Accurate Diagnosis Of Glioma Under The Guidance Of CHO PET

Background:Glioma remains the most common malignant central nervous system tumor and is associated with diverse prognosis.World Health Organization(WHO)grading,isocitrate dehydrogenase(IDH)mutation,telomerase reverse transcriptase(TERT)promoter mutation and O6-methylguanine-DNA methyltransferase(MGMT)promoter methylation status provide diagnostic,prognostic,predictive and therapeutic information of primary diffuse glioma.This study aimed to investigate the relationship between quantitative 18F-fluoroethylcholine(18F-CHO)positron emission tomography(PET)characteristics and these pathological and molecular features.Methods:Thirty-nine patients who underwent 18F-CHO PET with histopathologically confirmed primary diffuse glioma were prospectively enrolled and retrospectively analyzed for the differentiation of lower grade glioma(LGG)from glioblastoma(GBM),and 28 patients who were tested for IDH,TERT and MGMT alteration were included for the analysis of molecular alterations.The three-dimensional region of interest(ROI)was semiautomatically defined based on the standardized uptake value(SUV)threshold,and a total of 74 quantitative imaging biomarkers,including 13 shape features,31 SUV-based features,and 30 normalized SUV-based features,were defined.Five traditional CHO biomarkers,namely,SUVmax,SUVmean,metabolic tumor volume(MTV),total lesion CHO uptake(TLC)and tumor-to-normal contralateral cortex activity ratio(T/N ratio)were included in the SUV-based,SUV-based,SUV-based,shape and normalized classes,respectively.Wilcoxon rank sum test,receiver operating characteristic(ROC)curves and correlation coefficient analyses were applied to evaluate the performances of CHO features and to select the independent representative biomarkers.Kaplan-Meier curve and log-rank test were utilized to stratify patient prognosis.Results:Among the 74 quantitative imaging biomarkers,89.2%were significantly different between LGG and GBM,and the SUV-based features displayed higher area under the ROC curve(AUC)values than the other feature groups.Among the 5 traditional biomarkers,SUVmax and TLC were the most distinguishing,with AUCs of 0.880 and 0.860(0.938 and 0.927 after reclassification of two outliers),respectively,both of which could also stratify patient prognosis better than WHO grade.Five alternative imaging biomarkers,including 2 shape features and 3 SUV-based features,were considered representative to distinguish LGG and GBM,with AUCs ranging from 0.754 to 0.854.Due to the limited scale of the included cohort,only traditional imaging biomarkers were investigated regarding their relationship with molecular alterations,and all 5 features were significantly higher in IDH-wildtype gliomas than in IDH-mutant gliomas,with SUVmax,SUVmean,TLC and the T/N ratio exhibited AUCs of 0.856-0.918 in distinguishing the IDH status.Although the differences and performances of the traditional biomarkers in distinguishing diverse TERT and MGMT status were moderate in the whole population,T/N ratio and TLC displayed certain predictive value in discriminating the TERT status in the IDH-mutant and IDH-wildtype subgroups,with AUCs of0.860 and 0.857,respectively.Conclusions:Imaging biomarkers from 18F-CHO PET are reliable in determining the WHO grade of primary diffuse gliomas and suggested that GBM has a larger volume,a more spherical shape,higher choline activity in most interval segments and a more symmetrical distribution than LGG.The traditional 18F-CHO PET biomarkers are also correlated with IDH but not TERT or MGMT alterations in the whole population.In accordance with the clinical meaning of the TERT promoter mutation,T/N ratio and TLC can also discriminate the TERT status in IDH subgroups.