Abstract Background Lung adenocarcinoma (LUAD) is one of the most common subtypes of lung cancer. Finding prognostic biomarkers is helpful in stratifying LUAD patients with different prognosis. Methods We explored the correlation of LUAD prognosis and genes associated with chemotherapy in LUAD and obtained data of LUAD patients from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Drug sensitivity data were acquired from the Genomics of Drug Sensitivity in Cancer (GDSC) database. Differential and enrichment analyses were used to screen the target genes utilizing limma and “clusterProfiler” packages. Then univariate and LASSO Cox analyses were used to select the prognosis‐related genes. Survival analysis was used to estimate the overall survival (OS) of different groups. Results Twenty‐three differentially expressed genes (DEGs) were screened between LUAD samples and healthy samples, and BTK, FGFR2, PIM2, CHEK1, and CDK1 were selected to construct a prognostic signature. The OS of patients in the high‐risk group (risk score higher than 0.69) was worse than that in the low‐risk group (risk score lower than 0.69). Conclusion The risk score model constructed by five genes is a potential prognostic biomarker for LUAD patients.