Role of Combination Antimicrobial Therapy for Vancomycin-Resistant Enterococcus faecium Infections: Review of the Current Evidence

Role of Combination Antimicrobial Therapy for Vancomycin-Resistant Enterococcus faecium Infections: Review of the Current Evidence Juwon Yim, 1a Jordan R. Smith, 2 and Michael J. Rybak 1,3 * 1 Anti-Infective Research Laboratory, Eugene Applebaum College of Pharmacy and Health Sciences, Wayne State University, Detroit, Michigan; 2 Fred Wilson School of Pharmacy, High Point University, High Point, North Carolina; 3 School of Medicine, Wayne State University, Detroit, Michigan Enterococcus species are the second most common cause of nosocomial infections in the United States and are particularly concerning in critically ill patients with preexisting comorbid conditions. Rising resistance to antimicrobials that were historically used as front-line agents for treatment of enterococcal infections, such as ampicillin, vancomycin, and aminoglycosides, further complicates the treatment of these infections. Of particular concern are Enterococcus faecium strains that are associated with the highest rate of vancomycin resistance. The introduction of antimicrobial agents with specific activity against vancomycin-resistant Enterococcus (VRE) faecium including daptomycin, linezolid, quinupristin-dalfopristin, and tigecycline did not completely resolve this clinical dilemma. In this review, the mechanisms of action and resistance to currently available anti-vre antimicrobial agents including newer agents such as oritavancin and dalbavancin will be presented. In addition, novel combination therapies including b-lactams and fosfomycin, and the promising results from in vitro, animal studies, and clinical experience in the treatment of VRE faecium will be discussed. KEY WORDS vancomycin-resistant Enterococcus faecium, combination therapy, double b-lactam combinations, daptomycin, linezolid, quinupristin-dalfopristin, tigecycline. (Pharmacotherapy 2017;37(5):579 592) doi: 10.1002/phar.1922 Enterococcal species are the second leading cause of hospital-acquired infections in the United States, with Enterococcus faecalis and Enterococcus faecium accounting for 6.8% and 4.1% of annual Funding: This review article was not funded through external support. Disclosures: Michael Rybak has received funding support from, consulted for, or was on the speakers bureau for Allergan, Bayer, Cempra, Merck, The Medicine Company, Sunovian, and Theravance and is partially supported by the National Institutes of Health (R01-AI121400 and R21-109266-01). Juwon Yim and Jordan Smith have nothing to disclose. a Current address: Department of Pharmacy, St. John Hospital and Medical Center, Detroit, Michigan. *Address for correspondence: Michael J. Rybak, Pharmacy Practice 4148, Eugene Applebaum College of Pharmacy and Health Sciences, Wayne State University, 259 Mack Avenue, Detroit, MI 48201; e-mail: m.rybak@wayne.edu. Ó 2017 Pharmacotherapy Publications, Inc. infections, respectively. 1 Enterococci can cause a wide variety of infections including urinary tract infections, bloodstream infections, and infective endocarditis. 1 Infections with Enterococcus species are associated with a significant burden of illness, and previous studies of patients with enterococcal bloodstream infections have demonstrated mortality rates up to 43%. 2 This is not surprising considering that individuals of advanced age with comorbid conditions such as renal impairment, diabetes mellitus, and malignancy are more likely to acquire enterococcal bacteremia. 3 Among patients with enterococcal bacteremia, infective endocarditis occurs in 8 32% of cases, likely due to the innate ability of enterococci to cause deepseated infections and evade host defenses. 4 Enterococcal infections are difficult to treat because monotherapy often provides only bacteriostatic activity, even with agents that are

580 PHARMACOTHERAPY Volume 37, Number 5, 2017 typically bactericidal such as b-lactams. Accordingly, combination therapy has been the mainstay of treatment of enterococcal infections for decades, and guidelines recommend combination therapy as first line for enterococcal endocarditis. 5 Against susceptible organisms, the combination of ampicillin and an aminoglycoside has been frequently used to achieve bactericidal activity. However, high-level resistance to aminoglycosides is now prevalent in several countries. 6 8 Furthermore, the presence of multiple comorbid conditions that are common among patients with enterococcal infections makes use of highly nephrotoxic agents such as aminoglycosides challenging. Recent data have demonstrated the successful combination of ampicillin and ceftriaxone against susceptible E. faecalis, improving safety and expanding the arsenal available for therapy. 9 However, mortality up to 22% while on this antimicrobial combination and 8% at 3-month follow-up were reported, warranting investigation of alternative therapies. 9 Resistance to early-generation cephalosporins is intrinsic among enterococci, although two new agents, ceftaroline and ceftobiprole, demonstrate limited in vitro activity. 10 E. faecalis, accounting for roughly 80% of enterococcal infections, is infrequently resistant to ampicillin. Similarly, vancomycin resistance in E. faecalis is relatively uncommon, although rates are increasing, occurring in 9% of national isolates and up to 12.5% of isolates in southeastern Michigan. 1, 2 E. faecium, however, is frequently resistant to ampicillin and vancomycin, and national surveillance data suggest that up to 83% of isolates are vancomycin resistant. 