Integrating genomic features for prognosis in Chinese patients  with B-cell lymphoma following R-CHOP therapy

Abstract Background: Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma. Despite therapeutic advancements, 30–40% of patients develop relapsed or refractory disease, underscoring the need for reliable genomic biomarkers to predict responses to first-line R-CHOP therapy. Methods Tumour and matched oral control DNA samples from 147 Chinese patients with DLBCL treated with R-CHOP were subjected to targeted sequencing to identify genomic predictors of treatment response and survival. Kaplan–Meier survival analyses and Cox proportional hazards models were employed to assess the associations of genetic mutations and clinical factors with progression-free survival (PFS), overall survival (OS), and relapse risk. Results Mutations in TP53 (HR=2.78,P=0.002), SRP72 (HR=2.79,P=0.010), MYC (HR=1.97,P=0.033), BCL2 (HR=2.54,P=0.021, and ASXL2 (HR=2..20,P=0.048) were significantly associated with poor PFS in both univariate and multivariate analyses. A risk score incorporating these five mutations was developed, which, when combined with baseline LDH levels and the presence of bone marrow tumour cells, formed a comprehensive prognostic model. This integrated model outperformed the conventional International Prognostic Index (IPI) in predicting PFS. Additionally, TP53 (HR=4.45,P=0.003), SRP72 (HR=4.16,P=0.014), and MYC (HR=2.64,P=0.047) mutations were strongly associated with increased relapse risk. A novel model combining these three genes with baseline LDH demonstrated superior predictive power for relapse compared to the IPI. Conclusions Mutations in TP53, SRP72, MYC, BCL2, and ASXL2 serve as robust genomic markers of poor prognosis in DLBCL patients receiving R-CHOP. A newly developed combined model incorporating genetic and clinical variables offers improved prognostic accuracy over traditional models and may inform personalised risk stratification strategies.