# Machine learning assisted immune profiling of COPD identifies a unique emphysema subtype independent of GOLD stage

**Authors:** Natalie Bordag, Katharina Jandl, Ayu Hutami Syarif, Jürgen Gindlhuber, Diana Schnoegl, Ayse Ceren Mutgan, Vasile Foris, Konrad Hoetzenecker, Panja Maria Boehm, Robab Breyer-Kohansal, Katarina Zeder, Gregor Gorkiewicz, Francesca Polverino, Slaven Crnkovic, Grazyna Kwapiszewska, Leigh Matthew Marsh

PMC · DOI: 10.1016/j.isci.2025.112966 · 2025-06-19

## TL;DR

This study uses machine learning to identify a new COPD subtype linked to severe emphysema, independent of traditional disease staging.

## Contribution

A novel COPD immune subtype (EIS) is identified using machine learning and multiomics data, independent of GOLD stage.

## Key findings

- COPD immune profiles show a shift toward adaptive immune cells.
- A new COPD subtype (EIS) is associated with severe emphysema and distinct immune markers.
- EIS features increased antigen-presenting cells, mast cells, and CD8+ T cells.

## Abstract

Chronic obstructive pulmonary disease (COPD) is a severe, progressive, and heterogeneous disease with a poor outcome. Inflammation plays a central role in disease pathogenesis; however, the interplay between immune changes and disease heterogeneity has been difficult to unravel. We performed a multilevel immunoinflammatory characterization of patients with COPD using flow cytometry, cytokine profiling, single-cell, or spatial transcriptomics in combination with machine learning algorithms. Our cross-cohort analysis demonstrated shared skewing of immune profiles in COPD lungs toward adaptive immune cells. We furthermore identified a subgroup of patients with COPD with a distinct immune profile, characterized by increased antigen-presenting cells, mast cells, and CD8+ cells, and circulating IL-1β, IFN-β, and GM-CSF, that were associated with increased emphysema severity and decreased gas exchange parameters independent of their GOLD-stage. Our findings suggest that unbiased immune profiling can refine disease classification and reveal inflammation-driven disease subtypes with potential relevance for prognosis and treatment strategies.

•Data driven multiomics reveals distinct immune profiles in COPD lungs•COPD immune signatures are dominated by adaptive immune cells and cytokines•A COPD severe immune-subtype is associated with emphysema severity, termed EIS•COPD EIS is marked by pronounced mast cells, DCs, and CD8+ T cells

Data driven multiomics reveals distinct immune profiles in COPD lungs

COPD immune signatures are dominated by adaptive immune cells and cytokines

A COPD severe immune-subtype is associated with emphysema severity, termed EIS

COPD EIS is marked by pronounced mast cells, DCs, and CD8+ T cells

Respiratory medicine; Machine learning

## Linked entities

- **Proteins:** IL1B (interleukin 1 beta), IFNB1 (interferon beta 1), CSF2 (colony stimulating factor 2)
- **Diseases:** COPD (MONDO:0005002), emphysema (MONDO:0004849)

## Full-text entities

- **Genes:** CSF2 (colony stimulating factor 2) [NCBI Gene 1437] {aka CSF, GMCSF}, CD8A (CD8 subunit alpha) [NCBI Gene 925] {aka CD8, CD8alpha, IMD116, Leu2, p32}, IFNB1 (interferon beta 1) [NCBI Gene 3456] {aka IFB, IFF, IFN-beta, IFNB}, IL1B (interleukin 1 beta) [NCBI Gene 3553] {aka IL-1, IL1-BETA, IL1F2, IL1beta}
- **Diseases:** Inflammation (MESH:D007249), emphysema (MESH:D004646), COPD (MESH:D029424)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Figures

7 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12274847/full.md

---
Source: https://tomesphere.com/paper/PMC12274847