Early lung surveillance of cystic fibrosis: what have we learnt?

KEYWORDS:

• Respiratory infections
• bronchiectasis
• bronchoalveolar lavage
• chest computed tomography
• lung function

Newborn screening (NBS) for cystic fibrosis (CF) provides an opportunity to commence management and therapeutic interventions significantly earlier than indication by symptoms alone. While NBS has provided strong benefits in terms of nutritional status, related improvements in lung health have not been as clear [1]. Defining the natural history of CF lung disease following NBS and identifying the most appropriate outcome measures to track disease progression and to serve as end points for clinical trials are critical for clinical care to maximize the benefit of an NBS diagnosis. The two most prominent early surveillance programs are the Australian Respiratory Early Surveillance Team for Cystic Fibrosis (AREST CF) and the London Cystic Fibrosis Collaboration (LCFC). The LFCF initially recruited clinically diagnosed infants from 1999 to 2002 prior to the introduction of NBS in the UK and a second cohort of infants diagnosed by NBS from 2009 onwards. The LCFC performed lung function at 3, 12, and 24 months with bronchoalveolar lavage (BAL) and chest computed tomography (CT) at 1 year in the NBS cohort [2]. The AREST CF program [3] commenced its early surveillance program in 1999, initially using annual BAL and infant lung function before subsequently expanding to include preschool lung function in 2003 and chest CT in 2005, with bronchial brushing added in 2007 to obtain primary airway epithelial cells. Data from these early surveillance programs have provided important insights into drivers of development and progression of early CF lung disease.

1. BAL and bronchial brushings

The primary challenge quantifying airway pathogens and inflammation in young children is an inability to noninvasively obtain lower airway samples for analysis. Therefore, despite some risks of complications in the very young, sampling the lung by BAL remains the gold standard to study the pathology of early CF lung disease [4]. Clinical benefits of this approach over less invasive techniques remain controversial [5]; however, longitudinal BAL sampling by AREST CF in a repeatable, systematic manner has generated unique findings regarding early CF lung disease.

The key intractable problem in CF clinical care remains the chronically progressive inflammatory milieu of the airways. The AREST CF findings have highlighted that even in very young children with CF who are symptomatically well, airways feature elevated numbers of neutrophils and associated proteases such as neutrophil elastase (NE) and the tissue-modifying matrix metalloproteinases (MMPs) [6], which are not seen in healthy non-CF airways [7,8]. Furthermore, while Staphylococcus aureus, Haemophilus influenzae, Aspergillus species, and Pseudomonas aeruginosa have been linked to increased inflammatory indicators in BAL [9], data from AREST CF have also clearly emphasized that a significant number of children with CF have detectable inflammation in the absence of overt infection detected using standard microbiological techniques [10]. More sensitive metagenomic approaches that improve quantification of bacteria and viruses in BAL, combined with more detailed neutrophil characterization [11], will provide insight into initial pathogen exposures and the development of chronic neutrophilic inflammation in the CF airways. However, in light of the data demonstrating NE is a significant risk factor for bronchiectasis [3] and associated with dysregulated activation of MMPs [6], therapies targeted at controlling protease inflammation in early life may complement current antibacterial treatment protocols.

Expanding observational BAL to obtain lower airway epithelial cells by bronchial brushing for functional in vitro studies allows direct mechanisms to be tested. Using this approach, AREST CF has elucidated that CF epithelium mounts an abnormal interferon and apoptotic response to rhinovirus [12]. A more recent study by AREST CF [13] demonstrated in vitro that NE levels observed in BAL are likely to prevent epithelial homeostasis and repair post-injury [3]. Future opportunities include combining bronchial brushing with conditional reprogramming, a technique of culturing epithelial cells with an irradiated fibroblast feeder stock that permits almost limitless generation of air–liquid interface models from patient’s bronchial epithelial cells [14]. As the current gold standard in vitro model of airway epithelium, this development will expand opportunities to investigate personalized cystic fibrosis transmembrane conductance regulator (CFTR) therapeutics and answer key questions on airway surface fluid biology and the contribution of epithelial immune responses to driving chronic inflammatory disease.

2. Structural lung assessments

Greatly improving the context of findings from BAL studies has been the application of CT imaging in early CF lung disease. Structural lung abnormalities are an early sign of lung disease in CF and increase in prevalence and extent throughout childhood [15]. Although bronchiectasis is the most important facet of CF structural lung disease, abnormalities such as bronchial wall thickening, mucous plugging, and trapped air develop earlier than bronchiectasis, and unlike bronchiectasis may be reversible [16]. The extent of these abnormalities measured in 1 year olds is associated with both pulmonary inflammation and the rate of bronchiectasis progression over the subsequent 2 years [17], suggesting that early intervention, guided by imaging, may lead to ongoing improvements in lung health.

