Highlights
1. Baseline ctDNA levels independently predict progression-free survival in untreated follicular lymphoma.
2. Mutant molecules per milliliter (MMPM) outperformed FLIPI, FLIPI-2, and SPD scores in identifying high-risk patients.
3. The prognostic value of ctDNA was consistent across different chemotherapy regimens (CHOP/CVP and bendamustine).
Background
Follicular lymphoma (FL), the most common indolent non-Hodgkin lymphoma, typically has favorable outcomes with anti-CD20-based immunochemotherapy. However, approximately 20% of patients experience progression of disease within 24 months (POD24), associated with significantly worse 5-year overall survival (50% vs 90%). Current clinicogenetic risk scores like FLIPI and FLIPI-2 have suboptimal predictive accuracy for early progression, creating an urgent need for better stratification tools.
Study Design
This ctDNA analysis utilized baseline plasma samples from the global Phase-3 GALLIUM trial (NCT01332968), which compared obinutuzumab-chemotherapy versus rituximab-chemotherapy in untreated FL patients. Key endpoints included:
– POD24 prediction (primary)
– Progression-free survival (PFS)
– Comparison with FLIPI, FLIPI-2, and simplified POD24 (SPD) scores
Key Findings
Prognostic Performance
Baseline ctDNA quantification achieved:
– AUROC 0.69 for POD24 prediction (vs 0.57-0.59 for FLIPI/FLIPI-2)
– HR 2.2 (95% CI 1.8-2.6) for PFS using cutoff ≥168.57 MMPM
Multivariate Analysis
ctDNA maintained prognostic significance after adjusting for:
– Treatment arm (obinutuzumab vs rituximab)
– Chemotherapy backbone (CHOP/CVP vs bendamustine)
– Traditional risk scores
Clinical Implications
The 168.57 MMPM cutoff identified:
– 35% of patients as high-risk (median PFS 2.1 years)
– 65% as low-risk (median PFS not reached)
Expert Commentary
“This study provides the most robust evidence to date for ctDNA’s role in FL risk stratification,” notes Dr. Weigert (co-author). The findings align with emerging data in diffuse large B-cell lymphoma, suggesting pan-lymphoma applicability of liquid biopsies. Limitations include lack of serial ctDNA monitoring and validation in non-trial populations.
Conclusion
Baseline ctDNA measurement offers superior prognostication compared to clinical scores in untreated FL. These results support:
1. Incorporation of ctDNA in clinical trial stratification
2. Development of risk-adapted treatment strategies
3. Further exploration of ctDNA-guided surveillance
Funding and Registration
Funded by F. Hoffmann-La Roche Ltd. ClinicalTrials.gov identifier: NCT01332968. The original GALLIUM trial was published in NEJM (2017;377:1331-1344).
References
1. Lutterbeck C, et al. Haematologica. 2026;111(4):420-431.
2. Marcus R, et al. NEJM. 2017;377:1331-1344.
3. Schuster SJ, et al. Blood. 2019;134(Suppl_1):6.
