肺癌脑转移瘤术后预后因素分析

Analysis of prognostic factors after surgical operation of lung cancer brain metastases

目的对肺癌脑转移行手术治疗的患者进行预后相关因素分析。方法回顾性收集2016年1月至2018年11月在中国医学科学院北京协和医学院肿瘤医院神经外科行手术治疗的肺癌脑转移患者的临床资料,获得完整的资料共83例,单因素分析采用log-rank法,多因素分析采用Cox风险比例模型,应用R软件将影响因素绘制成列线图并进行验证。结果83例患者的中位总生存期(OS)为27.0个月。单纯手术患者中位OS为15.7个月,手术联合放疗、化疗或靶向治疗患者的中位OS为27.7个月,差异有统计学意义(χ^2=8.735,P=0.003)。单因素分析结果显示,性别(χ^2=4.652,P=0.031)、有无吸烟史(χ^2=8.239,P=0.004)、术后是否靶向治疗(χ^2=13.697,P<0.001)、术后有无辅助治疗(χ^2=8.735,P=0.003)、转移灶病理(χ^2=11.799,P=0.001)、肺肿瘤相关分子分级预后评估指数(Lung-molGPA)评分(χ^2=11.333,P=0.004)可影响患者OS。多因素分析结果显示,吸烟史(HR=0.311,95%CI为0.107~0.901,P=0.031)、术后靶向治疗(HR=3.563,95%CI为1.286~9.868,P=0.015)、转移灶病理(HR=0.364,95%CI为0.137~0.965,P=0.042)、Lung-molGPA评分(HR=0.595,95%CI为0.374~0.946,P=0.028)是影响肺癌脑转移手术患者OS的独立预后因素。为了进一步个体化评估患者预后,采用以上4个独立的预后因素绘制了列线图,该模型准确度较高,能较好地评估患者的预后生存情况。结论有手术指征的肺癌脑转移患者能从手术中受益,术后积极辅助治疗可进一步延长患者OS。采用吸烟、靶向治疗、转移灶病理、Lung-molGPA评分4个因素构建的列线图可以很好地个体化评估患者预后并指导临床治疗。

Objective To analyze the prognosis-related factors of patients with surgical treatment of lung cancer brain metastases.Methods From January 2016 to November 2018,the clinical data of the patients with lung cancer brain metastases received surgical treatment in Department of Neurosurgery,Cancer Hospital of Chinese Academy of Medical Sciences and Peking Union Medical College were retrospectively collected,and completed data of 83 patients were obtained.The single factor analysis was based on the log-rank method,and the multi-factor analysis was used by the Cox proportional hazard model,and the R software was used to map the influencing factors into the nomogram and verify them.Results The median overall survival(OS)of 83 patients was 27.0 months.The median OS of patients with surgery alone was 15.7 months,the median OS of patients with surgery combined with radiotherapy,chemotherapy or targeted therapy was 27.7 months,and the difference was statistically significant(χ^2=8.735,P=0.003).The results of single factor analysis showed that gender(χ^2=4.652,P=0.031),smoking history(χ^2=8.239,P=0.004),postoperative targeted treatment(χ^2=13.697,P<0.001),postoperative adjuvant therapy(χ^2=8.735,P=0.003),pathology of metastatic tumor(χ^2=11.799,P=0.001),and lung cancer molecular graded prognostic assessment(Lung-molGPA)sore(χ^2=11.333,P=0.004)affected patients′OS.The results of multivariate analysis showed that smoking history(HR=0.311,95%CI:0.107-0.901,P=0.031),postoperative targeted therapy(HR=3.563,95%CI:1.286-9.868,P=0.015),pathology of metastatic tumor(HR=0.364,95%CI:0.137-0.965,P=0.042),Lung-molGPA score(HR=0.595,95%CI:0.374-0.946,P=0.028)were independent prognostic factors for OS of patients with lung cancer brain metastases.In order to further evaluate the prognosis of patients,nomogram was drawn using these four independent prognostic factors.The model had high accuracy and could better evaluate the prognosis of patients.Conclusion Lung cancer brain metastases patients with operative indication can benefit from surgery,and active adjuvant therapy after operation can further prolong the OS of patients.The nomogram constructed by smoking,targeted therapy,pathology of metastatic tumor and Lung-molGPA score can be used to evaluate individual patient outcomes and guide clinical treatment.