Optimizing the prognostic capacity of baseline 18F-FDG PET/CT metabolic parameters in extranodal natural killer/T-cell lymphoma by using relative and absolute thresholds.
Heliyon
; 10(3): e25184, 2024 Feb 15.
Article
en En
| MEDLINE
| ID: mdl-38322946
ABSTRACT
Objectives:
To investigate the prognostic capacity of baseline 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) metabolic parameters in extranodal natural killer/T-cell lymphoma (ENKTCL), and the influence of relative thresholds (RT) and absolute thresholds (AT) selection on prognostic capacity. Materials andmethods:
Metabolic tumor volume (MTV)-based parameters were defined using RTs (41 % or 25 % of maximum standardized uptake value [SUVmax]), ATs (SUV 2.5, 3.0, 4.0, or mean liver uptake) in 133 patients. Metabolic parameters were classified into avidity-related parameters (SUVmax, mean SUV [SUVmean], standard deviation of SUV [SUVsd]), volume-related parameters (RT-MTV), and avidity- and volume-related parameters (total lesion glycolysis [TLG] and AT-MTV). The prognostic capacity of the metabolic parameters and the effects of different threshold types (RT vs. AT) were evaluated.Results:
All metabolic parameters were moderately associated with prognosis. However, the area under the receiver operating characteristic curve of MTV and TLG was slightly higher than that of avidity-related parameters for predicting 5-year progression-free survival (PFS) (0.614-0.705 vs. 0.563-0.609) and overall survival (OS) (0.670-0.748 vs. 0.562-0.593). Correlations of MTV and avidity-related parameters differed between RTs (r < 0.06, P = 0.324-0.985) and ATs (r 0.56-0.84, P ≤ 0.001). AT-MTV was the optimal predictor for PFS and OS, while RT-TLG was the optimal predictor for PFS, and the combination of RT-MTV with SUVmax was the optimal predictor for OS.Conclusion:
The incorporation of volume and avidity significantly improved the prognostic capacity of PET in ENKTCL. Composite parameters that encompassed both avidity and volume were recommended.
Texto completo:
1
Banco de datos:
MEDLINE
Tipo de estudio:
Prognostic_studies
Idioma:
En
Revista:
Heliyon
Año:
2024
Tipo del documento:
Article
País de afiliación:
China