Therapeutic approach to difficult-to-treat multidrug-resistant enterococcal infections

Authors: Adrianna M. Turner, Paul Kinsella, William R. Miller, Glen P. Carter, Truc T. Tran, Benjamin P. Howden, Cesar A. Arias

Abstract

Difficult-to-treat (DTR) enterococcal infections, particularly those caused by multidrug-resistant Enterococcus faecium and Enterococcus faecalis, pose significant clinical challenges due to limited treatment options and high rates of treatment failure, compounded by a paucity of new antimicrobial agents in the development pipeline. Despite advances in understanding resistance mechanisms and in vitro synergistic antibiotic combinations, robust clinical data to guide therapy for severe or DTR enterococcal infections remain limited. This review synthesizes available evidence to inform optimal management strategies, including drug selection and dosing, while highlighting areas needing further research. Given the ongoing threat posed by multidrug-resistant enterococci, we emphasize the importance of gathering robust clinical data to guide best practices for managing these difficult-to-treat infections.

INTRODUCTION

Healthcare-associated infections caused by multidrug-resistant (MDR) bacteria are a critical public health threat associated with high rates of treatment failure, morbidity, and mortality. Enterococci are commensals of the human gastrointestinal tract (GIT) that have emerged as leading causes of nosocomial infections, particularly in critically ill and immunocompromised patients (1). Enterococci have been implicated in a diverse range of infections, but bloodstream infections, endocarditis, intra-abdominal, and urinary infections are the most commonly seen, often requiring aggressive targeted therapy.

One of the most challenging issues that clinicians encounter when faced with an invasive enterococcal infection is the limited choice of antimicrobial agents since many hospital-associated isolates (e.g., Enterococcus faecium) are resistant to most anti-enterococcal antibiotics. Indeed, resistance to aminopenicillins, vancomycin, daptomycin, and linezolid is now reported among nosocomial isolates of E. faecium, leaving extremely limited alternatives for treatment. Moreover, recent studies on the epidemiology of enterococcal bacteremia indicate that failure to eradicate the bloodstream infection early in the course of the disease is associated with increased likelihood of death (2), emphasizing the fact that successful antimicrobial therapy is key for patient survival, particularly in those with immunocompromised conditions or with multiple co-morbidities.

Another major issue when dealing with enterococcal infections is the paucity of clinical data supporting a particular treatment. Despite the increasing number of patients infected by these organisms, robust clinical data to guide therapy in severe enterococcal infections are lacking. Here, we provide an updated overview (2022–2024) of the therapeutic options and approaches for drug optimization in difficult-to-treat (DTR) vancomycin-resistant enterococcal (VRE) infections. Although low-level vancomycin resistance is intrinsic to some Enterococcus species beyond Enterococcus faecalis and E. faecium, this review will focus on the latter two species, as they are the most frequently encountered in clinical infections.

ENTEROCOCCI IN THE HEALTHCARE SETTING

Enterococci can survive for long periods on environmental surfaces like medical equipment, bed rails, and doorknobs (3). They are tolerant to different temperatures, chlorine, and some alcohol preparations (4, 5), likely contributing to their success in the hospital setting. E. faecalis is the most common species isolated from human infections, particularly those of community-onset. However, in the last two decades, a major increase in hospital-associated infections caused by E. faecium has occurred, particularly in specific patient populations such as those with hematological malignancies, bone marrow, and liver transplants (6). Although the rise in E. faecium infections is likely multifactorial and remains under investigation, epidemiological studies suggest that certain problematic lineages of E. faecium are particularly well-adapted to the hospital environment (7, 8). These strains are more likely to acquire resistance to clinically relevant antimicrobials and possess survival advantages in the GIT, potentially due to other protective mechanisms such as bacteriocin production (9). Transmission of VRE has been documented through direct contact with colonized or infected patients, or by indirect contact via contaminated environmental surfaces or the hands of healthcare workers (4). Risk factors for VRE nosocomial transmission include prolonged hospitalization, use of broad-spectrum antibiotics, surgery, immunocompromised state, or occupying a room with a previously infected or colonized patient, among others (10). Indeed, one of the intrinsic characteristics of enterococci is their tolerance to antimicrobial agents that are often used as empiric therapy in hospitalized patients (e.g., β-lactams, vancomycin). This resistance phenotype promotes increased colonization of the GIT, which seems to be associated with an increased likelihood of developing an infection and increases the risk of transmission. Indeed, reported colonization rates vary by population type from 16.2% in solid organ transplant to up to 20% in hematopoietic cell transplant (11, 12). Infection rates are much higher in colonized patients compared to those who are not colonized (relative risk 24.15 [95% CI = 10.27–56.79]) (12). Active surveillance, contact precautions for colonized and infected individuals, strict hand hygiene practices for healthcare workers, and judicious use of antimicrobial agents can effectively curtail VRE transmission chains. More recently, microbiome manipulation to reduce VRE in the GIT is a strategy that is gaining traction to reduce the risk of infection and, possibly, transmission of VRE (13).

