Autophagy-related prognostic signature characterizes tumor microenvironment and predicts response to ferroptosis in gastric cancer

Haoran Li; Bing Xu; Jing Du; Yunyi Wu; Fangchun Shao; Yan Gao; Ping Zhang; Junyu Zhou; Xiangmin Tong; Ying Wang; Yanchun Li

doi:10.3389/fonc.2022.959337

Autophagy-related prognostic signature characterizes tumor microenvironment and predicts response to ferroptosis in gastric cancer

Front Oncol. 2022 Aug 16:12:959337. doi: 10.3389/fonc.2022.959337. eCollection 2022.

Authors

Haoran Li¹, Bing Xu², Jing Du¹, Yunyi Wu¹, Fangchun Shao³, Yan Gao¹, Ping Zhang¹, Junyu Zhou¹, Xiangmin Tong¹, Ying Wang⁴, Yanchun Li⁴

Affiliations

¹ Laboratory Medicine Center, Department of Clinical Laboratory, Zhejiang Provincial People's Hospital (Affiliated People's Hospital, Hangzhou Medical College), Hangzhou, China.
² Department of Clinical Laboratory, Hangzhou Women's Hospital, Hangzhou, China.
³ Department of Pulmonary and Critical Care Medicine, Zhejiang Provincial People's Hospital (Affiliated People's Hospital, Hangzhou Medical College), Hangzhou, China.
⁴ Department of Central Laboratory, Affiliated Hangzhou First People's Hospital, Zhejiang University School of Medicine, Hangzhou, China.

Abstract

Background: Gastric cancer (GC) is an important disease and the fifth most common malignancy worldwide. Autophagy is an important process for the turnover of intracellular substances. Autophagy-related genes (ARGs) are crucial in cancer. Accumulating evidence indicates the clinicopathological significance of the tumor microenvironment (TME) in predicting prognosis and treatment efficacy.

Methods: Clinical and gene expression data of GC were obtained from The Cancer Genome Atlas and Gene Expression Omnibus databases. A total of 22 genes with differences in expression and prognosis were screened from 232 ARGs. Three autophagy patterns were identified using an unsupervised clustering algorithm and scored using principal component analysis to predict the value of autophagy in the prognosis of GC patients. Finally, the relationship between autophagy and ferroptosis was validated in gastric cancer cells.

Results: The expression of ARGs showed obvious heterogeneity in GC patients. Three autophagy patterns were identified and used to predict the overall survival of GC patients. These three patterns were well-matched with the immunophenotype. Kyoto Encyclopedia of Genes and Genomes and Gene Ontology enrichment analyses showed that the biological functions of the three autophagy patterns were different. A scoring system was then set up to quantify the autophagy model and further evaluate the response of the patients to the immunotherapy. Patients with high autophagy scores had a more severe tumor mutation burden and better prognosis. High autophagy scores were accompanied by high microsatellite instability. Patients with high autophagy scores had significantly higher PD-L1 expression and increased survival. The experimental results confirmed that the expression of ferroptosis genes was positively correlated with the expression of autophagy genes in different autophagy clusters, and inhibition of autophagy dramatically reversed the decrease in ferroptotic cell death and lipid accumulation.

Conclusions: Autophagy patterns are involved in TME diversity and complexity. Autophagy score can be used as an independent prognostic biomarker in GC patients and to predict the effect of immunotherapy and ferroptosis-based therapy. This might benefit individualized treatment for GC.

Keywords: autophagy; ferroptosis; gastric cancer; microsatellite instability; prognosis; tumor microenvironment.