Transcriptomic analysis of transformed small-cell lung cancer from EGFR-mutated lung adenocarcinoma reveals distinct subgroups and precision therapy opportunities

Abstract

BACKGROUND: Small-cell lung cancer (SCLC) transformation is one of the major mechanisms of resistance to Epidermal Growth Factor Receptor tyrosine kinase inhibitors (EGFR-TKIs). Chemotherapy is typically the recommended treatment for transformed SCLC (T-SCLC), similar to primary SCLC. However, the benefits of chemotherapy alone are minimal. Prior research highlights differences between the biological traits of T-SCLC and primary SCLC or EGFR-mutated lung adenocarcinoma (LUAD). This study aims to elucidate the molecular characteristics of T-SCLC and identify potential treatment modalities. METHODS: We retrospectively collected tissue samples from LUAD, T-SCLC post-EGFR-TKI resistance, and primary SCLC. Genomics, transcriptomics, and proteomics were performed to clarify the differences between T-SCLC, LUAD, and primary SCLC. Hierarchical clustering analysis was then used to categorize the molecular subtype of T-SCLC, followed by a survival analysis based on these subtypes. RESULTS: A study involving 61 patients investigated differences between LUAD, SCLC, and primary SCLC. RNA sequencing revealed distinct gene expression variations, particularly up-regulation in PPM1E, INSM1, and KCNC1 genes in T-SCLC. Pathway analysis linked T-SCLC to the cell cycle and neural differentiation. By conducting Hierarchical clustering analysis on RNA-seq data, the entire population can be categorized into two distinct groups. While certain T-SCLC showed similarities and differences compared to SCLC, with subtypes: LUAD-like and Non-LUAD-like. Notably, the LUAD-like subtype had significantly higher NKX2-1 expression (mean 371.8 vs. 41.8, P < 0.0001). T-SCLC treatment approaches were categorized into matched and unmatched groups, delineated by the alignment of specific therapies with corresponding pathologies. The matched group (13 cases) exhibited a significantly prolonged median progression-free survival compared to the unmatched group (10 cases) (5.4 months vs. 3.6 months, P = 0.02). CONCLUSIONS: T-SCLC exhibits marked molecular distinctiveness, setting it apart not only from LUAD but also from classical SCLC. This distinction extends to its classification into two discernible molecular subtypes: LUAD-like and Non-LUAD-like. Customizing therapeutic protocols to align with these specific subtypes have the potential to identify the most appropriate treatment for T-SCLC.