Screening of Predictive Markers of Neoadjuvant Chemoradiotherapy in Esophageal Cancer Based on Weighted Gene Co-Expression Network Analysis

Abstract

Objective: This study aimed to identify gene markers that can predict the response to neoadjuvant chemoradiotherapy (neo-CRT) in esophageal cancer. Methods: Three datasets were used. After pre-processing, weighted gene co-expression network analysis (WGCNA) was used to screen the key module. Differentially expressed genes (DEGs) were selected, followed by screening of chemoradiotherapy (CRT) response markers by least absolute shrinkage and selection operator (LASSO) regression analysis. Results: Pink and yellow modules were screened using WGCNA. In total, 763 DEGs were identified. Ninety-eight common genes were identified after Venn analysis. Finally, LASSO regression analysis revealed 12 predictive gene markers, including LCE3D, PPP4R4, CTNNA2, ALOX12b, GLIS3, LINC00592, RIBC2, IQCF5-AS1, WIF1, MRAP2, ZIC1, and AkR1C1. The model constructed using these 12 genes accurately predicted the CRT response. Conclusions: The 12 screened genes, such as CTNNA2, WIF1, and GLIS3 may serve as predictive markers of neo-CRT response in esophageal cancer.