Biomarkers in COPD: Is Mortality the Holy Grail?

Keywords :

• COPD
• biomarkers
• mortality
• calprotectin
• exacerbation

Chronic obstructive pulmonary disease (COPD) is a complex disease with an increasing burden of morbidity and mortality worldwide (1). The diagnosis, assessment of severity, and therapy of COPD are guided primarily by the presence of symptoms and the evaluation of future risk, as expressed by the degree of airflow limitation and the previous history of exacerbations (2). It has been a long-standing belief that forced expiratory volume in 1 sec (FEV₁), the most widely accepted biomarker of COPD severity today, does not adequately address the complexity and heterogeneity of this disorder (3, 4). Several systemic (serum or plasma) biomarkers have been measured in patients with COPD in studies with different designs; however only a minority of those biomarkers have been properly validated and there is still little evidence to suggest that any of the existing biomarkers may be used in clinical practice in the near future (5, 6). An ideal biomarker for COPD should reflect the pathophysiological process of the disease, should be reproducible and only vary with events related to disease progression (e.g., exacerbations), should be easy to assess and be validated in large cohorts of patients and should predict disease progression and be sensitive to interventions that are known to be effective (7). In that direction, the need for biomarkers in COPD has been stressed from several authors, not only for the understanding of the diversity and natural history of COPD, but also for the development and evaluation of new drugs (5, 8, 9).

In the present issue of the Journal of Chronic Obstructive Pulmonary Disease, Holmgaard and colleagues have evaluated plasma levels of calprotectin, a protein biomarker of neutrophilic activity, as a predictor of mortality in a population of 460 COPD patients participating in a randomized controlled trial of azithromycin in patients with moderate to very severe COPD (10). Calprotectin is a heterodimer of S100A8 and S100A9, two of the three available calgranulins, that is highly expressed in the cytoplasm of neutrophils and may therefore serve as a surrogate marker of systemic neutrophilic inflammation. Calprotectin has been previously evaluated in studies of patients with various chronic inflammatory disorders, including inflammatory bowel disease (11) and rheumatoid arthritis (12), as well as in patients with respiratory diseases, including COPD, acute respiratory distress syndrome, cystic fibrosis and obstructive sleep apnea (13–15). Recent data support calprotectin as a candidate biomarker in cystic fibrosis, since its levels present a significant reduction after the management of an exacerbation with significant correlations with functional and physiological parameters (16).

The important merits of the study by Holmgaard et al. (this issue) are the impressive consistency in the follow-up data (over 99% in 10 years), the relatively long-term period of 5 years reported, and the high mortality rate (236 out of 460 patients or 51%) in this population of severe COPD patients that allows for proper survival analysis (10). These data are rather solid in the direction that higher levels of a single measurement of calprotectin in stable COPD patients are associated with increased all-cause mortality during 5 years of follow-up, using both the arbitrary cut-points of 100 and 200 ng/mL and the (more statistically sound) distribution by tertiles (10).

There are, however, several issues that need to be considered before suggesting the implementation of calprotectin in clinical practice based on the aforementioned findings. This study has evaluated a rather selected population of predominantly severe and very severe COPD patients (FEV₁ < 60% predicted, median value 0.90 L) that are mainly classified in the “high-risk” population according to the recent GOLD guidelines, having experienced at least one hospitalization in the previous 2 years. There is also no evidence on the reproducibility of calprotectin measurements in COPD, on the changes of this biomarker during the course of an exacerbation or on the effects of treatment. Finally, the authors do not provide any data on the co-morbidities of their COPD patients.

Besides its possible role as a surrogate biomarker of neutrophilic inflammation, calprotectin has recently gained a widespread interest in cardiovascular biology and disease (17). The proinflammatory role of S100A8/A9 has been associated with adipose tissue inflammation and the development of atherosclerosis (18), and increased plasma levels of calprotectin are a sensitive marker of future cardiovascular events in humans (19). Additionally, calprotectin represents a marker of obesity in subjects without type 2 diabetes (20), whereas increased systemic calprotectin levels have been associated with higher cardiovascular complication rate in type 2 diabetics (21) and increased cardiovascular risk in patients with obstructive sleep apnea (13).

Holmgaard and co-workers have not provided any evidence about cardiovascular co-morbidities that may have affected the results of the present study. This is an extremely important aspect in the evaluation of calprotectin as a candidate biomarker, given the fact that the presence of cardiovascular disease is associated with worse outcomes in COPD patients (22), whereas the use of statins and other cardiovascular medication is associated with reduced morbidity and mortality (23, 24).

In the Holmgaard study, significant independent predictors of mortality, besides the increased levels of calprotectin, included older age, COPD severity and neutrophil granulocyte counts, parameters that are readily available to every clinician. An important question about every novel biomarker in COPD is what does it add to our clinical judgment? We have now evidence from the ECLIPSE cohort that interleukin-6 and a panel of blood biomarkers improve the prediction of mortality when used in addition to clinically relevant parameters (25). In the recently published PROMISE-COPD study, the combination of plasma proadrenomedullin and BODE index improves the prediction of mortality in 2 years compared to BODE alone and may substitute for 6-minute-walk distance in BODE when 6-minute-walk testing is unavailable (26). Important additional endpoints for a candidate biomarker besides mortality include the risk of subsequent exacerbations or ­hospitalizations, events that are critical in the natural history of COPD (27). Among several other studies, we have recently proposed that serum fetuin-A (28) and serum uric acid (29) may serve as potential candidate biomarkers of future exacerbations and hospitalizations in COPD patients.

The study by Holmgaard and colleagues is a step forward in the evaluation of calprotectin as a candidate biomarker in COPD, providing evidence for its role as a predictor of 5-year mortality. Several other steps need to follow, in order to evaluate the reproducibility, the variation during exacerbations, the response to treatment and its ability to predict future exacerbations on top, or even instead, of clinical judgment alone. Proteomic, metabolomics and genomic studies may provide novel candidate biomarkers, but single biomarker studies evaluating clinically important parameters are also needed. The quest for surrogate biomarkers in COPD is an important part of the research of this heterogeneous disorder and, despite the fact that none of the available biomarkers is suitable to fulfill all of our expectations today, we are currently at a more advanced point compared to a few years ago.

Declaration of Interest