Biomarkers to monitor exacerbations in cystic fibrosis

KEYWORDS:

• Exacerbation
• biomarker
• inflammation
• cystic fibrosis

1. Introduction

Cystic fibrosis (CF) is a life-limiting genetic disease characterized by progressive lung destruction and pulmonary function decline. Pulmonary exacerbations (PEx) are a major clinical feature in CF and contribute to the overall decline in lung function and mortality [1]. In spite of this, PEx remain a poorly defined and understudied feature of the disease. The evolving field of translational medicine has driven the search for new ways with which to monitor disease, and many techniques such as proteomics have been utilized in the search for a CF biomarker that will be informative of both disease status and therapeutic response. In spite of a growing number of proposed biomarkers of PEx, few have progressed to clinical practice although a number of them show promise.

2. Definition of CF exacerbation

Although a major clinical event in CF, it is surprising that there is a lack of a universally accepted set of criteria to diagnose and classify PEx. Definitive diagnostic criteria would be useful on a number of levels. First, the accurate classification of exacerbations would allow the collection of clinically meaningful patient data that could be used to accurately track disease, for example, by recording the number of criteria meeting exacerbations/year. Second, universal criteria would allow a more accurate interpretation of biomarker studies and how they may be translated into clinically useful tests. Fuchs criteria have been used in a number of studies since being first described [2], and although useful in standardizing the terms of a PEx for clinical studies, are not commonly used to guide clinical therapy although modified versions have now been suggested [3].

3. Physiological biomarkers in exacerbation

Forced expiratory volume in 1 second (FEV₁) has been extensively utilized as a clinical tool in exacerbation and also as a major comparator for biomarker studies. Newer more specific physiological tests such as lung clearance index have been demonstrated to change informatively in PEx [4] but require specific equipment and expertise to perform and interpret. Therefore, biomarker measurement in samples such as sputum and blood may offer a more universal approach.

4. Sputum biomarkers of exacerbation

Sputum has been investigated in numerous studies to yield biomarkers in CF. The most extensively studied sputum biomarkers in PEx are neutrophil elastase and interleukin (IL)-8 which have been shown to decrease following exacerbation treatment [5,6] and also correlate with lung function over longitudinal monitoring [7]. Sputum IL-8, however, has not been found to consistently change in all studies of exacerbation [4,8], possibly due to differences in the studied populations of children and adults. Sputum biomarker measurement requires specialist sample preparation, which is not readily available in most clinical centers, and following preparation, further biochemical testing (using techniques such as ELISA and mass spectrometry) is required to measure the biomarker in the clinical sample. Significant time and effort are required in the investigation of sputum biomarkers and that may be perceived as a barrier to their eventual adoption. However, the ability to process sputum in universal way, that results in a more usable matrix, may result in sputum being an ideal biological fluid to assay for biomarkers. Urine may also serve as a useful fluid for the measurement of protein biomarkers, but this technology is still in development. Therefore, the development of biomarkers measured in blood samples from patients seems the most realistic prospect for early translation into the clinic.

5. Blood biomarkers of exacerbation

The measurement of biomarkers in blood to assess exacerbation offers great promise and was the subject of a recent meta-analysis [9], which concluded that there was reasonable evidence for the application of biomarkers to the study of PEx, but further studies were required to demonstrate the clinical usefulness of this approach. The most commonly studied blood biomarker in CF is C-reactive protein (CRP), an acute-phase protein produced by the liver and utilized in clinical practice in a wide range of inflammatory and infectious diseases including pneumonia and sepsis to assess the severity of illness and response to treatment [10]. In our own studies, we have demonstrated a temporal change in CRP with exacerbation treatment [4,8]. Conversely, CRP has been demonstrated not to be informative of the severity of exacerbation when compared to clinical scoring [11]. Sagel et al. conducted the largest blood biomarker study of systemic inflammation in exacerbation, showing that a number of biomarkers changed informatively with treatment including CRP, serum amyloid A, and IL-1ra, among other markers [12]. Even more interesting perhaps was their finding that higher levels of specific markers such as IL-8 and neutrophil elastase antiprotease complexes at the onset of exacerbation were associated with increased response to therapy. Nick and colleagues recently demonstrated that a panel of microRNA (mRNA) biomarkers could be measured in blood and changed temporally with exacerbation treatment, with the marker CD64 (expressed by neutrophil activation) being most highly associated with a response to treatment. Other potential biomarkers for measurement include mRNAs, a number of which have been found to be altered in various CF cell types [13]. Further work will be required to determine if altered mRNA levels are related to PEx although a recent study has identified altered circulating mRNAs in relation to CF liver disease, highlighting the potential of these mRNAs to predict early hepatic fibrosis [14]. Other potential markers of PEx include cleavage products of endogenous respiratory tract proteins produced during episodes of infection including secretory leukocyte protease inhibitor, elafin, and LL-37 [15]. Most recently, sophisticated mass spectrometry has been employed to generate a protein panel for the prediction of exacerbation which was demonstrated to be superior to the measurement of other more established panels of biomarker, although they did not assess some previously published biomarkers in their analysis such as calprotectin [16].

