Dual-utility ctDNA in diffuse large B-cell lymphoma: integrated genotyping unveils minimal residual disease dynamics and subtype-specific clonal evolution
摘要
Circulating tumor DNA (ctDNA) profiling offers a noninvasive approach to monitor minimal residual disease (MRD) and clonal evolution in diffuse large B-cell lymphoma (DLBCL).
MethodsIn this study, we analyzed 164 newly diagnosed DLBCL patients from the First Affiliated Hospital of Nanjing Medical University undergoing R-CHOP-like therapy. Tissue and serial plasma samples were sequenced using a 475-gene lymphoma-specific panel.
ResultsWe defined end-of-treatment MRD (EOT-MRD) positivity based on detectable tissue-informed variants or published driver mutations, which identified 37.2% (61/164) of patients as EOT-MRD( +). EOT-MRD status significantly predicted progression-free and overall survival, complementing both International Prognostic Index (IPI) and positron emission tomography-computed tomography response assessment. Integration of EOT-MRD with IPI into a composite “IPI-M” model improved risk stratification. CtDNA dynamics revealed that 77.1% (37/48) of patients acquired new gene alterations (GAs) at progression, of which most were enriched in cell cycle regulation, p53 pathway, PI3K/AKT signaling pathway, and epigenetic regulation. In addition, primary refractory patients exhibited a higher proportion of shared mutations from baseline to progression, while relapsed patients gained more emergent mutations at progression. Different genetic subtypes manifest divergent progression and distinct evolutionary patterns. TP53-disrupted subtype drove primary refractoriness via persistence of TP53 mutation, MCD subtype was prone to relapse despite high remission rate with frequent baseline mutation clearance and propensity for branched evolution, and BN2 showed mixed refractoriness and relapse with predominance of shared alterations.
ConclusionsOur findings underscore the dual utility of ctDNA in enhancing prognostic stratification and elucidating subtype-specific evolutionary dynamics, supporting personalized treatment strategies in DLBCL.
Graphical Abstract