DIFFICULT-TO-TREAT MULTIDRUG-RESISTANT ENTEROCOCCAL INFECTIONS

There is no agreed universal definition of what constitutes a DTR enterococcal infection. For ease of discussion, here we are using the term “difficult-to-treat” to include infections that are either “persistent” (failure to eradicate the organism despite seemingly appropriate antimicrobial therapy) or “recurrent” (defined as patients who had clearance of the bacteremia during hospitalization, with a second positive culture also during this period) (2). While these terms, persistence and recurrence, describe microbiological outcomes following an adequate course of treatment, we use the term DTR to capture a broader concept encompassing not only the microbiological characteristics of the infecting strains but also the clinical context, including patient-specific factors and antimicrobial treatment challenges, as outlined below. Consequently, DTR clinical scenarios typically exclude patients who are not significantly immunocompromised, have modifiable sources of infection, are not at risk for GIT colonization and/or selection of MDR enterococci, or are infected with organisms that remain broadly susceptible to available antimicrobial agents. Although these definitions are not standard, they capture the spectrum of challenging enterococcal infections encountered during routine clinical practice.

A cohort study of patients with enterococcal bacteremia (VENOUS) by Contreras et al. examined several outcomes, including lack of clearance of the organisms on day 4 while receiving at least 48 hours of active anti-enterococcal therapy (described as “microbiological failure”) (2). The cut-off of 4 days was derived from a previous study in bloodstream infections that suggested that day 4 of bacteremia was associated with increased mortality (14). Based on the VENOUS cohort, we recently (15) defined recurrent bacteremia as patients who had clearance of the bacteremia during hospitalization, with a second positive culture also during this period, or two negative cultures at least 1 day apart, followed by at least one positive culture. Overall, persistence and recurrence, described as recalcitrant bacteremia in this study, occurred in 3.2% of patients, although reports suggest that up to 12% of patients may exhibit this phenotype (16).

Clinically, DTR bacteremia is influenced by many factors, including source control of deep-seated infections (17), presence of infected central venous access devices, and the pharmacokinetics/pharmacodynamics (PK/PD) of the antimicrobial agents used against enterococci, among others. It is important to note that these DTR infections are often reported in patients with important degrees of immunosuppression who have been on antimicrobial therapies for a prolonged period of time and, often, associated with “domination” of the GIT microbiome by enterococci (18, 19). Historically, endocarditis is a disease that often leads to persistent bacteremia. E. faecalis is, by far, the most common cause of enterococcal endocarditis, and cases of multidrug-resistant E. faecium are rare but still occur (20). Indeed, in a recent multicenter prospective observational cohort study of enterococcal endocarditis in Spain, 7.0% of cases were attributed to E. faecium (21). However, it is unclear what proportion of these organisms were multidrug-resistant. In one of the largest cohort studies (n = 244 cases) to examine persistent enterococcal bacteremia, Bussini et al. demonstrated a higher mortality among patients with persistent enterococcal bacteremia than without (defined as positive blood cultures at least 72 hours after commencement of antibiotics) (22).

There are limited therapeutic options for the treatment of VRE bacteremia, as well as a paucity of new antimicrobial agents in the development pipeline. In the next sections, we describe the recent evidence for available treatment options for VRE infections, particularly focused on the importance of dose optimization and the emerging role of combination therapies.

