Combination of FOXD1 and Plk2: A Novel Biomarker for Predicting Unfavourable Prognosis of Colorectal Cancer

Colorectal cancer (CRC) is a worldwide disease with worse survival. Our objective is to identify previously unrecognized prognostic factors to better evaluate disease progression. Seven GEO datasets were collected and analysed using R software, followed by KEGG enrichment analysis and TFs network construction. LASSO-COX analysis was performed to select the most useful prognostic features. COX model was used to analyse prognostic factors associated with OS. The survival curve was constructed using Kaplan-Meier analysis. A Nomogram model was also constructed to predict prognosis. A total of 3559 differentially expressed genes (DEGs) and 66 differentially expressed transcription factors were identified. FOXD1 was identified as the most differentially expressed factor of TFs covering the most downstream DEGs and independent risk prognostic factor. Next, FOXD1 expression was detected using immunohistochemical staining in 131 CRC patients' tissue and the association between FOXD1 expression and clinicopathologic features was analysed. High expression of FOXD1 was correlated with TNM stage and pathological differentiation. Multivariate COX regression analyses confirmed that FOXD1 high-expression, TNM stage and tumour differentiation were independent prognostic risk factor of OS and DFS. Patients with high expression of FOXD1 were more likely to have poor overall survival and disease-free survival. The combination of FOXD1 and Plk2 which we have previously reported allowed us to predict the survival of post-surgical CRC patients more accurately, adding to the former prognostic model based on the TNM Stage. The results showed that patients with high expression of both FOXD1 and Plk2 have the worst survival. A combination of FOXD1 and Plk2 can better evaluate patients' survival.