Research Summary

Immune Profiling Identifies Emphysema-Linked COPD Subtype Independent of GOLD Stage

Anthony Calabro, MA

Key Highlights

  • Immune profiling revealed adaptive lymphocyte dominance and reduced innate immune cells in COPD lungs.
  • A severe emphysema inflammatory subgroup (EIS) was defined by elevated mast cells, dendritic cells, CD8⁺ T cells, and cytokines IL-1β, IFN-β, and GM-CSF.
  • The EIS pattern correlated with emphysema severity and impaired oxygen diffusion, independent of Global Initiative for Obstructive Lung Disease stage.
  • Results were consistent across flow cytometry, cytokine profiling, single-cell and spatial transcriptomics, and machine-learning analysis.

A study published in iScience applied advanced immune profiling and computational modeling to better understand the biological heterogeneity of chronic obstructive pulmonary disease (COPD). The researchers identified a distinct inflammatory pattern that differentiates a subgroup of patients with emphysema-predominant disease. This immune signature was independent of spirometry-based Global Initiative for Obstructive Lung Disease (GOLD) staging, suggesting that current classification methods may overlook key biologic differences.

The study analyzed explanted lungs from 20 patients with end-stage COPD and 23 healthy donor lungs using 23-marker flow cytometry. Lung and plasma cytokine levels were also assessed, and results were integrated with single-cell RNA sequencing and spatial transcriptomic datasets spanning GOLD stages I–IV.

Machine learning algorithms—including random forest and unsupervised clustering—were used to identify cell populations and cytokine profiles that best predicted disease subtype and severity.

Study Findings

COPD lungs showed a dominant adaptive immune profile with increased CD4⁺ and CD8⁺ T cells and B cells, and a relative decrease in macrophages, monocytes, and neutrophils. The random forest model accurately distinguished COPD from control lungs with 93% accuracy.

Cytokine analysis revealed elevated inflammatory mediators, including CCL5, CXCL9, and CXCL10 in the lungs and CCL5 and CXCL5 in plasma. Correlation analyses linked these cytokines with adaptive immune activation and reduced lung function.

Within the COPD group, a machine learning–defined cluster—termed the “emphysema inflammatory subgroup” (EIS)—was characterized by heightened mast cells, dendritic cells, and CD8⁺ T cells, along with increased IL-1β, IFN-β, and GM-CSF. Clinically, patients in this subgroup demonstrated more severe emphysema, lower oxygen levels, and decreased gas exchange capacity. Spatial transcriptomic data confirmed that these immune changes were detectable even in earlier GOLD stages.

Clinical Implications

The authors noted that unbiased immune profiling may improve how COPD is classified, revealing biologically distinct subtypes that could have prognostic and therapeutic relevance. The identification of an EIS, independent of GOLD stage, highlights the potential for integrating immune signatures into future disease stratification frameworks. These findings underscore the need for additional research to determine how such immune-defined subtypes might inform prognosis or treatment strategies.

Expert Commentary

“The characterization of the local immunoprofile in COPD revealed a severe subtype, which associates with emphysema severity as well as decreased gas exchange parameters,” the researchers concluded. “These insights help elucidate the complex role of immune cells in COPD pathogenesis.”

Reference
Bordag N, Jandl K, Syarif AH, et al. Machine learning assisted immune profiling of COPD identifies a unique emphysema subtype independent of GOLD stage. iScience. 2025;28(7).