Eur J Clin Microbiol Infect Dis DOI 10.1007/s10096-015-2347-4

REVIEW

Azithromycin use in patients with cystic fibrosis N. Principi & F. Blasi & S. Esposito

Received: 4 January 2015 / Accepted: 1 February 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Rational antimicrobial administration is still considered to be the most effective therapeutic approach in cystic fibrosis (CF), and long-term treatment with azithromycin (Az) is included in the current guidelines for CF patients aged ≥6 years. Az has microbiological, immunomodulatory and anti-inflammatory properties that can reduce some of the biological problems that are among the causes of the progressive lung damage associated with CF. Moreover, although it is not active against Pseudomonas aeruginosa (the most important bacterial pathogen responsible for infectious exacerbations), it can be efficiently used in the case of P. aeruginosa infections because sub-inhibitory concentrations can reduce their pathogenic role by interfering with some bacterial activities and increasing their susceptibility to antibiotics. Az also has antiviral activity that limits the risk of the bacterial pulmonary exacerbations that frequently occur after apparently mild viral infections. The available data seem to indicate that it is effective during its first year of administration, but the impact of longer treatment is debated. Other still undefined aspects of the use of Az include the possible emergence of antibiotic resistance in the other bacterial pathogens that usually colonise CF patients, the real incidence of adverse events and the drug’s potential interference with other routine therapies.

N. Principi : S. Esposito (*) Pediatric Highly Intensive Care Unit, Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Via Commenda 9, 20122 Milan, Italy e-mail: [email protected] F. Blasi Pneumology Unit, Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy

Introduction Cystic fibrosis (CF) has an incidence of 1 in 2,500 newborns, thus making it the most frequently inherited disease in white populations [1]. It is an autosomal recessive disease caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene located in chromosome 7, a gene that encodes for a protein that regulates liquid volume on epithelial surfaces by means of chloride secretion and the inhibition of sodium absorption. As CFTR protein is expressed in many cell types, CF may manifest in various organs but particularly affects the upper and lower airways, pancreas, bowel and reproductive tracts. However, lung disease is the most important problem for the majority of patients, and has the greatest impact on their quality of life and survival. Alterations in epithelial cell ion transport lead to increased sputum viscosity, the stasis of secretions, impaired mucociliary clearance, recurrent respiratory infections and chronic lung damage [1]. In the 1980s, children with CF rarely lived beyond adolescence, but they can expect to live until they are in their early forties as a result of the introduction of innovative management strategies based on daily chest physiotherapy aimed at facilitating mucus clearance, correct nutrition and early antimicrobial treatment in order to stem infections promptly [2]. A definite solution for CF can only come from drugs designed to correct cell processing and/or potentiate chloride movement across the ion channel, or from gene therapy capable of delivering functional CFTR directly to the lungs. At present, rational antimicrobial administration is still the most effective means of reducing the clinical signs and symptoms of disease in most children with CF [3], and long-term treatment with azithromycin (Az), a macrolide antibiotic, is included in the current guidelines for CF patients aged ≥6 years [4]. The aim of this paper is to discuss the reason for this

Eur J Clin Microbiol Infect Dis

recommendation, and the advantages and limitations of chronic Az treatment.

