Background and Aims Thyroid cancer is the disease with the fastest increase in incidence in recent years. Papillary thyroid carcinoma (PTC) is the most common subtype of thyroid cancer. There is an urgent need to find relevant biomarker molecules of PTC, so as to improve the prognosis diagnosis and provide highly specific therapeutic targets.Methods The microarray data sets (GSE60542, GSE33630 and GSE3467) related to PTC in GEO database were retrieved and analyzed, and the differentially expressed genes between PTC and normal thyroid tissue were screed using the GEO2R tool of the GEO database. The genome-wide enrichment analysis of differential genes was performed, and the interactions of proteins of these differential genes were analyzed by online database tools and visualized by Cytoscape software. The prognostic values of the hub genes for PTC were evaluated by Cbioportal analysis tool and further verified by validation experiments.Results A total of 62 up-regulated and 40 down-regulated differential genes were identified, and 10 hub genes with high connectivity were selected. Among the hub genes, the decrease of KIT was associated with poor prognosis of PTC (P<0.01). The expression of KIT was detected in 52 specimens of PTC tissue and normal adjacent tissue by qRT-PCR. The results showed that the expression of KIT in PTC tissue was significantly lower than that in normal adjacent tissue (P<0.001); the expression of KIT was significantly related to the clinical stage (P=0.008) and lymph node metastasis (P=0.023).Conclusion The expression of KIT in PTC is lower than that in normal tissues. It is probably a key gene for poor prognosis of PTC patients, and is expected to serve as a therapeutic target and molecular biomarker.