Predictive Value Of ¹⁸F-FDG PET/CT On EGFR Gene Mutation And ALK Gene Rearrangement In Non-small Cell Lung Cancer

Background and objectiveEpidermal growth factor receptor(EGFR)mutation and anaplastic lymphoma kinase(ALK)fusion are the two most common targets for drug therapy in non-small cell lung cancer(NSCLC).However,it is found in clinical practice that gene testing is sometimes impossible.The purpose of this study is to combine clinical risk factors with PET/CT parameters to initially establish and verify a prediction model for EGFR gene mutation and ALK fusion positive in patients with non-small cell lung cancer.MethodsWe retrospectively included 87 newly diagnosed NSCLC patients confirmed by pathology department of**hospital affiliated to Xiamen university.These patients underwent EGFR or ALK gene detection and PET/CT examination before treatment.The differences between relevant clinical risk factors and 18F-FDG PET/CT parameters were collected and compared,including the Maximum Standardized Uptake Value of primary tumor,lymph node and distant metastasis.Respectively defined as pSUVmax,nSUVmax,mSUVmax,tumor metabolic volume(MTV),total glycolysis(TLG)and other metabolic parameters.Univariate and multivariable Logistic regression analysis was conducted to screen out independent predictors of EGFR gene mutation and ALK gene rearrangement.And clinical-PET/CT nomogram,a decision aid tool for gene detection,was established and validated.ResultsEGFR mutation was found in 44 patients(50.6%),and ALK fusion gene expression was found in 8 patients(17.8%).EGFR mutation is more common in nonsmokers,adenocarcinoma and TNM stage IV.Lower pSUVmax,nSUVmax,mSUVmax,MTV,TLG,TLR are significantly correlated with EGFR mutations.Multivariate logistic regression showed that pSUVmax,nonsmokers and adenocarcinoma were predictive factors of EGFR mutation.AUC values obtained from the working characteristic curve(ROC)of the subjects obtained from the low pSUVmax alone and the combination of three factors were 0.784 and 0.860 respectively.ALK positive patients tend to be women and patients with higher pSUVmax and nSUVmax.Younger age and lower MTV are two independent predictors of ALK positivity.The models for predicting the two genes are visualized respectively,and clinical-PET/CT nomograms are preliminarily established and verified.ConclusionThis study demonstrates that low pSUVmax is associated with EGFR gene mutations and can be combined with other clinically relevant indicators to enhance the accuracy of predicting EGFR mutation status in some NSCLC patients who cannot be tested for EGFR genes.At the same time,it was found that another parameter was initially integrated and validated the clinical-PET/CT nomogram.As a risk assessment of genetic mutations for NSCLC patients and the slender guidance method,the clinical-PET/CT nomogram is expected to become a valuable predictive tool in clinic.