The Association Between ¹⁸F-FDG PET/CT Metabolic Parameters And Egfr Mutation In Non-small Cell Lung Cancer

Purpose:Epidermal growth factor receptor（EGFR）mutation targeted therapy is of great significance to treatments of non-small cell lung cancer（NSCLC）.Previous studies revealed that ¹⁸F-fluorodeoxyglucose(¹⁸F-FDG)positron emission computed tomography（PET/CT）parameters are associated with EGFR mutation status,while the results are controversial.Thus,in this study,we aim to investigate the association between ¹⁸F-FDG PET/CT metabolic parameters and EGFR mutation in NSCLC patients,as well as the value of combining PET/CT metabolic parameters,patient clinical characteristics and serum tumor markers in predicting EGFR mutation.Methods:122 NSCLC patients with surgical or biopsy pathology results were retrospectively reviewed for the pre-treatment PET/CT parameters of the primary tumor,clinical features and serum tumor markers.PET/CT metabolic parameters include maximum normalized uptake value（SUVmax）,mean normalized uptake value（SUVmean）,maximum SUV corrected for lean body mass（SULmax）,mean SUV corrected for lean body mass（SULmean）,metabolic tumor volume（MTV）and total lesion glycolysis（TLG）.Tumor markers include:carcinoembryonic antigen（CEA）,cytokeratin 19 fragment 21-1（CYFRA21-1）,carbohydrate antigen 19-9（CA19-9）,carbohydrate antigen 125（CA125）,and neuron specific enolase（NSE）.Univariate analysis was used to analyze the relationship between EGFR mutation status with the factors above,and further analysis were performed on patient subgroups with different pathological types,tumor stages and mutation types.Multivariate logistic regression analysis was used to identify the independent predictors of EGFR mutation.Receiver operating characteristic（ROC）curve was used to evaluate the value of PET/CT parameters and their combination with clinical features and tumor markers in predicting EGFR mutation in NSCLC patients.Results:Univariate analysis demonstrated that female,non-smoker,adenocarcinoma,CYFRA21-1<2.08ng/ml,CA19-9>35.00U/ml were associated with EGFR mutations.SUVmax（mean±SD:8.2±5.0 vs.13.4±8.3,p<0.001）,SUVmean（4.5±1.5 vs.5.6±2.3,p=0.003）,SULmax（6.2±3.9 vs.10.9±6.7,p<0.001）,SULmean（3.5±1.2vs.4.5±1.9,p<0.001）,MTV（26.98±38.11 vs.64.53±80.82,p=0.003）and TLG（138.15±181.98 vs.418.95±558.74,p=0.001）were significantly lower in EGFR mutant patients than wild-type patients.Multivariate analysis revealed that gender,pathology,SUVmax（or SULmax）,and CA19-9 were independent predictors of EGFR mutations.ROC curve shows that a cutoff of SUVmax<12.7 or SULmax<10.2yields the highest value of area under the curve（AUC）,which is 0.688（95%CI:0.613-0.762）and 0.694（95%CI:0.620-0.769）,respectively.When combining the 4factors(i.e.female,adenocarcinoma,SUVmax<12.7 or SULmax<10.2,and CA19-9>35.00U/ml,the values AUC was 0.864（95%CI:0.793-0.935）and 0.864（95%CI:0.794-0.935）respectively.Conclusion:The PET/CT parameters of EGFR mutant NSCLC patients are lower than that of EGFR wild-type patients.Female,adenocarcinoma,SUVmax<12.7（or SULmax<10.2）and CA19-9>35.00U/ml are independent predictors of EGFR mutation statu.PET/CT parameters may be used to predict EGFR mutation status in NSCLC patients.