Senescence-related signatures predict prognosis and response to immunotherapy in colon cancer

Xiaoshan Hu,Xiongjie Zhu,Yifan Chen,Wenkai Zhang,Laiqing Li,Huankun Liang,Bekzod B Usmanov,Matteo Donadon,Abrorjon A Yusupbekov,Yanfang Zheng
DOI: https://doi.org/10.21037/jgo-24-339
2024-06-30
Abstract:Background: Colorectal cancer (CRC) is one of the most common cancers. Cellular senescence plays a vital role in carcinogenesis by activating many pathways. In this study, we aimed to identify biomarkers for predicting the survival and recurrence of CRC through cellular senescence-related genes. Methods: Utilizing The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases, RNA-sequencing data and clinical information for CRC were collected. A risk model for predicting overall survival was established based on five differentially expressed genes using least absolute shrinkage and selection operator-Cox regression (LASSO-Cox regression), receiver operating characteristic (ROC), and Kaplan-Meier analyses. The study also delved into both the tumor microenvironment and the response to immunotherapy. Moreover, we gathered clinical sample data from our center in order to confirm the findings of public database analysis. Results: Through ROC and Kaplan-Meier analyses, a risk model was developed using five cellular senescence-related genes [i.e., CDKN2A, SERPINE1, SNAI1, CXCL1, and ETS2] to categorize patients into high- and low-risk groups. In the TCGA-colon adenocarcinoma (COAD) and GEO-COAD cohorts, the high-risk group was associated with a bleaker forecast (P<0.05), immune cell inactivation, and insensitivity to immunotherapy in IMvigor210 database (http://research-pub.gene.com/IMvigor210CoreBiologies/). Clinical samples were then used to confirm that ETS2 and CDKN2A could serve as independent prognostic biomarkers in CRC. Conclusions: Gene signatures related to cellular senescence, specifically involving CDKN2A and ETS2, are emerging as promising biomarkers for predicting CRC prognosis and guiding immunotherapy.
What problem does this paper attempt to address?