Current challenges in treating MRSA: what are the options?

Current challenges in treating MRSA: what are the options? Expert Rev. Anti Infect. Ther. 6(5), 601 618 (2008) Natasha VDV Ratnaraja and Peter M Hawkey Author for correspondence Regional HPA Microbiology Laboratory, Heart of England NHS Foundation Trust, Bordesley Green East, Birmingham, B9 5SS, UK Tel.: +44 121 424 1248 Fax: +44 121 772 6229 peter.hawkey@ heartofengland.nhs.uk This review looks at the challenges facing the worldwide community with the increasing problem of methicillin resistance in Staphylococcus aureus. The epidemiology and natural history of community-associated methicillin-resistant Staphylococcus aureus and the challenge of control is discussed. Options for treatment and review of key antimicrobial agents acting against methicillin-resistant S. aureus, both currently in use and in development, are addressed. There are a number of new agents, the place of which in therapeutic regimens is yet to emerge. The review attempts to inform the reader of the probable position of these agents. Key w o r d s : ceftaroline ceftobiprole community-associated methicillin-resistant Staphylococcus aureus dalbavancin daptomycin iclaprim linezolid methicillin-resistant S. aureus treatment oritavancin quinupristin dalfopristin telavancin tigecycline tomopenem The development of resistance to antimicrobials has been regarded as a consequence of their use since their introduction nearly 70 years ago and is worsening [1]. The incidence of methicillin-resistant Staphylococcus aureus (MRSA) has significantly increased since the first reports emerged in the 1960s. In 2006, the percentage of invasive S. aureus infections in the UK hospital setting due to MRSA was 43% [201]. The European Antimicrobial Resistance Surveillance System (EARSS) annual report for 2006 showed that for 15 out of 31 participating countries, the proportion of invasive S. aureus isolates that, were MRSA was greater than 25%. Both the UK and Ireland were included in those 15 countries. Statistically significant increases in the proportion of S. aureus isolates that were MRSA were also seen in Hungary, Portugal and Malta [201]. A meta-analysis by Cosgrove et al. comparing outcomes of MRSA and methicillin-susceptible S. aureus (MSSA) bacteremia showed that there was a higher mortality associated with MRSA bacteremia compared with bacteremias due to MSSA [2]. It is, therefore, important to try and prevent infections with MRSA and to treat them appropriately when they arise. Previously, MRSA was almost exclusively associated with healthcare facilities; however, MRSA has now begun to emerge within the community in some countries [2,3]. The dissemination of MRSA means that new methods of control of infection need to be sought. This may be difficult in the community setting. Treatment of MRSA infection, both hospital- and community-associated, needs to be effective. The ability to treat in an outpatient setting is of paramount importance to prevent large numbers of patients being admitted to hospital for what are often relatively minor and/or chronic infections. This review will discuss the following: Differences between community-associated MRSA (CA-MRSA) and hospital-associated MRSA (HA-MRSA) and the challenges in overcoming them; New agents to overcome MRSA in both the community and hospital settings. Emerging challenges in the hospital & the community Although occasional infections have previously occurred in the community, acquisition of MRSA has traditionally been associated with healthcare facilities. Infections with MRSA within the community were rare, although sometimes seen among intravenous drug users. An outbreak of CA-MRSA among indigenous Australians in Western Australia between 1989 and 1991 was one of the first reports of an outbreak within the community setting [4], with subsequent reports of similar infections emerging from the South Western Sydney Area Health Service region [5]. Between 1997 and 1999, the US CDC reported four cases of www.expert-reviews.com 10.1586/14787210.6.5.601 2008 Expert Reviews Ltd ISSN 1478-7210 601

Ratnaraja & Hawkey rapidly fatal MRSA infections in children, again occurring within the community [6]. Since the late 1990s, an increasing number of cases of CA-MRSA have been reported worldwide [7], to the extent that it is being considered in some parts of the world to be epidemic [8,9]. Many countries have reported cases of CA-MRSA, including the USA, Europe, New Zealand, Australia, Samoa, Canada and Finland. The USA has particularly been affected, with CA-MRSA (predominantly the USA300 clone but also the USA400 and USA500 clones) comprising 60 75% of all isolates of S. aureus in some regions [10]. There are some notable differences between HA-MRSA and CA-MRSA. Risk factors for the acquisition of HA-MRSA include older age, prolonged hospitalization, antimicrobial therapy, urinary catheterization, diabetes mellitus, intravenous drug use, skin conditions such as eczema and psoriasis, being on hemodialysis or continuous ambulatory peritoneal dialysis, previous surgery and the insertion of invasive devices. The aging population, with more contact with healthcare services and often more antimicrobial use, represents a significant population at risk of colonization and subsequent infection. By contrast, CA-MRSA differs from HA-MRSA in that it tends to affect younger, nonwhite or indigenous populations. These patients usually have no significant prior medical conditions and no prior contact with healthcare settings. Any situation where there is close physical contact may result in outbreaks or clusters of CA-MRSA, for example, within the military, in prisons, among children in institutions, among indigenous populations and in athletic teams [11 13]. Although CA-MRSA is typically seen in the community, transmission within the hospital setting has been described [14,15]. Familial transmission has also been described [16]. There is a diverse range of HA-MRSA infections, ranging from superficial infections to life-threatening bacteremias and endocarditis. The presence of MRSA on prosthetic materials, such as central venous catheters and prosthetic heart valves, may cause dilemmas over treatment due to poor penetration of some anti-mrsa antibiotics and persistence of infection because of an inability to remove some prostheses. Patients may