ABSTRACT Programmed cell death (PCD) pathways hold significant influence in the etiology and progression of a variety of cancer forms, particularly offering promising prognostic markers and clues to drug sensitivity for lung adenocarcinoma (LUAD) patients. We employed single‐cell analysis to delve into the functional role of PCD within the tumour microenvironment (TME) of LUAD. Employing a machine learning framework, a PCD‐related signature (PCDS) was constructed utilising a comprehensive data set. The PCDS exhibited superior prognostic performance compared with the 140 previously established prognostic models for LUAD. Subsequently, patients were stratified into high‐risk and low‐risk groups based on their risk scores derived from the PCDS, with the high‐risk group exhibiting significantly lower overall survival (OS) rates than the low‐risk group. Furthermore, the risk subgroups were compared for differences in pathway enrichment, genomic alterations, tumour immune microenvironment (TIME), immunotherapy and drug sensitivity. The low‐risk group displayed a more inflamed TIME, potentially leading to a more favourable response to immunotherapy. For the high‐risk group, potential effective small molecule drugs were identified, and the drug sensitivity were evaluated. Immunohistochemistry and quantitative real‐time polymerase chain reaction assays (qRT‐PCR) confirmed notable upregulation of the expression levels of PCD‐associated genes MKI67, TYMS and LYPD3 in LUAD tissues. In vitro experimental findings demonstrated a marked decrease in the proliferative and migratory capacities of LUAD cells upon knockdown of MKI67. Conclusively, we successfully constructed the PCDS, providing important assistance for prognosis prediction and treatment optimisation of LUAD patients.

Unveiling Varied Cell Death Patterns in Lung Adenocarcinoma Prognosis and Immunotherapy Based on Single‐Cell Analysis and Machine Learning