Chung-Chih Liao, Ke-Ru Liao, Jung-Miao Li
Chronic lung diseases, including asthma, chronic obstructive pulmonary disease, bronchiectasis, and idiopathic pulmonary fibrosis, are clinically distinct but share epithelial injury, host-defense, inflammatory, and remodeling processes. We integrated European-ancestry genome-wide association study (GWAS) summary statistics for these four diseases using genomic structural equation modeling to construct a multivariate chronic lung disease (mvCLD) factor. Variant-level association testing was combined with genomic control assessment, locus annotation, GWAS-by-subtraction, fine-mapping, transcriptomic prioritization, pathway enrichment, single-cell spatial mapping, and heritability partitioning. For discovery, across 6,255,777 autosomal variants, mvCLD identified 2067 genome-wide significant variants, 30 loci, and 53 lead variants. Five lead variants were genome-wide significant for mvCLD but not for any component disease and showed high-confidence fine-mapping support. For gene prioritization, transcriptomic and gene-level analyses prioritized 21 candidate genes, including ORMDL3, GSDMB, IL18RAP, IL18R1, IL1R1, SMAD3, and CLEC16A. In exploratory functional analyses, enrichment analyses converged on interleukin, cytokine-receptor, JAK-STAT, interleukin-4/interleukin-13, thymic stromal lymphopoietin, asthma, and lung fibrosis pathways. Exploratory single-cell spatial mapping, based on a mouse embryonic atlas, showed the strongest overall enrichment in the lung annotation, although cross-dataset cell-type and tissue analyses did not reach significance after false discovery rate correction; heritability partitioning implicated conserved and active regulatory elements. These findings support a shared epithelial-immune and cytokine-regulatory genetic architecture across major chronic lung diseases and nominate biologically coherent candidate genes and pathways for future functional and translational studies.