also be asymptomatic, with colonization of the skin, wound or ulcer. It is this group of patients who represent a risk both to themselves and to others, as asymptomatic colonization may go unnoticed, allowing both for infection to develop in a patient and for transmission to close contacts. The clinical presentation of CA-MRSA infection varies, with the most common presentation being skin and soft-tissue infections, such as folliculitis and other pustular lesions [3,8,11,13,17,18]. However, virulence of this pathogen varies and more serious infections, such as necrotizing pneumonia, necrotizing fasciitis, pyomyositis, septic thrombophlebitis, the pelvic syndrome (pelvic abscesses and septic arthritis of the hips) and ocular infections, can occur. There have been reports of bacteremia and endocarditis due to CA-MRSA [12,13,19]. It is the meca gene that is associated with methicillin resistance. This is carried on the staphylococcal cassette chromosome (SCC) mec, a large mobile element found in a range of staphylococci consisting of the mec and ccr gene complexes. There are five main types of SCCmec: type I SCCmec, carrying class B mec and type 1 ccr; type II SCCmec, carrying class A mec and type 2 ccr; type III SCCmec, carrying class A mec and type 3 ccr; type IV SCCmec, carrying class B mec and type 2 ccr; and type V SCCmec, carrying class C2 mec and type 5 ccr [20]. In contrast to HA-MRSA, which is commonly associated with types I, II and III SCCmec, CA-MRSA is usually associated with the smaller, types IV and V SCCmec [3,7,18,21]. However, there have been reports of CA-MRSA strains harboring SCCmec types I, II or III [22,23] and a report of HA-MRSA strains containing SCCmec type IV [24]. This suggests that the differences between HA-MRSA and CA-MRSA are not as clear-cut as previously thought. There is a strong association between CA-MRSA strains and the Panton Valentine leukocidin (PVL) toxin [25 27], with many strains (up to 80%) containing the gene encoding this [10]. The PVL toxin is encoded by two genes, luks-pv and lukf-pv, and has previously been seen with some strains of MSSA. The current rise in PVL-producing CA-MRSA strains may be due to de novo SCCmec elements spreading among known PVLproducing CA-MRSA, as suggested by Monecke et al. [28], or it could be due to PVL phages disseminating among strains of MRSA. Multilocus sequence typing has shown that there are discreet clonal complexes containing PVL-positive MRSA (both HA-MRSA and CA-MRSA) [25,26]. Clonal expansion and diversification within a clonal complex can produce differences in virulence and pathogenicity, as shown by Kennedy et al. in the USA300 clone [29]. Diep et al. showed through genomic sequencing that the USA300 clone had horizontally acquired the type I arginine catabolic mobile element (ACME) from Staphylococcus epidermidis [30]. It is thought that type I ACME promotes both the growth and survival of, and enhances colonization of, human skin, with USA300. Type I ACME is physically linked to type IV SCCmec, suggesting that there may be a link between pathogenicity and antimicrobial resistance. Dissemination of the USA300 clone from the community into the hospital setting may be aided by both the presence of type I ACME and type IV SCCmec [31,32]. Historically, CA-MRSA strains tend to be more sensitive to antimicrobials than HA-MRSA strains [3,7,33,34]. Strains of CA-MRSA tend to retain susceptibility to non-β-lactam antimicrobials, such as trimethoprim/sulfamethoxazole, clindamycin, gentamicin and tetracyclines. However, recent reports have suggested that multidrug-resistant strains are emerging. Diep et al. described a high incidence of multidrug-resistant CA-MRSA caused by clone USA300 in San Francisco, CA, USA [35]. Men who have sex with men (MSM) sex appeared to be a risk factor for acquisition, suggesting that sexual transmission of this strain may occur. These new data are worrying, as we could be seeing the rise of multidrug-resistant, virulent and easily transmissible strains of CA-MRSA to the epidemic proportions seen with HA-MRSA. Without the development of new agents, treatment options will be limited for a condition that was often previously 602 Expert Rev. Anti Infect. Ther. 6(5), (2008)

Current challenges in treating MRSA Review treated within the community with oral antibiotics [36]. Clonal diversity and transmission will mean that there will no longer be any real distinction between HA-MRSA and CA-MRSA. The increasing prevalence of CA-MRSA is of concern, not only because of its transmissibility but also because currently there are no effective decolonization strategies for it [37]. Typical decolonization regimes used to eradicate HA-MRSA have not been studied extensively for use in CA-MRSA; however, the utility of mupirocin in such regimens may be limited by the emergence of CA-MRSA strains with high-level resistance to mupirocin, especially in subtypes of the USA300 clone, as described by Han et al. at a health center in Boston in 2007 [38]. The same study also showed clindamycin- and tetracycline-resistant USA300 clones. A 2008 Cochrane review of antimicrobial drugs for treating HA-MRSA found insufficient evidence to support use of topical or systemic antimicrobial therapy for eradicating nasal or extranasal MRSA [39]. Due to a lack of clinical trials, currently there is not enough evidence to support decolonization therapy for CA-MRSA. With repeated skin infections, increased risk of shedding of the organism from infected skin coupled with a lack of an effective decolonization regimen could mean persistence and/or transmission of MRSA within the community. The organism can be present on fomites such as personal possessions and towels, again resulting in transmission to other individuals. Inadequate hand hygiene can also facilitate transmission. Several members of a household may be colonized/ infected, making eradication or reduction of MRSA within the community very difficult. Decolonization can only work if there is an effective screening policy in place. Rapid tests are now available that allow for the detection of MRSA in carriers on admission to healthcare facilities, enabling prompt patient isolation and the institution of decolonization regimes. Currently, there is an overall lack of supportive evidence for universal screening [40]. The reason for this may be multifactorial, with