骨肉瘤是最常见的原发性骨肿瘤，预后不良。本研究的目的是建立一个竞争风险模型的示意图，来预测患有骨肉瘤的患者的肿瘤特异性存活率。患者数据来源于美国的国家癌症登记、流行病学和终点数据库。我们使用亚分布比例风险模型来分析影响骨肉瘤患者肿瘤特异性死亡率（CSM）的独立危险因素。基于这些危险因素，建立了一个竞争风险模型，用于预测骨肉瘤患者的1年、3年和5年的肿瘤特异性存活率（CSS）。我们使用一致性指数（C-index）、接收者操作特征曲线下的面积（AUC）以及校准曲线对示意图的可靠性和准确性进行了评估。总共包括了2900例骨肉瘤患者。分析结果显示年龄、原发肿瘤部位、M分期、手术、化疗和家庭年收入中位数是影响患者CSM的独立危险因素。竞争风险模型被构建出来，用于预测骨肉瘤患者的CSS。在训练集和验证集中，该模型的C-index分别为0.756（95％CI 0.725-0.787）和0.737（95％CI 0.717-0.757），而AUC值均大于0.7。校准曲线也显示了预测的存活率与实际存活率之间的高一致性，证实了该模型的准确性和可靠性。我们建立了一个竞争风险模型，用于预测骨肉瘤患者的1年、3年和5年CSS。该模型表现出良好的预测性能，并可以帮助临床医生和患者进行临床决策和制定随访策略。© 2023. 作者(们)专有许可给Springer-Verlag GmbH Germany，属于Springer Nature的一部分。

Osteosarcoma is the most common primary bone tumor with a poor prognosis. The aim of this study was to establish a competitive risk model nomogram to predict cancer-specific survival in patients with osteosarcoma.Patient data was obtained from the Surveillance, Epidemiology, and End Results database in the United States. A sub-distribution proportional hazards model was used to analyze independent risk factors affecting cancer-specific mortality (CSM) in osteosarcoma patients. Based on these risk factors, a competitive risk model was constructed to predict 1-year, 3-year, and 5-year cancer-specific survival (CSS) in osteosarcoma patients. The reliability and accuracy of the nomogram were evaluated using the concordance index (C-index), the area under the receiver operating characteristic curve (AUC), and calibration curves.A total of 2900 osteosarcoma patients were included. The analysis showed that age, primary tumor site, M stage, surgery, chemotherapy, and median household income were independent risk factors influencing CSM in patients. The competitive risk model was constructed to predict CSS in osteosarcoma patients. In the training and validation sets, the C-index of the model was 0.756 (95% CI 0.725-0.787) and 0.737 (95% CI 0.717-0.757), respectively, and the AUC was greater than 0.7 for both. The calibration curves also demonstrated a high consistency between the predicted survival rates and the actual survival rates, confirming the accuracy and reliability of the model.We established a competitive risk model to predict 1-year, 3-year, and 5-year CSS in osteosarcoma patients. The model demonstrated good predictive performance and can assist clinicians and patients in making clinical decisions and formulating follow-up strategies.© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.