〔Abstract〕 Objective A prognostic risk model for lung adenocarcinoma patients was established based on the cancer genome atlas(TCGA)database to explore the prognostic performance of autophagy related gene risk model for lung adenocarcinoma patients and its correlation with immune microenvironment. Methods Clinical information and transcriptome data of lung adenocarcinoma patients were downloaded and extracted from TCGA database, and 232 autophagy-related genes were screened from the human autophagy database. cox regression analysis was used to screen out four autophagy genes independently associated with prognosis. The prognostic prediction model of lung adenocarcinoma was constructed by risk score, and the performance of prediction model was evaluated by ROC curve. The relationship between risk scores and tumor immune microenvironment was explored using ESTIMATE(estimation of stromal and immune cells in malignant tumour tissues using expression data) and CIBERSORT algorithms. Results Thirty differentially expressed autophagy-related genes were identified in lung adenocarcinoma, of which four autophagy genes(BIRC5,ERO1A,ITGB4,NLRC4) could predict the prognosis of the patients. Grouped by risk score, the Kaplan-Meier analysis demonstrated that the survival rate of high-risk group was significantly lower than that of low-risk group(P<0.000 1). The ROC curve proved the accuracy of the model in predicting the prognosis of lung adenocarcinoma(AUC=0.757). The ESTIMATE and CIBERSORT analyses revealed that the risk scoring model was associated with multiple immune cells and immune infiltrates in the tumor microenvironment. Conclusion Compared with clinical data, the autophagy gene prognostic risk model can better predict the prognosis of patients with lung adenocarcinoma. In the high-risk group, CD4+memory quiescent cells can improve prognosis in lung adenocarcinoma patients.