1 In both species of enterococci, vancomycin resistance has been associated with 2.5- times higher mortality, which might be attributable to delayed appropriate therapy. Cur- 11, 12 rently available options to treat vancomycinresistant Enterococcus (VRE) infections are limited and include linezolid, quinupristinvdalfopristin, tigecycline, daptomycin, and, potentially, the newly approved agents tedizolid and oritavancin. However, enterococcal resistance to these agents has also been reported, and monotherapy with any of these agents potentially presents similar issues to monotherapy with other antibiotics, so creative and novel approaches are necessary. In this review, we describe the agents currently available to combat VRE infections as well as resistance mechanisms that render them ineffective. The primary focus will be to review the currently available data regarding innovative combination therapies that demonstrate in vitro efficacy or have been used to successfully treat infections caused by VRE faecium. Resistance to Vancomycin and Ampicillin in Enterococci Vancomycin Vancomycin is a glycopeptide antibiotic that inhibits bacterial cell wall synthesis by binding to the D-alanine (Ala)-D-Ala C-terminus of peptidoglycan precursors and preventing transpeptidation and transglycosylation. 13 VRE frequently possess mutations in the vana and vanb regions of their genotypes. 14 Expression of these gene clusters mediates altered biosynthesis of peptidoglycan precursors, producing D-Ala-D-lactate (Lac) instead of normal D-Ala-D-Ala, resulting in up to 1000-fold less affinity for vancomycin. 14 Ampicillin Ampicillin resistance in Enterococcus species occurs through two known mechanisms, either through the production of penicillinase or the expression of a mutated, low-affinity penicillinbinding protein (PBP) 5. Penicillinase production is rare and easily overcome by the addition of a b-lactamase inhibitor. 15 The PBPs are membrane-bound enzymes that mediate cross-linking of bacterial peptidoglycan by catalyzing transpeptidation and are the target sites of b-lactam antibiotics. 16 Increased expression of lowmolecular-weight PBP5, which confers resistance through reduced affinity for b-lactams, virtually guarantees resistance to ampicillin. 17 Resistance to ampicillin and, more importantly, vancomycin, necessitate therapy with alternative antibiotic agents. These therapies require careful clinical consideration, including dosage selection, efficacy and safety concerns, and potentially useful combination regimens. Currently Available Anti-VRE Antibiotics and Resistance Daptomycin Daptomycin, a lipopeptide antibiotic, is the only currently available bactericidal agent with activity against VRE. It forms complexes with calcium ions and disrupts the cellular membranes of actively dividing cells, creating porous

ANTIMICROBIAL COMBINATIONS FOR VRE FAECIUM Yim et al 581 channels that lead to leakage of intracellular ions and cell death. 18 Recent evidence suggests that daptomycin also causes membrane defects, or patches, which result in significant alteration of cell wall synthesis. 19 As a bactericidal agent, it has a potential advantage as a treatment option for deep-seated infections such as infective endocarditis. Daptomycin dosing for enterococcal infections is variable, and recent data suggest that higher doses may be beneficial for invasive VRE infections. 20 Due to variability in dosing, frequently used regimens for daptomycin, along with other antimicrobial agents commonly used to treat enterococcal infections, are summarized in Table 1. Daptomycin Nonsusceptibility Epidemiologic data demonstrate that 99.9% of E. faecalis and 99.8% of E. faecium isolates remain susceptible to daptomycin. 22 Resistance, however, has been reported and is associated with immunosuppression, multiple comorbidities, and previous antimicrobial exposure. 23 The emergence of resistance may be associated with suboptimal daptomycin dosing as well. Given the current elevated Clinical and Laboratory Standards Institute (CLSI) minimum inhibitory concentration (MIC) susceptibility breakpoint for enterococci of 4 lg/ml, 24 it stands to reason that high-dose therapy is warranted for enterococcal infections, and the currently approved regimen of 6 mg/kg/day may be insufficient to eradicate infection. Several in vitro studies and limited clinical data have demonstrated that reduced daptomycin susceptibility is attributable to suboptimal dosing and potential mutations in Table 1. Usual Dosages for Antibiotics Commonly Used for Treatment of Enterococcal Infections 21 Antibiotic Dosage Ampicillin Vancomycin Daptomycin Linezolid Tedizolid Quinupristindalfopristin Tigecycline 1 2 g i.v. every 4 6 hrs Initially 15 20 mg/kg i.v. every 8 12 hrs; subsequently dosages adjusted based on vancomycin trough concentrations Approved dosing: 4 6 mg/kg/day i.v.; alternatively, 8 12 mg/kg/day i.v. has been recommended for invasive infections 600 mg p.o./i.v. every 12 hrs 200 mg/day p.o./i.v. 7.5 mg/kg i.v. every 8 12 hrs 100 mg i.v. once, followed by 50 mg i.v. every 12 hrs isolates with daptomycin MICs in the 3 4 lg/ml range. 25 28 In vitro pharmacokinetic/pharmacodynamic (PK/PD) dose response experiments have demonstrated that daptomycin at a minimum of 8 mg/kg/day is required for sufficient bactericidal kill, and 10 mg/kg/day maybe needed to prevent the emergence of resistance in high-inoculum infections caused by VRE faecalis and faecium. 20, 23 Recently, a multicenter retrospective cohort study that included 62 E. faecium bloodstream infections showed that initial isolates with elevated daptomycin MICs at 3 4 lg/ml were predictive of microbiologic treatment failure with daptomycin and suggested that the current CLSI susceptibility breakpoint for enterococci should be reevaluated. 29 The mechanism of daptomycin nonsusceptibility in Enterococcus species is not fully understood. Most studies have focused on mutations in various metabolic pathways leading to daptomycin nonsusceptibility. 