Department of Orthopaedics, China-Japan Union Hospital of Jilin University, Changchun, China.
Wenzhou Medical University, Wenzhou, China.
Front Public Health. 2022 Jun 30;10:901680. doi: 10.3389/fpubh.2022.901680. eCollection 2022.
Chondrosarcoma is the most common primary bone sarcoma among elderly population. This study aims to explore independent prognostic factors and develop prediction model in elderly patients with CHS.
This study retrospectively analyzed the clinical data of elderly patients diagnosed as CHS between 2004 and 2018 from the Surveillance, Epidemiology, and End Results (SEER) database. We randomly divided enrolled patients into training and validation group, univariate and multivariate Cox regression analyses were used to determine independent prognostic factors. Based on the identified variables, the nomogram was developed and verified to predict the 12-, 24-, and 36-month overall survival (OS) of elderly patients with CHS. A k-fold cross-validation method (=10) was performed to validate the newly proposed model. The discrimination, calibration and clinical utility of the nomogram were assessed using the Harrells concordance index (C-index), receiver operating characteristic (ROC) curve and the area under the curve (AUC), calibration curve, decision curve analysis (DCA), the integrated discrimination improvement (IDI) and net reclassification index (NRI). Furthermore, a web-based survival calculator was developed based on the nomogram.
The study finally included 595 elderly patients with CHS and randomized them into the training group (419 cases) and validation group (176 cases) at a ratio of 7:3. Age, sex, grade, histology, M stage, surgery and tumor size were identified as independent prognostic factors of this population. The novel nomogram displayed excellent predictive performance, which can be accessible by https://nomoresearch.shinyapps.io/elderlywithCHS/, with a C-index of 0.800 for the training group and 0.789 for the validation group. The value AUC values at 12-, 24-, and 36-month of 0.866, 0.855, and 0.860 in the training group and of 0.839, 0.856, and 0.840 in the validation group, respectively. The calibration curves exhibited good concordance from the predicted survival probabilities to actual observation. The ROC curves, IDI, NRI, and DCA showed the nomogram was superior to the existing AJCC staging system.
This study developed a novel web-based nomogram for accurately predicting probabilities of OS in elderly patients with CHS, which will contribute to personalized survival assessment and clinical management for elderly patients with CHS.
软骨肉瘤是老年人中最常见的原发性骨肉瘤。本研究旨在探讨老年人软骨肉瘤（CHS）的独立预后因素，并建立预测模型。
本研究回顾性分析了 2004 年至 2018 年期间来自监测、流行病学和最终结果（SEER）数据库的诊断为 CHS 的老年患者的临床数据。我们将入组患者随机分为训练组和验证组，使用单因素和多因素 Cox 回归分析确定独立的预后因素。基于确定的变量，制定并验证了预测 CHS 老年患者 12、24 和 36 个月总生存（OS）的列线图。采用 k 折交叉验证方法（k=10）验证新模型。使用 Harrell 一致性指数（C 指数）、接受者操作特征（ROC）曲线和曲线下面积（AUC）、校准曲线、决策曲线分析（DCA）、综合判别改善（IDI）和重新分类指数（NRI）评估列线图的判别、校准和临床实用性。此外，还基于该列线图开发了一个在线生存计算器。
本研究最终纳入 595 例 CHS 老年患者，按 7:3 的比例随机分为训练组（419 例）和验证组（176 例）。年龄、性别、分级、组织学、M 分期、手术和肿瘤大小被确定为该人群的独立预后因素。新的列线图显示出优异的预测性能，可通过 https://nomoresearch.shinyapps.io/elderlywithCHS/ 访问，在训练组中的 C 指数为 0.800，在验证组中的 C 指数为 0.789。在训练组中，12、24 和 36 个月时的 AUC 值分别为 0.866、0.855 和 0.860，在验证组中，12、24 和 36 个月时的 AUC 值分别为 0.839、0.856 和 0.840。校准曲线显示出从预测生存概率到实际观察的良好一致性。ROC 曲线、IDI、NRI 和 DCA 表明该列线图优于现有的 AJCC 分期系统。
本研究开发了一种新的基于网络的列线图，可准确预测 CHS 老年患者的 OS 概率，有助于对 CHS 老年患者进行个性化生存评估和临床管理。
