Predictive Value of Sequential 18F-FDG PET/CT Scan in Concurrent Chemoradiotherapy of Non-Surgical Esophageal Carcinoma
李夷民,林勤,王丽琛,孙龙,戴明明,罗作明,郑华,赵龙,吴华
DOI: https://doi.org/10.3760/cma.j.issn.2095-2848.2013.06.002
2013-01-01
Abstract:Objective To investigate the metabolic changes during the concurrent chemoradiotherapy (CCRT) of non-surgical esophageal squamous cell cancer (ESCC) and to explore the predictive value of sequential 18F-FDG PET/CT scan in CCRT.Methods From May 2009 to October 2011,28 patients with pathologically confirmed ESCC were prospectively enrolled into this study.All patients received definitive treatment with CCRT.18F-FDG PET/CT scans were performed at the following 4 time points:before therapy,at the time when radiation dose reached 50 Gy and 60 Gy,and one month after treatment completion.SUVmax of the 4 time points were recorded as SUVmax1--4.The volume of the area with SUV greater than 40％SUVmax in primary tumor and adjacent lymph node was summed and named as MTV.The starting point of all survival data was from primary disease documentation.All patients had regular follow-up (median time:18.5 months) to record the disease status,including progress free survival (PFS),recurrence free survival (RFS) and overall survival (OS).Patients were divided into different groups according to the following status:progress free or progress,no recurrence or recurrence and death or living.The Mann-Whitney u test was used to compare SUVmax and the rate of △SUVmax in different groups.The AUC of ROC curve was calculated to find the optimal threshold of MTV.Kaplan-Meier and Cox regression were utilized to analyze the differentiation of survival.Results Among the 28 patients,19 showed disease progression,17 showed recurrence and 15 died during follow-up.In the groups of recurrence or no recurrence,there were statistical differences for SUVmax2 and SUVmax3,(SUVmax2:7.4±3.3 vs 4.8±2.5 ; SUVmax3:5.5±2.1 vs 3.8±2.1 ; u =46 and 47,both P＜0.05).As for SUVmax4,statistical differences could be found at every survival group (progress vs progress free:5.3±3.9 vs 2.4± 1.7 ; recurrence vs no recurrence:5.6±4.0 vs 2.4± 1.5 ; death vs living:5.8±4.2 vs 2.6± 1.5; u =31,26 and 29,all P＜0.05).The analysis of the derived parameters of SUVmax showed that △R14 ((SUVmax 1-SUVmax4)/SUVmax 1) had a statistically significant difference between any comparable groups (progress vs progress free:0.49±0.57 vs 0.76±0.22; recurrence vs no recurrence:0.48±0.60 vs 0.73±0.22; death vs living:0.44±0.64 vs 0.73±0.19,u=39,50,53,all P＜0.05).ROC curve was used to obtain the optimal threshold of MTV for predicting the recurrence or not,and it showed that 12.4 ml was the optimal value with the AUC of 0.93.Using this cut-off value,the survival analysis showed a poor prognosis with MTV more than 12.4 ml.The median survival time of MTV less or more than 12.4 ml was 31.4 and 13.1 months respectively (x2 =9.0,P＜0.05).The optimal threshold of △R14 for patients' status of progress free or progress and death or living was 0.75.Multivariate Cox regression survival analysis indicated that the pre-treatment MTV was an independent prognostic factor for PFS and RFS (Wald values:10.80 and 10.30,hazard ratios:1.13 and 1.14 respectively).Conclusions In this group of nonsurgical ESCC patients with CCRT,pre-treatment MTV could independently predict the survival.SUVmaxax2,SUVmax3 and SUVmax4 could be able to predict the RFS.The patients with reduction of SUVmax4 to SUVmax1less than 75％ had a poor PFS,RFS and OS.