRC. The new model suggests directions and ideas for evaluating prognosis and provides guidance to choose better treatment strategies for CRC.