Identification of the molecular subtypes and signatures to predict the prognosis, biological functions, and therapeutic response based on the anoikis‐related genes in colorectal cancer
Xiang Zhai,Baoxiang Chen,Heng Hu,Yanrong Deng,Yazhu Chen,Yuntian Hong,Xianghai Ren,Congqing Jiang
DOI: https://doi.org/10.1002/cam4.7315
IF: 4.711
2024-05-26
Cancer Medicine
Abstract:Background Tumors that resist anoikis, a programmed cell death triggered by detachment from the extracellular matrix, promote metastasis; however, the role of anoikis‐related genes (ARGs) in colorectal cancer (CRC) stratification, prognosis, and biological functions remains unclear. Methods We obtained transcriptomic profiles of CRC and 27 ARGs from The Cancer Genome Atlas, the Gene Expression Omnibus, and MSigDB databases, respectively. CRC tissue samples were classified into two clusters based on the expression pattern of ARGs, and their functional differences were explored. Hub genes were screened using weighted gene co‐expression network analysis, univariate analysis, and least absolute selection and shrinkage operator analysis, and validated in cell lines, tissues, or the Human Protein Atlas database. We constructed an ARG‐risk model and nomogram to predict prognosis in patients with CRC, which was validated using an external cohort. Multifaceted landscapes, including stemness, tumor microenvironment (TME), immune landscape, and drug sensitivity, between high‐ and low‐risk groups were examined. Results Patients with CRC were divided into C1 and C2 clusters. Cluster C1 exhibited higher TME scores, whereas cluster C2 had favorable outcomes and a higher stemness index. Eight upregulated hub ARGs (TIMP1, P3H1, SPP1, HAMP, IFI30, ADAM8, ITGAX, and APOC1) were utilized to construct the risk model. The qRT‐PCR, Western blotting, and immunohistochemistry results were consistent with those of the bioinformatics analysis. Patients with high risk exhibited worse overall survival (p
oncology