作者：张晓杰，赵东兵

单位：国家癌症中心 国家肿瘤临床医学研究中心 中国医学科学院 北京协和医学院 肿瘤医院 胰胃外科，北京100021

Authors: Zhang Xiaojie, Zhao Dongbing

Unit: Department of Pancreatic and Gastric Surgery, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, China

摘要：

目的 通过列线图的方式构建初诊为胃癌肝转移患者预后模型并进行风险分层。方法 通过对美国国立癌症研究所数据库(Surveillance,Epidemiology and End Results database, SEER数据库)2010—2015年初诊为胃癌肝转移的患者预后危险因素进行分析,构建列线图模型并依据列线图评分进行风险分层。结果 共1721例初诊为胃癌肝转移的患者被纳入研究,其1年、2年、3年总生存率分别为26.3%、12.1%和8.2%。通过单因素及多因素分析发现性别、年龄、分化程度、T分期、骨转移、肺转移、原发部位手术以及化疗是影响预后的独立因素(P<0.05)。列线图模型内部验证一致性指数分别为0.728(95%CI:0.715～0.741),校准曲线显示模型预测能力良好。通过列线图评分将初诊为胃癌肝转移患者分为三组:低危组(950例),中危组(381例)及高危组(390例)。三组预后具有统计学差异(P<0.001)。结论 列线图模型及风险分层对初诊为胃癌肝转移患者预后具有良好的预测能力。

关键词：  慢性便秘; 菌群丰度; 肠道菌群; 16S rDNA

Abstract：

Objective Constructing the prognosis model and risk stratification through nomogram for gastric cancer with liver metastasis. Methods The prognostic factors of newly diagnosed gastric cancer with liver metastasis from 2010-2014 years in the Surveillance, Epidemiology and End Results database (SEER database)were analyzed. The nomogram model was used to predict its prognosis. Finally, risk stratification was conducted based on the nomogram score. Results A total of 1721 patients with newly diagnosed GCLM were included in the study. The 1-year, 2-year, and 3-year overall survival rates were 26.3%, 12.1%, and 8.2%, respectively. Univariate and multivariate analysis revealed that gender, age, degree of differentiation, T stage, bone metastasis, lung metastasis, primary site surgery, and chemotherapy were independent factors affecting prognosis (P<0.05). The internal verification demonstrated that C-index of the nomogram model was 0.728 (95% confidence interval: 0.715~0.741), and the calibration curve showed that the model had good predictive ability. The patients with newly diagnosed GCLM were divided into three groups by the nomogram score: low-risk group (950 people), middle-risk group (381 people) and high-risk group (390 people). The prognosis of the three groups was statistically different (P<0.001). Conclusion The nomogram model and risk stratification have a good predictive ability for the prognosis of patients with newly diagnosed GCLM, but we still need more data to further verify.

Key Words: Gastric cancer; Liver metastasis; Nomogram; Prognosis