Uterine sarcoma patients without distant metastases between 2004 and 2015 were identified from the Surveillance, Epidemiology, and End Results (SEER) database. The LASSO Cox regression was performed to identify essential prognostic predictors and a nomogram was built to predict the 1-, 3-, and 5-year OS. Receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA) were used to validate the nomogram. Finally, prognostic stratification was performed by decision tree analysis based on the total points of the nomogram.
2871 patients with uterine sarcoma were included. Preliminary analysis suggested that adjuvant radiotherapy failed to provide an OS benefit for the total population without our nomogram. The built nomogram showed good discrimination and calibration abilities to predict the OS in uterine sarcoma patients and the patients were stratified into three risk groups based on the nomogram. For patients in the high-risk group, adjuvant radiotherapy significantly improved the 5-year OS and median survival time by 26.4% and 17 months, respectively (P < 0.001); while radiotherapy failed to improve the survival outcomes of patients in the low- and intermediate-risk groups.
The nomogram-based prognostic stratification provides preliminary characterization of uterine sarcoma patients who may benefit from radiotherapy. The newly defined high-risk patients may gain significant OS benefit from adjuvant radiotherapy.
方法:从监测中确定了2004年至2015年间无远处转移的子宫肉瘤患者,流行病学,和结束结果(SEER)数据库。进行LASSOCox回归以确定必要的预后预测因子,并建立列线图来预测1-,3-,5年OS接收机工作特性(ROC),校正曲线,和决策曲线分析(DCA)用于验证列线图。最后,通过基于列线图总点的决策树分析进行预后分层.
结果:包括2871例子宫肉瘤患者。初步分析表明,没有我们的列线图,辅助放疗未能为总人口提供OS益处。构建的列线图显示出良好的辨别和校准能力,可以预测子宫肉瘤患者的OS,并根据列线图将患者分为三个风险组。对于高危人群,辅助放疗使5年OS和中位生存时间分别提高26.4%和17个月,(P<0.001);而放疗未能改善低危组和中危组患者的生存结局。
结论:基于列线图的预后分层提供了可能受益于放疗的子宫肉瘤患者的初步特征。新定义的高危患者可能从辅助放疗中获得显著的OS益处。