Abstract
Background: Idiopathic pulmonary fibrosis (IPF) leads to excessive fibrous tissue in the lungs, increasing the risk of lung cancer (LC) due to heightened fibroblast activity. Advances in nucleotide point mutation studies offer insights into fibrosis-to-cancer transitions. Methods: A two-sample Mendelian randomization (TSMR) approach was used to explore the causal relationship between IPF and LC. A weighted gene co-expression network analysis (WGCNA) identified shared gene modules related to immunogenic cell death (ICD) from transcriptomic datasets. Machine learning selected key genes, and a multi-layer perceptron (MLP) model was developed for IPF prediction and diagnosis. SMR and PheWAS were used to assess the expression of key genes concerning IPF risk. The impact of core genes on immune cells in the IPF microenvironment was explored, and in vivo experiments were conducted to examine the progression from IPF to LC. Results: The TSMR approach indicated a genetic predisposition for IPF progressing to LC. The predictive model, which includes eight ICD key genes, demonstrated a strong predictive capability (AUC = 0.839). The SMR analysis revealed that the elevated expression of MS4A4A was associated with an increased risk of IPF (OR = 1.275, 95% CI: 1.029-1.579; p = 0.026). The PheWAS did not identify any significant traits linked to MS4A4A expression. The rs9265808 locus in MS4A4A was identified as a susceptibility site for the progression of IPF to LC, with mutations potentially reprogramming lung neutrophils and increasing the LC risk. In vivo studies suggested MS4A4A as a promising therapeutic target. Conclusions: A causal link between IPF and LC was established, an effective prediction model was developed, and MS4A4A was highlighted as a therapeutic target to prevent IPF from progressing to LC.