RT Journal Article SR Electronic T1 Prognostic value of 18F-FDG PET tumor texture analysis in non-small cell lung cancer JF Journal of Nuclear Medicine JO J Nucl Med FD Society of Nuclear Medicine SP 1391 OP 1391 VO 56 IS supplement 3 A1 Jihyun Park A1 Byung Hyun Byun A1 Hansol Moon A1 Joon Ho Choi A1 Ilhan Lim A1 Byung Il Kim A1 Chang Woon Choi A1 Sang Moo Lim YR 2015 UL http://jnm.snmjournals.org/content/56/supplement_3/1391.abstract AB 1391 Objectives This study investigated whether 18F-FDG PET tumor texture analysis could predict overall survival (OS) in patients with non-small cell lung cancer (NSCLC).Methods A total of 55 patients with stage Ⅰ to Ⅲ NSCLC (35 squamous cell carcinoma, 20 adenocarcinoma) were retrospectively analyzed. All patients underwent pretreatment 18F-FDG PET/CT, and were followed up for at least 2 years. To analyze metabolic parameters, volumes of interest including the whole tumor region were drawn, and the maximum SUV (SUVmax), the metabolic tumor volume (MTV) and the total lesion glycolysis (TLG) at the cutoff SUV 2.5 were respectively measured. To analyze heterogeneity parameters, regions of interest were drawn on one transaxial PET slice containing SUVmax. First- and second-order texture analyses were performed, and significant texture parameters were selected on the basis of the receiver operating characteristic (ROC) curve analysis. The relationships between covariates and OS were analyzed using multivariate Cox proportional-hazard regression method.Results The median OS of 55 patients were 25 months (IQR, 12-43 months). Based on ROC curve analysis, none of the first-order texture parameters predicted OS. Among second-order texture parameters, entropy, contrast, correlation, and variance predicted OS. On univariate analysis, older age (p = 0.038), higher SUVmax (p = 0.030), higher MTV (p = 0.0002), higher TLG (p = 0.0002), and higher entropy (p = 0.0039) were significantly associated with poorer OS. On multivariate analysis, age (p = 0.025; HR = 2.70), TLG (p = 0.011; HR = 3.02), and entropy (p = 0.045; HR = 2.90) were found as independent predictors of OS.Conclusions The combination of the 18F-FDG PET tumor texture analysis and TLG predicts survival more accurately than TLG alone in patients with NSCLC.Research Support None.