Plasma proteomic profiles of lung volume–based phenotypes in tobacco-exposed individuals without spirometric chronic obstructive pulmonary disease

BACKGROUND: Among individuals with a history of smoking but preserved spirometry (tobacco exposed with preserved spirometry, or TEPS), lung volume-based stratification identifies 2 phenotypes at increased risk for chronic obstructive pulmonary disease (pre-COPD): those with a relatively elevated total lung capacity ([TLC]high) and those with a relatively elevated functional residual capacity-to-TLC ratio ([FRC/TLC]high). These subgroups exhibit distinct respiratory symptoms, radiographic abnormalities, and clinical trajectories. OBJECTIVE: This study aimed to determine whether these lung volume-based pre-COPD phenotypes have distinct biological features reflected in their circulating proteomes. METHODS: We analyzed peripheral blood proteomic profiles (SomaScan v4.0; 4979 aptamers) from 1959 TEPS participants at the 5-year follow-up visit (visit 2) of the Genetic Epidemiology of COPD study cohort. Participants with [TLC]high and [FRC/TLC]high (based on computerized tomography scan-derived supine lung volumes) were compared with a low-COPD-risk reference group (without high TLC or high FRC/TLC). Analyses included covariate-adjusted regression, machine learning, and pathway enrichment modeling, with adjustment for age, sex, height, weight, smoking status and burden, leukocyte and platelet counts, forced expiratory volume in 1 second (percent predicted), and study site (random effect). RESULTS: Using visit 2 data and visit 3 (10-year) follow-up outcomes, we confirmed the reproducibility and prognostic validity of the lung volume-based phenotypes in 1232 participants with longitudinal data. Over a mean (SD) of 5.3 (1.1) years, spirometric COPD developed in 17% (133 out of 761) of pre-COPD TEPS vs 8% (37 out of 471) of low-risk TEPS (adjusted odds ratio [aOR], 2.51; 95% confidence interval [CI], 1.69-3.75; P < .001). Among pre-COPD subgroups, [FRC/TLC]high TEPS showed greater progression to a Global Initiative for Chronic Obstructive Lung Disease stage 2 or higher (aOR, 2.90; 95% CI, 1.62-5.18; P < .001) and preserved ratio and impaired spirometry (aOR, 3.29; 95% CI, 1.41-7.69; P = .005). At baseline (n = 1959), plasma proteomic analysis identified 165 upregulated and 145 downregulated proteins in [TLC]high TEPS compared with low-COPD-risk TEPS, whereas only 22 proteins were differentially expressed in [FRC/TLC]high TEPS vs low-risk group. Comparison between the 2 pre-COPD phenotypes identified 269 differentially expressed proteins (116 upregulated and 153 downregulated in [FRC/TLC]high vs [TLC]high), including previously described COPD-related mediators (eg, soluble receptor for advanced glycation end products, insulin-like growth factor-binding protein) and novel candidates (eg, zymogen granule membrane protein 16). Pathway analysis highlighted immune signaling, cellular trafficking, and apoptotic pathways relevant to COPD pathogenesis. CONCLUSIONS: Lung volume-based stratification in TEPS identifies biologically distinct subgroups with differing plasma proteomic signature and COPD risk, underscoring the heterogeneity of early disease and revealing potential circulating biomarkers of pre-COPD states. REGISTRATION: COPDGene study is registered with ClinicalTrials.gov: ID NCT00608764.Keywordspre-COPD, lung volumes, proteomics, biomarkers, chronic obstructive pulmonary disease.