Demystifying programmed cell death in lung adenocarcinoma: combined prognostic model construction.

Background: Lung adenocarcinoma (LUAD) presents significant treatment challenges due to late-stage diagnosis and limited therapeutic options. The prognostic value of programmed cell death (PCD) gene signatures in LUAD remains inadequately investigated. This study aims to develop and validate prognostic models based on PCD-related genes, particularly necroptosis and pyroptosis, to improve risk stratification and identify potential therapeutic targets for LUAD patients. Methods: Multiple prognostic models and a combined prognostic model (CPM) for LUAD were developed using least absolute shrinkage and selection operator (LASSO) regression, based on 12 PCD-related prognostic genes. Model performance was assessed using time-dependent receiver operating characteristic (ROC) curves. Multivariate Cox regression identified key genes and led to the construction of a new CPM. Validation was performed using The Cancer Genome Atlas and Gene Expression Omnibus datasets, as well as machine learning techniques. Immune infiltration and single-cell RNA sequencing analyses were employed to correlate gene expression with immune responses. Results: The LASSO models for necroptosis [area under the curve (AUC) =0.717] and pyroptosis (AUC =0.713) demonstrated strong predictive performance, while the CPM model (AUC =0.735) successfully identified high-risk patients with LUAD experiencing poorer overall survival [hazard ratio (HR) =3.05, P<0.001]. Multivariate Cox regression identified 13 key genes, including those involved in necroptosis and pyroptosis, which were used to develop the new CPM. Validation across eight Gene Expression Omnibus (GEO) datasets confirmed the robustness of the model (average HR =2.52, P<0.001). Machine learning analyses identified Conclusions: This study highlights the prognostic relevance of PCD in LUAD, particularly the combined roles of necroptosis and pyroptosis.