28, 30, 31 In gram-positive bacteria such as Staphylococcus aureus, a proposed phenotypic mechanism of daptomycin nonsusceptibility involves a more positively charged cell surface via alteration in phospholipid content, resulting in reduced binding of cationic daptomycin to its target site. 32, 33 Genetically, there are several modifications proposed to promote daptomycin nonsusceptibility in VRE. Concomitant mutations in liaf and genes encoding enzymes responsible for phospholipid metabolism, such as glycerophosphoryl-diester-phosphodiesterase (gdpd) and cardiolipin synthetase (cls), may lead to daptomycin nonsusceptibility in a stepwise manner. Initially, activation of the LiaFSR system that regulates the response of the cell envelope to antimicrobial agents may induce transcription of several genes that help minimize damage by the antibiotic. Daptomycin nonsusceptibility may then be potentiated by subsequent modification in composition or distribution of transmembrane phospholipid that occurs as a result of mutations in gdpd. 30 Mechanistically, researchers demonstrated that daptomycin-nonsusceptible strains possessed lower daptomycin binding to the cell membrane and that those same strains had thicker cell walls than their daptomycin-susceptible parent strains. 28 In support of this hypothesis, another group 31 recently reported that binding of calcium-daptomycin complex to the septal structures, a vital structure for antimicrobial activity, was suppressed in daptomycin-nonsusceptible strains compared to daptomycin-susceptible strains. Despite the information presented here, the definitive mechanisms of daptomycin

582 PHARMACOTHERAPY Volume 37, Number 5, 2017 resistance in enterococci are yet to be determined. More genetic and biochemical studies are needed to ascertain the mechanism of resistance in the near future. Oxazolidinones Linezolid is an oxazolidinone approved for the treatment of uncomplicated and complicated skin and soft structure infections, communityacquired and nosocomial pneumonia, and VRE faecium infections, including bacteremia. It binds to the bacterial 23S ribosomal portion of the 50S ribosomal subunit, providing bacteriostatic activity against enterococci. 34 The most welldescribed mechanism of resistance to linezolid is target modification via a mutation in domain V of 23S ribosomal ribonucleic acid (rrna) genes in which guanine at position 2576 is replaced with uracil (G2576U). Both E. faecalis and E. faecium carry several copies of these rrna genes, and the level of linezolid resistance is determined by the percentage of rrna with a G2576U mutation. 15 Another mechanism responsible for transferrable linezolid resistance involves the cfr (chloramphenicol-florfenicol resistance) gene, which encodes an rrna methyltransferase that methylates the C-8 position of 23S rrna nucleotide A2503. It is speculated that methylation of A2503 leads to impaired positioning and binding of linezolid. 35 Tedizolid is the second oxazolidinone antibiotic, after linezolid, approved for use in the United States. It possesses a similar mechanism of action to linezolid, but enterococcal MICs are 4 8 times lower than those of linezolid due to enhanced binding to the target. Surveillance data demonstrate tedizolid activity in > 99% of enterococcal isolates. 36 Although it appears that tedizolid maintains activity against cfr-mediated enterococcal resistance, chromosome-mediated ribosomal mutations present in linezolid-resistant isolates are also likely to demonstrate increased tedizolid MICs, conferring cross-resistance via this mechanism by MIC breakpoint criteria. Clinical data evaluating tedizolid use in serious enterococcal infection are currently lacking, and it possesses the same limitation of bacteriostatic activity inherent to linezolid. Quinupristin-Dalfopristin Quinupristin-dalfopristin is a combination of streptogramin antibiotics with activity against E. faecium, although nearly all E. faecalis are resistant. 37 Mechanistically, the two agents bind to the 50S ribosome and act synergistically to halt protein synthesis. The bacteriostatic nature of the agent, along with frequent adverse events such as arthralgia and myalgia, limits its utility as a single agent in deep-seated infections. 37 Among E. faecalis, resistance to quinupristin-dalfopristin is nearly ubiquitous and secondary to presence of the lsa gene that possesses a similar sequence to known ABC transporters and is associated with efflux of quinupristin-dalfopristin. 38 Within E. faecium, resistance to quinupristin-dalfopristin is far less common and frequently involves mutations within the erm gene, a gene that encodes methyltransferases that modify 23S rrna and confer macrolide-lincosamide-streptogramin B resistance. 39 Tigecycline Tigecycline is a glycylcycline antibiotic with excellent in vitro activity against E. faecalis and E. faecium. Its mechanism is similar to tetracycline, as it binds to the 30S ribosomal subunit and inhibits protein synthesis, producing bacteriostatic activity. 40 Similar to the previous agents, tigecycline s bacteriostatic activity may hinder its ability to be used as monotherapy against enterococcal infections. Furthermore, it possesses a large volume of distribution with low serum concentrations, limiting its role in bacteremia. In addition, tigecycline was noted to have up to a 6.2% discontinuation rate in clinical trials, most often due to nausea, potentially limiting its use. 40 Lipoglycopeptides Oritavancin and dalbavancin are two lipoglycopeptide antimicrobials active against vancomycin-susceptible enterococci that have 41, 42 recently been approved in the United States. Dalbavancin possesses little activity against VRE, and only VRE isolates with mutations in the vanb gene have proven susceptible, likely limiting its viability as an anti-vre agent. Oritavancin, however, retains in vitro activity against VRE, and the MIC that inhibits growth in 90% of VRE isolates (MIC 90 ) even with the vana mutation present is 0.06 lg/ml. 42 The discrepancy in the two agents activity may be due to unique mechanistic properties of oritavancin. Oritavancin shares the glycopeptide mechanism of vancomycin in which it disrupts cell wall synthesis through blocking of transpeptidation and transglycosylation.

