Lung adenocarcinoma (LUAD), the most common subtype of non-small cell lung cancer (NSCLC), often presents with mild or absent symptoms in its early stage, leading to delayed diagnosis and poor outcomes. This study aimed to elucidate the molecular mechanisms underlying early-stage LUAD and to identify effective biomarkers for early detection and therapeutic intervention. A machine-learning framework integrating random forest (RF) and least absolute shrinkage and selection operator (LASSO) was used to identify candidate diagnostic biomarkers. The robustness of the identified marker was evaluated in an independent external dataset. Single-cell RNA sequencing was employed to localize gene expression within the tumor microenvironment. Additional analyses—including cell–cell communication inference, copy number variation (CNV) profiling, and pseudotime trajectory reconstruction—were performed to investigate the functional role of the identified biomarker in LUAD progression. IGSF9 emerged as a promising diagnostic biomarker and potential therapeutic target for early-stage LUAD, with its diagnostic value validated in an external dataset. Single-cell RNA sequencing located IGSF9 expression primarily in alveolar cells within the tumor microenvironment. Cell–cell communication analyses suggested that IGSF9 contributes to immune evasion and promotes tumor cell migration and invasion. CNV analysis revealed substantial genomic instability in the alveolar compartment. Pseudotime trajectory inference indicated that IGSF9 may drive the differentiation of a stem-like AT2 subpopulation toward a more malignant state. This study identifies IGSF9 as a robust diagnostic biomarker for early-stage LUAD and elucidates its multifaceted role in malignant transformation. These findings provide valuable insights for early diagnosis and precision oncology in LUAD.
Li et al. (Sun,) studied this question.