6. Calprotectin as a biomarker in exacerbation

Calprotectin, a neutrophil protein, has been associated with CF since the 1970s and recently has been demonstrated in a number of studies to change following exacerbation treatment [4,8]. Furthermore, it has been shown to predict further PEx and lung function decline [17]. Interestingly, in our studies, we have consistently demonstrated that calprotectin is a superior biomarker in terms of PEx prediction than CRP [8,17]. Calprotectin has also been demonstrated to change in serum following treatment of patients with azithromycin [18], suggesting that small changes in inflammation may be monitored by the measurement of calprotectin in blood. Numerous point-of-care assays have now been developed for calprotectin measurement in the context of inflammatory bowel disease, and therefore, we would suggest that measurement of calprotectin as a near-patient assay offers a great deal of promise.

6.1. Single vs. combined biomarkers for clinical application

Many studies have focused on the measurement of a panel of biomarkers and focused on the information provided by comparing biomarkers with each other and assessing which biomarker may be ‘best.’ The creation of biomarker panels is an alternative approach and has been previously suggested [19]. Combining a number of biomarkers may increase the significance of an association with exacerbation, which may be useful as a surrogate end point in clinical studies. For universal adoption into clinical practice, a biomarker panel would need to be practical to both measure and interpret and as such this approach would seem to be a way off from clinical application.

7. The development of biomarkers as a clinical tool

To date, the majority of studies assessing biomarkers in PEx have utilized laboratory testing to assess the change in biomarkers while observing normal clinical protocols of therapy, with many of these studies using specialized laboratory techniques not universally available in all centers. The real challenge will be to test the application of PEx biomarkers in clinical practice. This should be considered in the context that many clinical decisions are already made in CF utilizing lung function as a biomarker, with most ‘novel’ biomarkers only being considered informative if they show an association with FEV₁. Therefore, the adoption of new biomarkers will require a demonstration of superiority or ‘added value’ to universally available (and cheap) measures of lung function such as FEV₁. Utilizing biomarkers to aid the therapeutic decision of when to start antibiotics or using biomarkers to dictate the length of antibiotic treatment for PEx may be an appropriate niche for exploitation. There has been a frustrating lack of studies that take the assessment of biomarkers in CF and PEx closer to the clinic. We would therefore propose that future studies are designed to address this. These may take the form of biomarker-guided duration of therapy versus standard clinical protocol for exacerbation treatment, with outcomes such as time to next exacerbation and total number of exacerbations over a given time being measured. Pragmatically, these would have to be single-blind studies with patients being randomized to a protocol of standard therapy (e.g. guided by FEV₁ response) or biomarker guided (e.g. a predetermined change in a given marker or markers). This approach has previously been used to assess inflammatory biomarkers in asthma such as sputum eosinophils and exhaled nitric oxide and yielded clinically meaningful data that have influenced practice [20,21]. The development of point-of-care assays for use in PEx will also drive the clinical application of markers, particularly if interfaced with emerging technology such as smart phones to provide patients and clinicians with real-time feedback on their exacerbation status. Although still in its infancy, such technology is being utilized to improve diabetes care in children [22]. Therefore, it will require considerable investment from researchers, industry, and the CF patient population to ensure that biomarker studies are translated from interesting observations into useful clinical tools.

8. Conclusion

In conclusion, biomarker research in CF and CF exacerbation has led to a number of potential clinical biomarkers being detected. The real challenge now is how to translate these research findings into clinically useful tests that will help to stratify patients and guide treatment decisions. Pragmatic and well-designed clinical trials of biomarkers in these settings are now required to make the next significant step toward clinical application.

Declaration of interest

RD. Gray received funding from a Wellcome Trust grant 093767/Z/10/Z. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

This paper was not funded.