Bone and Soft Tissue Tumours Research Centre of Yunnan Province, Department of Orthopaedics, The Third Affiliated Hospital of Kunming Medical University (Yunnan Cancer Hospital), Kunming, 650118, Yunnan, 119 Kunzhou Road, China.
Independent Investigator, No. 22 Xi guan North Road, Kaifeng, 475000, Henan, China.
Sci Rep. 2021 Sep 6;11(1):17728. doi: 10.1038/s41598-021-97090-0.
This study aimed to construct a widely accepted prognostic nomogram in Chinese high-grade osteosarcoma (HOS) patients aged ≤ 30 years to provide insight into predicting 5-year overall survival (OS). Data from 503 consecutive HOS patients at our centre between 12/2012 and 05/2019 were retrospectively collected. Eighty-four clinical features and routine laboratory haematological and biochemical testing indicators of each patient at the time of diagnosis were collected. A prognostic nomogram model for predicting OS was constructed based on the Cox proportional hazards model. The performance was assessed by the concordance index (C-index), receiver operating characteristic curve and calibration curve. The utility was evaluated by decision curve analysis. The 5-year OS was 52.1% and 2.6% for the nonmetastatic and metastatic patients, respectively. The nomogram included nine important variables based on a multivariate analysis: tumour stage, surgical type, metastasis, preoperative neoadjuvant chemotherapy cycle, postoperative metastasis time, mean corpuscular volume, tumour-specific growth factor, gamma-glutamyl transferase and creatinine. The calibration curve showed that the nomogram was able to predict 5-year OS accurately. The C-index of the nomogram for OS prediction was 0.795 (range, 0.703-0.887). Moreover, the decision curve analysis curve also demonstrated the clinical benefit of this model. The nomogram provides an individualized risk estimate of the 5-year OS in patients with HOS aged ≤ 30 years in a Chinese population-based cohort.
本研究旨在构建一个被广泛认可的中国 30 岁以下高级别骨肉瘤（HOS）患者的预后列线图，以深入了解预测 5 年总生存率（OS）的方法。回顾性收集了 2012 年 12 月至 2019 年 5 月期间我中心的 503 例连续 HOS 患者的数据。收集了每位患者诊断时的 84 项临床特征和常规实验室血液学及生化检测指标。基于 Cox 比例风险模型构建了预测 OS 的预后列线图模型。通过一致性指数（C-index）、受试者工作特征曲线和校准曲线评估性能。通过决策曲线分析评估效用。无转移患者和转移患者的 5 年 OS 分别为 52.1%和 2.6%。多因素分析后，该列线图纳入了 9 个重要变量：肿瘤分期、手术类型、转移、术前新辅助化疗周期、术后转移时间、平均红细胞体积、肿瘤特异性生长因子、γ-谷氨酰转移酶和肌酐。校准曲线表明，该列线图能够准确预测 5 年 OS。该列线图预测 OS 的 C-index 为 0.795（范围 0.703-0.887）。此外，决策曲线分析曲线也表明了该模型的临床获益。该列线图为中国人群中 30 岁以下 HOS 患者提供了个体化的 5 年 OS 风险估计。
