Abstract
BACKGROUND: Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma, characterized by significant clinical and molecular heterogeneity, which leads to considerable variability in patient prognosis. Programmed cell death (PCD) plays a critical role in the development and progression of various cancers. A comprehensive analysis of PCD-related gene expression in DLBCL could enhance risk stratification and inform personalized treatment strategies. METHODS: This study integrated five DLBCL datasets with 18 PCD-related gene expression profiles to identify differentially expressed genes (DEGs) associated with PCD. Patients were stratified into two subgroups (C1 and C2) using consensus clustering analysis. We further performed immune infiltration analysis, GSVA enrichment analysis, and WGCNA to uncover significant differences in the immune microenvironment and signaling pathways between the subgroups. Additionally, 12 machine learning algorithms were employed to construct predictive models for DLBCL, with performance evaluated using AUC and F-score metrics. Finally, transcriptome sequencing of the DLBCL cell line VAL and the normal human B lymphocyte cell line IM-9 was conducted to validate potential biomarkers. RESULTS: A total of 1074 PCD-related DEGs were identified. Unsupervised clustering revealed two distinct molecular subtypes of DLBCL. The C2 subgroup exhibited upregulation of pathways involved in DNA repair, cell cycle, and energy metabolism, alongside significant downregulation of immune evasion-related pathways, indicating its classification as a high-risk group. Machine learning algorithms and transcriptome sequencing validation identified five potential biomarkers for DLBCL, including CTSB, DPYD, SCARB2, STOM, and GBP1. CONCLUSIONS: This study identifies two distinct DLBCL subtypes based on PCD-related gene expression, with the C2 subtype characterized as high-risk due to enhanced DNA repair and cell cycle pathways. Five key biomarkers (CTSB, DPYD, SCARB2, STOM, GBP1) may improve risk stratification and understanding of DLBCL heterogeneity. These findings lay the groundwork for further exploration of DLBCL progression and potential prognostic improvements.