© 2009 American Thoracic Society doi: 10.1164/rccm.200812-1927UP
Update in Cystic Fibrosis 20081 University of Toronto, Division of Respiratory Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada Correspondence and requests for reprints should be addressed to Felix Ratjen, M.D., Ph.D., University of Toronto, Division of Respiratory Medicine, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada. E-mail: felix.ratjen{at}sickkids.ca Cystic fibrosis (CF) is caused by mutations in the cystic fibrosis transmembrane regulator (CFTR) gene, and more than 1,500 distinct mutations with different functional consequences have been reported in the CFTR database (1). Multiple organs are involved in the disease process, but lung disease continues to account for the majority of morbidity and mortality in patients with CF. Knowledge about the pathophysiology of cystic fibrosis has greatly increased in recent years and CF has evolved into a model demonstrating how a better understanding of the underlying defect can lead to novel therapeutic approaches. However, complex interactions exist between CFTR and other regulatory proteins and all of the relevant components that cause disease manifestations both in the lung and in other organ systems are still incompletely understood. A large body of evidence supports the concept that airway surface liquid depletion is a key component in the development of CF lung disease. This is thought to be the consequence of an imbalance in defective chloride secretion and enhanced sodium absorption that has been consistently observed in studies of CF epithelium. The net fluid loss on the airway surface leads to the collapse of cilia and impaired mucociliary transport. However, regulation of mucociliary clearance is a dynamic process involving other epithelial channels that are thought to play an important role in stress-induced increases in airway surface liquid (ASL) height (2). These CFTR independent pathways are functional in CF and can be up-regulated by both external and internal stimuli, such as ATP, which may explain why mucociliary clearance is reduced but not absent in patients with CF (2). Improving airway surface hydration either with pharmacological agents or osmotic agents, such as hypertonic saline, has become an important therapeutic target in CF (3, 4). ADDRESSING THE BASIC DEFECT Multiple treatment strategies are currently under development which, rather than aiming to ameliorate the clinical sequelae of CFTR dysfunction, address the molecular basis of the disease. These treatment strategies focus on underlying pathophysiological abnormalities by either improving CFTR expression/function or by altering the imbalance between reduced chloride secretion and enhanced sodium absorption through CFTR independent pathways. An example of CFTR specific therapy is PTC124, which can cause readthrough of premature stop codons in patients carrying class I (stop) mutations (5). PTC124 has been shown to improve protein expression and CFTR function in an unblinded single center study (6). These positive effects on ion channel function were paralleled by improvements in some clinical parameters but need to be confirmed in a randomized controlled trial, which is currently being initiated. Alternatively, airway surface liquid could be increased through CFTR independent pathways. Denufosol, which activates an alternative chloride channel, has been shown to improve lung function in a phase III trial, and a second phase III study is under way (7, 8). Another strategy is to improve airway hydration in CF through inhibition of excessive sodium absorption. Amiloride has been studied in the past and was found to be ineffective when administered to patients with established lung disease (9). The lack of efficacy of amiloride was, among other factors, attributed to its very short half life, but an alternative explanation was recently suggested. Using mice overexpressing the β subunit of the epithelial sodium channel that exhibit airway surface liquid depletion similar to CF, Zhou and colleagues demonstrated that treatment with amiloride was ineffective when started after lung disease had developed, but could prevent pathology when introduced immediately after birth (10). Although this concept is intriguing, the mouse model used in this study does not display all of the CF-related phenotypes and its long term morphology is more representative of chronic obstructive pulmonary disease rather than CF (11, 12). These issues highlight the limitations in murine models of CF lung disease, as none of the cftr–/– mouse strains have a pulmonary phenotype that reflects the abnormalities seen in patients with CF. Pigs carrying a null allele of CFTR have been developed as a large animal model for CF (13). Potentially they could represent a more suitable model to study therapeutic interventions aimed at restoring mucociliary transport, but it is yet unclear whether CF pigs will develop lung disease (13). EARLY LUNG DAMAGE Early lung disease is thought to start in the small airways, but the exact timing of initial events remains poorly defined. It is currently believed that CF lungs are essentially normal at birth. This has been challenged by studies using infant pulmonary function performed by the London cystic fibrosis collaboration, as they have shown reduced forced expiratory flows in a significant proportion of infants at the time of diagnosis (14). Longer term follow-up into preschool years has now demonstrated that reduced lung function persists over time despite aggressive therapy (15). These studies were performed in infants diagnosed clinically who may already have manifestations of lung disease and could therefore not clarify whether reduced lung function is a primary or secondary event (16). A recent longitudinal study from Australia conducted in a cohort of infants diagnosed within a newborn screening program has helped to put these data into perspective. In that analysis, lung function in infants diagnosed by newborn screening was found to be within the normal range in infants with CF younger than 6 months of age and became abnormal thereafter (17). Although this