Recent advances have reduced radiation dose to the point where regular surveillance using CT is feasible and safe in young children with CF [18]. Volumetric CT scans in unsedated, free-breathing infants and young children are now feasible and reduce the requirement for sedation or anesthesia. However, the lack of lung volume standardization comes at the cost of reduced sensitivity to detect disease, and pressure-controlled inspiratory and expiratory scans remain the optimal approach in young patients with CF [19].

Due to the mild disease present in early life, traditional CT outcome measures such as the CF–CT scoring system may be less suitable for use in young children. The LCFC reported low prevalence of structural abnormalities and poor repeatability of CF–CT scores in newborn screened, 1-year-old infants [20], and it is likely that CF–CT underreports early structural lung changes. Recently, the Perth-Rotterdam Annotated Grid Morphometric Analysis for CF (PRAGMA-CF) scoring system was developed as the first quantitative measure of structural lung disease designed specifically for this age group [17]. PRAGMA-CF is more sensitive than the traditional CF–CT scores in this age group and more accurately reflects both the extent and progression of lung disease [17]. Novel measures of airway geometry and lung density show promise in objectively measuring lung disease but need further validation [21]. While lung magnetic resonance imaging is being explored in the context of CF lung disease, it is technically challenging in young children and infants [22] and has not entered routine research or clinical practice in infants and young children with CF. Due to these limitations, CT scans, using age-appropriate outcomes, are ideally suited for both early disease surveillance and determining the efficacy of prophylactic treatments to prevent the onset and progression of structural lung disease in young children.

3. Lung function

Measuring lung function in early life is challenging but allows a noninvasive means to monitor the progression of CF lung disease. The main lung function tests that are performed in young children with CF include the multiple breath washout (MBW) test and forced expiratory maneuvers, such as spirometry and the raised volume rapid thoracoabdominal compression technique (RVRTC) in infants. Using both RVRTC and MBW measurements, the LFCF reported that both forced expiratory volume (FEV) and the lung clearance index (LCI) at 3 months could identify a high-risk group with persistent lung function abnormalities at 1 year of age [2]. Data from AREST CF suggest that FEV measured in infants with CF is reduced compared to healthy infants and declines with neutrophilic inflammation in BAL and P. aeruginosa and S. aureus infection [23]. A recent longitudinal study further showed that infection with P. aeruginosa and S. aureus as well as other bacterial pathogens that elicit a pro-inflammatory response, such as H. influenza and Aspergillus species during infancy, was associated with lower lung function compared with uninfected infants with CF with this diminished lung function persisting into school age [24]. Collectively, these data support the use of both FEV and LCI in infancy to identify CF infants at risk of poorer outcomes and therefore benefit from treatment intensification.

Spirometry remains the mainstay for monitoring CF lung disease. However, it is difficult to obtain reliable measurements in preschool children. Despite the ability of spirometry to identify abnormal lung function in infants and young children with CF, the LCFC showed that LCI measured by MBW was more sensitive than spirometry for the detection of structural lung disease. AREST CF data show that infants who had a prior infection had a progressive increase in ventilation inhomogeneity compared to infants who were never infected [25]. While increased LCI was associated with infections, it was insensitive to detect structural abnormalities in infants [26]. The same study however showed that LCI correlated strongly with structural lung disease in preschool and school-aged children, and it is possible that the reduced sensitivity in infants may be due to the inability to detect mild disease in this age group. There are some limitations to the widespread incorporation of LCI into clinical practice, including optimizing data collection and analysis and full characterizing the short- and long-term repeatability of LCI in individuals with CF.

4. Conclusions

Early lung surveillance has enabled us to understand CF lung disease in unparalleled detail, including that neutrophilic inflammation rapidly increases with age independent of infection, and is associated with increased structural lung damage and reduced lung function, although bacterial pathogens remain a major contributor to early CF lung disease. Despite advances in imaging and lung function techniques, BAL remains the gold standard for monitoring early infection. The development of low-dose CT imaging and new scoring systems now makes chest CT scans feasible and represents the most sensitive modality for assessing structural lung disease in very young children with CF. The lung function data suggest that despite current best practice following NBS, pulmonary infections and inflammation remain significant contributors to a decline in lung function in early life. For noninvasive, frequent monitoring of CF lung disease, MBW offers complementary information on ventilation inhomogeneity, is highly feasible in young children, and is sensitive to structural lung disease. Critically, no one surveillance tool has been demonstrated to be superior in isolation, but offer complimentary information on lung disease in CF, and should not be used in isolation. Challenges yet to be addressed by early surveillance programs include the role of pulmonary exacerbations on the progression of lung disease. The outcomes from the various early surveillance programs should be incorporated into early intervention trials to maximize the preservation of lung health in individuals with CF.

Declaration of interest

The authors have no 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. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.