Background: Anoikis is an apoptotic cell death, which is resulting from the loss of interaction between cells and the extracellular matrix, and has served a prominent role in metastasis. The aim of the present study was to identify an anoikis-revelant genes (ARGs) signature for Lung Adenocarncinoma (LUAD) patients’ prognosis and explore the underlying molecular mechanisms.
Methods: In the training cohort, LUAD patients from The Cancer Genome Atlas (TCGA) were used, and Gene Expression Omnibus (GEO) cohort GSE72094 was used for validation. A total of 508 anoikis-revelant genes downloaded from the GeneCards. Univariate Cox analysis was applied for preliminary screening of anoikis-revelant genes with potential prognostic capacity in the training cohort. These genes were then applied into an overall survival-based LASSO regression model, building a gene signature. The discovered gene signature was then evaluated via Kaplan–Meier (KM), Cox, and ROC analyses in both cohorts. To better explore the functional annotation of the gene signature and the character of tumor microenvironment, the GSEA enrichment and CIBERSORT algorithm were performed.
Results: A thirteen-gene signature was built in the TCGA-LUAD cohort and further validated in GSE72094 cohort, revealing its independent prognosis value in LUAD. Next, the signature's predictive ability for LUAD prognosis was confirmed through ROC analysis. Moreover, analyses of gene enrichment and immune infiltrating detailed exhibited cell adhesion and VEGF pathways related with the thirteen-gene signature, also showing that M0 macrophages, mast cells, dendritic cells and CD4+ memory T cells involved in the prognosis of the thirteen-gene signature.
Conclusions: An inventive anoikis-revelant thirteen-gene signature (ABHD4, CDCP1, CDK1, CENPF, EIF2AK3, FADD, FYN, HGF, OGT, PIK3CG, PPP2CA, RAC1, and XRCC5) was generated through this study. It could accurately predict LUAD prognosis and was related to M0 macrophages, mast cells, dendritic cells, and CD4+ memory T cells.