The Efficacy And Prognosis Of Standardized PET/CT Parameters In Patients With Locally Advanced NSCLC

Objective:The purpose of this study was to investigate the prognostic value of the different pretreatment parameters of 18-Fluorodeoxy-Glucose Positron Emission Tomography-Computed Tomography（18F-FDG PET/CT）and the related clinical factors in patients with locally advanced non-small-cell lung cancer（NSCLC）treated with concurrent chemoradiotherapy（CCRT）.The different parameters of 18F-FDG PET/CT include TLR（tumor SUVmax-to-liver SUVmax ratio）,TBR（tumor SUVmax-to-blood SUVmax ratio）,TLMR（tumor SUVmax-to-liver SUVmean ratio）and TBMR（tumor SUVmax-to-blood SUVmean ratio）.Secondly,we assessed the metabolic parameters of ¹⁸F-FDG PET/CT to explore the value in predicting the prognosis of concurrent chemoradiotherapy in patients with NSCLC.Methods:We collected the clinicopathological characteristics and the PET/CT metabolic parameters of 107 patients with pathologically confirmed stage III NSCLC in Yantai Yuhuangding Hospital affiliated to Qingdao University and Shandong Cancer Hospital from June 2014 to June 2017.The clinical characteristics included age,gender,smoking history,drinking history,tumor location,pathological type,clinical stage,carcinoembryonic antigen（CEA）level,neuron specific enolase（NSE）level,cytokeration 19 fragment（Cyfra21-1）level,treatment response and the survival outcomes.The PET/CT metabolic parameters include maximum standardized uptake value（SUVmax）and mean standardized uptake value（SUVmean）of primary tumors,as well as the SUVmax and SUVmean of liver and blood pool as background tissues.The standardized PET/CT parameters of TLR,TBR,TLMR and TBMR were obtained by calculation,the ratio of the SUVmax of the primary tumor to the SUVmax and SUVmean of the background tissues.The cut-off values were calculated by plotting receiver operating characteristic（ROC）curves and the continuous variables were classified into two categories.The influencing factors of the objective response rate（ORR）were analyzed by univariate and multivariate Logistic regression analysis.The Kaplan-Meier method with Log-rank test and the Cox regression model was used to calculate the overall survival（OS）and estimate survival curve of patients.Results:1.A total of 107 patients with locally advanced NSCLC and CCRT were enrolled in the study,including 77 males and 30 females;74 cases aged≤65 years,33cases>65 years old;65 cases with a history of smoking,42 cases with no history of smoking;59 cases with a history of drinking,48 cases with no history of drinking;59cases of central lung cancer,48 cases of peripheral lung cancer;36 cases of squamous cell carcinoma,71 cases of adenocarcinoma;46 cases in stage IIIA and 61 cases in stage IIIB.2.The optimal cut-off values of CEA level,NSE level,Cyfra21-1 level,SUVmax of primary tumor,SUVmean of primary tumor,TLR,TBR,TLMR and TBMR were obtained by ROC curve,which were 7.60,15.30,3.95,12.00,4.35,4.27,6.73,4.48 and6.69,respectively.Further comparative analysis showed that the areas under the ROC curve（AUCs）of each PET/CT parameter ranged from 0.720 to 0.843.Among them,primary tumor SUVmax（AUC=0.814）and TLMR（AUC=0.843）could predict the prognosis of ORR better.3.Univariate and multivariate logistic regression analysis indicated that TLMR（HR:0.085;95%CI:0.012–0.610;P=0.014）,SUVmax（HR:0.139;95%CI:0.026–0.734;P=0.020）and CEA level（HR:0.030;95%CI:0.006–0.149;P<0.001）were independent predictors of ORR.4.The Kaplan-Meier method showed that the continuous variables SUVmax,SUVmean,TLR,TBR,TLMR and TBMR which were classified into two categories according to the optimal cut-off values had significant differences in OS.Univariate analysis showed that tumor clinical stage（P=0.007）,tumor treatment response（P=0.001）,SUVmax（P=0.017）,SUVmean（P=0.021）,TLR（P<0.001）,TBR（P=0.018）,TLMR（P<0.001）and TBMR（P=0.003）were the important predictors of OS.5.Multivariate analysis showed that TLMR（HR:0.344;95%CI:0.141–0.841;P=0.019）and tumor clinical stage（HR:0.605;95%CI:0.387–0.946;P=0.028）were independent prognostic factors for patients’OS which analyzed by COX proportional hazards model.Among them,the median OS of TLMR above the critical value group and below the critical value group was 12.5 months and 26 months,respectively.Conclusions:1.The standardized PET parameter TLMR is a better index than other parameters such as the SUVmax in predicting the efficacy of treatment.2.The standardized PET parameter TLMR is an independent predictor for ORR and OS in patients with locally advanced non-small cell lung cancer.Patients with a lower TLMR would have a better efficacy and prognosis.