cost, availability and implementation of isolation/ decontamination and timing of screening all being important considerations. It is difficult to control for so many variables in a study. There are three important limitations of screening. First, MRSA decolonization therapy is not wholly effective (50 60% effective for long-term clearance) and long-term and recurrent therapy, as well as mass use, with mupirocin may lead to resistance. Second, isolation and barrier nursing is essential in the prevention of transmission of MRSA. Current UK guidelines recommend that in new hospitals, at least 50% of all beds should be in side rooms with en suites to prevent transmission of healthcare-associated infections (HCAIs) [41]. However, many hospitals in the UK are old, with nowhere near this number of side rooms available. A similar situation exists in hospitals in many other countries. The additional problems of Clostridium difficile and other multidrug-resistant organisms means prioritization of side rooms for patients with the most transmissible conditions. Understaffing of nursing and medical personnel, bed occupancy often above the recommended 85% and patients with increasingly complex medical conditions all act to impede successful barrier nursing and strict infection-control practices. Eradication of MRSA can only be successful if there are adequate resources to implement safe practices. Third, while PCR is an effective and time-saving method of screening for MRSA, its cost means that it may not be suitable for mass screening. Once universal screening is in place, no doubt studies will be undertaken to determine if this is a costeffective exercise and a useful tool in the fight against MRSA. The problem of CA-MRSA may mean that screening will need to be extended to household contacts of known carriers. This will have huge cost implications as well as presenting practical issues of screening so many patients. Studies will need to be undertaken to see if there is any benefit to be obtained though screening this population. It is likely that the controversy surrounding policies of universal MRSA screening as a tool to help reduce rates of MRSA infection and transmission may persist until further robust evidence is available. Other approaches to prevent MRSA infection in the future include the development of a vaccine against staphylococci. Problems encountered with developing a vaccine include a failure of whole-cell live or killed vaccines, as well as conjugate polysaccharide vaccines, to elicit an adequate and long-lasting immune response [42,43]. The large number of virulence factors possessed by the organism also hinders the development of an effective vaccine, with single immunological targets having limited use for this purpose [44]. Despite these limitations, the widespread problem of MRSA means that there continues to be work into the development of a suitable and long-lasting vaccine [44,45]. The many problems surrounding MRSA mean that newer strategies are required, both for the rapid diagnosis and for the successful management of both CA-MRSA and HA-MRSA, to prevent further dissemination into both the community and hospital environments in epidemic proportions. New agents to overcome MRSA Traditionally, there has been a limited antimicrobial armamentarium for use against MRSA, with glycopeptides being the main agents used. Known shortcomings of glycopeptides include poor tissue penetration, poor oral bioavailability and a narrow therapeutic window, necessitating the monitoring of serum drug levels. More recently, there have been reports of increasing MICs over time, a phenomenon termed vancomycin creep [46]. Although the MICs remain within the susceptible range, the higher MICs may be associated with higher rates of clinical failure [47]. The British Society for Antimicrobial Chemotherapy Bacteraemia Resistance Surveillance Programme has recently reported a rise in vancomycin MICs for MRSA at various centers in the UK and Ireland [48]. Problems in detecting this creep using routine methods of antimicrobial susceptibility testing (e.g., disk diffusion and automated testing) means that this phenomenon may go unnoticed [49]. There have been rare reports of vancomycin-resistant S. aureus (VRSA), glycopeptide-intermediate S. aureus (GISA) and heterogeneous glycopeptide-intermediate S. aureus (hgisa) strains [50] www.expert-reviews.com 603

Ratnaraja & Hawkey and it seems possible that the increase in MICs may be associated with the emergence of these strains. In the case of hgisa, glycopeptide therapy kills the dominant glycopeptide-susceptible population, allowing the glycopeptide-resistant subpopulation to predominate. hgisa infection is associated with a poorer outcome than MRSA infection [51]. A recent report from Detroit (MI, USA) has shown a significant increase in the prevalence of heterogeneous vancomycin-intermediate S. aureus (VISA) over a 20-year period (from 2.27% between 1986 and 1993 to 8.2% between 2003 and 2006) [52]. This time period correlates with the start of the MRSA epidemic and the subsequent increase in vancomycin use. Glycopeptides are inadequate to treat these strains and the development of alternative, effective agents is crucial for the management of VRSA, GISA and hgisa infections. In the 1990s, concern over the potential of MRSA to increase to endemic levels and the known shortcomings associated with glycopeptides encouraged the development of new agents to overcome this problem. We are now beginning to see the results of this research. These new agents have improved efficacy against MRSA, with greater tolerability and/or bioavailability than vanco mycin, which may help in reducing the mortality and morbidity associated with these resistant strains. A slow rate of emergence of resistance and lack of equally cheap, alternative agents has meant that vancomycin has been the most commonly used drug for the treatment of MRSA infections. However, there are a number of shortcomings regarding vancomycin. First, it may not have sufficient tissue penetration and may not achieve adequate levels in cerebrospinal fluid (CSF) or in hte lung. In mechanically ventilated patients, it has been shown that concentrations of vancomycin in epithelial lining fluid of the lung is only 14 16% of that in serum [53,54]. Continuous infusion of vancomycin also shows similar limitations; Moise-Broder