Objective Using the TCGA database to explore the prognostic effects of DNA methylation sites on lung adenocarcinoma. Methods TCGA database was used to collecting DNA methylation data, gene expression data and clinical data of lung adenocarcinoma patients. The population were divided into the exploratory group and the validation group. The LASSO regression analysis was used to screen the methylation sites associated with the prognosis of lung adenocarcinoma in the exploratory group. Receiver operating characteristic curve was used to evaluate the prediction effect of the model, and further verified in the validation population. Results A total of 158 methylation sites with differential expression and significant correlation with the mRNA expression of the corresponding gene were screened from the TCGA database. With LASSO regression analysis, the DNA methylation sites associated with prognosis of lung adenocarcinoma were cg19378330. Those patients with levels above the median methylation site were assigned to the high-risk group, while those with levels below the median methylation site were assigned to the low-risk group. Patients' death risk in the high-risk group were 38% higher than those in the low-risk group (OR=1.38, 95%CI=1.16-2.69). The area under the curve was 0.80 (95%CI=0.73-0.88) and the sensitivity was 86.2% in the exploratory stage population.Similar results were seen in the validation population. Conclusions The DNA methylation site cg19378330 was significantly associated withthe prognosisof lung adenocarcinoma, and could effectively predict the risk of lung adenocarcinoma.