18F-FDG PET/CT for Prognostic Stratification of Patients With Extranodal Natural Killer/T-Cell Lymphoma

Purpose The prognostic value of PET/CT for extranodal natural killer/T-cell lymphoma (ENKTL) patients is controversial. We sought to investigate the use of pretreatment and interim PET/CT in this disease. Methods Sixty-five patients diagnosed with ENKTL who underwent pretreatment (n = 53) and/or interim PET/CT scans (after 2–4 cycles of chemotherapy and/or radiotherapy, n = 33) were retrospectively enrolled. Interim PET/CT images were interpreted according to the Deauville 5-point scale. PET/CT results were assessed for their predictive value of progression-free survival (PFS) and overall survival (OS). Kaplan-Meier estimates were generated to evaluate the predictive value of clinical parameters and PET/CT scans for prognosis evaluation. The Cox proportional hazards model was performed to assess the potential independent predictors for PFS and OS. Results Patients with a high score (4/5) according to the Deauville scale had heavier tumor burdens and tended to have elevated serum LDH and β2-microglobulin (β2-MG). In univariate analysis, the following parameters were predictive of PFS: age (≤60 vs > 60 years), stage (I/II vs III/IV), lesion location (nasal vs extranasal), LDH (normal vs elevated), β2-MG (normal vs elevated), SUVmax before treatment (≤8.3 vs > 8.3), and interim PET/CT results. For OS, the predictive factors were composed of stage, lesion location, LDH, β2-MG, and interim PET/CT results. However, in multivariate analysis, only interim PET/CT scan interpreted by the Deauville scale remained as an independent predictor for both PFS and OS. Conclusions Interim PET/CT scan displayed potential predictive value for ENKTL patients. Response assessment according to the Deauville 5-point scale may help to improve the accuracy of prediction. Patients with advanced stage (III/IV), elevated LDH or β2-MG, and a high Deauville score of 4 to 5 on the interim PET/CT scan were more likely to have reduced PFS and OS.

http://bit.ly/2TCmRlM