et al. showed that in patients undergoing cardiac surgery, tissue-toplasma ratios of vancomycin were only 0.3 in non diabetic patients and only 0.1 in diabetic patients [55]. In CSF, these levels are enhanced by the presence of inflammation; however, a recent study has shown that adequate levels of vancomycin may be achieved when concomitant steroids are given [56]. Nevertheless, there remains some concern about the reliability of vancomycin in the treatment of CNS infections and pneumonia. There is also a small but increasing number of strains of MRSA that have either full or intermediate susceptibility to vanco mycin (VRSA, hgisa and VISA) [57,58]. In the case of VRSA, it is thought that the organism acquires the vana resistance operon from vancomycin-resistant enterococci (VRE) in the presence of concurrent infection with these organisms [59]. VISA infections tend to occur with prolonged vancomycin therapy and are associated with resistant bacteria which possess thick cell walls [60]. As discussed, hgisa has been described, involving a small subpopulation that has a MIC of vancomycin within the intermediate or resistant range [61]. Vancomycin tolerance can also be seen. This is a state where the bacteria are suppressed but not killed by vancomycin therapy. This can be determined in the laboratory by a minimum bactericidal concentration:mic ratio of greater than 32 [62]. The shortcomings of vancomycin means that newer agents are needed to overcome MRSA. The problem of developing resistance among strains of CA-MRSA means that treatment options within the community are becoming limited. Over recent years, a number of novel agents have been marketed, with other agents also being developed. These agents vary in their pharmacodynamic and pharmacokinetic profile, as well as tissue penetration and distribution. This allows different infections at a range of body sites to be treated, both in the in-patient and outpatient settings. Linezolid (Zyvox, Pharmacia/Pfizer, Inc.) This oxazolidinone, marketed by Pfizer, is the first of its class; by binding at the 23S subunit of the 50S ribosome to prevent formation of the 70S ribosomal complex, it inhibits the initiation of protein synthesis. It is bacteriostatic against MSSA, MRSA, coagulase-negative staphylococci (CoNS), enterococci (including VRE strains) and streptococci (including penicillin-resistant and multidrug-resistant pneumococci). It demonstrates in vitro activity against some Gram-negative organisms (Bacteroides spp., Moraxella catarrhalis and Pasteurella spp.) but is not generally thought to be efficacious against most other Gram-negative bacteria due to its elimination via efflux pumps [63]. Linezolid has antianaerobic activity, being active against C. difficile, Clostridium perfringens, Bacteroides and Fuseobacterium spp. Linezolid was first approved for use in the UK in 2001 and is currently licensed for pneumonia and skin and soft-tissue infections (both complicated and noncomplicated) caused by Gram-positive organisms. It has been shown to be at least as efficacious as (and superior to in many studies) vancomycin for the treatment of hospital-acquired and nosocomial pneumonia [63 67]. Linezolid is not currently licensed in the UK for the treatment of MRSA bacteremia; however, there have been reports of successful treatment of this condition with linezolid, either used alone or in combination with rifampicin or fusidic acid [68,69]. Schorr et al. pooled and analyzed five randomized trials comparing linezolid with vancomycin for the treatment of secondary S. aureus bacteremia and found it to be noninferior for this indication [70]. It has excellent tissue penetration. Linezolid has reasonable penetration into the CSF [71] and studies have shown it to be adequate for the treatment of CSF shunt infections caused by CoNS and other Gram-positive organisms [71,72]. There have been reports of both treatment failures (monotherapy) [73] and successes (linezolid used alone and in combination with either rifampicin, gentamicin, fusidic acid or amikacin) [74 76] for MRSA endocarditis associated with linezolid use. Further data are required regarding the use of a bacteriostatic agent for the treatment of endocarditis, as bactericidal drugs are preferential treatment for this condition. Linezolid has 100% bioavailability at the standard dose of 600 mg twice daily, making it an effective oral as well as intravenous agent against MRSA. After only 1 2 h of oral dosing, maximal plasma levels are achieved. The half-life is 4.26 5.4 h [77]. The drug is distributed well into tissues, demonstrating only 31% protein binding [78,79]. However, reversible bone 604 Expert Rev. Anti Infect. Ther. 6(5), (2008)

Current challenges in treating MRSA Review marrow suppression may occur as a side effect, most commonly following prolonged therapy (greater than 14 days) [78,80], and necessitates regular monitoring of full blood counts. If myelosuppression occurs, discontinuation of the drug should be considered. Peripheral and optic neuropathies have also been described in association with treatment durations greater than 28 days. Lactic acidosis is a rare but serious side effect, not associated with either duration or dosage of this agent. Linezolid is also a weak reversible, nonselective monoamine oxidase inhibitor and has been associated with the serotonin syndrome [78,81 83]. Concomitant therapy with other drugs should be reviewed carefully before prescribing this antibiotic. Linezolid is excreted mainly via the urine (with 30% of the drug excreted unchanged), as well as nonrenal routes. It does not require therapeutic drug monitoring and there is no need for dosage adjustment in renal or hepatic impairment [71]. In 2007, Pfizer and the US FDA released an alert because of new safety concerns following the results of an as yet unpublished openlabel, randomized trial comparing linezolid with comparator antibiotics (vancomycin, oxacillin and dicloxacillin) for the treatment of critically ill patients with intravascular catheter-related bloodstream and catheter-site infections. The study showed that treatment with linezolid resulted in higher mortality compared with the comparator antibiotics. This difference in mortality was not seen in pure Gram-positive infections, but was seen in those patients with pure Gram-negative infections, in those with Gram-negative/ Gram-positive infections and in those treated with linezolid who did not have an infection at the time of entering the study. The new safety alert has stated that linezolid is not approved for the treatment of catheter- related bloodstream infections, catheter-site infections, or for the treatment of infections caused by Gram-negative bacteria [202]. Linezolid should not be used as monotherapy in mixed MRSA and Gram-negative infections. As with clindamycin, linezolid has been shown to reduce the production of Panton Valentine leukocidin (PVL), α-hemolysin and toxic-shock-syndrome toxin [84], making it an important agent in the treatment of Gram-positive causes of toxic shock and also of necrotizing fasciitis and pneumonia. Resistance to linezolid has been described in VRE [85,86], with the mechanism being a G2576U ribosomal mutation at the binding site [87]. Rare cases of linezolid-resistant strains of S. aureus emerging during therapy have been reported [88], and an intrinsic linezolid-resistance gene in a strain of MRSA has also been described. Transferable resistance has also been noted in one case [89]. Nevertheless, overall, the high activity of linezolid (MIC 90 2 mg/l) means that it remains an excellent choice in the treatment of MRSA infections. Quinupristin dalfopristin (Synercid, King Pharmaceuticals) This streptogramin consists of a 30:70 ratio of two semisynthetic pristinamycin derivatives, quinupristin and dalfopristin. This combination allows for synergistic inhibition of bacterial protein synthesis at the 50S ribosome. It is bactericidal against MRSA, MSSA and Streptococcus pyogenes but is bacteriostatic against vanco mycin-resistant Enterococcus faecium (VREF); most strains of Enterococcus faecalis are inherently resistant to it [82,83]. The half-life of synercid is 0.70 0.85 h [90]. It is approved for use in the UK for serious Gram-positive infections, including hospital-acquired pneumonia, skin and soft-tissue infections and infections due to VREF. Quinupristin dalfopristin is occasionally used for the treatment of VREF bacteremia but its use in the treatment of endocarditis and pneumonia is limited by concerns over its safety profile and potential drug interactions. Resistance to quinupristin dalfopristin in MRSA is rare [87]. The SENTRY antimicrobial surveillance program has reported resistance to quinupristin dalfopristin in 10.0% of all E. faecium isolated in Europe and in 0.6% of all E. faecium isolated in North America [91]. A further limitation of quinupristin dalfopristin is that it has poor bioavailability and can only be given by intravenous infusion through a central vein [79,83], precluding its use in less serious infections in patients who otherwise would not need a central venous catheter. The drug is mainly eliminated through bile in the feces. Renal impairment may reduce clearance. Quinupristin dalfopristin has a considerable toxicity profile, including hepatotoxicity, hyper bilirubinemia, myalgias, arthralgias and, in up to 74% of patients, thrombophlebitis [79,81,83]. The British National Formulary recommends dose reduction in moderate hepatic impairment and avoidance in severe hepatic impairment or if the plasma bilirubin concentration is or becomes greater than three-times the upper limit of reference range [92]. It is also contraindicated in breastfeeding. Quinupristin dalfopristin inhibits the activity of cytochrome P450 3A4 and, thus, can interact with other drugs, including ciclosporin. Quinupristin dalfopristin does not itself induce QTc prolongation but can interfere with the metabolism of other drugs that may cause this, necessitating close monitoring. The presence of other agents with greater efficacy, less troublesome side effects and greater ease of administration means that quinupristin dalfopristin is not usually considered to be a first-line agent for MRSA infections. Daptomycin (Cubicin, Cubist Pharmaceuticals) Daptomycin is a semisynthetic cyclic lipopeptide derived from the fermentation products of Streptomyces roseosporus. This rapidly bactericidal antibiotic has a novel mechanism of action, requiring physiological levels of Ca 2+. The lipophilic tail of daptomycin inserts into the bacterial cell membrane. This results in rapid membrane depolarization and potassium efflux, which stops DNA/RNA and protein synthesis. Daptomycin has a long postantimicrobial effect and its activity is concentration dependent [93 95]. The half-life of daptomycin is 7 9 h [96]. The current recommended doses are 6 mg/kg once daily for bacteremia and endovascular infections and 4 mg/kg once daily for complicated skin and skin-structure infections (csssis). Daptomycin is only effective against Gram-positive organisms, including MRSA, MSSA and VRE. It is also active against strains of linezolid- and Synercid (quinopristin dalfopristin)-resistant strains of VRE [93,94,97,98]. www.expert-reviews.com 605

Ratnaraja & Hawkey Daptomycin is efficacious in, and currently indicated for use in csstis [99,100,101]. Fowler et al. recently published the results of an open-label trial that showed a nonsignificant increase in successful treatment of bacteremia and endocarditis (including right-sided cases) due to MRSA and MSSA [102]. The authors used gentamicin plus either an antistaphylococcal penicillin or vancomycin as comparator agents. Many of the cases of persistent or relapsed bacteremia were attributed to complicated bacteremia associated with the presence of indwelling devices and/ or osteomyelitis. The findings of a systematic review by Falagas et al. in 2006 confirm that further studies on the use of daptomycin for bacteremia and/or endocarditis are warranted [103]. Microbiological failures have been described during prolonged therapy with daptomycin and were associated with an increase in MIC during the treatment phase [98,104,105]. The underlying resistance mechanism for this is poorly understood. Daptomycin is inhibited by pulmonary surfactants, precluding its use in pneumonia [106]. It has been successfully used for the treatment of bone and joint infections, although clinical trials are lacking [98,107]. Adverse effects include a dose-dependent rise in creatinine kinase, which may be accompanied by myopathy, and derangement of hepatic transaminases [100,102]. Dose adjustment is required for renal impairment. There are insufficient data regarding the use of daptomycin