Abstract
BACKGROUND: Survival prediction accuracy of fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography ((18)F-FDG PET/CT) in extra-nodal natural killer/T-cell lymphoma (ENKTL) is controversial. This study aimed to evaluate the prognostic value of (18)F-FDG PET/CT parameters including maximum standardized uptake value (SUVmax), total metabolic tumor volume (TMTV) and total lesion glycolysis (TLG), and to develop a new prognostic model for ENKTL. METHODS: We analyzed 390 ENKTL patients with comprehensive clinical and survival data. All patients received asparaginase-based chemotherapy with or without radiotherapy, or radiotherapy alone. Metabolic tumor volume (MTV) was calculated using a 41% SUVmax threshold, and TLG was computed as MTV multiplied by the average SUV. Progression-free survival (PFS) and overall survival (OS) were assessed using Kaplan-Meier curves and compared with log-rank tests. Optimal cut-off values were determined using the Youden' index. Cox regression analysis identified significant prognostic factors. A nomogram predicting 1-, 3-, and 5-year survival was developed and validated using the C-index and calibration curves. Statistical significance was set at p < 0.05. RESULTS: Of the 390 patients, 262 (67.2%) were included in the training set and 128 (32.8%) in the validation set. (18)F-FDG PET-CT parameters with cutoff values of SUVmax > 12.8, TMTV > 16.4 cm(3), and TLG > 137.0, were significantly associated with poorer OS (p = 0.009) and PFS (p = 0.003). Multivariable Cox regression identified the following as independent predictors of worse OS: age > 60 years (HR = 1.923, 95% CI: 1.001-3.693), presence of B symptoms (HR = 1.861, 1.132-3.059), ECOG score ≥ 2 (HR = 2.076, 1.165-3.699), extranodal involvement ≥ 2 sites (HR = 2.349, 1.384-3.988), bone marrow involvement (HR = 4.884, 2.137-11.163), SUVmax > 12.8 (HR = 2.226, 1.260-3.930), and TMTV > 16.4 cm(3) (HR = 1.854, 1.093-3.147). The new prognostic model achieved a C-index of 0.772 for OS and 0.750 for PFS in the training set, and 0.777 for OS and 0.696 for PFS in the validation set. Area under the curve (AUC) values for 1-, 3-, and 5-year OS were 0.841, 0.804, and 0.767 in the training set, and 0.718, 0.786, and 0.893 in the validation set. Risk stratification divided patients into four groups with significant differences in survival (p < 0.001). CONCLUSION: SUVmax, TMTV, and TLG are independent prognostic factors in ENKTL. Our new model, which integrates (18)F-FDG PET/CT metrics with clinical data, enhances survival prediction and may support personalized treatment strategies, though further validation is required.