Daptomycin and combinations

Mechanisms of action and resistance

Daptomycin (DAP) is a cyclic lipopeptide that was first approved for clinical use in the United States (2003) for complicated skin and skin structure infections (cSSTIs) (4 mg/kg of body weight) and later for bacteremia and right-sided endocarditis caused by Staphylococcus aureus (6 mg/kg). DAP has a distinct mechanism of action that disrupts cell membrane and cell wall homeostasis by forming a tripartite complex with phosphatidylglycerol and undecaprenyl-coupled intermediates (lipid II) at the division septum (23). The antibiotic also appears to delocalize important peripheral proteins involved in both peptidoglycan and phospholipid metabolism (23). DAP does not have a specific FDA-approved indication for infections caused by VRE but is often used “off-label” for this indication due to the potent in vitro bactericidal activity against these organisms. The European Committee on Antimicrobial Susceptibility Testing (EUCAST) does not provide breakpoints due to pharmacological (PK) evidence suggesting that high-dose DAP therapy (10–12 mg/kg/day) may be suboptimal to treat infections caused by isolates at the upper end of the wild-type distributions for E. faecalis (MIC 4 mg/L) or E. faecium (MIC 4 or 8 mg/L) (24). The Clinical and Laboratory Standards Institute (CLSI) breakpoints address this issue with two sets of breakpoints, one for E. faecium (susceptible dose dependent at ≤4 mg/L and resistant at MICs of ≥8 mg/L) and another for other enterococcal species (susceptible ≤2 mg/L, intermediate at 4 mg/L, and resistant at MICs of ≥8 mg/L) (25, 26).

DAP resistance is most commonly associated with two distinct sets of mutations involving cell envelope stress response systems (such as liaFSR, walKR, and madRS) and phospholipid biosynthetic pathways (cls, gdpD, dak, and cfa) (27, 28), and more recently, mutations in rpoB (29). Evolutionary selection experiments have shown (30) that mutations in genes encoding proteins that function in cell envelope stress signaling occur first (i.e., Ile177del LiaF in E. faecalis; W73C LiaR and T120A LiaS in E. faecium), followed most frequently by mutations in genes encoding enzymes involved in phospholipid metabolism, including the gene coding for cardiolipin synthase cls, leading to a fully resistant phenotype (28, 31). Importantly, changes in LiaFSR have been associated with DAP tolerance, or impaired antibiotic killing in vitro, and this phenomenon is postulated to contribute to enterococcal survival and the emergence of resistance at lower drug exposures (32). Furthermore, DAP bears similarities to several antimicrobial peptides of the innate immune system (33). Thus, exposure to the host environment may prime the pathways involved in resistance. Additionally, resistance has been documented to develop during the course of DAP therapy, particularly in immunocompromised patients and during severe invasive infections, with prior DAP exposure linked to the risk of isolating a subsequent DAP-resistant isolate (34). An important consideration is that the development of resistance to DAP, particularly via the LiaFSR system, is associated with increased susceptibility to β-lactams, a phenomenon designated as the seesaw effect (35, 36) whose mechanistic bases are still unknown. Both ex vivo PK/PD models and in vivo studies support using the synergistic interaction between DAP and β-lactams as a treatment strategy for high-inoculum infections caused by DAP-tolerant or -resistant strains (see below).

Clinical use of DAP monotherapy

The body of evidence suggests that the FDA-approved dose of DAP for S. aureus bacteremia (6 mg/kg/day) is insufficient to treat invasive enterococcal infections (37). Indeed, an analysis of the mutant prevention concentration for DAP dosages between 4 and 12 mg/kg/day found that the previously recommended FDA doses at 4 to 6 mg/kg/day fell into a mutant selection window, suggesting the risk of developing resistance was increased, especially in severe infections where the concentration of DAP is lower at infection sites due, in part, to the high protein binding of the antibiotic (38). The notion that doses of ≥10 mg/kg are needed for therapeutic efficacy is also supported by several in vitro PK/PD studies (32). In a recent multi-center prospective study, patients who received higher doses of DAP (≥11 mg/kg) had lower mortality than patients who received lower doses (8 to <11 mg/kg) (adjusted OR = 0.85; 95% CI = 0.73–0.99; P = 0.03), although higher doses were associated (P = 0.04) with an increased frequency of elevated creatine kinase (>2,000 U/L; 3.9%) (37). Similarly, lower doses of DAP (<8 mg/kg/day) were associated with higher rates of patient mortality in post-liver transplant patients with VRE bloodstream infection (BSI) (<60 days after transplantation) (P = 0.024) (39). Thus, in vitro, animal, and clinical data support the use of higher doses of DAP for the treatment of VRE BSI isolates with susceptible DAP MICs.

