Dual-Phase ¹⁸F-FDG PET/CT For Predicting EGFR Mutations In Patients With Non-Small Cell Lung Cancer

ObjectiveThe aim of this study was to investigate the relationship between glucose metabolic changes in dual-phase 18F-FDG PET/CT and epidermal growth factor receptor(EGFR)mutation in patients with non-small cell lung cancer(NSCLC).MethodsThis retrospective study included 60 NSCLC patients(from August 2013 to November 2018)who were received dual-phase 18F-FDG PET/CT(early phase:40 minutes after injection of 18F-FDG;delay phase:120 minutes after injection of 18F-FDG)scanning at Zhejiang University Medical PET Center and had the result of EGFR mutation assay.The patients were divided into two groups,including EGFR mutant group(32 cases)and EGFR wildtype group(28 cases).Statistical analysis was performed by using univariate logistic regression and Fishery exact test to analyze the value of clinical characteristics,early and delay phases semi-quantitative parameters of the primary tumor(pSUVmax,pSUVmean,pSUVpeak,pMTV,pTLG)and the variation of semi-quantitative parameters of the primary tumor(%change of pSUVmax,pSUVinc,ASUVmean,ASUVpeak)in predicting EGFR mutations in NSCLC patients.Multivariate logistic regression analysis was used to investigate the independent predictor for EGFR mutations in patients with NSCLC.The cutoff value for predicting EGFR mutations was analyzed using the receiver operating characteristic curve(ROC)and the area under the ROC curve(AUC).P value less than 0.05 was considered as statistically significant difference.ResultsThere were 28 males(46.7%)and 32 females(53.3%)with a median age of 64.5 years(range,35-80 years;average 62.3 ± 11.1 years).Among these 60 patients,32 patients(53.3%)were EGFR mutations,and 28 patients(46.7%)were EGFR wildtype.1.Univariate analysis(EGFR mutations vs EGFR wildtype)Compared with patients with EGFR wildtype,patients with EGFR mutations had significantly lower early phase semi-quantitative parameters(epSUVmax,epSUVpeak,epMTV,epTLG),delay phases semi-quantitative parameters(dpSUVmax,dpSUVmean,dpMTV,dpTLG)and the variation of semi-quantitative parameters(%change of pSUVmax,pSUVinc and △SUVpeak)(P=0.001～0.043).In patients with NSCLC,EGFR mutations were markedly correlated with females(P<0.001),non-smokers(P<0.001),adenocarcinomas(P=0.003)and primary tumor diameter less than 3.25cm(P=0.013).EGFR mutations had no significant correlation with age(P=0.77),family history(P=0.201),TNM staging(P=0.701),primary tumor location(P=0.362),△SUVmean(P=0.083),epSUVmean(P=0.094)and dpSUVpeak(P=0.123).2.Multivariate analysis(EGFR mutations vs EGFR wildtype)Non-smokers(P=0.01)and pSUVinc less than 1.885(P=0.03)were independent predictors of EGFR gene mutation in NSCLC patients.Conclusion(1)Compared with patients with wildtype,patients with EGFR mutations have lower glucose metabolism in two(early and delay)phases and lower value change of glucose metabolism;(2)Smoking and the variation of glucose metabolism are strong factors in predicting EGFR mutations of NSCLC patients;(3)Dual-phase 18F-FDG PET/CT has a potential in noninvasively detecting EGFR mutations in NSCLC patients.