Fang Qian, Cai Guojing, Chen Guizeng, Xu Xiang, Zeng Haifeng, He Yulong, Cai Shirong, Wu Hui
Department of Gastrointestinal Surgery, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou 510080, China.
Heliyon. 2023 Jun 1;9(6):e16867. doi: 10.1016/j.heliyon.2023.e16867. eCollection 2023 Jun.
Considering the rarity and aggressive nature of retroperitoneal leiomyosarcoma (RLMS), several prognostic factors might contribute to the cancer-specific mortality of these patients. This study aimed to construct a competing risk-based nomogram to predict cancer-specific survival (CSS) for patients with RLMS. In total, 788 cases from the Surveillance, Epidemiology, and End Results (SEER) database (2000-2015) were included. Based on the Fine & Gray's method, independent predictors were screened to develop a nomogram for predicting 1-, 3-, and 5-year CSS. After multivariate analysis, CSS was found significantly associated with tumor characteristics (tumor grade, size, range), as well as surgery status. The nomogram showed solid prediction power and was well calibrated. Through decision curve analysis (DCA), a favorable clinical utility of the nomogram was demonstrated. Additionally, a risk stratification system was developed and distinctive survival between risk groups was observed. In summary, this nomogram showed a better performance than the AJCC 8th staging system and can assist in the clinical management of RLMS.
考虑到腹膜后平滑肌肉瘤（RLMS）的罕见性和侵袭性，多种预后因素可能导致这些患者的癌症特异性死亡。本研究旨在构建一种基于竞争风险的列线图，以预测RLMS患者的癌症特异性生存（CSS）。总共纳入了监测、流行病学和最终结果（SEER）数据库（2000 - 2015年）中的788例病例。基于Fine & Gray方法，筛选出独立预测因素以开发用于预测1年、3年和5年CSS的列线图。多因素分析后发现，CSS与肿瘤特征（肿瘤分级、大小、范围）以及手术状态显著相关。该列线图显示出强大的预测能力且校准良好。通过决策曲线分析（DCA），证明了列线图具有良好的临床实用性。此外，还开发了一种风险分层系统，并观察到风险组之间存在明显的生存差异。总之，该列线图表现优于美国癌症联合委员会（AJCC）第8版分期系统，可协助RLMS的临床管理。