in pregnancy or breastfeeding. It has been shown that the MIC of daptomycin to MRSA increases in cases of prior exposure of MRSA to vancomycin. It is thought that heterogenous vancomycin resistance develops, producing a thickening of the bacterial cell wall that impedes the entry of the large daptomycin molecule into the organism [108 110]. However, Wooton et al. demonstrated that the bactericidal activity of daptomycin is not significantly affected by this phenomenon [109]. Daptomycin has a relatively low toxicity profile and its once-daily dosing means that it has a place in prolonged and/or outpatient therapy. However, in mixed Gram-negative and -positive infections, another agent with a broader spectrum may be more appropriate. Tigecycline (Tygacil, Wyeth Pharmaceuticals) Tigecycline, a derivative of minocycline, is a bacteriostatic glycylcycline, a new class of antimicrobial agent. Modification of the side chain enhances binding to the 30S ribosomal subunit, with subsequent inhibition of protein synthesis. This modification also makes it less susceptible to the Tet(A-E) and Tet(K) efflux pumps, the latter of which are predominantly associated with resistance to tetracycline in staphylococci [111,112]. Tigecycline has a half-life of 36 h and less than 15% of the unchanged drug is excreted in the urine. It exhibits extensive protein binding (~68%) that increases with increasing dose and has a large volume of distribution (>10 l/kg). The recommended dose of tigecycline is 100 mg then 50 mg intravenously twice daily [113]. Tigecycline has a broad spectrum of activity, being active against MRSA, penicillin-resistant pneumococci, vancomycin-susceptible and -resistant enterococci, b-hemolytic streptococci and Listeria monocytogenes, as well as tetracycline-resistant strains of MRSA. It has the advantage over some of the newer anti-mrsa agents of having Gram-negative activity, notably against Haemophilus influenzae, Neisseria spp., Acinetobacter spp. (including multidrugresistant strains) and Enterobacteriacae [112 114]. It has variable activity against Providencia and Burkholderia spp. but tends to be active against other nonfermenting Gram-negative organisms. In Pseudomonas aeruginosa and Proteus spp., resistance mainly occurs via MexAB-OprM and AcrAB efflux pumps, respectively [115 117]. It also shows good in vitro activity against some atypical pathogens, such as Mycobacterium chelonae, Mycobacterium abscessus, the Mycobacterium fortutium group, Mycobacterium marinum, Mycoplasma spp. (Mycoplasma hominis and Mycoplasmapneumoniae) and Ureaplasma urealyticum [118]. Tigecycline is currently indicated for the treatment of csssis and complicated abdominal infections caused by multiple-antibacterial-resistant organisms. It has been shown to be an efficacious drug for the treatment of both skin and skin-structure infections and intra-abdominal infections, and is as efficacious as vancomycin/aztreonam and imipenem cilastin in randomized, double-blind studies [119 122]. Trials comparing tigecycline and meropenem for the treatment of intra-abdominal infection are ongoing, as are trials for the treatment of bloodstream infections. Early results of a Phase III multicenter, double-blind trial comparing tigecycline against imipenem cilastin for the treatment of hospital-acquired pneumonia showed that it did not achieve noninferiority in ventilator-associated pneumonia [123]. Further studies are warranted to assess its utility in the treatment of hospital-acquired and ventilator-associated pneumonia. To date, tigecycline has not been associated with any significant adverse effects, dose-dependent nausea and vomiting being the most common [113]. Vouillamoz et al. recently showed that no adverse interactions occur when tigecycline is used in combination with other antimicrobials, which may make it a useful drug in the treatment of mixed multidrug-resistant infections [124]. To date, there have not been any reports of tigecycline resistance in MRSA isolates. Ceftobiprole (Zeftera, Johnson & Johnson Pharmaceuticals/Basilea Pharmaceutica) This novel broad-spectrum bactericidal cephalosporin has been shown to be effective against MRSA. It is currently the most studied of its class for activity against this organism. It is highly active against MRSA, with a high affinity for PBP2 [125 130]. It also has a high affinity for PBP2x, the penicillin-binding protein which is associated with resistance in pneumococci [130,131]. It is available in intravenous formulation and has a half-life of 3 4 h. The dosing regimen is 500 mg three-times daily. Conversion of the water-soluble prodrug, ceftobiprole medocaril, to the active drug, ceftobiprole, in plasma is rapid. It distributes into the extracellular compartments. The drug is excreted renally, almost entirely unchanged, necessitating dose adjustment in renal impairment [125,128,130]. Ceftobiprole is potent against MRSA, VISA and penicillinresistant pneumococci, with poor activity against E. faecium. However, unusually for a cephalosporin, it has bactericidal 606 Expert Rev. Anti Infect. Ther. 6(5), (2008)

Current challenges in treating MRSA Review Table 1. Profile of new and current agents for the treatment of MRSA. Drug name Class Bactericidal or bacteriostatic for MRSA Site of action Linezolid Oxazolidinone Bacteriostatic 23S subunit of 50S ribosome Half-life (h) Stage of development 4 5 Currently licensed for use worldwide Synercid Streptogramin Bactericidal 50S ribosome 0.7 0.8 Currently licensed for use worldwide Daptomycin Cyclic lipopeptide Bactericidal Cell membrane 7 9 Currently licensed for use worldwide Tigecycline Glycylcycline Bacteriostatic 30S ribosome 36 Currently in use, ongoing clinical trials for HAP/VAP Ceftobiprole Ceftaroline Dalbavancin Oritavancin Novel cephalosporin (β-lactam) Novel cephalosporin (β-lactam) Secondgeneration glycopeptide Secondgeneration glycopeptide Bactericidal PBP (especially PBP2 and PBP2x) 3 4 Under review by regulatory authorities in Canada, the EU and Switzerland Resistance mechanism G2576U ribosomal mutation at binding site Streptograminresistance genes vatd, vate and vgba and the macrolideresistance gene ermb (E. faecium) Possibly thickening of bacterial cell wall with vancomycin therapy No reports of resistance so far Undefined at present Bactericidal PBP2 0.6 2.4 Phase III trials Undefined at present Bactericidal Bactericidal Bacterial cell wall Bacterial cell membrane Telavancin Lipoglycopeptide Bactericidal Bacterial cell wall and membrane 123 210 US FDA fast-track status Phase III trials Not described Clinical use csssis including diabetic foot infections due to MRSA and uncomplicated skin and skin-structure infections; HAP/CAP; infections associated with vancomycinresistant Enterococcus faecium csssis; HAP, vancomycin-resistant E. faecium bacteremia csssis; right- and left-sided endocarditis; staphylococccal bacteremia csssis; complicated abdominal infection HAP/VAP; csssis csssis; HAP/VAP csssis 100 Phase III trials Not described csssis 7 11 Phase III trials Not described csssis CAP: Community-acquired pneumonia; csssi: Complicated skin and skin-structure infection; HAP: Hospital-acquired pneumonia; MRSA: Methicillin-resistant Staphylococcus aureus; PBP: Penicillin-binding protein; VAP: Ventilator-associated pneumonia. www.expert-reviews.com 607

