Total metabolic tumor volume (TMTV) on [18F]FDG PET/CT is a potent prognostic factor in diffuse large B-cell lymphoma (DLBCL). However, its clinical application is limited by complex measurement methods. This study aimed to develop a simple, practical...
Total metabolic tumor volume (TMTV) on [18F]FDG PET/CT is a potent prognostic factor in diffuse large B-cell lymphoma (DLBCL). However, its clinical application is limited by complex measurement methods. This study aimed to develop a simple, practical PET-based scoring system that reflects both anatomical extent and metabolic activity and to evaluate its prognostic value.
A total of 93 patients with DLBCL were retrospectively analyzed. A scoring system was developed based on the sum of lymph node (LN) and extranodal (EN) lesion scores. The LN score was derived from the involvement extent and Deauville score across six regions. The optimal weight factor for EN involvement was determined via 5-fold internal cross-validation. The prognostic performance of the score for progression-free survival was evaluated using receiver operating characteristic (ROC) curve and Cox proportional hazard ratio regression analyses.
A weight factor of 15 was identified as optimal, yielding the highest correlation with TMTV. The score significantly correlated with TMTV (ρ = 0.7815) and demonstrated fair prediction performance in the time-dependent ROC curve analysis (AUC = 0.7826). At an optimal cutoff of 25, the high-score group (≥ 25) exhibited significantly inferior PFS compared to the low-score group. Subgroup analysis demonstrated that the scoring system could further stratify patients within the low International Prognostic Index (IPI 0–2) group. In multivariable analysis, the score remained a significant independent predictor of PFS, even after adjusting ECOG performance status.
The novel PET-based scoring system is a simple and reliable tool for risk stratification in DLBCL. By providing an accessible alternative to TMTV, this system could be easily integrated into routine clinical practice and combined with other predictive clinical factors to improve prognostic accuracy.