Department of Orthopaedic Surgery, Division of Orthopaedic Oncology, Harvard Medical School, Massachusetts General Hospital, Boston, Massachusetts, USA.
Department of Orthopaedic Surgery, Beth Israel Lahey Health, Burlington, Massachusetts, USA.
J Surg Oncol. 2024 Mar;129(3):531-536. doi: 10.1002/jso.27514. Epub 2023 Nov 16.
Leiomyosarcoma (LMS) is associated with one of the poorest overall survivals among soft tissue sarcomas. We sought to develop and externally validate a model for 5-year survival prediction in patients with appendicular or truncal LMS using machine learning algorithms.
The Surveillance, Epidemiology, and End Results (SEER) database was used for development and internal validation of the models; external validation was assessed using our institutional database. Five machine learning algorithms were developed and then tested on our institutional database. Area under the receiver operating characteristic curve (AUC) and Brier score were used to assess model performance.
A total of 2209 patients from the SEER database and 81 patients from our tertiary institution were included. All models had excellent calibration with AUC 0.84-0.85 and Brier score 0.15-0.16. After assessing the performance indicators according to the TRIPOD model, we found that the Elastic-Net Penalized Logistic Regression outperformed other models. The AUCs of the institutional data were 0.83 (imputed) and 0.85 (complete-case analysis) with a Brier score of 0.16.
Our study successfully developed five machine learning algorithms to assess 5-year survival in patients with LMS. The Elastic-Net Penalized Logistic Regression retained performance upon external validation with an AUC of 0.85 and Brier score of 0.15.
平滑肌肉瘤（LMS）是软组织肉瘤中总体生存率最差的肿瘤之一。我们试图利用机器学习算法为肢体或躯干 LMS 患者建立 5 年生存率预测模型，并进行外部验证。
本研究利用监测、流行病学和最终结果（SEER）数据库建立和内部验证模型；利用本机构数据库进行外部验证。开发了 5 种机器学习算法，并在本机构数据库中进行了测试。使用接收者操作特征曲线下面积（AUC）和 Brier 评分来评估模型性能。
SEER 数据库共纳入 2209 例患者，本机构数据库纳入 81 例患者。所有模型的 AUC 为 0.84-0.85，Brier 评分 0.15-0.16，校准度均良好。根据 TRIPOD 模型评估性能指标后，我们发现弹性网络惩罚逻辑回归模型优于其他模型。本机构数据的 AUC 分别为 0.83（插补）和 0.85（完全案例分析），Brier 评分分别为 0.16。
本研究成功开发了 5 种机器学习算法来评估 LMS 患者的 5 年生存率。弹性网络惩罚逻辑回归模型在外部验证中具有良好的性能，AUC 为 0.85，Brier 评分为 0.15。
