基于MRI增强的体素减影技术对复发胶质母细胞瘤患者预后评估的价值

Value of MRI enhanced voxel subtraction technique in prognostic assessment for recurrent glioblastoma

目的探讨复发性胶质母细胞瘤(recurrent glioblastoma,rGBM)二线治疗前的肿瘤增强体积(V)、总体积(Vn)以及相对非增强体积比(relative non-enhancing tumor ratio,rNTR)对总生存时间(overall survival,OS)以及复发后生存期(post-recurrence survival,PRS)的影响。方法回顾性分析2012年7月至2022年12月确诊和治疗的85例rGBM患者。T1加权成像(T1-weighted imaging,T1WI)减影模式下定量测量V、Vn(含肿瘤强化、肿瘤坏死和囊变体积)以及rNTR(FLAIR高信号体积/增强体积)。以Kaplan-Meier生存曲线和log-rank检验分析各指标的高、低水平组间预后差异;采用Cox比例风险回归模型进行单因素、多因素预后分析。结果患者中位PRS为238 d(95%CI:197.472~278.528),中位OS为449 d(95%CI:348.755~549.245)。Kaplan-Meier曲线和log-rank检验显示V、Vn和rNTR高、低水平组间PRS和OS差异有统计学意义(P均<0.05)。单因素分析中,V、Vn及rNTR均是PRS、OS的影响因素(P<0.05)。多因素分析中,V、Vn是PRS的独立危险因素(HR=1.019,P<0.001),Vn是OS的独立危险因素(HR=1.008,P=0.042),Vn-PRS模型在所有模型中-2倍对数似然值(-2 log likelihood,-2LL)最小。结论体素减影技术能准确地评估rGBM患者复发肿瘤大小,V、Vn及rNTR可用于预测预后,其中Vn对于PRS的预测效能最佳,能为临床对病变范围和患者生存时间的评估提供可靠的参考依据。

Objective To investigate the values of enhancing tumor volume(V),total volume(Vn)and relative non-enhancing tumor ratio(rNTR)before second-line treatment on overall survival(OS)and post-recurrence survival(PRS)in recurrent glioblastoma(rGBM).Methods Clinical data of 85 patients with confirmed rGBM admitted in our medical center from July 2012 to December 2022 were collected and retrospectively analyzed.Quantitative measurements of V,Vn(including tumor enhancement,necrotic and cystic volume)and rNTR(FLAIR high signal volume/enhancing tumor volume)were performed with T1-weighted imaging(T1WI)subtraction mode.Kaplan-Meier survival curve and log-rank test were used to analyze the differences of prognosis between high and low level groups of above indicators.Cox proportional risk regression model was employed for univariate and multivariate prognostic analyses.Results The median PRS was 238 d(95%CI:197.472~278.528)and the median OS was 449 d(95%CI:348.755~549.245)in these patients.Kaplan-Meier curve and log-rank test showed significant differences in PRS and OS between high and low levels groups of V,Vn and rNTR(P<0.05).Univariate analysis showed that V,Vn and rNTR were all influencing factors for PRS and OS(P<0.05).Multivariate analysis indicated that V and Vn were independent risk factors for PRS(HR=1.019,P<0.001),and Vn was an independent risk factor for OS(HR=1.008,P=0.042).The Vn-PRS model had the lowest-2 log likelihood(-2LL)among all the models.Conclusion Voxel subtraction technique can accurately assess the size of recurrent tumors in rGBM patients.V,Vn and rNTR can be used for prognosis,and Vn is the most valuable predictor for PRS,which can provide a reliable reference for clinical evaluation of the extent of lesions and survival time of patients.