The aim of this study was to find disease-associated genes and gene functions in esophageal squamous cell carcinoma (ESCC) with DNA microarrays. We downloaded the gene expression profile GSE20347 from the Gene Expression Omnibus database including 17 ESCC and 17 matched normal adjacent tissue samples. Compared with normal samples, the probe level data were pre-processed and the differentially expressed genes (DEGs) were identified (FDR <0.05, and |logFC|>2) with packages in R language. The selected DEGs were further analyzed with bioinformatic methods. After an interaction network of DEGs was constructed by STRING, we selected the most important hub gene through network topological analysis (including node degree, clustering coefficient and path length) and analyzed functions and pathways of the hub gene network. A total of 538 genes were filtered as DEGs between normal and disease samples, and we selected the gene TSPO as the most important hub gene. Among its interactors, the CTSK gene and the IL8 gene participated in the toll-like receptor signaling pathway which is closely related to tumor occurrence. The TSPO gene and its interactors may affect the cancer-specific gene expression by participating in the toll-like receptor signaling pathway. Our discovery may be useful in investigating the complex interacting mechanisms underlying the disease. However, further experiments are still needed to confirm our result.