Abstract
Lung adenocarcinoma (LUAD) is the most prevalent subtype of lung cancer, presenting significant challenges in treatment and prognostic prediction. Despite advancements in therapeutic approaches in recent years, personalized medicine has not yet achieved a notable breakthrough. Given the poor prognosis of patients, there is an urgent need to enhance the ability for precise prediction. Mitochondria play a crucial role in the metabolism and energy production of cancer cells, yet their specific impact in lung adenocarcinoma warrants further investigation. This study leveraged data from the TCGA and GEO databases to stratify 515 lung adenocarcinoma patients into two distinct subtypes based on mitochondrial-related genes. We systematically evaluated survival outcomes and biological pathway activities between subtypes, characterized their immune infiltration profiles, and developed a prognostic model using subtype-specific differentially expressed genes. Drug sensitivity disparities were further assessed. Single-cell RNA sequencing data were analyzed using an XGBoost classifier to delineate cell-type heterogeneity across subtypes at single-cell resolution. In LUAD, we identified two distinct subtypes. One subtype exhibited active mitochondrial metabolism, which was associated with poor prognosis and higher tumor purity. Moreover, this subtype showed greater sensitivity to Osimertinib. Further single-cell analysis revealed that this subtype was characterized by substantial macrophage infiltration, potentially promoting tumor progression through the NF-κB signaling pathway. Overall, our study identified novel LUAD subtypes and provided new insights into the clinical treatment of LUAD.