Abstract
Chronic obstructive pulmonary disease (COPD), the third leading cause of death worldwide, lacks effective disease-modifying therapies, partly because of complex gene-environment interactions and extensive missing heritability. Here, we applied a multiomics Mendelian randomization (MR) framework - integrating proteome- and transcriptome-wide association analyses (pQTLs/eQTLs) with genome-wide association summary statistics, sensitivity analyses, and colocalization - to assign evidence levels to genes and prioritize those with higher causal likelihoods across diverse cohorts. We identified serpin family G member 1 (SERPING1) as a robust causal candidate, with consistent pQTL associations with COPD (β = -0.038 to -0.006) and with lung function measures, including FEV₁ (β = 0.008 to 0.015) and FEV₁/FVC% (β = 0.014 to 0.026). Longitudinal analyses in the UK Biobank (n = 46,369) and ECOPD cohort (n = 576) revealed that higher circulating SERPING1 protein levels were causally linked to slower FEV₁ decline during early follow-up (UKB: adjusted difference = -22.1 mL/year per standardized unit; ECOPD: -0.73 mL/year per ng/mL), accompanied by marked expression differences between European (higher) and Asian (lower) smokers and COPD patients. In a murine model exposed to cigarette smoke, AAV-mediated SERPING1 overexpression improved lung function, reduced alveolar destruction, and upregulated the expression of fibroblast elastic fiber-related genes. Collectively, these findings identify SERPING1 as a complement pathway regulator that may function both as a short-term biomarker of lung function decline and as a population specific, disease-modifying therapeutic target for COPD.</p>