Lung adenocarcinoma (LUAD) is a malignant tumor of the respiratory system that has a poor 5-year survival rate. Anoikis, a type of programmed cell death, contributes to tumor development and metastasis. The aim of this study was to develop an anoikis-based stratified model, and a multivariable-based nomogram for guiding clinical therapy for LUAD. Through differentially expressed analysis, univariate Cox, LASSO Cox regression, and random forest algorithm analysis, we established a 4 anoikis-related genes-based stratified model, and a multivariable-based nomogram, which could accurately predict the prognosis of LUAD patients in the TCGA and GEO databases, respectively. The low and high-risk score LUAD patients stratified by the model showed different tumor mutation burden, tumor microenvironment, gemcitabine sensitivity and immune checkpoint expressions. Through immunohistochemical analysis of clinical LUAD samples, we found that the 4 anoikisrelated genes (PLK1, SLC2A1, ANGPTL4, CDKN3) were highly expressed in the tumor samples from clinical LUAD patients, and knockdown of these genes in LUAD cells by transfection with small interfering RNAs significantly inhibited LUAD cell proliferation and migration, and promoted anoikis. In conclusion, we developed an anoikisbased stratified model and a multivariable-based nomogram of LUAD, which could predict the survival of LUAD patients and guide clinical treatment.