ANTIMICROBIAL COMBINATIONS FOR VRE FAECIUM Yim et al 583 However, the structure of oritavancin allows it to bind to D-Ala-D-Lac effectively, unlike vancomycin, preserving its activity against VRE. Oritavancin is also able to successfully anchor to lipid II in the cellular membrane through its hydrophobic tail, and some evidence demonstrates that it may depolarize gram-positive cell membranes. Each of these mechanisms may contribute to the unique ability of oritavancin among lipoglycopeptides to retain activity against VRE. Clinical data are currently lacking, but oritavancin demonstrates interesting potential for therapy of serious enterococcal infections. Combination Antibiotic Therapy The emergence of resistance to available antibiotics necessitates novel therapeutic approaches. The dearth of antibiotics on the horizon that are active against resistant enterococci, along with the presence of resistance to currently available agents, necessitate novel combination therapies with existing agents to most effectively combat VRE infections. In addition, combination antibiotic regimens are recommended and frequently used because single-antibiotic regimens are often bacteriostatic against enterococcal infections and because the clinical situation is often serious. Several dual-agent regimens reported to be active against VRE faecium infections are reviewed; these combinations are summarized in Table 2. Daptomycin Combination Therapy The successful combination of daptomycin and b-lactams is well documented against S. aureus. 43 Recent data have demonstrated the successful combination of daptomycin and b-lactams against enterococci as well. Researchers 44 described a patient treated with the combination of high-dose daptomycin and ampicillin for ampicillin-resistant VRE faecium. A patient receiving hemodialysis who had persistent endocarditis and bacteremia failed to respond to 7 days of daptomycin 6 mg/kg every 48 hours in combination with linezolid 600 mg twice/day. At that point, therapy was changed to daptomycin 12 mg/kg every 48 hours in combination with ampicillin 1 g every 6 hours, with subsequent clearance of bacteremia within 24 hours. Of interest, the authors provided in vitro data demonstrating that ampicillin significantly lowered the daptomycin MIC and worked synergistically with daptomycin as demonstrated in time-kill assays, possibly by reducing cell surface charge and increasing daptomycin binding. Ampicillin also enhanced the antimicrobial efficacy of several endogenous cationic peptides that share a similar mechanism of action to daptomycin. The authors also simulated the combinations of daptomycin 4 10 mg/kg/day and ampicillin 2 g every 4 hours in an in vitro onecompartment PK/PD model over 48 hours and demonstrated therapeutic enhancement with this combination. Another group corroborated this mechanism by demonstrating that fluorescently labeled daptomycin more effectively binds to the daptomycin-nonsusceptible E. faecium membrane when in the presence of b-lactams (Figure 1). 45 Each of these findings suggests that ampicillin may counteract some daptomycin nonsusceptibility mechanisms used by E. faecium, and the clinical case demonstrates that ampicillin may be added on to high-dose daptomycin in the situation of refractory bacteremia. 44 Similar work with other b-lactams has demonstrated that synergy with daptomycin against Enterococcus species extends beyond ampicillin. Continuing their research into the combinations of daptomycin and b-lactams against VRE, the investigators 46 published additional in vitro data suggesting synergistic activity between daptomycin and ceftaroline against an isogenic pair of a daptomycin-susceptible and daptomycinnonsusceptible E. faecium strains. In their time-kill analyses, synergy was observed with daptomycin plus ampicillin and daptomycin plus ceftaroline against the daptomycin-susceptible strain, but only ceftaroline in combination was synergistic against the daptomycin-nonsusceptible strain. Ancillary studies demonstrated that ceftaroline increased membrane fluidity, increased daptomycin binding, and decreased cell surface charge, indicating that ceftaroline was able to reverse several previously established daptomycin nonsusceptibility mechanisms. Even though the definite mechanism of these phenomena has not yet been determined, the authors suggested that alteration in surface physiology and subsequent changes in daptomycin activity might be attributable to high affinity of ceftaroline to PBP5. 46 Another in vitro PK/PD model of simulated endocardial vegetations over a 96-hour period 47 evaluated the combination of daptomycin at either 6 or 12 mg/kg/day and ceftriaxone 2 g every 24 hours against the previously defined daptomycin-susceptible/daptomycin-nonsusceptible E. faecium strain pair from the earlier

584 PHARMACOTHERAPY Volume 37, Number 5, 2017 Table 2. Summary of Combination Therapies Employed Clinically and In Vitro Against VRE Infections Antibiotic Combination/Study Isolates Design Results DAP + AMP 44 DS VRE faecium (DAP MIC 1 lg/ml) DAP + b-lactams 45 DAP + b-lactams 46 DAP + CRO 47 DAP + b-lactams 48 2DSE. faecalis, 1DS VRE faecium, and 1 DNS VRE faecium (DAP MICs 2 lg/ ml, 2 lg/ml, 4 lg/ ml, and 32 lg/ml, respectively) DS/DNS VRE faecium strain pair (DAP MIC 0.38 lg/ml/ DAP MIC 10 lg/ml) DS/DNS VRE faecium strain pair (DAP MIC 4 lg/ml, DAP MIC 32 lg/ml), DS E. faecalis (DAP MIC 2 lg/ml) 2 DS VRE faecium and 1 DS E. faecalis (DAP MICs 4 lg/ ml, 4 lg/ml, and 2 lg/ml, respectively) Case report of E. faecium bacteremia; in vitro studies of DAP + AMP 24-hr, in vitro, timekill studies using 0.5 9 MIC DAP + CPT, ERT, AMP, CRO, CFZ, FEP, or CTX In vitro studies of LL37 and DAP + CPT, AMP, PIP, CRO, CFZ 96-hr, in vitro, PK/ PD study of simulated endocardial vegetations using DAP 6 mg/kg/day and DAP 12 mg/kg/ day + CRO 2 g/day 96-hr, in vitro, PK/ PD study using DAP 10 mg/kg/ day + CPT 600 mg q12 hr, AMP 2 g q4 hr, or ERT 1 g q24 hr DAP + FOF 49 4 DS VRE faecium 24-hr, in vitro, timekill studies using 0.5 9 MIC DAP + FOF DAP + FOF 50 LZD + GENT 52 LZD + GENT 53 DS/DNS VRE faecium strain pair and 2 DS E. faecalis 1 VRE faecium (LZD MIC 2 lg/ml) 1 VRE faecium (LZD MIC 2 lg/ml) 72-hr, in vitro, PK/ PD study using DAP 8 mg/kg/day and DAP 12 mg/kg/ day + FOF 40 mg/ kg q8 hr Case report of 62-yrold man with meningitis caused by VRE faecium Case report of pregnant, 23-yr-old woman with refractory E. faecium bacteremia DAP 12 mg/kg q48 hr + AMP 1 g q6 hr resulted in clearance of bacteremia in 24 hrs; DAP + AMP synergistic