Immunomodulatory Effects of Macrolide Antibiotics - Part 2: Advantages and Disadvantages of Long-Term, Low-Dose Macrolide Therapy

The available evidence for long-term, low-dose treatment with 14and 15-membered ring macrolides in non-cystic fibrosis (CF) bronchiectasis, COPD, chronic sinusitis, and asthma is reviewed with special attention to possible adverse effects and the emergence of resistance during long-term macrolide treatment. Macrolide maintenance therapy has been proven to be of benefit in diffuse panbronchiolitis and CF, presumably due to an anti-inflammatory mechanism of action in addition to its direct antimicrobial effect. Solid evidence to justify this treatment regimen for non-CF bronchiectasis, asthma, or sinusitis is still lacking, although a beneficial effect of long-term macrolide therapy has been found in small clinical trials on these subjects. Data from randomized trials of long-term macrolide treatment in COPD are conflicting. A sufficiently long duration of treatment and the careful selection of patients appears to be crucial. Aside from its beneficial effects, possible side effects of macrolide treatment should be taken into account, the most important of these being gastrointestinal upset and cardiac arrhythmias. Development of macrolide resistance among respiratory pathogens is very common during long-term macrolide treatment. Whether this finding is clinically significant is a matter of debate. Copyright © 2010 S. Karger AG, Basel Received: November 23, 2009 Accepted after revision: June 10, 2010 Published online: August 21, 2010 Drs. J. Altenburg, MD Department of Pulmonary Diseases Medical Centre Alkmaar, Wilhelminalaan 12 NL–1812 JD Alkmaar (The Netherlands) Tel. +31 72 548 2750, Fax +31 72 548 2167, E-Mail j.altenburg @ mca.nl © 2010 S. Karger AG, Basel 0025–7931/11/0811–0075$38.00/0 Accessible online at: www.karger.com/res D ow nl oa de d by : 54 .7 0. 40 .1 1 11 /1 /2 01 7 12 :2 7: 10 A M Altenburg /de Graaff /van der Werf / Boersma Respiration 2011;81:75–87 76 nusitis [2, 3] . High numbers of neutrophils and lymphocytes, together with high levels of IL-8 and other proinflammatory cytokines and chemokines, are found in the bronchoalveolar lavage (BAL) fluid of DPB patients [4, 5] . This indicates a chronic inflammatory process which is further deteriorated by the presence of pathogenic microorganisms. In the early course of the disease, sputum cultures of DPB patients mainly yield Haemophilus influenzae , which is in many cases replaced by Pseudomonas aeruginosa (PA) in a more advanced stage of DPB [4, 6] . In addition, there appears to be a genetic component contributing to the disease mechanism. In approximately 60% of DPB patients the human leukocyte antigen (HLA)-B54 haplotype is found; it is only present in 11% of the healthy Japanese population [7] . After the introduction of EM as standard therapy for DPB in 1987, an impressive increase in 10-year survival was reported, i.e. from 10–20% to over 90% [2, 8–10] . The unexpected success was attributed to a previously unknown anti-inflammatory effect in addition to the antimicrobial potency of EM. This theory is supported by more recent studies which demonstrated that serum levels of EM in DPB patients were well below minimal inhibitory concentrations for the detected pathogens [9] . Furthermore, Nakamura et al. [11] showed that the beneficial effect of macrolides on pulmonary function and general well-being is frequently found without a change in the number or type of bacterial isolates. Macrolides have been demonstrated to reduce IL-8 and IL-1 levels in the BAL fluid of DPB patients [12] . By influencing the production of these and other cytokines, they have a dampening effect on the proinflammatory response. Furthermore, the majority of cells involved in both the innate and adaptive immune response are, in one way or another, influenced when macrolide antibiotics are administered. Their antibacterial effect consists of the inhibition of bacterial protein synthesis, impaired bacterial biofilm synthesis, and the attenuation of other bacterial virulence factors. The effect on biofilm formation is suggested to be especially important in DPB patients who are frequently colonized with biofilm-forming PA. Results of Japanese in vitro studies indicate that azithromycin and clarithromycin change the structure of bacterial biofilms via the inhibition of polysaccharide synthesis, a major biofilm component [13, 