Lung damage in cystic fibrosis It was long thought that the repeated respiratory infections and consequent progressive lung damage observed in most CF patients were mainly due to the presence of desiccated secretions and impaired mucociliary clearance [5]. By preventing the elimination of infectious agents, these conditions favour the persistence of pathogens in the airways and the development of chronic inflammation, which can perpetuate the risk of new infections and cause irreversible respiratory tract destruction and fibrosis. The chronic inflammation was attributed to bacteria, particularly Staphylococcus aureus in early stages, and Pseudomonas aeruginosa and Stenotrophomonas maltophilia in later stages of the disease [6–9]; viruses, with rhinovirus as the most frequently found in patients with pulmonary exacerbations regardless of bacterial infections [10, 11], that can also favour bacterial infections and contribute to bacteria-induced damage by transferring antibioticresistant genes through phages [12]; and, albeit rarely, fungi such as Aspergillus spp. [13]. Neutrophils were considered the predominant cause of the inflammation because they produce large amounts of pro-inflammatory cytokines, and, thus, induce the further recruitment of inflammatory cells and perpetuate the inflammatory response [14]. It was also shown that they are the main cause of direct damage to respiratory tissues because they release large amounts of proteases and oxidants that can digest structural proteins, thus leading to tissue destruction and declining lung function [15, 16]. Other extra-cellular triggers have been attributed a role in conditioning excessive inflammation and tissue damage. These triggers were severe gastro-intestinal reflux and pulmonary microaspiration, which were associated with declining lung function in infants and children [17], and reactive oxygen species-induced damage probably due to the depletion of glutathione, the major component of cell defences against oxidative injury, in the epithelial lining fluid and plasma of adults with CF [18–20]. However, although an increasing amount of data suggest the role of extra-cellular triggers in CF-related airway alterations, a number of questions remained unanswered. The most important issues were why host defences failed to eradicate infections despite significant neutrophil recruitment, and why this recruitment took place even in the absence of apparent infection [21]. These have now been at least partially answered by the finding that CFTR dysfunction can directly affect airway immunity by increasing the production of proinflammatory mediators and impairing the immune response to pathogens, thus suggesting that airway inflammation may precede bacterial infection and play an independent role in

favouring airway dysfunction and damage [22]. It has been found that bronchoalveolar lavage (BAL) fluid in young CF infants without bacterial infection contains inflammatory markers such as interleukin 8 (IL-8), peroxidases and their oxidants [23], and that high BAL neutrophil counts and increased neutrophil elastase activity are associated with respiratory symptoms but not with infection [24]. Furthermore, the findings of a number of in vitro studies suggest that CF epithelial cells tend to show an exaggerated response to external stimuli that is, per se, sufficient to induce chronic problems [5]. It has been found that, in comparison with normal subjects, innate and adaptive immune responses are dysregulated in CF patients and that, even in the absence of infection, the BAL of children with CF contained increased levels of alveolar macrophages and their related chemokines [25]. Furthermore, the expression of toll-like receptor 4 (a factor that recognises pathogens and initiates a immune response) is increased in children without acute infective exacerbations [26], and that other immune system cells such as T lymphocytes are dysregulated in very young CF children and induce, for example, lower levels of interleukin (IL)-8 and higher levels of IL-12 than those found in healthy subjects [27]. In conclusion, lung damage is probably initiated by intrinsic CF cell defects that increase immune responses to beyond normal limits and are themselves the cause of chronic inflammation, which is made worse by the chronic inflammatory responses triggered by external factors, mainly infections. The use of antibiotics to control bacterial infections can reduce (but not eliminate) the degree of inflammation and, therefore, the progression to permanent lung damage.

The characteristics of azithromycin Antimicrobial activities Azithromycin (Az) is a macrolide and, like all members of this class of antibiotics, a bacteriostatic drug. It inhibits bacterial growth by binding the 23S rRNA in the 50S subunit of the bacterial ribosome, thus preventing the transfer of tRNA from the A to the P site of the ribosome and aborting polypeptide growth [28]. In vitro, Az plays a marginal role against most of the pathogens usually involved in causing lung infections in CF patients, as a number of S. aureus and all of the P. aeruginosa and S. maltophilia strains are resistant to it. It is, therefore, not approved for the treatment of infections caused by P. aeruginosa and S. maltophilia, and there are no published breakpoints for these species. However, some of its properties seem to indicate that it can have beneficial effects at least against P. aeruginosa [29, 30]. It has been found that P. aeruginosa exerts its pathogenic effects by means of various virulence factors and by producing cell-surface components with pro-inflammatory and/or

