Establishment of a prognostic model for diffuse large B-cell lymphoma based on mitochondrial energy metabolism-related genes

Abstract Mitochondrial energy metabolism has emerged as a critical regulator of tumor progression and immune dynamics. However, its prognostic relevance in diffuse large B-cell lymphoma (DLBCL) remains poorly defined. This study aimed to develop a prognostic model based on mitochondrial metabolism-related genes (MMRGs) and elucidate its relationship with the tumor immune microenvironment (TIME). Consensus clustering based on MMRG expression identified two distinct metabolic subtypes characterized by divergent clinical outcomes and immune profiles. Functional enrichment analyses revealed differential enrichment in pathways such as glucose metabolism, oxidative phosphorylation, and PI3K-Akt signaling. To develop a robust prognostic tool, we constructed a seven-gene risk model (IDH3A, CPT1A, ACADL, ATP12A, ADH1A, ECI1, and ACSL5) using LASSO-Cox regression. This model effectively stratified patients into high- and low-risk groups, with the latter exhibiting significantly prolonged overall survival (P < 0.001). The predictive performance of this model was validated in both internal and external cohorts and further supported by time-dependent ROC analysis and nomogram calibration. To explore the biological mechanisms underlying prognostic stratification, we evaluated immune infiltration using ESTIMATE, ssGSEA, and CIBERSORT. High-risk patients exhibited lower immune and stromal scores, elevated tumor purity, and greater infiltration of immunosuppressive cells, including regulatory T cells and plasmacytoid dendritic cells. Notably, risk scores correlated positively with the expression of immune checkpoint genes such as DOT1L, MCL1, PLK1, and ASXL1, suggesting potential therapeutic targets. Drug sensitivity prediction further revealed that high-risk patients may respond better to cisplatin and oxaliplatin but show reduced sensitivity to venetoclax. Additionally, WGCNA identified ADH1A and ACADL as hub genes with strong prognostic and immunological relevance, and protein expression validation using the Human Protein Atlas supported their differential expression in lymphoma tissues. In summary, this study established a novel MMRG-based prognostic model that enables precise risk stratification and reveals key links between mitochondrial metabolism and the immune landscape in DLBCL. These findings provide valuable insights for individualized prognosis evaluation and the development of metabolism-immune targeted therapies.