摘要
目的胸中段食管癌手术或放化疗疗效相似,最佳治疗模式存在争议。本研究分析可能影响胸中段食管癌放化疗患者预后的因素,建立起可个体化预测患者远期生存的Nomogram模型。方法回顾性分析汕头市中心医院2009-01-21-2015-12-29接受放化疗的胸中段食管癌患者413例,用Kaplan-Meier法计算生存率,分析远期疗效,并采用Cox单因素和多因素分析方法分析影响患者预后的因素,进一步采用R软件建立Nomogram预测模型。结果全组中位随访时间为15.7个月,1、2、3、5年生存率分别为62.5%、36.3%、26.8%和17.7%。单因素分析显示,患者性别、肿瘤直径、长度、T分期、N分期和TNM分期、KPS评分、放疗剂量、辅助化疗以及近期疗效可能影响患者放化疗后远期生存,均P<0.05;进一步Cox多因素分析显示,性别、N分期、放疗剂量和近期疗效是影响患者远期生存的独立预测因素,均P<0.05。基于这些预后因素采用R软件建立了Nomogram模型,一致性指数C-index为0.851(95%CI:0.756~0.946),可以个体化预测患者远期生存概率。结论基于本中心常治患者人群临床多因素分析结果建立的Nomogram模型可以个体化预测患者远期生存情况,对指导患者个体化治疗选择具有一定临床意义。
OBJECTIVE Previous studies show that chemoradiotherapy is not inferior to operation in treatment of middle thoracic esophageal cancer,however,which is the optimal treatment remains unclear.This study was to investigate important factors affecting the prognosis and establish an individual prognostic Nomogram that could predict survival outcome after chemoradiotherapy in middle thoracic esophageal cancer.METHODS A total of 413 patients with middle thoracic esophageal cancer who received radiotherapy and chemoradiotherapy from January 21 st,2009 to December 29 th,2015 were retrospective analyzed.The survival rates were calculated by Kaplan-Meier method.Cox regression model was used for invariant analysis and multivariate analysis.A prognostic Nomogram model was then established according to the important variables that may affect prognosis by R.RESULTS The median follow up time was 15.7 months.The 1-,2-,3-and 5-year survival rates were 62.5%,36.3%,26.8% and 17.7%,respectively.In univariate analysis,gender,diameter,length,T stage,N stage,TNM stage,KPS score,radiotherapy dose,whether treated with adjuvant chemotherapy and short-term response were related to the prognosis of the patients（all P〈0.05）.The Cox regression model showed that gender,N stage,radiotherapy dose and short-term response were independent prognostic factors that affected survival outcome.According to theses four prognostic factors,a visual prognostic Nomogram with a high C-index of 0.851（95%CI：0.756-0.946）that could predict the survival outcome was established.CONCLUSION The Nomogram model construction based on the Cox regression of the data from the common patients in our center could predict the survival outcome and play an important role in selection of individual treatment for middle thoracic esophageal cancer.
作者
吴盛喜
罗何三
张晓平
许鸿鹞
黄河澄
林连兴
WU Sheng-xi;LUO He-san;ZHANG Xiao ping;XU Hong-yao;HUANG He-cheng;LIN Lian -xing(Department of Radiation Oncology , Shantou Central Hospital, Shantou 515031, P. R. China)
出处
《中华肿瘤防治杂志》
CAS
北大核心
2018年第6期417-421,427,共6页
Chinese Journal of Cancer Prevention and Treatment
基金
广东省科技计划(2016ZC0250)