Eur J Clin Microbiol Infect Dis

adhesion activity. Furthermore, it survives by switching to the biofilm mode of growth, which enables it to tolerate the inflammatory defence mechanism, the aerobic respiratory zone, the conductive pulmonary zone containing anaerobic sputum and antibiotic therapy [31]. Transcriptomic and proteomic analyses have shown that sub-inhibitory concentrations of Az can reduce the production of some quorum-sensing signal molecule-dependent genes by P. aeruginosa and, thus, limit the synthesis of many virulence factors. The most important data in this regard concern 3-oxo-C12-homoserine lactone and butyryl-homoserine lactone, which activate the expression of many pathogenic proteins, such as proteases, pyocyanin, exotoxin A, phospholipase C and extracellular polysaccharides. It has also been reported that sub-inhibitory concentrations of Az inhibit alginate production [32], increase susceptibility to serum bactericidal activity [33], kill stationaryphase and biofilm-forming cells [34], and make P. aeruginosa 2–4 times more susceptible to antibiotics such as aztreonam, tetracycline, carbenicillin, chloramphenicol and novobiocin [35]. In addition to this specific antibacterial effect, Az seems to have significant anti-viral activity. Schögler et al. measured viral RNA, interferon (IFN) and IFN-stimulated gene and pattern recognition receptor expression in primary bronchial epithelial cells taken from CF and control children infected with rhinovirus after Az pre-treatment [36]. They found that the antibiotic reduced rhinovirus replication seven-fold in CF cells without inducing cell death. Furthermore, it increased the levels of rhinovirus-induced pattern recognition receptor, IFN and IFN-stimulated gene mRNA, thus suggesting that its anti-viral effect was possibly due to the amplification of the anti-viral response mediated by the IFN pathway. Immunomodulatory activities In addition to the potentially positive impact of sub-inhibitory concentrations on P. aeruginosa, Az may benefit CF patients by favourably modifying the immune system response against infections and, above all, by reducing the inflammatory response triggered by internal and external factors. It modulates host defences and reduces inflammation by interacting with structural cells such as epithelial cells, smooth muscle cells and fibroblasts, as well as with neutrophils and mononuclear leukocytes [28]. In epithelial cells, it assures integrity and transepithelial resistance against the permeability induced by P. aeruginosa virulence factors [37], reduces mucin secretion [38] and attenuates the expression of inflammatory cytokines [39]. In airway smooth muscle cells, it relaxes pre-contracted cells [40], inhibits the release of IL-8 [41] and attenuates the fibroblast growth factors induced by vascular endothelial growth factor [42]. The particular pharmacokinetic characteristics of Az mean that therapeutic doses of Az accumulate and persist in

neutrophils for several days [43], which contributes to its antibacterial activity by stimulating neutrophil degranulation and phagocytosis-associated oxidative bursts [44]. It also inhibits the neutrophil release of a large number of cytokines and chemokines, including IL-1β, IL-6, IL-10, tumour necrosis factor (TNF)-α, chemokine ligand (CCL) 1, CCL3, CCL5, CCL20 and CCL22, and granulocyte colony-stimulating factor (G-CSF), increases phagocytosis, attenuates the secretion of IL-12 by macrophages (thus reducing Th-1 responses and subsequent chronic inflammatory disease) and modulates the differentiation and maturation of dendritic cells [45–48]. Furthermore, Az readily penetrates the phospholipid bilayers of cell membranes by decreasing the fluidity and elasticity of macrophage membranes [49], accumulates in lysosomes and induces phospholipidosis, thus affecting endocytosis and phagocytosis [50]. Moreover, by deregulating toll-like receptor 4 (TLR4), recycling modifies downstream signalling [51] and this may explain its inhibition of prostaglandin production, the indirect inhibiting cytokine secretion and, undoubtedly, may contribute to its overall anti-inflammatory activities. Finally, it can affect cell autophagy and apoptosis, and regulate the expression of a number of genes involved in mucin production and lipid metabolism [28]. However, the immunomodulatory effects of Az depend on the presence and phase of inflammation. The findings of both animal and human studies seem to indicate that it promotes host defences when administered early during bacterial infection, whereas it inhibits inflammation and promotes its resolution—with potentially negative effects on defences—when administered after inflammation has been present for a long time [52]. The promotion of host defences in the first hours after Az administration was evidenced by Culić et al., who found that, in healthy human subjects, the administration of Az was followed by enhanced neutrophil activation after 2.5 h, an effect that was significantly reduced after 1–28 days [44]. On the contrary, pre-treatment with Az in experimental animals reduced inflammatory cell infiltration, and its administration 72 h after the establishment of inflammation further limited infiltration and increased the concentration of the antiinflammatory M2 macrophage phenotype at the inflamed site [53].

