Development and Validation of a Prognostic Model of RNA-Binding Proteins in Colon Adenocarcinoma: A Study Based on TCGA and GEO Databases

Background: Previous studies reported that dysregulation of RNA-binding proteins (RBPs) is significantly associated with the development of cancer. However, there are few studies to date on the role of RBPs in colon adenocarcinoma (COAD).

Methods: RNA sequencing and clinical data for COAD patients were downloaded from The Cancer Genome Atlas (TCGA) database to identify differentially expressed (DE) RBPs between COAD tissue and normal colon tissue, and then the expression and prognostic significance of these RBPs were investigated in detail by systematic bioinformatics analysis. qRT-PCR was used to validate the expressions of prognosis-related RBP-encoding genes.

Results: Seven RBPs (RPL10L, ERI1, POP1, CAPRIN2, TDRD7, SNIP1 and PPARGC1A) were identified as hub genes associated with prognosis by a series of regression analyses, and were then used to construct a prognostic model. Further analysis based on this model indicated that the overall survival (OS) of the high-risk groups was lower than that of the low-risk groups. In this prognostic model, the area under the ROC curve (AUC) was 0.694, 0.709 and 0.665 for the TCGA cohort at 1, 3 and 5 years, respectively, while the AUC was 0.671, 0.633 and 0.601 for the GEO combined cohort at 1, 3 and 5 years, respectively, indicating the good predictive ability of the model. We also built a nomogram based on the 7 RBPs in the TCGA cohort, and the model showed good discriminatory ability for COAD.

Conclusion: We screened seven prognosis-related genes in COAD patients based on RBP-related genes, validated the expressions of the seven prognosis-related RBP-encoding genes by qRT-PCR and constructed a prognosis-related nomogram for patients with COAD.