in vitro CFZ and CTX demonstrated little synergy, but all other b-lactams were synergistic with DAP against all strains, with ERT being most synergistic against the DNS E. faecium CPT increased DAP binding, was most synergistic against DNS isolate, and enhanced LL37 activity most compared with AMP, PIP, CRO, CFZ DAP 12 mg/kg/day + CRO was bactericidal against DS E. faecium and E. faecalis, but not against DNS E. faecium CPT and ERT enhanced DAP activity and were bactericidal against all strains; AMP enhanced DAP activity against only 1 strain of DS E. faecium DAP + FOF was synergistic against all strains and bactericidal against 3 of the 4 strains DAP 12 mg/kg/day + FOF resulted in statistically significant bactericidal improvement over either agent alone against all DS isolates, but enhancement was not observed against the DNS E. faecium LZD for 21 days + GENT for first 5 days cleared CSF cultures at days 5 and 10, but patient ultimately died due to underlying malignancy LZD + GENT cleared bacteremia on day 1, and blood cultures remained clear 30 days after 14- day course of LZD + GENT (continued)

ANTIMICROBIAL COMBINATIONS FOR VRE FAECIUM Yim et al 585 Table 2 (continued) Antibiotic Combination/Study Isolates Design Results LZD + GENT, RIF 54 1 E. faecalis and 1 VRE faecium 96-hr, in vitro, PK/ PD study of simulated endocardial vegetations using LZD, LZD + RIF, LZD + GENT, DAP 6 mg/kg/day + RIF or GENT, and DAP 10 mg/kg/day + RIF or GENT LZD + RIF 55 1 VRE faecium Case report of meningitis caused by VRE faecium Q-D + LZD 56 Q-D + AMP 56 1 VRE faecalis (LZD MIC 2 lg/ml, Q-D MIC 4 lg/ml) and 1 VRE faecium (LZD MIC 2 lg/ml, Q-D MIC 0.25 lg/ml) 1 VRE faecalis (AMP MIC 1 lg/ml, Q-D MIC 4 lg/ml) and 1 VRE faecium (AMP MIC 128 lg/ml, Q- D MIC 0.25 lg/ml) 48-hr, in vitro, PK/ PD study using Q-D 7.5 mg/kg q8 hr, LZD 600 mg q12 hr, Q-D + LZD 48-hr, in vitro, PK/ PD study using Q-D 7.5 mg/kg q8 hr, AMP 1 g q6 hr, Q-D + AMP LZD + DOX 58 1 DNS VRE faecium Case report of 22-yrold man with refractory VRE faecium bacteremia Q-D + DOX 60 2 VRE faecium 96-hr, in vitro, PK/ PD study of simulated endocardial vegetations using Q-D 7.5 mg/kg q8 hr, Q-D continuous infusion, DOX 200 mg q24 hr Q-D + DOX + RIF 61 1 VRE faecium Case report of 76-yrold man with refractory endocarditis caused by VRE faecium, along with in vitro studies Q-D + AMP 64 1 VRE faecium Case report of 38-yrold woman with refractory VRE faecium bacteremia GENT enhanced LZD activity against VRE faecium, but the combination was not synergistic against E. faecalis; RIF antagonized LZD against VRE faecium; DAP combinations were superior to LZD combinations in this study LZD + RIF cleared CSF cultures in this patient, and the patient was bacteriologically cured 1 mo after treatment completion Q-D + LZD demonstrated enhanced activity against E. faecalis, whereas Q-D alone was bactericidal against E. faecium and could not be evaluated for enhancement with LZD Q-D + AMP demonstrated enhanced activity against E. faecalis, whereas Q-D alone was bactericidal against E. faecium and could not be evaluated for enhancement with AMP LZD + DOX cleared blood cultures after previous blood cultures grew DNS E. faecium Q-D + DOX was superior to Q-D 7.5 mg/kg q8 hr alone against one E. faecium strain, but Q-D continuous infusion was superior to Q-D + DOX; Q-D + DOX was not superior to Q-D or DOX alone against the second E. faecium strain Q-D + DOX + RIF for 8 wks cleared aortic valve endocarditis and blood cultures after failure of Q-D alone, and patient remained healthy 1 yr after therapy; in vitro studies demonstrated synergy with Q-D + DOX, Q-D + RIF, and DOX + RIF Q-D + AMP resolved fever and cleared VRE bacteremia after failure of several other therapies, including LZD, and blood cultures remained clear until the patient died of other causes (continued)

586 PHARMACOTHERAPY Volume 37, Number 5, 2017 Table 2 (continued) Antibiotic Combination/Study Isolates Design Results Q-D + AMP 65 1 VRE faecium Case report of 56-yrold man with refractory VRE faecium endocarditis TGC + DAP 67 1 LR VRE faecium Case report of 62-yrold man with refractory endocarditis caused by VRE faecium TGC + DAP 68 1 LR VRE faecium Case report of 39-yrold woman with refractory endocarditis caused by VRE faecium TGC + VAN, GENT, 2 VRE faecium 24-hr, in vitro, timekill RIF, DOX 70 studies using 0.5 9 MIC TGC + VAN, GENT, RIF, DOX DAP + CPT 74 DS VRE faecium (DAP MIC 2 lg/ml) Case report of 63-yrold man with refractory E. faecalis endocarditis; in vitro studies of DAP + CPT, DAP + AMP 63 days of Q-D + AMP led to cure of VRE faecium endocarditis, which was still cured 2 yrs out from completion of therapy 10 wks of TGC + DAP 6 mg/kg/ day eliminated bacteremia and cleared endocarditis without valve replacement after initial therapy with VAN and LZD 8 wks of TGC + DAP 8 mg/kg/ day eliminated fever and cleared bacteremia due to VRE faecium previously treated with DAP 6 mg/kg/day, RIF, and GENT VAN enhanced TGC activity against both strains, DOX enhanced TGC activity against 1 strain, and GENT and RIF did not enhance TGC activity DAP 8 mg/kg/day + CPT 600 mg q8 hr resulted in clearance of bacteremia out to 6 wks after completion of therapy; DAP + CPT was more synergistic in vitro than DAP + AMP AMP = ampicillin; CFZ = cefazolin; CPT = ceftaroline; CRO = ceftriaxone; CSF = cerebrospinal fluid; CTX = cefotaxime; DAP = daptomycin; DNS = daptomycin-nonsusceptible; DOX = doxycycline; DS = daptomycin-susceptible; ERT = ertapenem; FEP = cefepime; FOF = fosfomycin; GENT = gentamicin; LL-37 = human cathelicidin antimicrobial peptide; LR = linezolid-resistant; LZD = linezolid; MIC = minimum inhibitory concentration; PIP = piperacillin; PK/PD = pharmacokinetic/pharmacodynamic; Q-D = quinupristin-dalfopristin; RIF = rifampin; TGC = tigecycline; VAN = vancomycin; VRE = vancomycin-resistant enterococci. work. 46 Against the daptomycin-susceptible parent E. faecium strain, the addition of ceftriaxone to daptomycin 6 and 12 mg/kg/day not only enhanced bactericidal activity compared to daptomycin alone but also prevented the emergence of daptomycin nonsusceptibility. However, the combination exhibited no appreciable activity against the daptomycin nonsusceptible E. faecium strain (daptomycin MIC of 32 lg/ml) even at the higher daptomycin dose of 12 mg/kg/day. Despite its inactivity against enterococci as monotherapy, ertapenem is another b-lactam antibiotic that has exhibited synergistic activity combined with daptomycin against enterococcus in an in vitro study. In this PK/PD study comparing synergistic activity of three b-lactams ampicillin, ceftaroline, and ertapenem with daptomycin against two VRE faecium strains, ertapenem demonstrated enhanced bactericidal activity in combination with daptomycin to a similar or greater extent as ceftaroline and ampicillin. 