Back

Extracellular vesicle surface markers inform on COPD severity and mortality in COSYCONET

Martin, R.; Laakmann, K.; Pott, H.; Bertrams, W.; Hinz, L.; Burhorst, I.; Bals, R.; Herr, C.; Jung, A. L.; Alter, P.; Vogelmeier, C. F.; Rohde, G.; Schmeck, B.; Heider, D.

2026-07-02 respiratory medicine
10.64898/2026.06.30.26356923 medRxiv
Show abstract

Background: Chronic obstructive pulmonary disease (COPD) is a leading cause of global morbidity and mortality, and its heterogeneity demands better biomarkers of severity and progression risk. Extracellular vesicles (EVs) are promising blood-based biomarkers, but have not been examined for COPD severity and outcomes in a large multicentre cohort. Methods: We analysed 600 COSYCONET participants (up to 54 months of follow-up). EV surface markers were profiled with the MACSPlex EV Kit IO. Cross-sectional associations with severity (GOLD, FEV1) were primary (ordinal and linear regression); longitudinal trajectories and all-cause mortality were prespecified exploratory endpoints. Results: Six EV markers showed robust associations with cross-sectional severity: CD29, CD49e and CD31 increased with severity (a cell-adhesion/matrix-remodelling signal), whereas CD81 and CD8 decreased; HLA-ABC (increasing) was less specific. No marker was linked to FEV1 decline. After FDR correction, lower levels of three markers with higher 54-month mortality (all HR<1): CD25 (HR 0.77, 95% CI 0.65-0.90, q=0.018), CD56 (HR 0.75, 95% CI 0.63-0.89, q=0.018) and CD142 (HR 0.74, 95% CI 0.60-0.90, q=0.024). CD25 and CD142 also improved reclassification, CD56 did not; a CD25 + CD69 combination showed the largest incremental signal ({Delta}C 0.017, 95% CI 0.002-0.032, p=0.027). Conclusion: Circulating EV surface markers are associated with cross-sectional COPD severity. Exploratory analyses nominate CD25, CD142 and CD25 + CD69 as candidate prognostic markers requiring external validation, suggesting minimally invasive EV profiling could complement clinical assessment in COPD.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.