Identification and validation of a novel stress granules-related prognostic model in colorectal cancer

Aims: A growing body of evidence demonstrates that Stress granules (SGs), a nonmembrane cytoplasmic compartments, are important to colorectal development and chemoresistance. However, the clinical and pathological significance of SGs in colorectal cancer (CRC) patients is unclear. The aim of this study is to propose a newprognosticmodel related to SGs forCRC on the basis of transcriptional expression. Main methods: Differentially expressed SGs-related genes (DESGGs) were identified in CRC patients from TCGA dataset by limma R package. The univariate and Multivariate Cox regressionmodelwas used to construct a SGs-related prognostic prediction gene signature (SGPPGS). The CIBERSORT algorithm was used to assess cellular immune components between the two different risk groups. ThemRNA expression levels of the predictive signature from 3 partial response (PR) and 6 stable disease (SD) or progress disease (PD) after neoadjuvant therapy CRC patients' specimen were examined. Key findings: By screening and identification, SGPPGS comprised of four genes (CPT2, NRG1, GAP43, andCDKN2A) fromDESGGs is established. Furthermore, we find that the risk score of SGPPGSis an independent prognostic factor to overall survival. Notably, the abundance of immune response inhibitory components in tumor tissues is upregulated in the group with a high-risk score of SGPPGS. Importantly, the risk score of SGPPGS is associated with the chemotherapy response in metastatic colorectal cancer. Significance: This study reveals the association between SGs related genes and CRC prognosis and provides a novel SGs related gene signature for CRC prognosis prediction.