Combination of Systemic Inflammatory Biomarkers in Assessment of Chronic Obstructive Pulmonary Disease: Diagnostic Performance and Identification of Networks and Clusters.
Iva HlapčićDaniela BelamarićMartina BosnarDomagoj KiferAndrea Vukić DugacLada RumoraPublished in: Diagnostics (Basel, Switzerland) (2020)
Interleukin (IL)-1α, IL-1β, IL-6, IL-8 and tumor necrosis factor (TNF)α contribute to inflammation in chronic obstructive pulmonary disease (COPD). We wanted to investigate their interrelations and association with disease severity, as well as to combine them with other inflammation-associated biomarkers and evaluate their predictive value and potential in identifying various patterns of systemic inflammation. One hundred and nine patients with stable COPD and 95 age- and sex-matched controls were enrolled in the study. Cytokines' concentrations were determined in plasma samples by antibody-based multiplex immunosorbent assay kits. Investigated cytokines were increased in COPD patients but were not associated with disease or symptoms severity. IL-1β, IL-6 and TNFα showed the best discriminative values regarding ongoing inflammation in COPD. Inflammatory patterns were observed in COPD patients when cytokines, C-reactive protein (CRP), fibrinogen (Fbg), extracellular adenosine triphosphate (eATP), extracellular heat shock protein 70 (eHsp70) and clinical data were included in cluster analysis. IL-1β, eATP and eHsp70 combined correctly classified 91% of cases. Therefore, due to the heterogeneity of COPD, its assessment could be improved by combination of biomarkers. Models including IL-1β, eATP and eHsp70 might identify COPD patients, while IL-1β, IL-6 and TNFα combined with CRP, Fbg, eATP and eHsp70 might be informative regarding various COPD clinical subgroups.
Keyphrases
- chronic obstructive pulmonary disease
- lung function
- end stage renal disease
- oxidative stress
- newly diagnosed
- ejection fraction
- rheumatoid arthritis
- chronic kidney disease
- heat shock protein
- peritoneal dialysis
- prognostic factors
- cystic fibrosis
- machine learning
- physical activity
- single cell
- mass spectrometry
- single molecule