The use of azithromycin in cystic fibrosis patients The microbiological, immunomodulatory and antiinflammatory properties of Az explain why it has long been used to treat respiratory diseases (including CF) that are characterised by chronic lung inflammation leading to fibrosis, bronchiectasis and recurrent infections mainly due to P. aeruginosa [54–60]. Since the first studies of Az and CF about 15 years ago, a number of trials have investigated the effect of chronic Az administration on the evolution of the

Eur J Clin Microbiol Infect Dis

disease. A meta-analysis in a recent Cochrane review of the findings published in the scientific literature up to February 2012 [54] used six apparently bias-free studies involving a total of 836 patients aged mainly >6 months: five placebocontrolled trials [55–59] and one comparing two dose regimens [60]. Regarding the dosage, in some studies, Az was administered every day with a single dose of 250 mg (500 mg if weight >40 kg) [55, 59]; in other trials, the drug was given at the same daily dose but only three times a week [56–58]. The period of administration was 6 months in five and 12 months in one [56]. Although there were differences between the studies, the aggregate data showed that 6 months’ treatment with Az was effective [54] in improving respiratory function, as shown by the changes in forced expiratory volume in one second (FEV1) and forced vital capacity (FVC). During this period, the patients receiving Az were more likely to be free of exacerbations, at less risk of requiring of oral or intravenous antibiotics, gained more weight and had a better quality of life. Better results were apparently obtained in patients infected by P. aeruginosa because, in uninfected subjects, Az only reduced the number of exacerbations and did not improve in lung function [58], thus suggesting that its beneficial effects were mainly due to its microbiological activity. However, the difference may only be apparent because the vast majority of patients recruited in this study had very good lung function (FEV1 nearly 100 % of predicted), thus reducing the chances of any improvement. The positive findings of the meta-analysis were subsequently confirmed by a double-blind, randomised, controlled trial of oral Az in CF patients aged 6–18 years [61], in which the treated and untreated patients had similar neutrophil counts and similar levels of serum myeloperoxidase (MPO), high-sensitivity C-reactive protein (hsCRP), intracellular adhesion molecule 1, IL-6, calprotectin, serum amyloid A (SAA) and G-CSF. The levels of all of these inflammatory markers significantly decreased in the Az group during the first 28 days, and neutrophil counts and SAA and hsCRP levels remained significantly lower at the 6-month evaluation. However, the small number of patients treated for 12 months prevented the authors from drawing definite conclusions concerning the duration of the beneficial effects after 6 months. More favourable data concerning the persistence of a positive effect of Az in the second semester of treatment have recently published by Saiman et al., who enrolled some of the children aged 6–18 years previously included in a placebo-controlled trial specifically designed to evaluate the efficacy of 6 months’ treatment with Az in a new study [62]. By administering Az for 6 months to the previously treated and untreated subjects, the authors were effectively able to compare 12 and 6 months of treatment, and found that the response to treatment (as measured on the basis of pulmonary exacerbations and continued weight gain) lasted for

12 months, although these children required greater use of oral antibiotics [62]. On the basis of these results, the American Pulmonary Clinical Practice Guidelines Committee recommended the administration of Az to all subjects with CF aged ≥6 years [28]. However, there are still a number of open questions regarding the use of Az in CF patients, particularly its posology and duration of treatment. Various dosing schemes have been used in clinical trials but, as there is a lack of comparative data for most of them, it is not possible to establish which was most effective. However, this limitation may be overcome considering the data published by Wilms et al. [63], who tried to prepare generally applicable dosing recommendations by combining what is known about Az pharmacokinetics and the dosing schedules used in clinical trials. They concluded that a dose of 22–30 mg/kg/week was the lowest with proven efficacy and, given the extended half-life of Az in CF patients, indicated that the weekly dose can be divided into 1–7 administrations, depending on patient preference and gastrointestinal tolerance [63]. In terms of treatment duration, it is not clear whether Az remains efficacious if given for more than 12 months. Two studies have evaluated the impact of longer periods of treatment, and both found that efficacy was poor after the first year [64, 65]. An open retrospective study by Tramper-Stranders et al. [64] found that Az administered for 3 years had a positive effect on the percentage of predicted FEV1 in the first year, but this percentage subsequently returned to pretreatment levels. Similarly, Willekens et al. [65], who studied a group of CF patients aged 13–47 years treated with Az for 4.5–8.6 years, found that, although there was no change in pulmonary function parameters or the incidence of severe exacerbations requiring intravenous antibiotic treatment, the first year of Az administration was associated with a significant reduction in the number of orally treated pulmonary exacerbations per patient. However, this reduction was not maintained in the second and third treatment years. The authors, therefore, concluded that Az therapy should be limited to 6– 12 months because, after this period, the risk of drug-related problems could outweigh the advantages [65]. Finally, it is not known whether a 1-year course of Az administration can be periodically administered, whether further courses can have the same impact as the first period of therapy and what time interval has to pass between subsequent courses in order to have positive results without risks of adverse events. In relation to this, it is worth pointing out that, although the tolerability and safety of Az are generally considered to be very good as the global evaluation of the studies included in the Cochrane review showed that the incidence of adverse events was similar in subjects treated with Az and those treated with placebo [54], tolerance can be affected by the schedule of administration. McCormack et al. found that patients taking Az 1,200 mg once a week were more likely to experience

