淋巴结转移率对胃癌患者的预后价值

Prognostic value of metastatic lymph node ratio for patients with gastric cancer

目的探讨淋巴结转移率(缩写为"rN")在判断胃癌患者预后中的价值。方法回顾性分析2014年1月至2019年7月期间在兰州大学第二医院普外一科收治的255例行胃癌根治术的胃癌患者的临床病理资料。采用Kaplan-Meier生存曲线和Cox比例回归分析进行生存分析,并使用受试者操作特征(ROC)曲线来比较rN分期和N分期对胃癌患者预后判断的准确性,同时采用Spearman相关性分析方法对rN和淋巴结转移数目与淋巴结获取数目的相关性进行分析。结果单因素生存分析结果显示,脉管浸润、神经浸润、肿瘤直径、分化程度、T分期、N分期、rN与胃癌患者的累积总生存率有关(P<0.05);多因素分析结果显示,基于rN分期的预后模型较N分期的预后模型有着较高的HR值(1.756比1.610)及较小的–2倍似然函数值的自然对数值(648.548比649.469)。相关性分析结果显示,在淋巴结检出数目<15枚和≥15枚时,rN与淋巴结获取数目均无关(rs=0.275,P=0.058;rs=0.075,P=0.285);此外,在淋巴结检出数目<15枚和≥15枚时,rN分期均可准确地对胃癌患者的预后进行分层(χ2=11.24,P=0.009;χ2=30.25,P<0.001)。ROC曲线分析结果显示,淋巴结检出数目<15枚时,相较于N分期,r N分期的ROC曲线下面积更大[0.863,95%CI(0.752,0.974)和0.813, 95%CI(0.687,0.938)]。结论当淋巴结检出数目<15枚时,rN分期较N分期在预测胃癌患者预后时可能更为准确。

Objective To investigate the value of metastatic lymph node ratio(abbreviation: rN) in the prognosis of patients with gastric cancer. Methods The clinical data of 255 patients with gastric cancer who underwent the radical gastrectomy in the First Department of General Surgery of Lanzhou University Second Hospital from January 2014 to July 2019 were retrospectively reviewed. The survival analysis was performed using Kaplan-Meier survival curves and Cox regression model. The receiver operating characteristic(ROC) curve was used to compare the accuracy of rN stage and N stage for prognosis judgment of patients with gastric cancer, and the Spearman correlation analysis method was used to analyze the correlation between rN or number of metastatic lymph node and total number of lymph nodes detected. Results The univariate survival analysis showed that the vascular invasion, nerve invasion, histologic grade, tumor diameter, T stage, N stage, and rN stage were significantly associated with the prognosis of patients with gastric cancer(P<0.05);The multivariate survival analysis demonstrated that the prognosis model based on the rN stage had a higher HR value(1.756 versus 1.610) and a smaller –2 log likelihood value(648.548 versus 649.469) than the N stage. Correlation analysis results showed that rN was independent of the number of lymph nodes detected when the number was <15 and ≥15(rs=0.275, P=0.058;rs=0.075, P=0.285). In addition, when the number of lymph nodes detected was <15 and ≥15, the rN staging could accurately stratify the prognosis of gastric cancer patients(χ~2=11.24, P=0.009;χ~2=30.25, P<0.001). ROC curve analysis results showed that when the number of lymph nodes detected was less than 15, rN stage had a higher area under ROC curve value [0.863, 95%CI(0.752, 0.974) and 0.813, 95%CI(0.687, 0.938)] as compared with N stage. Conclusion rN stage might be more accurate than N stage in predicting prognosis of patients with gastric cancer when number of lymph nodes harvested is less than 15.