Chronic obstructive pulmonary disease (COPD) : from biomarkers to clinical phenotypes

Research output: ThesisDoctoral Thesis


Chronic obstructive pulmonary disease (COPD) is a common, preventable and treatable chronic airway disease (CAD) and is the fourth leading cause of illness and death worldwide. CAD represents a group of diseases with different clinical phenotypes, airway inflammation, obstruction and destruction of the lung parenchyma. COPD is a heterogeneous disease, which is characterized by persistent respiratory symptoms and airflow limitation and its prevalence varies considerably across populations. Asthma-COPD overlap syndrome (ACOS) had been reported in 2009 and recognised by Global Strategy for Asthma Management and Prevention (GINA) and Global Initiative for Chronic Obstructive Lung Disease (GOLD) in 2014. ACOS is identified by the features that it shares with both asthma and COPD. Inhaling tobacco smoke is the main risk for CAD, thus smoking cessation is not only the key step for prevention and maintenance therapy for COPD (GOLD 2017), but also it is encouraged in the patient with asthma and their families (GINA 2017). Smoking cessation, pulmonary rehabilitation, pharmacologic therapy and the other therapies for COPD may improve symptoms, health status and exercise ability and reduce frequency of exacerbations (GOLD 2017). At present, COPD is generally diagnosed by lung function measurement using spirometry, but not all individuals presenting with airflow limitation follow the paradigm. On the individual level spirometry may poorly correlate with symptoms, risk of exacerbation, prognosis and response to treatment in COPD. Due to disease heterogeneity, limitations of spirometry and lack of predictive molecular markers, the mechanisms underlying different COPD phenotypes are still unclear. Therefore, identification of the similarities and differences of COPD clinical phenotypes is important, as these phenotypes require different therapeutic approaches and have distinct clinical outcomes. We hypothesised that the potential biomarkers may play an important role and even predict the development of COPD. Accordingly, patients with asthma, COPD and ACOS would exhibit different biomarker profiles. Furthermore, distinguishing COPD phenotypes via distinct molecular and cellular pathways may facilitate the development of individualised diagnosis and precision medicine. The first goal of this study was to identify COPD-specific proteomic changes involved in disease onset and severity. The second goal was to identify novel biomarkers for the clinical COPD phenotypes in smokers and in patients with COPD, asthma and ACOS in cross-sectional and longitudinal studies. Those studies included the 6-year longitudinal cohort of adult smokers for early diagnosis of COPD studies, and the Finnish discovery cohort and Japanese validation cohort for ACOS studies. The third goal was to investigate whether the levels of these markers are associated with variables such as pack-years, lung function and sputum cell profiles. Finally, we aimed to evaluate the association of the biomarker levels with a longitudinal decline of lung function. We selected biomarkers identified in our non-hypothesis-based proteomic approaches for early COPD diagnosis, such as transglutaminase 2 (TGM2), bactericidal/permeability-increasing (BPI) fold-containing protein B1 (BPIFB1) and soluble receptor for advanced glycation end-products (sRAGE). Based on our previous results and non-hypothesis-based proteomic approaches, we developed two panels of potential biomarkers for ACOS: COPD-related biomarkers and asthma-related biomarkers. The four COPD-related biomarkers are neutrophil gelatinase-associated lipocalin (NGAL), myeloperoxidase (MPO), surfactant protein A (SP-A) and sRAGE. The three asthma-related biomarkers are chitinase-like protein (YKL-40), interleukin-6 (IL-6) and interleukin IL-13 (IL-13). For the early COPD diagnosis studies, induced sputum and plasma samples were categorised into 3 groups: non-smokers, smokers without COPD and smokers with COPD (stable stage I and stage II-III). Lung tissue samples were obtained from non-smokers, smokers without COPD, smokers with COPD and patients with idiopathic pulmonary fibrosis (IPF). For the ACOS studies, induced sputum and plasma samples were categorised into 5 groups: non-smokers, healthy smokers, patients with asthma, COPD and ACOS. Sputum cell counts were evaluated from sputum cytospins. TGM2 and BPIFB1 were assessed by Western blot analysis and by cysteine-specific two-dimensional difference gel electrophoresis (2D-DIGE) coupled with mass spectrometry (MS). BPIFB1 was further detected by immunohistochemistry and functional enzyme-linked immunosorbent assay (ELISA). TGM2, sRAGE, NGAL, MPO, SP-A, YKL-40, IL-6 and IL-13 levels were measured by commercial ELISA or magnetic human high sensitivity luminex assay (luminex assay). Our results showed that TGM2 levels in sputum and plasma were elevated in mild-moderate COPD and associated with lung function. Sputum BPIFB1 levels were elevated in smokers with COPD, whereas plasma sRAGE levels were decreased in smokers without COPD and smokers with COPD. Sputum BPIFB1 and plasma sRAGE levels were significantly associated with reduced airflow limitation over time in smokers with COPD. In the Finnish discovery cohort of the ACOS study, sputum MPO and plasma SP-A levels were significantly elevated in ACOS, whereas plasma sRAGE levels were decreased in ACOS. Sputum IL-13 levels were increased in ACOS when compared with controls. Sputum YKL-40 and IL-6 differentiated ACOS from COPD and asthma based on receiver operating characteristic (ROC) curve analysis. However, only sputum NGAL results could be repeated from the Finnish discovery cohort to Japanese validation cohort. Importantly, NGAL differentiated ACOS from COPD and independently correlated with forced expiratory volume in one second % (FEV1%-predict) in both cohorts. In conclusion, our findings support the hypothesis that monitoring of these biomarkers may be useful for differential diagnosis between the clinical phenotypes of COPD. TGM2 may be a potential diagnostic and therapeutic marker in COPD, which may relate to COPD onset and severity. BPIFB1 and sRAGE play vital roles in the pathogenesis of smoking-related COPD. High concentrations of sRAGE might be involved in the protective role of the airway. Patients with COPD and ACOS exhibit different NGAL profiles in sputum. These findings suggest that these potential biomarkers could be a novel diagnostic and therapeutic target for COPD. On the other hand, sputum biomarkers reflect both airway inflammation and tissue remodelling, and monitoring these biomarkers may be useful for differential diagnosis between asthma, COPD and ACOS.
Original languageEnglish
  • Pulkkinen, Ville, Supervisor
  • Mazur, Witold, Supervisor
Award date9 Mar 2018
Place of PublicationHelsinki
Print ISBNs978-951-51-4038-8
Electronic ISBNs978-951-51-4039-5
Publication statusPublished - 2018
MoE publication typeG5 Doctoral dissertation (article)

Fields of Science

  • 3121 General medicine, internal medicine and other clinical medicine

Cite this