Type of study

Randomised, placebocontrolled, crossover trial

Randomised, placebocontrolled trial

Multicentre, randomised placebo-controlled trial

Multicentre, randomised, double-blind, placebocontrolled trial

Multicentre, parallel randomised, controlled trial

Open-label study

Multicentre, placebocontrolled trial

Equi et al. (2002) [55]

Wolter et al. (2002) [59]

Saiman et al. (2003) [57]

Clement et al. (2006) [56]

McCormack et al. (2007) [60]

Tramper-Stranders et al. (2007) [64]

Saiman et al. (2010) [58]

Az patients had a significant increase in FEV1, and significantly more nausea, diarrhoea and wheezing but less PEx and better growth development

Relative change in FEV1 (% predicted), adverse events, selfreported symptoms, audiology and laboratory tests, respiratory cultures, relative change in FVC (% predicted), body weight, PEx (number and time to), hospitalisation rate, use of nonquinolones antibiotics, inflammatory markers and QoL Relative change in FEV1 and FCV% predicted, number of pulmonary exacerbations, additional antibiotic treatment, lung microbiology and adverse events

Az 250 mg 3 days a week (>40 kg, 500 mg) vs. placebo for 168 days

Az 250 mg 3 days a week (>40 kg, 500 mg) vs. placebo for 12 months

Az 250 mg once a day vs. 1,200 mg once-weekly for 6 months

185 patients (6–XX years) with chronic P. aeruginosa chest infection (30 % predicted

72 young people (6–21 years), 35 with Az

208 patients (6–58 years)

Children (6–18 years) without chronic P. aeruginosa airway infection (≥2 negative cultures) for 12 months

100 children (6–18 years)

FEV1 improved in the first year but decreased later. S. aureus resistance increased to 83 % in the first year, 97 % in the second and 100 % in the third No improvement in pulmonary function, no difference in height, use of intravenous or inhaled antibiotics, and hospitalisations. Az patients had significantly less exacerbations, greater increase in body weight, less cough

Relative change in FEV1, S. aureus colonisation

Relative change in FEV1, respiratory exacerbations, treatment requirements and adverse events, acquisition of resistant bacteria Az 250 mg 3 days a week (>36 kg, 500 mg) vs. placebo for 6 months

No differences in improvements in lung function, CRP, days spent in hospital, admission rates and nutrition. In children, the daily group had significantly better growth. Gastrointestinal adverse effects were more common with weekly therapy

Az 5–10 mg/kg once-daily for 3 years

Change in FEV1 (%) at 1, 3 and 6 months from baseline, time to PEx, adverse effects, days in hospital

Az patients maintained FEV1 and FCV baseline values, had fewer courses of intravenous antibiotics, a decline in CRP and QoL unchanged. The opposite occurred in placebo patients (p=0.035)

% change in FEV1, FVC, weight, QoL, inflammatory markers, microbiology, respiratory exacerbations

Az 250 mg once a day for 3 months vs. placebo for 6 months

No difference in FEV1 between the groups. The number of PEx, the time elapsed before the first PEx and the number of additional courses of oral antibiotics were significantly reduced in the Az group. No severe adverse events were reported

