中期18F-FDG PET/CT显像不同评价方法对弥漫大B细胞淋巴瘤患者预后判断价值

摘要

Objective:To explore the prognostic value of interim 18F-FDG PET/CT (i-PET/CT) scan for the patients with newly diagnosed diffuse large B-cell lymphoma (DLBCL).Methods:A total of 70 cases of initially diagnosed of DLBCL by 158 18F-FDG PET/CT scans in our hospital were retrospectively analyzed.The 5-point scale,the Lugano classification and maximum standardized uptake value induction (△ SUVmax) criteria were used respectively to assess i-PET/CT scans.Receiver-operating characteristics (ROC) analysis was used to determine an optimal cutoff for △ SUVmax.Progression-free survival (PFS) and overall survival (OS) times were estimated as prognostic indicators using the Kaplan-Meier method and Cox regression.Results:Optimal cutoff to predict progression or death was 62％ for △SUVmax.The positive predictive value (PPV) for 2-year PFS and OS of i-PET/CT diagnosed by 5-point scale was low,and could be improved by using the Lugano classification with decreased sensitivity or △ SUVmax criteria.Kaplan-Meier survival curve analysis showed that the Lugano classification and △ SUVmax were good predictors for PFS and OS,respectively,while the 5-point scale could only predict OS.Cox regression univariate analysis showed that the International Prognostic Index (IPI) score was better to predict PFS than 5-point scale,but worse than the three assessments in predicting OS.COX regression multivariate analysis showed that △ SUVmax ＜ 62％ was an independent risk factor of prognosis,while the Lugano classification was only the OS independent prognostic predictor.Conclusion:Assessing i-PET/CT by 5-point scale is a limited value for predicting PFS and OS in DLBCL patients.The Lugano classification is recommended to discriminate the patients with poorer outcomes.The △ SUVmax criteria for i-PET/CT of DLBCL patients is an independent prognostic predictor for PFS and OS,better than the IPI score.%目的:探究中期18F-FDG PET/CT(i-PET/CT)显像对弥漫大B细胞淋巴瘤(DLBCL)患者预后判断价值.方法:回顾分析我院70例初治DLBCL患者的158次18 F-FDG PET/CT结果,分别采用5分法、Lugano分类和最大标准摄取值减少率(△SUVmax)法评价i-PET/CT.利用受试者操作特性(ROC)曲线确定△SUVmax最佳界值.应用无进展生存(PFS)时间及总生存(OS)时间作为随访指标.运用Kaplan-Meier方法和Cox回归模型进行生存分析.结果:经ROC曲线计算△SUVmax最佳界值点为62％.5分法诊断2年PFS和OS阳性预测值(PPV)低,Lugano分类和△SUVmax法均可提高PPV,但采用Lugano分类敏感度下降.Kaplan-Meier生存曲线分析显示,Lugano分类和△SUVmax法可预测PFS和OS时间,而5分法仅可预测OS时间.COX回归单因素分析示,国际预后指数(IPI)评分对PFS预测仅优于5分法,对OS预测则逊于三者.COX回归多因素分析显示,△SUVmax法是判断预后的独立危险因素,采用Lugano分类仅对OS有独立预后价值.结论:5分法评价i-PET/CT对于DLBCL患者判断预后价值有限,建议结合Lugano分类标准.采用△SUVmax法评价i-PET/CT对DLBCL患者具有独立预后价值,优于IPI评分.