摘要
目的 探讨改良格拉斯哥预后评分(mGPS)与胆囊癌患者切除术后预后的相关性.方法 回顾性分析2017年1月至2022年12月于宁波市医疗中心李惠利医院肝胆胰外科行手术切除治疗的137例胆囊癌患者的临床资料,其中男性58例,女性79例,年龄(67.7±10.1)岁.收集患者的肿瘤最大径、血管侵犯情况、神经侵犯情况、肿瘤分化程度、肿瘤TNM分期等临床资料,计算术前mGPS,并基于mGPS将患者分为3组:mGPS 0分组(n=78)、mGPS 1分组(n=39)和mGPS 2分组(n=20).以门诊复查和电话沟通的方式随访患者的生存情况.Kaplan-Meier法进行生存分析,生存率比较采用log-rank检验.Cox回归分析影响术后无复发生存期的因素.构建并验证胆囊癌患者术后无复发生存的列线图模型.结果 mGPS 0分组、mGPS 1分组和mGPS 2分组胆囊癌患者的肿瘤最大径、肿瘤分化程度、TNM分期、术前肿瘤糖类抗原19-9以及R0切除比例间差异均具有统计学意义(均P＜0.05).mGPS 0分组、mGPS 1分组和mGPS 2分组胆囊癌患者术后累积生存率(x2=28.13)和无复发生存率(x2=25.39)间差异均具有统计学意义(均P＜0.001).多因素Cox回归分析显示,肿瘤低分化(HR=2.433,95％CI:1.396～4.242,P=0.002)、血管侵犯(HR=2.809,95％CI:1.598～4.941,P＜0.001)、神经侵犯(HR=1.980,95％CI:1.188～3.300,P=0.009)、TNM 分期 Ⅲ～Ⅳ 期(HR=2.689,95％CI:1.069～6.762,P=0.036)、mGPS 2 分(HR=2.496,95％CI:1.372～4.541,P=0.003)的胆囊癌患者,术后无复发生存期短的风险高.基于上述5个危险因素建立的预测胆囊癌患者无复发生存的列线图模型,C指数为0.810(95％CI:0.769～0.851),决策曲线分析显示该列线图模型具有明显的正向净收益,校准曲线显示该列线图模型的预测结果与实际结果具有良好的符合度.结论 术前mGPS与胆囊癌患者切除术后预后相关,mGPS高是预后不良的风险因素,基于mGPS建立的列线图模型可用于预测胆囊癌患者切除术后的无复发生存情况.
Abstract
Objective To analyze the correlation between the modified Glasgow prognostic score(mGPS)and the prognosis of patients undergoing surgery for gallbladder cancer.Methods Clinical data of 137 patients undergoing surgery for gallbladder cancer in the Department of Hepatobiliary and Pancreatic Sur-gery at Ningbo Medical Center Lihuili Hospital from January 2017 to December 2022 were retrospectively analyzed,including 58 males and 79 females,aged(67.7±10.1)years old.According to mGPS,patients were divided into the mGPS 0 group(n=78),mGPS 1 group(n=39)and mGPS 2 group(n=20).Clini-copathological data,such as maximum tumor diameter,vascular invasion,perineural invasion,tumor differen-tiation and TNM stage,were compared between the groups.Survivals of patients were followed-up via outpa-tient follow-ups and telephone reviews,analyzed using the Kaplan-Meier method,and compared between the groups using the log-rank test.Univariate and multivariate Cox regression analysis were performed to identify prognostic factors for recurrence-free survival.Based on the results of multivariate analysis,a nomogram model of recurrence-free survival of gallbladder cancer patients was established and validated respectively.Results The maximum tumor diameter,tumor differentiation,TNM stage,preoperative CA19-9 level and R0 resection rate differed statistically among the mGPS 0,mGPS 1 and mGPS 2 groups(all P＜0.05).Postop-erative cumulative survival rate(x2=28.13)and recurrence-free survival rate(x2=25.39)of gallbladder cancer patients also differed among the mGPS groups(all P＜0.001).Multivariate Cox regression analysis showed that the poor differentiation of tumor(HR=2.433,95％CI:1.396-4.242,P=0.002),vascular invasion(HR=2.809,95％CI:1.598-4.941,P＜0.001),perineural invasion(HR=1.980,95％CI:1.188-3.300,P=0.009),TNM stage Ⅲ-Ⅳ(HR=2.689,95％CI:1.069-6.762,P=0.036)and mGPS 2(HR=2.496,95％CI:1.372-4.541,P=0.003)were independent risk factors for poor recur-rence-free survival in gallbladder cancer.Based on the above risk factors,a nomogram of prediction model for recurrence-free survival in patients with gallbladder cancer was established,with a C-index value of 0.810(95％CI:0.769-0.851).The decision curve analysis findings demonstrated that the nomogram model had a significant positive net benefit,and the calibration curve demonstrated that the predicted results of the nomogram model correlates well with the actual results.Conclusions The preoperative mGPS is asso-ciated with the overall prognosis of patients undergoing surgery for gallbladder cancer,and a high mGPS is a risk factor for poor prognosis.The mGPS-based nomogram of prediction model showed a good predictive value of the recurrence-free survival of patients undergoing surgery for gallbladder cancer.
NETL
NSTL
万方数据