Improvement of FEV1 in 13/41 and deterioration in 5/41 of Az-treated patients. FVC, MEF and other studied variables did not change. No noticeable side effects

% change in FEV1, FVC and MEF (average of 4 and 6 months values), hearing, sputum bacterial densities, inflammatory markers, exercise tolerance, subjective well-being

Az 250 mg once a day (>40 kg, 500 mg) for 6 months vs. placebo. After 2 months of washout, the treatments were crossed over

41 children (8–18 years), with FEV1 of 61 %

60 adults

Results

Outcomes

Interventions

Patients

Main studies regarding the use of azithromycin in cystic fibrosis (CF) patients

Reference

Table 1

Eur J Clin Microbiol Infect Dis

No differences in respiratory function and BMI after the first year. In this period, reduction of orally treated PEx not maintained later Relative changes in FEV1, FVC, FEF25–75 % predicted, BMI Z-scores and number of PEx Open-label study Willekens et al. (2015) [65]

Az azithromycin; BMI body mass index; CRP C-reactive protein; FEV1 forced expiratory volume in one second; FEF25–75 forced expiratory flow at 25–75 % of the forced vital capacity; FVC forced vital capacity; MEF maximum expiratory flow; PEx pulmonary exacerbation; QoL quality of life

Results similar to those of the 2010 study but no differences in outcomes between 6 and 12 months of treatment Relative change in FEV1, respiratory exacerbations, treatment requirements and adverse events, acquisition of resistant bacteria Az 250 mg 3 days a week (>36 kg, 500 mg) vs. placebo for 6 months Open-label, follow-on study Saiman et al. (2012) [62]

146 of the children enrolled in the 2010 study received Az. Those previously treated for 6 months were globally treated for 1 year 21 patients (13–47 years)

Az 250 mg 3 days a week (>36 kg, 500 mg) vs. placebo for 3 years

Results Outcomes Type of study Reference

Table 1 (continued)

Patients

Interventions

Eur J Clin Microbiol Infect Dis

gastro-intestinal adverse events and discontinue treatment compared to those taking 250 mg/day [60]. Moreover, the use of Az can lead to the emergence of macrolide resistance among other significant CF pathogens, as it has been found that, although Az is not associated with any increase in the acquisition or eradication of S. aureus, the risk of emergence of macrolide-resistant strains may be about five times higher in Az-treated patients, and the risk of isolating macrolide-resistant Haemophilus influenzae may be ten times higher [58]. Finally, there is some concern that the chronic use of Az in subjects with occult or active nontuberculous mycobacteria (NTM) infection can lead to resistance and, thus, complicate NTM treatment [57]. These findings suggest that the microbiological impact of Az therapy on S. aureus and H. influenzae should be periodically monitored, and experts recommend screening patients for NTM before starting Az (withholding it from those with NTM infection) and then re-assessing the situation at intervals of 6–12 months [4]. Attention should also be given to the risk of cardiac arrhythmias during the course of chronic Az treatment. It is known that macrolides such as Az cause torsades de pointes and other ventricular arrhythmias leading to sudden death [66]. The findings of a recent observational study [67] have prompted the Food and Drug Administration to strengthen the warning and precautions section of the Az drug label [68]. Although no major heart problems have emerged during the long-term treatment of CF with Az, and the published data indicate that the great majority of subjects experiencing macrolide-induced cardiac arrhythmias have co-existing risk factors, caution is required and Az should not be prescribed for CF patients with comorbidities of concern. Furthermore, Nick et al. have reported that Az may antagonise the therapeutic benefit of inhaled tobramycin (Tb) [69] because, in comparison with patients treated with Tb alone, those undergoing combined Az and Tb therapy showed a significant decrease in the percentage of predicted FEV1 after one and three courses of inhaled Tb, needed earlier use of additional antibiotics, experienced a lesser improvement in their quality of life and showed a trend toward a lower reduction in sputum P. aeruginosa density. Finally, no solid data are available concerning children aged

Azithromycin use in patients with cystic fibrosis.

Rational antimicrobial administration is still considered to be the most effective therapeutic approach in cystic fibrosis (CF), and long-term treatme...
239KB Sizes 1 Downloads 14 Views