14] . An insufficient biofilm allows for enhanced phagocytosis and clearance of bacteria by alveolar macrophages [15, 16] . The effectiveness of macrolide maintenance therapy in order to reduce disease activity, exacerbations, and the decline in lung function has been well proven in CF [17] . Since 2002, five large randomized trials including a total of 608 patients have been published in which the role of macrolide maintenance treatment in CF is addressed [18– 22] . These studies all used azithromycin in different dosages (250 or 500 mg daily, 250 or 500 mg three times a week, or 1,200 mg once a week) with a mean duration of 200 days. All studies showed a significant increase in lung function [forced expiratory volume in 1 s (FEV 1 )]. Additional outcomes were a decrease in the frequency and duration of infectious exacerbations, improvement in physical condition, and body weight gain. The most recent trial was performed by McCormack et al. [22] . In their double-blind randomized study including 208 patients, improvement in lung function, Creactive protein, days spent in hospital, admission rate, and nutritional status was demonstrated after 6 months of treatment with azithromycin. Daily(250 mg) and weekly(1,200 mg) administered azithromycin showed similar outcomes although gastrointestinal adverse effects were more common with weekly therapy. Nowadays, macrolide maintenance therapy is considered common practice in the treatment of CF patients, especially those colonized with PA. Colonization with PA is associated with reduced survival and a faster decline in lung function in DPB and CF [4, 23, 24] . Saiman et al. [25] were the first to study the effect of azithromycin (250–500 mg 3 times weekly for 24 weeks vs. placebo) on pulmonary function in 260 children with CF and mild lung disease, uninfected by PA. Treatment with azithromycin did not result in improved pulmonary function in this relatively healthy patient group. On the other hand, a promising amelioration in secondary endpoints weight and frequency of infectious exacerbations was observed after treatment with azithromcyin. The improvement of pulmonary function, the reduction in exacerbation frequency, and the improvement of quality of life (QOL) shown in the aforementioned trials are often attributed to an anti-inflammatory effect of macrolide treatment [18, 19, 22] . Macrolides have a direct antimicrobial effect, but also modulate many aspects of the immune response. This so-called ‘immune-modulatory effect’ is exclusively demonstrated for 14and 15-membered ring macrolides (EM, clarithromycin, roxithromycin, and azithromycin, respectively) [26–30] . Their effect on both inflammation and bacterial colonization appears to make 14and 15-membered ring macrolides exceptionally suited to contribute to the treatment of chronic inflammatory diseases like COPD and bronchiectasis, having both distortion of the inflammatory D ow nl oa de d by : 54 .7 0. 40 .1 1 11 /1 /2 01 7 12 :2 7: 10 A M Immunomodulatory Effects of Macrolide Antibiotics – Part 2 Respiration 2011;81:75–87 77 response and bacterial infection as key features [31–33] . An overview of the biological mechanisms through which macrolides exert their immunomodulatory effect is provided in part 1 of this series of articles [34] . Considering the increasing popularity of macrolide maintenance therapy, concern regarding possible disadvantages seems appropriate. In the past, macrolides, especially EM, were known for their ability to cause cardiac arrhythmias and hearing loss when administered in high dosages [35, 36] . Finding out whether this also applies to long-term, low-dose macrolide therapy is relevant to clinical practice. In addition, a reduced susceptibility or resistance to macrolides is very likely to develop when these drugs are administered in low dosages during longer periods of time. We performed a literature search for studies exploring the clinical effectiveness of long-term macrolide therapy in COPD, chronic rhinosinusitis, non-CF bronchiectasis, and asthma. We focused primarily on including randomized controlled trials, if available. An additional search was performed for studies about macrolide safety and the emergence of macrolide resistance during long-term macrolide therapy. As none were found, the search was broadened to macrolide safety and resistance in general, with special attention to ototoxicity and cardiac toxicity. Non-CF Bronchiectasis and Long-Term, Low-Dose