Ratnaraja & Hawkey Table 1. Profile of new and current agents for the treatment of MRSA (cont.). Drug name Class Bactericidal or bacteriostatic for MRSA Site of action Iclaprim Diaminopyrimidine Bactericidal Bacterial folate pathway Ranbezolid Novel oxazolidinone Bacteriostatic 23S subunit of50s ribosome Half-life (h) Stage of development 4 Fast-track status Phase III trials for csssis, Phase II trials for HAP/VAP and intravenous-tooral switch Not available Completed Phase I trials Resistance mechanism Resistance is rare; mechanism has not been described Target modification Clinical use csssis; HAP Intravenous catheterrelated infections Tomopenem 1β methylcarbapenem Bactericidal PBPs 1, 2 and 4 2 Phase II trials Not described As yet undetermined CAP: Community-acquired pneumonia; csssi: Complicated skin and skin-structure infection; HAP: Hospital-acquired pneumonia; MRSA: Methicillin-resistant Staphylococcus aureus; PBP: Penicillin-binding protein; VAP: Ventilator-associated pneumonia. activity against E. faecalis (both vancomycin-resistant and -susceptible strains) at therapeutically achievable concentrations. Its broad spectrum means that it is active against many Gramnegative pathogens, including H. influenzae, Neisseria gonorrhoeae, Moraxella catarhalis and extended-spectrum-β-lactamase (ESBL)-negative Enterobacteriaceae. It is also active against nonfermenting Gram-negative pathogens, such as P. aeruginosa and Acinetobacter baumannii. Ceftobiprole is not stable to metalloβ-lactamases and is hydrolyzed by ESBL-producing organisms. It has some activity against Gram-positive (Actinomyces, Clostridium, Fuseobacterium and Veillonella spp., among others) but not Gram-negative anaerobes [125 129,132,133]. Ceftobiprole has been granted fast-track status by the FDA for the indications of healthcare-associated pneumonia and csssis. Two multicenter, randomized, double-blind trials have recently been completed. The first compared ceftobiprole with vancomycin plus ceftazidime for the treatment of csssis [134] and the second compared ceftobiprole with vancomycin for the treatment of csssis caused by Gram-positive bacteria [135]. Both studies showed noninferiority against the comparator antimicrobial regimens. Nausea and taste disturbances were the most common symptoms reported. A recent Phase III trial comparing ceftobiprole with ceftazidime plus linezolid showed noninferiority in both the clinically evaluable and intention-to-treat patient groups for the treatment of hospital-acquired pneumonia; however, noninferiority was not achieved in the ventilator-associated pneumonia subgroup [203]. Ceftobiprole has obtained regulatory approval from Health Canada for the treatment of csssis, including diabetic foot infections, and is currently being reviewed by regulatory authorities in the EU and Switzerland. The broad spectrum of activity of ceftobiprole makes it a welcome addition to the armamentarium of agents that can overcome MRSA. However, the three-times daily infusions, each lasting 1 2 h, and lack of oral formulation, may make it less amenable to use in outpatient therapy. Ceftaroline (Forest Laboratories, Inc.) Ceftaroline is another novel cephalosporin that is currently under development. It is administered clinically as an N-phosphono prodrug. As with ceftobiprole, it exhibits a high affinity for PBP2 and has potent activity against MRSA and vancomycin-resistant staphylococci, as well as many streptococci and Enterobacteriaceae [136 138]. It has variable activity against E. faecalis and minimal activity against anaerobes and nonfermenting bacteria (especially P. aeruginosa). Mustaq et al. found that this drug may be susceptible to hydrolysis by classical TEM and SHV β-lactamases as well as ESBLs, although this is reversible with clavulanate [136]. Owing to its poor bioavailability, ceftaroline must be given intravenously [79]. It has a half-life of 0.6 2.4 h, necessitating frequent dosing [79,139]. The optimal dosing regimen has yet to be determined. Experimental in vivo studies using a rabbit model have been performed, showing that ceftaroline is more efficacious than vancomycin or linezolid in the treatment of endocarditis. It was also shown to be the most efficacious agent against a strain of GISA [139]. A multicenter Phase II trial recently showed noninferiority of ceftaroline versus vancomycin plus aztreonam for csssis [140,141]. Phase III trials are in progress, including one comparing ceftaroline with ceftriaxone for the treatment of community-acquired pneumonia. Early results from this trial look promising, showing some superiority over ceftriaxone [142]. Earlier this year, Novexel and Forest Laboratories announced an agreement to develop Novexel s novel intravenous β-lactamase inhibitor, NXL 104, with Forest s ceftaroline [204]. Phase I trials of this ceftaroline/nxl 104 combination are expected to start in 2009. If successful, this combination may eliminate the major drawback associated with this otherwise promising antimicrobial. Dalbavancin (Zeven, Pfizer, Inc.) Dalbavancin is a semisynthetic derivative of A-40926, a teicoplanin-like glycopeptide. It is a bactericidal second-generation glycopeptide, formed by modifying the functional groups and 608 Expert Rev. Anti Infect. Ther. 6(5), (2008)

