We’ve understood for years that chronic obstructive pulmonary disease (COPD) is not just a respiratory disease, but a disease of systemic inflammation, with COPD patients displaying higher levels than healthy controls of inflammatory biomarkers, including C-reactive protein (CRP); tumor necrosis factor alpha (TNF-a); interleukin-8 (IL-8); fibrinogen, albumin, and nitrates/nitrites.1 This inflammation contributes to the significant comorbidities seen in patients with COPD, particularly cardiovascular disease and depression.
A new study in the American Journal of Respiratory and Critical Care Medicine now shows that adding a panel of inflammatory biomarkers to the clinical parameters typically used to predict mortality in patients with COPD can substantially improve the predictive value of those clinical markers.1
The complexity and heterogeneity of COPD limits the ability of traditional clinical parameters like forced expiratory volume (FEV), age, and hospitalizations to predict outcomes. More recent multi-dimensional tools like the BODE index (body mass index, FEV1, dyspnea, and 6-minute walk distance) may improve prediction of outcomes.
However, symptoms and surrogate markers are not direct indicators of the underlying pathophysiology, as biomarkers can be. In the diagnosis and management of other chronic inflammatory conditions, particularly cardiovascular disease and rheumatoid arthritis, biomarkers of physiology play an important role. Therefore they could also be useful for diagnosis and management in COPD.
Several biomarkers are associated with an increased mortality risk in patients with respiratory disease (not necessarily COPD), including CRP, chemokine ligand 18 (CCL-18/PARC), interleukin-6 (IL-6), and surfactant protein-D (S-PD). However, none has been definitively linked to COPD mortality.
In this study, researchers used serum and plasma samples collected from participants in the ECLIPSE (Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints) study. The study enrolled 2,164 patients with COPD (GOLD stage II to IV) who were evaluated at baseline, at 3 and 6 months after enrollment, and then every 6 months for a total of 3 years. The primary outcome, all-cause mortality, was compared with results for 337 smoking controls and 245 nonsmoking controls. Investigators obtained complete biomarker data (CCL-18/PARC, S-PD, IL-8, Clara cell secretory protein 16 [CC-16], and TNFα in serum samples, and fibrinogen and CRP in plasma samples) for 1,843 patients.
Patients whose levels of IL-6, CCL-18/PARC, fibrinogen, CRP, and SP-D were higher than the median in control subjects were less likely to be alive at 3 years than those with lower levels. After adjusting for age, previous hospitalizations, and the BODE index, abnormal levels of white blood cells, neutrophils, and IL-8 were also associated with an increased mortality risk.
With the exception of IL-6, adding any of these markers individually to the clinical markers of age, previous hospitalization, and BODE index score, however, did little to improve the clinical predictability of mortality. Yet adding a panel of IL-6, IL-8, white blood cell count, fibrinogen, CRP, and SP-D to these clinical markers significantly improved the predictive value of the clinical biomarkers. Only IL-6 improved the predictability of the clinical biomarkers when added alone.
While this study identified biomarkers associated with mortality; others have tried to identify inflammatory biomarkers that might be used to classify the severity of COPD. These include serum monocyte chemoattractant protein-1 (MCP-1), higher levels of which are found in COPD patients versus healthy smokers, and CRP, shown to correlate with airway obstruction.1,2
Although this panel of COPD biomarkers is not ready for clinical use at present, some coauthors of the study are from Glaxo SmithKline, which is working toward a marketable test.
lthough it's still too early to expect that clinicians will integrate the biomarker panel, said lead author Bartoleme R. Celli, MD, the study does highlight the fact that these markers can provide additional information beyond that available with clinical markers alone. "Perhaps the best use right now is to see if therapies can induce a decrease in the markers and if these precede and predict a clinical outcome response," he said. "Only time will tell whether these biomarkers can arrive to the bedside."