Objective Lung cancer is the leading cause of cancer-related death globally, and non-small cell lung cancer (NSCLC) is the most common type of lung cancer. However, the diagnosis and prognosis of NSCLC remain dim. Our team has focused on identifying differentially expressed genes (DEGs) between NSCLC tissues and adjacent tissues, which may be useful as effective diagnostic markers that can better explain the progression of NSCLC. Methods The Gene Expression Omnibus (GEO) database was used to screen the Gene Expression Omnibus series, which records the information of a large number of patients with primary NSCLC (n > 50). Then, the DEGs were validated using Student’s t -test. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using DAVID. The prognosis information was analyzed separately using data obtained from three databases, Human Protein Atlas, Kaplan–Meier Plotter, and SurvExpress. Results A series of 180 DEGs (33 upregulated and 147 downregulated genes), mainly involving genes associated with extracellular exosomes, focal adhesion, and cell adhesion, were identified via GO analysis. Subsequently, KEGG analysis demonstrated that focal adhesion, cell adhesion molecules, and PPAR signaling pathway were the most enriched pathways. Then, we paid particular attention to pleckstrin 2 (PLEK2) and sodium voltage-gated channel alpha subunit 7 (SCN7A), as they have not been investigated as cancer-related genes previously. Kaplan–Meier survival analysis illustrated that PLEK2 and SCN7A levels were significantly correlated with the prognosis of NSCLC. Conclusions Our research found that, as potential biomarkers, both PLEK2 and SCN7A are related to the development and prognosis of NSCLC. They may be used in disease screening and prognosis. The clinical significance of these two genes deserves further investigation.