48 Furthermore, exposure to ertapenem as well as ceftaroline or ampicillin enhanced bacterial killing by an endogenous antimicrobial peptide, LL37, demonstrating a role for potential synergy with host innate cationic peptide defense. Another in vitro study involving combination MIC testing and time-kill assays was conducted by the same group of investigators to evaluate various b-lactams for synergistic activity with daptomycin against 35 VRE isolates, of which 20 were VRE faecium. 45 Of the b-lactams evaluated, synergistic activity was consistently observed with ceftaroline, ertapenem, cefepime, ceftriaxone, and ampicillin against all strains, with ertapenem being most synergistic against the DNS E. faecium. The authors suggested that ertapenem in combination with daptomycin might be a viable therapeutic option for polymicrobial infections involving both VRE and resistant gram-negative bacteria and that the once-daily dosing strategy of ertapenem may be beneficial for lengthy outpatient antimicrobial therapy. These data also further demonstrate

ANTIMICROBIAL COMBINATIONS FOR VRE FAECIUM Yim et al 587 Figure 1. Binding of fluorescent daptomycin (DAPI) to daptomycin-nonsusceptible Enterococcus faecium. Cells were pretreated with b-lactam antibiotics ceftaroline (CPT), imipenem (IMI), or ceftriaxone (CRO) for 20 minutes prior to exposure to fluorescently labeled daptomycin. Untreated = no b-lactam exposure. (Reprinted with permission from reference. 43 ) that the combination of daptomycin and ceftaroline possesses promising potential for therapy VRE resistant to daptomycin. Another antibiotic that has demonstrated synergistic activity in combination with daptomycin is fosfomycin. In a time-kill experiment evaluating activity of daptomycin in combination with fosfomycin, amoxicillin, or linezolid against 32 VRE faecium urine isolates, fosfomycin in combination with daptomycin was the most potent and highly synergistic combination. 49 The authors hypothesized that the synergistic activity may be attributable to enhanced sensitivity of the bacterial cell envelope to daptomycin in the presence of fosfomycin. The combination of daptomycin and fosfomycin has also been studied in 72-hour, in vitro PK/PD models. Researchers evaluated four VRE strains, including an isogenic daptomycin-susceptible and daptomycin-nonsusceptible strain pair, against the combination of fosfomycin 40 mg/kg every 8 hours and daptomycin at either 8 or 12 mg/kg/day. 50 Fosfomycin in combination with both daptomycin 8 and 12 mg/kg/day resulted in significantly greater bactericidal activity than daptomycin alone against all daptomycin-susceptible isolates. However, the higher daptomycin dose of 12 mg/kg/ day was necessary to maintain bactericidal activity for the entire 72 hours. Against the daptomycin-nonsusceptible VRE strain, there was no synergistic or bactericidal activity with the combination, regardless of the daptomycin dose. The addition of fosfomycin did prevent development of daptomycin nonsusceptibility in two of three daptomycin-susceptible VRE isolates. The authors proposed that enhanced bactericidal activity from the combination might be attributable to reduction in cell surface charge by fosfomycin. Given the limited number of studies and conflicting study results, this combination warrants further investigation of its efficacy. In addition, unavailability of intravenous form of fosfomycin in the United States precludes its use at this time. The intravenous form of fosfomycin is currently being investigated for the treatment

588 PHARMACOTHERAPY Volume 37, Number 5, 2017 of complicated urinary tract infection and acute pyelonephritis, potentially allowing its use in the near future. 51 Linezolid Combination Therapy Linezolid, an anti-vre therapy with approval for VRE bacteremia, has also been used in combination with other antibiotics against this pathogen. A case report describes the successful use of linezolid in combination with gentamicin for treatment of VRE faecium meningitis. 52 A 62- year-old man had cerebrospinal fluid (CSF) culture positive for enterococci after a suboccipital craniectomy and placement of a right frontal ventriculostomy on postoperative day 10. The patient was initially treated with high-dose ampicillin and gentamicin. Two days later, the organism was identified as VRE faecium, and ampicillin was substituted with intravenous linezolid 600 mg every 12 hours. Within 48 hours, the patient became afebrile, and his CSF leukocyte count decreased to 8 cells/mm 3 from 250 cells/mm 3 at baseline, with improvement in mental status. Cultures from days 5 and 10 after initiation of linezolid remained negative, and his clinical status significantly improved. Gentamicin was stopped on day 5, whereas intravenous linezolid was continued for 3 weeks when the patient died due to progression of his underlying malignancy. The infecting isolate was susceptible to both linezolid and gentamicin with MICs of 2 and 500 lg/ml, respectively. In another case report, a 23-year-old woman with profound neutropenia secondary to chemotherapy (absolute neutrophil count of < 100 cells/ mm 3 ) had bacteremia with VRE faecium, which was resistant to almost all therapeutic options except for chloramphenicol, gentamicin (MIC < 500 lg/ml), and linezolid (MIC 2 lg/ml). 53 The patient underwent therapy with intravenous linezolid 600 mg every 12 hours in addition to gentamicin 1 mg/kg every 8 hours for 14 days, and blood cultures were cleared and remained negative at 14 and 30 days after completion of therapy. Although the combination appeared to be effective clinically, in time-kill experiments performed by the authors, linezolid and gentamicin failed to exhibit synergistic activity. In contrast, a recent in vitro study 54 using simulated endocardial vegetation models demonstrated synergistic activity of this combination. In this study, activity of linezolid in combination with either rifampin or gentamicin was compared with daptomycincontaining combinations against enterococcal strains, including a VRE faecium isolate susceptible to linezolid and daptomycin. The addition of gentamicin at 1.3 mg/kg every 12 hours to linezolid 600 mg every 12 hours improved the efficacy of linezolid and was bactericidal against VRE faecium at 72 hours. Of note, daptomycin-containing combinations, both daptomycin 6 and 10 mg/kg/day, achieved a greater reduction in bacterial colony count in vegetations at predefined time points than linezolid-based combinations. When rifampin was added to linezolid, it antagonized the activity of linezolid against the VRE faecium strain. Although linezolid and rifampin failed to demonstrate enhancement in this in vitro study, a case report was published on the successful use of linezolid 600 mg orally every 12 hours in combination with rifampin 600 mg/day for treatment of VRE faecium meningitis. 