Total metabolic tumor volume as a prognostic biomarker in high-burden follicular lymphoma: A systematic review and meta-analysis

Abstract

Abstract Background: Follicular lymphoma is an indolent disease characterized by clinical diversity. The Follicular Lymphoma International Prognostic Index and FLIPI2 lack precision in differentiating patients with a high chance of early progression. Total metabolic tumor volume on baseline [¹⁸F]FDG PET/CT provides a measure of total disease burden in a single value and is a promising imaging biomarker for outcomes. Objective: To assess the prognostic importance of baseline TMTV in high-tumor-burden FL research, specifically its association with progression-free survival, overall survival, and disease progression within 24 months, as well as its integration into clinical scores and biomarkers. Methods: We performed a systematic review and meta-analysis of studies reporting baseline TMTV in patients with high-burden FL (per GELF criteria). Databases searched included PubMed, Scopus, and Web of Science through July 2025. Extracted data included hazard ratios (HRs), odds ratios (ORs), TMTV cutoffs, segmentation methods (e.g., SUV4, 41% SUVmax), and integration with FLIPI, FLIPI2, or β2-microglobulin. Pooled estimates were calculated using random-effects models. Results: Eleven studies comprised 2,631 patients; high TMTV was significantly correlated with inferior results: pooled HR for PFS = 1.91 (95% CI: 1.59–2.29), OS = 1.78 (95% CI: 1.33–2.37), and OR for POD24 = 3.72 (95% CI: 2.22–6.23). The most validated cutoff was >510 cm³. PET methodology was generally standardized. In multivariable analyses, TMTV remained independent of FLIPI and PRIMA-PI. A combined TMTV/β2-microglobulin score stratified patients into distinct risk groups, with 5-year OS ranging from 96.1% to 73.7% (Zduniak et al., 2025). In the RELEVANCE and FOLL12 cohorts, integration of TMTV with FLIPI2 improved POD24 prediction and PFS discrimination (Cottereau et al., 2024; Durmo et al., 2025). Radiomic models combining TMTV with texture-based features (e.g., medPCD, TVSR) yielded even higher predictive accuracy (Draye-Carbonnier et al., 2024). These associations held across alternative TMTV thresholds (e.g., >144 mL; Jiang et al., 2025). Conclusion: Baseline TMTV is a reproducible and independent prognostic biomarker in high-burden FL, strongly associated with worse PFS, OS, and POD24. Combining TMTV with clinical or serum markers enhances risk stratification. These findings support incorporating TMTV into clinical workflows for risk-adapted treatment strategies; however, prospective validation and methodological standardization are crucial for broader implementation.