Abstract Introduction High-resolution computed tomography (CT)-defined emphysema remains the gold standard for identifying emphysema in patients with Chronic Obstructive Pulmonary Disease (COPD). However, clinical heterogeneity suggests that some patients progress to severe emphysema more rapidly than others, and the biological mechanisms driving this variability remain poorly understood. The objective of this study was to characterize lung transcriptomic signatures associated with fast versus slow emphysema progression. Methods We developed LobTe (Lobe-based Transformer Encoder), a machine-learning model trained on baseline and 5-year follow-up CT images from 4,612 smokers with or without COPD in the COPDGene cohort. LobTe stratified patients by progression rate in an independent in-house cohort. To profile the pulmonary immune landscape, we performed GeoMx spatial transcriptomics analyses of FFPE lung sections from never-smoker (NSC) and ever-smoker (SC) controls, as well as COPD patients, deconvoluting bulk spatial data using single-cell signatures to quantify immune cell proportions in parenchyma. Results A total of 48 subjects (8 never-smoker controls, 13 ever-smoker controls, and 27 COPD patients; 54 lobes) were analyzed. The LobTe model effectively differentiated fast and slow emphysema progressors across both emphysematous and nonemphysematous individuals, revealing distinct molecular signatures associated with disease trajectory. In fast progressors, 1,078 genes were upregulated, prominently including metallothioneins (MT1X, MT1A, MT1M, MT1E, MT1G). Pathway analysis demonstrated strong enrichment for processes related to metal-ion metabolism (copper, zinc, cadmium, iron), ion detoxification, and metal homeostasis (all FDR 1.5, p 0.001), suggesting an adaptive yet potentially maladaptive response to heavy-metal accumulation and oxidative stress. In contrast, slow progressors exhibited upregulation of 188 genes enriched in pathways associated with redox scavenging, oxidative balance, and epithelial-to-mesenchymal transition (all FDR 1.5, p 0.001), indicating a more controlled antioxidant and tissue-repair response. For comparison, emphysematous patients showed 1,710 upregulated genes with enrichment in Systemic Lupus Erythematosus (fold enrichment 2.8, p = 2.1 × 10−4) and immune activation pathways, including immunoglobulin class switching (GO:0042571, p = 1.9 × 10−6), consistent with a strong autoimmune and inflammatory profile. Conclusions This pilot analysis identifies a distinct alveolar transcriptomic signature associated with fast emphysema progression, characterized by metallothionein upregulation and activation of heavy-metal detoxification pathways. These findings suggest that metal overload may contribute to accelerated emphysema, consistent with the growing recognition of pollution-related metals (PM2.5-PM10) as risk factors for disease progression and exacerbation. Further functional studies are warranted to elucidate mechanisms of cellular detoxification and failure in fast-progressing emphysema. This abstract is funded by: RO1 HL149744 and R01 HL171622.
Adrish et al. (Fri,) studied this question.