Value Of ¹⁸F-FDG PET/CT Metabolic Parameters Of The Primary Tumor For Predicting Clinical Stage In Non-Small Cell Lung Cancer

Part 1 Study of different methods for delineating primary tumor on the measurement of metabolic parameters in non-small cell lung cancerObjective:To evaluate the impact and comparative reproducibility of different PET/CT-based methods for delineating primary tumor on the measurement of metabolic parameters in non-small cell lung cancer(NSCLC).Methods:The data of 79 pathologically proven non-small cell lung cancer(NSCLC)patients who underwent 18F-FDG PET/CT examination were retrospectively analyzed.Maximum standardized uptake value(SUV)max was measured and metabolic tumor volume(MTV)and total lesion glycolysis(TLG)were calculated by three methods:the absolute SUV threshold method(MTVp and TLGp,p=2.5,3,4);the fixed%SUVmax threshold method(MTVq%and TLGq%,q=40,50),and the adaptive interative algorithm method(ALA)(MTVALA and TLGALA).The differences of metabolic parameters measured by different PET/CT-based methods and interobserver variability on metabolic parameters were analyzed.Results:The difference between MTV measured based on different outlining methods was statistically significant(χ2=14.759,P=0.011),but the difference between measured TLG was not statistically significant(χ2=3.013,P=0.698).For the absolute SUV threshold method and the fixed%SUVmax threshold method,the higher the threshold value,the smaller the measured MTV and TLG,and MTV and TLG measured by the fixed%SUVmax threshold method were smaller.The reproducibility was high for all metabolic parameters using any method(inter-class correlation>0.99 each).The ring chart showed the fixed%SUVmax threshold method(MTVq%and TLGq%,q=40,50)required minimal manual adjustment times,with the percentage of without manual adjustment 18.32%.Its accuracy was significantly the largest(χ2=11.748,p=0.038),91.14%.Conclusions:High reproducibility of MTV and TLG is obtained by all of the methods used,whereas the measured MTV are different.The fixed%SUVmax threshold method(MTVq%and TLGq%,q=40,50)requires minimal manual adjustment times,with its largest accuracy(91.14%),providing methodological guidance for clinical setting.Part 2 Value of 18F-FDG PET/CT metabolic parameters of the primary tumor for predicting clinical stage in non-small cell lung cancerObjective:To investigate the value of 18F-FDG PET/CT metabolic parameters of the primary tumor for predicting clinical stage in non-small cell lung cancer.Methods:The data of 140 pathologically proven non-small cell lung cancer(NSCLC)patients who underwent 18F-FDG PET/CT examination were retrospectively analyzed.The maximum diameter,maximum standardized uptake value(SUVmax),average standardized uptake value(SUVmean)of the primary tumor were measured.Metabolic tumor volume(MTV)and total lesion glycolysis(TLG)were calculated by the fixed%SUVmax threshold method(MTVq%and TLGq%,q=40).The risk factors of advanced-stage lung cancer were determined through univariate and multivariate logistic regression analysis.The ROC curves evaluated the value of each parameter in predicting advanced-stage lung cancer.Results:1.A total of 140 patients were included in the study.54 of them were squamous-cell carcinoma(SCC)and 86 of them were adenocarcinoma(ADC).For both groups,age(t=4.278,P<0.001),gender(χ2=15.389,P<0.001),smoking history(χ2=8.038,P=0.005),tumor sites(χ2=16.119,P<0.001)and symptoms at lung cancer diagnosis(χ2=8.481,P=0.004)differed significantly.Significantly higher values of the maximum diameter,SUVmax,MTV,TLG of the primary tumor were found in SCC(Z=-3.183,-4.902,-3.226,-4.506,P≤0.001 for all).2.In SCC,the proportion of patient who had central primary lesion(χ2=15.368,P<0.001)and who had symptoms at lung cancer diagnosis(χ2=5.597,P=0.018)in advanced-stage lung cancer were significantly higher than those in early-stage lung cancer.Significantly higher values of the maximum diameter,SUVmax,MTV,TLG of the primary tumor were found in advanced-stage lung cancer(Z=-3.566,-3.436,-3.514,-3.895,P≤0.001 for all).Receiver operating characteristic(ROC)analysis showed that all parameters had important value for the evaluation of advanced-stage lung cancer.The best diagnostic efficiency was TLG with the optimal cut-off value of 109.83(AUC=0.809).The sensitivity and specificity of TLG were 84.6%and 78.6%.The multivariate analysis showed that central tumor was the independent risk factor in advanced-stage SCC(odds ratio,17.448,95%CI=1.613-188.787,P=0.019).3.In ADC,the proportion of patient who had central primary lesion(χ2=8.585,P=0.003)and who had symptoms at lung cancer diagnosis(χ2=14.512,P<0.001)and the proportion of CEA ≥5.OOng/ml(χ2=9.498,P=0.002)in advanced-stage lung cancer were significantly higher than those in early-stage lung cancer.Significantly higher values of the maximum diameter,SUVmax,MTV,TLG of the primary tumor were found in advanced-stage lung cancer(Z=-3.05 1,-3.985,-3.93 1,-4.360,P≤0.001 for all).Receiver operating characteristic(ROC)analysis showed that all parameters had important value for the evaluation of advanced-stage lung cancer.The best diagnostic efficiency was TLG with the optimal cut-off value of 14.27(AUC=0.782).The sensitivity and specificity of TLG were 90.6%and 61.1%.The multivariate analysis showed that symptoms at lung cancer diagnosis,CEA≥5.00ng/ml and SUVmax were the independent risks factor in advanced-stage ADC(odds ratio,6.401,4.650,1.256,95%CI=1.889-21.696,1.338-16.158,1.073-1.470,P=0.003,0.016,0.004).4.The diagnostic accuracy and AUC were 79.0%and 0.854(95%CI=0.773-0.934),respectively,for the logistic regression model,and the sensitivity and specificity for predicting advanced-stage adenocarcinoma were 78.1%and 79.6%,respectively.Conclusions:1.18F-FDG PET/CT metabolic parameters of the primary tumor are different according to the pathological type.2.18F-FDG PET/CT metabolic parameters of the primary tumor have important value for the evaluation of advanced-stage lung cancer.TLG shows the most powerful predictive performance for predicting mediastinal lymph node metastasis in both SCC and ADC with the optimal cut-off value of 109.83 and 14.27,suggesting great value for clinical setting.3.For ADC,18F-FDG PET/CT combined with clinical features can improve the diagnostic efficiency of clinical staging of lung cancer.Logistic regression analysis shows that symptoms at lung cancer diagnosis,CEA≥5.00ng/ml and SUVmax were the independent risks factor in advanced-stage ADC,and the model shows the most powerful predictive performance for predicting advanced-stage ADC,providing more accurate guidance for clinical setting.