would confirm the concept of normal lung physiology at birth, it is still not clear whether the raised volume rapid thoraco-abdominal compression technique, which was used in these studies, is sensitive enough to capture subtle early alterations in lung function, Other lung function parameters based on multiple breath washout techniques, such as the lung clearance index (LCI), hold considerable promise to evaluate early lung disease, as studies have detected abnormalities in a high percentage of CF patients with normal spirometry (18). Moreover, the LCI has been shown to predict the presence of structural abnormalities, such as bronchiectasis, in patients with CF (19). Regardless of whether CF newborns have completely normal lung function initially, the results of the Australian study demonstrate that even an early diagnosis by newborn screening leading to introduction of treatment before the appearance of symptoms is unable to prevent lung function decline and early lung damage in CF. Studies using therapies, such as hypertonic saline, that have proven efficacy in older patients are currently being evaluated as early intervention strategies in infants and young children and will hopefully increase the spectrum of therapeutic options in this vulnerable age group. BACTERIAL INFECTION Although progress is being made in therapies addressing the basic defect in CF, infection and inflammation remain key components of CFTR-related pathophysiology that lead to lung function decline. Pseudomonas aeruginosa continues to be the major pathogen in CF lung disease, and there is ongoing need to develop a better understanding how infection persists in patients with CF. Defective mucociliary clearance appears to be a relevant component, but other factors, such as host pathogen interactions, may also be important in this context. It has been long known that toxins produced by P. aeruginosa may contribute to lung damage, but their exact role in disease pathogenesis remains elusive. Bianchi and colleagues showed that the P. aeruginosa toxin pyocyanin impairs clearance of apoptotic neutrophils by macrophages, a process driven by reactive oxidant species and inhibited by antioxidant treatment (20). Along those lines, pyocyanin was also shown to inhibit the dual oxidase-base antimicrobial system in airway epithelial cells and to induce intracellular oxidative stress that impairs host defense, thereby potentially contributing to P. aeruginosa persistence in the airways (21). This not only broadens current understanding on how P. aeruginosa affects host response but also opens new mechanisms in which antioxidants could be beneficial in CF lung disease. There is evidence that oxygen tension is low in CF airway secretions. This information has led to the concept that P. aeruginosa may persist under anaerobic conditions in CF airways. It has also raised the question of whether anaerobic bacteria play a role in lower airway infection in patients with CF. These organisms do not grow in routine sputum cultures and may therefore be missed by diagnostic laboratories. Tunney and colleagues demonstrated that anaerobic bacteria are present in concentrations similar to those of P. aeruginosa in adult patients with CF (22). The high concentrations per se do not imply pathogenicity and supporting evidence, such as assessment of the inflammatory response to these organisms, could potentially help to better delineate their relevance (23). If confirmed to be relevant, the presence of anaerobic airway infections would have practical implications for choosing the adequate drug combination in patients with CF requiring antibiotic treatment: The major anaerobic bacteria described in this study were not necessarily sensitive in vitro to commonly used antipseudomonal drugs such as ceftazidime, whereas other antimicrobials such as meropenem showed a higher rate of efficacy. Many bacteria can be found in respiratory secretions of patients with CF and molecular techniques have demonstrated that the diversity of organisms in sputum and bronchoaveolar lavage is rather high (24). The clinical relevance for many of these bacteria is still unclear because transient infection is common. Methicillin-resistant Staphylococcus aureus (MRSA) is found with increasing frequency in patients with CF; this organism is not specific to cystic fibrosis and parallels the overall increase in prevalence in the population. Studies have provided evidence that the presence of MRSA in respiratory cultures is associated with worse lung function, but it is unclear whether this is caused by infection with the organism or whether this is the consequence of selective pressure by more intense antibiotic therapy in patients with advanced disease (25). Using a large CF database, Dasenbrook and colleagues showed that MRSA was not only associated with more severe lung disease, but that infection also affected subsequent lung function decline (26). Interestingly, this effect was found in patients with persistently positive cultures rather than in those with only intermittent infection, raising the question whether similar criteria should be applied to reassess the role of other emerging bacteria, such as Stenotrophomonas maltophilia, that so far have not been shown to affect the course of lung disease in patients with CF (27). Bacteria from the Burkholderia cepacia complex are a group of emerging pathogens that clearly have the potential to negatively affect the clinical course in patients with CF. Outbreaks of B. cenocepacia and dolosa have been associated with lung function decline and B. cenocepacia has been linked to increased mortality in patients with CF both before and after lung transplantation (28, 29). Less information is available for other B. cepacia species, an issue that is especially relevant in the decision process as to whether to accept patients infected with these organisms for lung transplantation. Alexander and colleagues demonstrated that species other than B. cenocepacia did not affect the 5-year outcome in patients with CF receiving a lung transplant; they also confirmed the negative impact of B. cenocepacia reported previously (30). This finding has subsequently been confirmed in an independent, albeit smaller, cohort in Europe and conveys an important message for transplant centers (31). Although it remains controversial whether patients with B. cenocepacia should be accepted as lung transplant candidates, the Toronto experience would indicate that these patients have a survival benefit from the procedure (32). TREATMENT OF INFECTION Maintenance treatment for patients with chronic P. aeruginosa infection includes regular administration of inhaled antibiotics, and inhaled tobramycin is considered to be part of recommended care (33). The spectrum of inhaled antibiotics has been rather limited, as sufficient evidence from clinical trials was not available for drugs other than tobramycin. Inhaled azthreonam, administered with a fast and efficient delivery system, the Pari eflow, was shown to improve lung function and quality of life as well as to reduce pulmonary exacerbations in patients with CF (34). No differences were observed between the two-times versus three-times daily dosing group within the placebo-controlled trial, but subsequent open-label data suggested superiority of the three-times daily dosing group, providing the first evidence for a dose-dependent efficacy of inhaled antibiotic therapy. Although these data look promising, as the increase in lung function was quite substantial, studies using direct comparisons to the current standard of care (inhaled tobramycin) will help to better define whether treatment with inhaled azthreonam is superior to current therapy. INFLAMMATION Inflammation is a key component of CF lung disease and many inflammatory pathways are primarily or secondarily altered in CF airway disease. To date we lack a clear understanding on the key factors contributing to the inflammatory response as well as the best way to monitor inflammation over time. Using airway samples from patients with CF, Rowe and colleagues demonstrated that high mobility group box 1 (HMBG1) is found in increased concentrations in CF and acts as a major chemoattractant for neutrophils (35). Isolated administration of HMBG1 to mice increased neutrophil influx into airways and led to collagen matrix degradation. This mechanism is likely not unique to CF lung disease, but opens the possibility of HMBG1 as a potential treatment target or, alternatively, as a marker protein of CF airway inflammation. Monitoring airway inflammation in patients with CF continues to be a challenge as sputum markers show considerable variability, and systemic markers of inflammation are often negative despite significant airway inflammation. Rather than aiming for a single marker of inflammation, Saavedra and colleagues used mRNA transcripts in peripheral blood leukocytes of patients with CF undergoing treatment of a pulmonary exacerbation to test their potential ability to predict treatment response (36). The study defined three genes that predicted lung function response to treatment; the authors subsequently confirmed these results in an independent cohort of patients with CF. Proteomic approaches have also been applied to sputum samples and have identified a number of new markers that potentially may better reflect the local inflammatory process in the lung (37). However, the utility of these markers has yet to be confirmed before they can be used in interventional trials targeting airway inflammation. Nitric oxide (NO) metabolism is altered in CF lung disease and NO augmentation therapy has been discussed as a potential treatment option to overcome NO deficiency (38). A number of potential treatment options that affect NO bioavailability are available, but some of the potential compounds appear to have pleotropic intracellular effects beyond influencing NO related pathways. S-nitrosoglutathione (GSNO) has been shown to increase CFTR expression and there is now evidence that phosphodiesterase 5 inhibitors such as sildenafil or vardenafil activate CFTR in both wild type and DF508 mice (39, 40). However, the mouse model used in those studies is known to have residual CFTR function and it needs to be demonstrated whether these results will be relevant for patients carrying this mutation and for whom CFTR function is more profoundly reduced. Although the exact intracellular mechanism responsible for CFTR rescue has not been elucidated, phosphodiesterase 5 inhibitors and their analogs could potentially offer a new therapeutic approach to improve CFTR function in CF airways (41). CFTR OUTSIDE THE EPITHELIUM The clinical phenotype in patients with CF is primarily apparent in organs that exhibit disease manifestations, such as the lung, pancreas, and liver, and epithelial and mucus-producing cells are considered to be the predominant cell types affected by CFTR dysfunction. Relatively little attention has been paid to the potential impact of altered CFTR expression in other tissues. Recent studies in CF children have demonstrated increased smooth muscle mass in airway biopsies, but the pathophysiological steps leading to these morphological changes are unclear (42). CFTR expression has been detected in smooth muscle and may regulate muscle contractility both in the airways and the vasculature (43, 44). CFTR may also play a role in the development of CF bone disease. This is supported by studies in cftr–/– mice, which show decreased bone mass not explained by reduced weight or reduced vitamin D concentrations alone (45). A functional role of CFTR in bone tissue could potentially explain the low bone mineral density that has been reported early in the disease process in children with CF (46, 47) and exemplifies the complex nature of the disease beyond the respiratory tract. It is likely that we will learn more about the functional consequences of CF in other organ systems in the near future. FOOTNOTES Conflict of Interest Statement: F.R. has acted as a consultant for Genentech, Gilead, Inspire, Novartis, and Vertex; he has received $10,000 for lectures sponsored by Genentech. Received in original form December 22, 2008; accepted in final form December 22, 2008 REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