55 At 1-month follow-up after completion of treatment, the patient achieved bacteriologic cure. Doxycycline is another antibiotic that has demonstrated enhanced activity in combination with linezolid against VRE faecium in vitro. 56 Addition of doxycycline to linezolid at subinhibitory doses in vitro seems to prevent selection of linezolid-resistant E. faecium. 57 This favorable in vitro activity of linezolid-doxycycline combination is consistent with a case report of its successful use for treatment of E. faecium bacteremia after bacteriologic and clinical failure of daptomycin therapy. 58 Antibiotics that have displayed potential benefit in combination with linezolid include gentamicin, rifampin, and doxycycline. Evidence in support of linezolid combinations are sporadic and sometimes even contradictory between in vitro studies and case reports, leaving the clinical question of which agent to choose still inconclusive. Even though linezolid-based combinations may be used as salvage therapies for VRE faecium infections, their use is perhaps better reserved for patients who have failed or cannot tolerate bactericidal antibiotics that have more clinical data available. Quinupristin-Dalfopristin Combination Therapy Quinupristin-dalfopristin also possesses U.S. Food and Drug Administration approval for the treatment of E. faecium, and these agents have been used in combinations both clinically and in vitro. 59 The antimicrobial agents that have accumulated the most evidence in support of their use for VRE faecium infections in combination with

ANTIMICROBIAL COMBINATIONS FOR VRE FAECIUM Yim et al 589 quinupristin-dalfopristin are doxycycline and minocycline, although the strength of evidence is low. In a PK/PD model of simulated endocardial vegetations, addition of doxycycline to quinupristin-dalfopristin has been demonstrated to be synergistic, and the combination prevented development of quinupristin-dalfopristin resistance in VRE faecium. 60 Similar findings were published in a case report on cure of endocarditis caused by VRE faecium with intravenous quinupristin-dalfopristin 500 mg every 8 hours, intravenous doxycycline 100 mg every 12 hours, and rifampin 300 mg orally every 12 hours. 61 In addition, time-kill studies using the patient s VRE faecium isolate displayed significant synergistic activity for quinupristin-dalfopristin in combination with doxycycline. It should be noted, however, that failure of the combination to cure VRE faecium bacteremia has also been published, 62 making interpretation of these data more challenging. Similarly, efficacy of minocycline in combination with quinupristin-dalfopristin was demonstrated in a small clinical trial that included 56 oncology patients for treatment of VRE infections, of which 91% were VRE faecium. 63 Infections included bacteremia, urinary tract infections, and wound infections, and they were treated for an average of 12 days (range 2 52 days). Overall, therapy with this combination led to clinical and microbiologic cure in 68% of patients but with the unfortunate occurrence of arthralgia or myalgia in more than one-third of patients. Eradication of VRE faecium bacteremia and endocarditis using high-dose ampicillin with quinupristin-dalfopristin has also been documented by several case reports, perhaps warranting further study of this combination. 64, 65 Although quinupristin-dalfopristin carries an approved indication for VRE faecium as monotherapy, clinical and in vitro data suggest that quinupristin-dalfopristin should be used in combination with other antimicrobial agents when possible. However, the optimal antibiotic chosen for the combination regimen remains unclear due to limited data available. Tigecycline Combination Therapy Use of tigecycline alone for treatment of invasive enterococcal infections has been discouraged due to attainment of low serum concentrations. 66 Nonetheless, its successful use as part of combination therapy has been reported for treatment of 67, 68 VRE faecium bacteremia and endocarditis. A case report described a patient with VRE faecium endocarditis whose infection was cleared by tigecycline in combination with high-dose daptomycin. 68 After treatment failure with daptomycin 6 mg/kg/day followed by continued failure at 8 mg/kg/day in combination with several antibiotics, the patient s clinical and microbiologic status dramatically improved with the addition of tigecycline to daptomycin. In this patient, blood cultures remained negative at 9 weeks after discharge. Tigecycline has also been evaluated in combination with gentamicin. In a rabbit model of endocarditis caused by VRE faecium, tigecycline produced a significant decrease in bacterial load of vegetation in comparison to no treatment, and addition of gentamicin led to further reduction in colony count. 69 Interestingly, sustainable antimicrobial activity of tigecycline alone and in combination with gentamicin was observed on day 5 after completion of treatment. Similarly, in an in vitro, time-kill analysis including two strains of VRE faecium evaluating tigecycline and tigecycline combinations, addition of gentamicin improved bactericidal activity of tigecycline against both strains, even though the degree of enhancement was minimal in one strain. 70 The clinical implication of these in vitro findings potentially warrants further investigation. However, due to the limited clinical efficacy of tigecycline and its significant adverse effect profile, tigecycline should be reserved for cases that absolutely necessitate its use. Conclusions and Opinion Prevalence of VRE has continued to rise since its emergence in the United States in the late 1980s. 71 It has been reported that up to 35.5% of hospitalacquired enterococcal infections are vancomycin resistant, with E. faecium strains accounting for the majority of resistance. E. faecium often possesses intrinsic resistance to ampicillin and highlevel resistance to aminoglycosides as well, limiting therapeutic options. For years, linezolid, daptomycin, quinupristin-dalfopristin, and tigecycline have been the mainstays of therapy for VRE faecium infections. 72 For refractory VRE faecium infections that fail the traditional treatment options, a variety of antimicrobial combinations have been investigated. Although a large portion of the data is derived from in vitro and animal studies, several case reports and case series have described successful treatment of VRE faecium infections with a myriad of combinations. As of

