Thoracic Surgery Division, The Ohio State University Wexner Medical Center, Columbus, Ohio, USA.
Department of Biomedical Informatics, Center for Biostatistics, College of Medicine, The Ohio State University, Columbus, Ohio, USA.
J Surg Oncol. 2021 Mar;123(4):881-890. doi: 10.1002/jso.26349. Epub 2020 Dec 17.
Locally advanced esophageal carcinoma is treated with neoadjuvant chemoradiation and esophagectomy. Patients may still experience recurrence and death despite undergoing potentially curative trimodality therapy. This study describes predictive nomograms for recurrence-free (RFS) and overall survival (OS) after the completion of trimodality therapy.
A total of 215 patients with esophageal adenocarcinoma underwent trimodality therapy from September 2010 to April 2018. Multivariate Cox proportional hazards regression models were used to create nomograms for OS and RFS. Kaplan-Meier survival curves were calculated for OS and RFS comparing high-risk and low-risk cohorts.
On multivariate analysis, clinical N-stage, tumor differentiation, tumor regression grade, anastomotic leak, body mass index, age, and number of lymph nodes removed were predictive variables for overall survival. Clinical N-stage, tumor differentiation, tumor regression grade, anastomotic leak, age, and positive lymph nodes were significant predictors of RFS in a multivariate model. The nomogram for OS had good predictive ability (Harrell's Concordance index [C-index]: 0.71 [95% confidence interval {CI}: 0.66-0.76]). The nomogram for RFS also performed well (C-index: 0.70 [95% CI: 0.65-0.74]).
Our nomograms can accurately predict OS and RFS after trimodality therapy and may provide guidance regarding adjuvant therapy and surveillance.
局部晚期食管癌采用新辅助放化疗和食管切除术治疗。尽管接受了潜在的根治性三联疗法，患者仍可能复发和死亡。本研究描述了完成三联疗法后无复发生存（RFS）和总生存（OS）的预测列线图。
2010 年 9 月至 2018 年 4 月，共有 215 例食管腺癌患者接受了三联疗法。使用多变量 Cox 比例风险回归模型为 OS 和 RFS 创建列线图。计算 OS 和 RFS 的 Kaplan-Meier 生存曲线，比较高风险和低风险队列。
多变量分析显示，临床 N 分期、肿瘤分化、肿瘤退缩分级、吻合口漏、体重指数、年龄和淋巴结清扫数目是总生存的预测变量。临床 N 分期、肿瘤分化、肿瘤退缩分级、吻合口漏、年龄和阳性淋巴结是多变量模型中 RFS 的显著预测因子。OS 的列线图具有良好的预测能力（Harrell 一致性指数[C 指数]：0.71[95%置信区间{CI}：0.66-0.76]）。RFS 的列线图也表现良好（C 指数：0.70[95%CI：0.65-0.74]）。
我们的列线图可以准确预测三联疗法后 OS 和 RFS，可能为辅助治疗和监测提供指导。
