Pan-Cancer Analysis of Endoplasmic Reticulum Stress-Related Molecular Features and Drug Resistance

Background: The tumor microenvironment (TME) including hypoxia, nutritional deficiency, and reactive oxygen species stress and endoplasmic reticulum stress (ERS) play a determinant role in cancer progression. ERS is critical for maintaining cell survival; however, its influence on the survival and resistance of patients undergoing anti-cancer therapies remains unknown. Methods: We estimated the ERS level of 10423 samples of 33 cancer types from The Cancer Genome Atlas (TCGA) using R package “GSVA”. Samples in each cancer type were classified into high-ERS, intermediate-ERS, and low-ERS status using unsupervised hierarchical clustering (K-means). Finding: We revealed that patients with high ERS scores to be more resistant to anti-cancer drugs. Then, we identified the genes that are abnormally expressed in distinct ERS statuses. We found different molecular features associated with ERS to correlate with anti-cancer drug resistance and sensitivity. We further analyzed the characteristics of these genes, including methylation, copy number, mutation status, transcription level, and pathways, using TCGA pan-cancer data. Interpretation: Our study provides a comprehensive molecular-level understanding of ERS and suggests its potential practical implications in clinical cancer treatment. Funding: This work was supported by the National Nature Science Foundation of China [grant numbers: 81773008, 81672756, 81872399, 81972897]. Declaration of Interests: The authors declare no competing financial interests.