The role of immunogenic cell death in the prognosis and development of treatment strategies for non-small cell lung cancer: a multiomics and machine learning approach for predictive and personalized treatment.

Background: Non-small cell lung cancer (NSCLC) is the predominant histological subtype of lung cancer, whose diverse genomic landscape complicates prognosis and outcome prediction. In this context, immunogenic cell death (ICD), a distinct mechanism of cell death, may exert important antitumor effects. However, the specific role of ICD in NSCLC has not been clarified, and there is no suitable method for using ICD to achieve the treatment and prognosis assessment of NSCLC. The purpose of the current research is to develop a new approach to predict the survival prognosis and response to chemotherapy and targeted therapy in patients with NSCLC. Methods: We used 101 combinations of 10 machine learning algorithms to construct an ICD-related signature (ICDRS). The predictive potential of this specific ICDRS for immune cell infiltration and therapeutic response was evaluated. We also examined the molecular mechanisms underlying the various responses of different ICDRS subpopulations, characterized the mutational landscape and tumor mutational burden (TMB), and assessed the applicability of the ICDRS in single-cell transcriptomic datasets. Gene expression patterns were subsequently validated with quantitative real-time polymerase chain reaction (qRT-PCR) and immunohistochemistry (IHC). Results: We screened out five key ICDRS genes ( Conclusions: The ICDRS exhibited excellent predictive performance and broad applicability, suggesting it as a powerful tool for prognosis and therapy response. The current study may contribute to more adequate patient selection in the context of tailored therapies.