基于SEER数据库绘制列线图分析60岁以上软骨肉瘤患者预后的影响因素

Analysis of prognostic factors influencing the prognosis of elderly chondrosarcoma patients by drawing column-line diagrams based on SEER database

目的 软骨肉瘤是第二常见的原发性骨恶性肿瘤,在老年人中最常见并且愈后不佳。本研究目的是构建列线图,用于研究高龄软骨肉瘤患者的特征和预后因素。方法 从监测、流行病学和最终结果(SEER) 数据库中提取 2004 至 2015 年,诊断的年龄 > 60 岁软骨肉瘤患者的数据。SEER 数据库共纳入 705 例 60 岁以上软骨肉瘤患者。按照 7∶3 的比例将数据分为训练集和内部验证集,训练集数据使用 Kaplan-Meier 生存曲线分析筛选出有意义的变量,然后纳入到多因素 Cox 回归分析,筛选出影响愈后的独立危险因素。将筛选出的独立危险因素建立列线图。收集 2012 年至 2021 年,新疆医科大学附属医院诊断年龄 > 60 岁的 68 例软骨肉瘤患者的数据作为外部验证集。在训练集、内部验证集和外部验证集中通过 C-index,接收机工作特性(ROC) 曲线、校准曲线和决策曲线分析 (DCA) 对结果进行了评估。结果 60 岁以上软骨肉瘤患者影响预后的独立危险因素包括年龄 (HR:2.452,95% CI:1.826~3.293)、肿瘤大小 (HR:1.498,95% CI:1.162~1.931)、肿瘤数目 (HR:0.704,95% CI:0.542~0.916)、组织学类型 (HR:1.151,95% CI:1.065~1.243)、级别 (HR:1.447,95% CI:1.246~1.681)、N 分期 (HR:3.930,95% CI:1.690~9.141)、M 分期(HR:2.297,95% CI:1.519~3.473)、手术 (HR:0.345,95% CI:0.233~0.511)。训练集、内部验证集和外部验证集中 C-index 为 0.763、0.723、0.766,预测 3 年生存率 ROC 曲线下面积分别为 0.842、0.765 和 0.809。预测 5 年生存率 ROC 曲线下面积分别为 0.822、0.757 和 0.812。校准曲线和 DCA 的结果也证实建立的列线图可以准确预测高龄软骨肉瘤患者预后因素。结论 对于高龄软骨肉瘤患者,年龄越大,肿瘤越大,数目越多,未分化型肿瘤,高级别肿瘤,有淋巴结转移,有远处转移都会产生较差的愈后,但是手术对于高龄软骨肉瘤患者是保护因素。

Objective Chondrosarcoma is the second most common primary malignant tumour of the bone,and the most common one in the elderly with poor prognosis.The aim of this study is to construct a column chart to study the characteristics and prognostic factors of elderly patients with chondrosarcoma.Methods Data were extracted from the Surveillance,Epidemiology,and End Results(SEER)database for patients with chondrosarcoma aged>60 years diagnosed between 2004 and 2015.A total of 705 patients with chondrosarcoma aged>60 years were included in the SSER database.The data were divided into a training set and an internal validation set according to a ratio of 7:3.The training set data were screened for meaningful variables using Kaplan-Meier survival curve analysis,and then included in multifactorial cox regression analyses to screen for independent risk factors affecting healing.Column line graphs were created for the screened independent risk factors.In addition,data from 68 patients with chondrosarcoma aged>60 years diagnosed in Xinjiang Medical University Hospital from 2012 to 2021 were collected as the external validation set.The results were evaluated by C-index,receiver operating characteristic(ROC)curve,calibration curve and decision curve analysis(DCA)in the training set,internal validation set and external validation set. Results Independent risk factors affecting prognosis in chondrosarcoma patients over 60 years of age included age( HR: 2.452, 95% CI: 1.826 - 3.293 ), tumour size ( HR: 1.498, 95% CI: 1.162 - 1.931 ), tumour number ( HR: 0.704,95% CI: 0.542 - 0.916 ), histological type ( HR: 1.151, 95% CI: 1.065 - 1.243 ), grade ( HR: 1.447, 95% CI: 1.246 -1.681 ), N-stage ( HR: 3.930, 95% CI: 1.690 - 9.141 ), M-stage ( HR: 2.297, 95% CI: 1.519 - 3.473 ) and surgery ( HR:0.345, 95% CI: 0.233 - 0.511 ). The C-index of the training set, internal validation set and external validation set were0.763, 0.723, 0.766, and the area under the ROC curve for predicting 3-year survival was 0.842, 0.765, and 0.809,respectively. The area under the ROC curve predicting 5-year survival was 0.822, 0.757, and 0.812, respectively. Theresults of the calibration curves and DCA also confirmed that the establishment of the of the column-line graphs couldaccurately predict prognostic factors in elderly chondrosarcoma patients. Conclusions For patients with advancedchondrosarcoma, older age, larger and more tumours, undifferentiated tumours, high grade tumours, presence of lymphnode metastasis, and distant metastasis produce poorer healing, and surgery is a protective factor for patients withadvanced chondrosarcoma.