Prognostic genes related to mitochondrial dynamics and mitophagy in diffuse large B-cell lymphoma are identified and validated using an integrated analysis of bulk and single-cell RNA sequencing

Qingjiao Chen^1,2,3Mingui Chen⁴Jizhen Wang^1,2,3Jinfeng Dong^1,2,3Apeng Yang^1,2,3Xiaolin Zhu^1,2,3Qiaoxian Lin^1,2,3Jinlong Huang^1,2,3Guilan Lai^1,2,3Meihong Zheng^1,2,3Zhiyong Zeng^1,2,3Junmin Chen^1,2,3Junfang Lin^1,2,3Xiaoqiang Zheng^1,2,3*

• ¹First Affiliated Hospital of Fujian Medical University, Fuzhou, China
• ²First Affiliated Hospital of Fujian Medical University Binhai Campus, Fuzhou, China
• ³Key Laboratory of Laboratory Medicine, Fuzhou, China
• ⁴Hongshan Town Community Healthcare Service Center, Fuzhou, China

Background: While the link between mitochondrial homeostasis, specifically dynamics and mitophagy, and the progression of diffuse large B-cell lymphoma (DLBCL) has been suggested, their prognostic significance and functional networks remain unclear. This study aimed to investigate the role of mitochondrial dynamics-related genes (MDRGs) in DLBCL patient outcomes. Methods: Candidate MDGRs were identified via Weighted Gene Co-expression Network Analysis (WGCNA) and differential expression analysis using public RNA-seq data. A prognostic signature was established via LASSO-Cox regression, followed by proportional hazards assumption validation. Functional pathways, regulatory networks (including miR-1252-5p/NEAT1), and a risk-scoring model were analyzed. Model assessment included nomograms, immune cell infiltration, m6A regulator, and pharmacogenomics. Single-cell mapping was employed to characterize B-cell differentiation and spatial gene expression. Finally, the findings were validated using RT-qPCR on clinical samples. Results: Six lysosomal-enriched genes (TCF7, CEBPA, BBC3, GALR3, BMP8B, and BAALC) were identified as independent prognostic indicators. A composite model integrating our risk score and clinical parameters showed superior predictive accuracy (AUC > 0.8). High-risk DLBCL was characterized by altered M0 macrophage infiltration, YTHDC1-mediated m6A dysregulation, and dihydrotestosterone sensitivity. Single-cell analysis revealed an association between stage-specific B-cell differentiation and gene expression gradients. RT-qPCR confirmed the upregulation of CEBPA, BBC3, GALR3, BMP8B, and BAALC in DLBCL clinical samples. Conclusion: TCF7, CEBPA, BBC3, GALR3, BMP8B, and BAALC were identified as novel lysosomal pathway-enriched prognostic genes in DLBCL. Our validated composite model demonstrated strong predictive power. These findings establish an association between high-risk disease and specific tumor microenvironment alterations (M0 macrophages), epitranscriptomic dysregulation (m6A), and therapeutic vulnerabilities, providing valuable insights for refining prognosis and advancing targeted therapies for DLBCL.