Comprehensive multi-omics analysis identifies BTG2 as a potential regulator of radiotherapy response in lung adenocarcinoma.

Lung adenocarcinoma (LUAD) is one of the most common malignancies worldwide. However, patients exhibit heterogeneous responses to radiotherapy, and a subset of individuals develops resistance, which severely compromises clinical outcomes. The underlying molecular basis of this variability remains unclear. This study aimed to investigate RSG-associated molecular features in radiotherapy-treated LUAD and to identify potential prognostic biomarkers. We integrated multi-omics data and applied multiple machine learning approaches to comprehensively analyze LUAD samples. In addition, single-cell data were incorporated to further characterize the tumor microenvironment. The results revealed two distinct RSG-associated prognostic subtypes with different survival outcomes and biological characteristics. The poor-prognosis subtype was associated with significant alterations in the immune microenvironment and related biological pathways. Furthermore, we developed a predictive model with stable performance for evaluating patient survival outcomes. Among the identified candidates, BTG2 emerged as a key gene in the RSG-based prognostic model. Functional experiments further showed that BTG2 overexpression inhibited tumor cell proliferation. Bidirectional clonogenic radiation dose-response assays showed lower post-irradiation survival after BTG2 overexpression and higher survival after BTG2 depletion, supporting BTG2 as a cellular modulator of radiation response. In summary, this study identified two RSG-associated prognostic subtypes among radiotherapy-treated patients with LUAD and highlighted BTG2 as a potential biomarker associated with prognosis and radiation-response-related biological features. These findings provide a framework for further prospective validation and may contribute to future risk stratification and individualized treatment strategies.