590 PHARMACOTHERAPY Volume 37, Number 5, 2017 yet, however, the optimal combination that would yield the best outcomes in the treatment of VRE faecium has not been elucidated. Of the combinations investigated, daptomycin regimens involving the combination of b-lactam agents such as ceftaroline or ertapenem appear to hold the most promise based on in vitro studies and case reports. Furthermore, it has been suggested that the addition of b-lactams not only potentiates the bactericidal activity of daptomycin but also prevents development of daptomycin nonsusceptibility in enterococci. Given the rising concerns of emergence of daptomycin nonsusceptibility in enterococci that may be associated with daptomycin monotherapy, use of high-dose daptomycin combined with a b-lactam antibiotic appears to be a potentially reasonable therapeutic choice. This is especially true in the treatment of VRE faecium that fails to respond to traditional therapeutic options. Recently, two studies corroborated that daptomycin may be a more appropriate choice than linezolid for the treatment of VRE bacteremia and that high-dose daptomycin, up to 10 mg/kg/day, is likely a better choice than lower doses. 20, 73 In these studies, the role of daptomycin combination therapy with b-lactams is not as easily defined. Daptomycin alone was just as effective as daptomycin in combination with b-lactams in the first study, demonstrating treatment failure rates of 50.6% and 56.3%, respectively. 73 In the second trial, daptomycin combinations with b-lactams were associated with better outcomes, but that effect was largely due to most combination therapy being present in the high-dose (10 mg/kg/day) daptomycin cohort. 20 Because these studies were not conducted in an attempt to compare combination therapy to daptomycin monotherapy, and in the presence of impressive in vitro combination data with b-lactams and daptomycin, it is the opinion of the authors that combination therapy with daptomycin be given strong consideration in cases of deep-seated VRE infections. Combination regimens that include linezolid, quinupristin-dalfopristin, or tigecycline could be viable salvage options for VRE infections not responsive to the therapies described above, although their appropriateness should be assessed on a case-by-case basis in consideration of preexisting evidence and clinical experience. If one of these combinations is warranted, it is the authors opinion that linezolid be given first based on data published regarding its use in VRE bacteremia. The recent approvals of tedizolid and oritavancin provide two more potential options for VRE infections, perhaps alone or in combination, although more data are needed before any recommendation can be made. Further clinical data are paramount to better understand various combination therapies against VRE faecium, reduce the emergence of resistance, and ultimately improve patient outcomes. References 1. Sievert DM, Ricks P, Edwards JR, et al. Antimicrobial-resistant pathogens associated with healthcare-associated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2009 2010. Infect Control Hosp Epidemiol 2013;1:1 14. 2. Hayakawa K, Martin ET, Gudur UM, et al. Impact of different antimicrobial therapies on clinical and fiscal outcomes of patients with bacteremia due to vancomycin-resistant enterococci. Antimicrob Agents Chemother 2014;7:3968 75. 3. Billington EO, Phang SH, Gregson DB, et al. Incidence, risk factors, and outcomes for Enterococcus spp. blood stream infections: a population-based study. Int J Infect Dis 2014;76 82. 4. Anderson DJ, Murdoch DR, Sexton DJ, et al. Risk factors for infective endocarditis in patients with enterococcal bacteremia: a case control study. Infection 2004;2:72 7. 5. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis in adults: diagnosis, antimicrobial therapy, and management of complications: a scientific statement for healthcare professionals from the American Heart Association. Circulation 2015;15:1435 86. 6. Chow JW. Aminoglycoside resistance in enterococci. Clin Infect Dis 2000;2:586 9. 7. Padmasini E, Padmaraj R, Ramesh SS. High level aminoglycoside resistance and distribution of aminoglycoside resistant genes among clinical isolates of Enterococcus species in Chennai, India. ScientificWorldJournal 2014;2014:329157. 8. Tsakris A, Pournaras S, Maniatis AN, Douboyas J, Antoniadis A. Increasing prevalence of high-level gentamicin resistance among Enterococci isolated in Greece. Chemotherapy 2001;2:86 9. 9. Fernandez-Hidalgo N, Almirante B, Gavalda J, et al. Ampicillin plus ceftriaxone is as effective as ampicillin plus gentamicin for treating Enterococcus faecalis infective endocarditis. Clin Infect Dis 2013;9:1261 8. 10. Duplessis C, Crum-Cianflone NF. Ceftaroline: a new cephalosporin with activity against methicillin-resistant Staphylococcus aureus (MRSA). Clin Med Rev Ther 2011;3:a2466. 11. DiazGranados CA, Zimmer SM, Klein M, Jernigan JA. Comparison of mortality associated with vancomycin-resistant and vancomycin-susceptible enterococcal bloodstream infections: a meta-analysis. Clin Infect Dis 2005;3:327 33. 12. Zasowski EJ, Claeys KC, Lagnf AM, Davis SL, Rybak MJ. Time is of the essence: the impact of delayed antibiotic therapy on patient outcomes in hospital-onset enterococcal bloodstream infections. Clin Infect Dis 2016;10:1242 50. 13. Courvalin P. Vancomycin resistance in gram-positive cocci. Clin Infect Dis 2006;S25 34. 14. Gold HS. Vancomycin-resistant enterococci: mechanisms and clinical observations. Clin Infect Dis 2001;2:210 9. 15. Arias CA, Murray BE. The rise of the Enterococcus: beyond vancomycin resistance. Nat Rev Microbiol 2012;4:266 78. 16. Macheboeuf P, Contreras-Martel C, Job V, Dideberg O, Dessen A. Penicillin binding proteins: key players in bacterial cell cycle and drug resistance processes. FEMS Microbiol Rev 2006;5:673 91. 17. Rice LB. Mechanisms of resistance and clinical relevance of resistance to beta-lactams, glycopeptides, and fluoroquinolones. Mayo Clin Proc 2012;2:198 208.