BACKGROUND: Osteosarcoma is the most common primary bone tumor. It has a high rate of early metastasis, and its treatment is one of the most challenging topics in the bone tumor field. Recent studies have shown that neutrophil extracellular traps play an important role in tumor metastasis and may provide new horizons for exploring metastasis in osteosarcoma. METHODS: OS data were downloaded from the TARGET database and Gene Expression Omnibus datasets. Univariate Cox regression was conducted to assess NETRGs. Patients were subsequently categorized into high- and low-risk groups on the basis of risk score values derived from multivariate Cox analysis, and prognostic models were established. The immune infiltration of relevant genes and drug sensitivity of key genes were also analyzed. RESULTS: A total of 15 NETs-related genes associated with osteosarcoma metastases were identified. Among them, a total of 4 genes were related to prognosis, namely, MAPK1, CFH, ATG7 and DDIT4, and a prognostic model based on these 4 genes was established. The prognosis was worse in the high-risk group, whose areas under the ROC curves (AUCs) were 0.857, 0.779, and 0.689 at 1, 3, and 5 years, respectively. The key genes were subsequently found to be associated with the infiltration of 20 types of immune cells. Finally, the small-molecule drug toxin c 10, an approximately 6700 mw protein, may target key genes. Finally, ATG7 was validated at the histological level by combining the results of the validation group dataset analysis. CONCLUSIONS: A risk model based on 4 NETRDEGs is a reliable prognostic predictor for OS patients, and CFH and ATG7 may serve as a new diagnostic and therapeutic target.