CC-16 as a Biomarker in Chronic Obstructive Pulmonary Disease

Clara cell secretory protein-16 (CC-16, CCSP, CC-10 or uteroglobin) is a small secreted disulphide-bridged dimeric protein that is a member of the secretoglobulin family [1]. It is secreted by non-ciliated Clara cells found predominantly in respiratory bronchioles [2] and from the non-ciliated columnar cells of the large and small airways [3]. CC-16 is also secreted from nasal and urogenital tract epithelial cells [4, 5]. Serum levels correlate significantly with lower respiratory tract levels and are not influenced by urogenital CC-16 release [6]; thus serum levels appear to be influenced only by i) pulmonary production, ii) lung epithelial permeability and iii) renal clearance [7]. CC-16 has therefore been postulated as a potential surrogate marker of non-ciliated epithelial dysfunction in lung disease, including COPD. It may represent a more specific marker of pulmonary dysfunction than other postulated biomarkers of COPD such as CRP, fibrinogen or IL-8, which all reflect the systemic inflammatory response and therefore may be influenced by co-morbid conditions such as cardiovascular or metabolic disease. CC-16 is an immunosuppressant and antioxidant protein, which modulates mediators of the inflammatory response including phospholipase A2, interferon γ and tumour necrosis factor α. Indeed knockout mice deficient in CC-16 display an exaggerated response to viral infection and oxidative stress [7]. Secretion of CC-16 increases transiently in response to acute cigarette smoke exposure as well as local inflammation [5].

An early study by Bernard and colleagues showed that CC-16 was decreased in both the serum and bronchoalveolar lavage (BAL) of individuals with COPD when compared with healthy controls [8]. These findings were supported in other small cohorts; by Pilette, who found significantly reduced CC-16 staining in lung resection specimens from COPD patients when compared with healthy controls [9] and by Braido who noted that those with severe COPD had lower sputum CC-16 levels than those with moderate disease [10]. In contrast, others found no correlation between a diagnosis of COPD and serum CC-16 levels [11] or correlation between forced expiratory volume in 1 second (FEV1) and CC-16 in individuals with COPD [12]. Thus a large-scale study was necessary to more comprehensively characterise any association.

Results from the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) cohort of over 2000 individuals have provided insight into the potential utility of CC-16 as a biomarker in COPD [13]. Importantly, CC-16 was found to be reproducible in stable disease, a characteristic required for any potential biomarker to be clinically useful. CC-16 was confirmed to be significantly lower in individuals with COPD when compared with controls, though differences in median values were too small to foresee clinical utility as a diagnostic screening tool.

It was evident in this cohort that CC-16 levels are reduced by current smoking and increased with male sex and advancing age (consistent with other reports and attributed to decreasing glomerular filtration rate and increased alveolar capillary leak [14, 15]). These factors would all require adjustment should CC-16 be used in clinical practice.

A biomarker such as CC-16, which reflects epithelial integrity may be expected to correlate with disease severity and/or progression. In the ECLIPSE cohort, a correlation was evident between baseline serum CC-16 and disease severity in former smokers, and a small but significant correlation was subsequently seen with rate of decline in FEV₁ over a 3-year follow-up period [16], though this did not translate to an association with mortality [17]. CC-16 may therefore have utility as a surrogate marker of therapies that slow loss of epithelial integrity. To date, there is no evidence that CC-16 is affected by administration of inhaled corticosteroids [13, 18], though a combination steroid/long-acting β agonist inhaler did reduce CC-16 after a preceding wash-out period. Intriguingly, CC16 was also lower in individuals taking long-acting β agonists and higher in former smokers with reversible airflow obstruction in the ECLIPSE cohort, raising the possibility that CC-16 may reflect the presence of reversible airflow obstruction. Other therapies studied to date include losmapimod, an experimental mitogen activated protein kinase (MAPK) inhibitor which did not affect CC-16 levels after a 12-week trial [18], and the phosphodiesterase-4 inhibitor roflumilast which prevented loss of CC-16 in an experimental mouse model [19].

The complexity of COPD pathogenesis and varied clinical presentation of COPD itself and its associated co-morbidities make it likely that no one biomarker will be powerful enough for use as a screening tool. However, CC-16 joins a growing list of biomarkers that are able to differentiate individuals with COPD from controls in large cohorts and may have utility when grouped together as a composite marker. Individual biomarkers may be more useful in defined disease sub phenotypes; in the case of CC-16 there is evidence that it may reflect loss of airway integrity and perhaps the presence of reversibility in airflow obstruction. Thus future trials, which must be a) large scale and b) longitudinal, should select appropriate individuals to analyse these aspects in further detail; firstly to confirm an association with lung function decline and then to use those with declining lung function to test existing and new therapies that are postulated to affect epithelial integrity. The development of a screening tool to define individuals with COPD at a population level is likely to include CC-16 and other serum biomarkers and will require large, multinational studies to determine its exact composition.