Current challenges in treating MRSA Review sugar moieties of A-40926 but maintaining the d-alanyl-d-alanine binding site. Dalbavancin acts by disrupting the formation of the bacterial cell wall. Dalbavancin has a unique pharmacokinetic profile with a half life of 123 210 h, which allows effective once-weekly dosing. Because of poor oral bioavailability, it is only available in an intravenous formulation. Dalbavancin has a volume of distribution of 10 l, high protein binding (>95%) and a systemic clearance of 0.05 l/h. A total of 33% of the intact drug is excreted via the urine [143,144]. Following administration of a first dose of 1 g, the weekly dose of dalbavancin is 500 mg. It has a similar spectrum of activity to teicoplanin and other glycopeptides, being active against staphylococci (MSSA, MRSA and CoNS) and enterococci (including vanb- and vanc- but not vana-positive strains). Dalbavancin is potent against GISA, linezolid-resistant strains of S. aureus, penicillin-resistant and -susceptible strains of pneumococci and viridans streptococci [144,145]. In vitro MIC 90 values for dalbavancin against MRSA and MSSA isolates, and methicillin-resistant and -sensitive isolates of CoNS, are similar or lower than that for vancomycin or teicoplanin [146]. Dalbavancin is also active against Gram-positive anaerobes, Clostridium spp. and many fastidious aerobes. It is inactive against Clostridium clostridioforme and some Lactobacillus spp. and lacks any activity against Gram-negative organisms [145]. Dalbavancin has been shown in vitro to have superior activity against MSSA and MRSA compared with vancomycin [147]. It has been shown to be as efficacious as linezolid for the treatment of csssis (including those due to MRSA) when given as a 14-day course; clinical success was seen in 90% of the dalbavancin arm and 92% of the linezolid arm at the test-of-cure visit in the 2005 randomized, double-blind study by Jauregui et al. [148]. In 2005, Raad et al. published the results of a Phase II openlabel randomized controlled multicenter study of 75 patients with catheter-related bloodstream infections caused by Gram-positive organisms, including MRSA, MSSA, CoNS and enterococci [149]. They compared a 14-day course of treatment of dalbavancin with twice-daily vancomycin. Treatment with dalbavancin was associated with a significantly higher success rates than treatment with vancomycin. Dalbavancin has only been associated with mild side effects, mainly headaches and gastrointestinal symptoms [144]. Hypokalemia and, rarely, hypotension have been reported [149]. Currently, there have been no reports of an association between dalbavancin use and C. difficile-associated diarrhea. Dalbavancin is not a substrate for the cytochrome P450 system and there are no known significant drug interactions [145]. The unique pharmacological profile and excellent safety profile make dalbavancin an attractive option for the treatment of MRSA infections, especially in the outpatient setting. In the long run, it may lead to significant savings due to shorter in-patient stay. Oritavancin (Targanta Therapeutics) Oritavancin is modified from vancomycin by substitution of the vancosamine of the disaccharide moiety by an epivancosamine [83,151]. It acts by disrupting transmembrane potential and is rapidly bactericidal, with concentration-dependent activity and a postantibiotic effect [82,150]. Oritavancin has a half-life of approximately 100 h [84]. Similar to vancomycin, oritavancin is active against many Gram-positive organisms, including MRSA, glycopeptide-resistant enterococci (including vana-positive strains), streptococci, Pepostreptococcus spp., Propionibacterium spp., Clostridium perfringens and Corynebacterium jeikeium, but is more potent and consistently exhibits lower MICs of less than 1 mg/l. It is active against penicillin-intermediate and -resistant strains of Streptococcus pneumoniae and is unaffected by methicillin resistance in both coagulase-positive staphylococci and CoNS. Oritavancin has no activity against Gram-negative organisms, including anaerobes [150 154]. It has been shown that in bacteremia, the percentage of time that the free oritavancin is above the MIC is important for the microbiological response [152,155]. Oritavancin achieves good concentrations in both plasma and blister fluid, making it a suitable candidate for the treatment of csssis [154]. It is slowly eliminated from the body, taking 7 days for 6% of a single dose to be eliminated. Oritavancin has a long half-life at approximately 100 h [153] and it is expected that the recommended dosing regimen will be once daily or once on alternate days. This will enable it to be used in the outpatient setting. Oritavancin has been shown to be as efficacious as either vancomycin or a β-lactam for the treatment of S. aureus bacteremia [156]. Unpublished Phase III trial data have shown equivalence with vancomycin and cephalexin for the treatment of csssis [157,158]. After a delay in the submission by Intermune for a new drug application for oritavancin due to adverse events (rash and phlebitis), further Phase III trials are ongoing. Resistance to oritavancin has not been demonstrated among strains of S. aureus, including VISA, although reduced susceptibility to oritavancin has been seen in vitro among VanA and VanB strains of enterococci [82,150]. If safety concerns are eliminated, oritavancin may be a promising alternative to traditional glycopeptides for the treatment of MRSA infection. Telavancin (Theravance, Inc.) Being a semisynthetic analog of vancomycin, telavancin is a rapidly bactericidal lipoglycopeptide. It blocks the transpeptidation and transglycosylation steps involved with peptidoglycan chain formation. It also acts directly on the bacterial cell membrane, changing its permeability [79,82,151,159]. Telavancin shows enhanced activity against streptococci (including penicillin-resistant and -sensitive strains of S. pneumoniae) compared with vancomycin and is active against enterococci, including vana-positive organisms [82,151,159 163]. It is also more potent against MRSA, MSSA (albeit to a lesser extent compared with that shown against streptococci) and GISA compared with vancomycin and linezolid [163]. This rapidly bactericidal drug that exhibits concentrationdependent killing has a half-life of 7 11 h and a long postantibiotic effect [79,163]. The recommended dose is 10 mg/kg intravenously once daily. It has poor oral bioavailability and so must be given intravenously. www.expert-reviews.com 609