Metabolic tumor volume derived from F-FDG PET/CT as a prognostic parameter for non-small cell lung cancer (NSCLC) patients.

To determine whether the prognostic stratification of non-small cell lung cancer (NSCLC) patients could be made by fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (F-FDG PET/CT)-derived parameters such as maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV) and total lesion glycolysis (TLG). A total of 106 patients who were initially diagnosed with NSCLC with clinical stage III or stage IV at our hospital from January 2015 to January 2018 were included. The metabolic and volumetric parameters of F-FDG PET/CT were systematically collected, and their optimal cut-off values were determined on the basis of the receiver operating characteristic (ROC) curves. Kaplan-Meier methods and log-rank test were used to evaluate the relationships between F-FDG PET/CT-derived parameters and overall survival (OS) of NSCLC patients. The univariate and multivariate Cox analysis were conducted to identify the independent predictors of OS. The optimal cut-off value of SUVmax was 8.94 and area under the curve (AUC) for identifying patients with mortality risk was 0.618 (95% confidence interval [CI]: 0.490-0.745), with a sensitivity of 78.6% and specificity of 53.3%. The optimal cut-off value of MTV40 was 12.44 and the AUC value was 0.785 (95%CI: 0.676-0.893), with a sensitivity of 85.7% and specificity of 71.7%. Furthermore, the ROC curves identified 71.95 as the optimal cut-off value of TLG40, and the AUC value, sensitivity and specificity were 0.782 (95%CI: 0.681-0.883), 78.6% and 70.4%, respectively. The Kaplan-Meier curves showed that SUVmax (HR for SUVmax >8.94: 3.501, 95%CI: 1.133-10.817, P=0.029), MTV40 (HR: 6.926 for MTV40 >12.44, 95%CI: 2.244-21.378, P=0.001) and TLG40 (HR: 4.314 for TLG40 >71.95, 95%CI: 1.503-12.381, P=0.007) were significantly associated with poor OS of NSCLC patients. However, only MTV40 (HR: 4.235, 95%CI: 1.324-13.526, P=0.015) was shown to have an independent role in the multivariate Cox analysis. Metabolic tumor volume had a superiority in predicting the prognosis of NSCLC patients compared with other metabolic and volumetric parameters, suggesting that it might be a valuable prognostic marker.