Combination of DAP plus other antibiotics

Due to the propensity of enterococci to develop resistance during therapy and the ability of these organisms to become tolerant to DAP after their cell envelope stress responses are activated by DAP and antimicrobial peptides produced by the innate immune system, there is an argument that DAP monotherapy (even at high doses) may still be inadequate to treat severe enterococcal infections, especially high inoculum infections (e.g., infective endocarditis [IE]) (40). Enterococci are intrinsically resistant to most β-lactams (particularly cephalosporins) due, at least in part, to the presence of a low-affinity class B penicillin-binding protein (PBP-4 in E. faecalis, PBP-5 in E. faecium) encoded in their core genome. Nonetheless, development of DAP resistance is associated with increased susceptibility to β-lactams (the see-saw effect), an observation that is supported by in vitro and in vivo data (35). While the specific mechanisms underlying the “see-saw” effect have yet to be fully elucidated, there is evidence that activation of the LiaFSR system and downstream proteins such as LiaX play a major role in the phenotype (36). LiaX exhibits a dual sensing and regulatory activity in the activation of the LiaFSR system and has been shown to be capable of binding both DAP and PBP-5 from E. faecium (41, 42). On the other hand, some studies have shown that exposure to β-lactams reduces the net positive bacterial surface charge and leads to an increase in DAP binding (43).

Using a PK/PD model of infective endocarditis, the combinations of DAP plus ampicillin, ertapenem, or ceftaroline exhibited enhanced efficacy against E. faecium isolates with LiaFSR substitutions (which are likely predisposed to develop DAP resistance) compared to DAP monotherapy, even at higher doses (44). Additionally, DAP plus ampicillin was effective against all strains, and no development of resistance was observed (44). This enhanced activity was also evident when using a DAP-resistant E. faecium strain (35). Using an in vivo rat model of IE, the most reliable combination was that of DAP plus ampicillin, showing a significant decrease in bacterial burden in vegetations compared to controls (35, 40). Nevertheless, it remains a challenge to demonstrate that the synergistic effect observed between DAP and β-lactam antimicrobials translates into improved clinical outcomes. A relatively recent multi-cohort study suggested that DAP plus β-lactam treatment was significantly associated with a higher rate of clinical success than DAP alone for treatment of VRE BSIs (adjusted OR = 3.19; 95% CI = 1.61–6.33; P = 0.001) (Table 1) (45). Furthermore, several case reports (46, 47) have described the successful use of the combination of DAP plus β-lactam, including in patients with difficult-to-treat VRE joint infection, bacteremia, and infective endocarditis. Of note, the combination showed synergism despite the fact that the isolates were resistant to ampicillin and had increased MICs for DAP.

Fosfomycin, an antibiotic that interferes with early stages of peptidoglycan synthesis, has synergism with DAP against methicillin-resistant Staphylococcus aureus (MRSA) in a rabbit model of endocarditis (52) and showed mixed results in a randomized trial as compared to DAP monotherapy in patients with MRSA BSI (53). Fosfomycin inhibits the MurA enzyme, which catalyzes the first committed step in peptidoglycan biosynthesis, and the mechanism of the synergistic relationship between DAP and fosfomycin is thought to be similar to that of β-lactam combinations. DAP in combination with fosfomycin has only been shown to be synergistic against VRE in vitro. Two separate patient cohort studies comparing DAP monotherapy versus DAP combination therapy with fosfomycin for treatment of VRE BSI indicated that combination therapy was associated with better clinical outcomes through associations with lower patient mortality. Both analyses indicated the combination was less effective for patients with fosfomycin-resistant isolates (MICs > 64 mg/L), with the highest fosfomycin MICs associated with increased mortality (48, 49). Furthermore, hypokalemia, a common adverse event associated with the use of fosfomycin, was frequently reported, which could be related to the interactions of multiple drugs administered to critically ill patients and limit the utility of this combination. Thus, further clinical data supporting the use of DAP with β-lactam and other antibiotics that interfere with peptidoglycan homeostasis are needed.

Linezolid and other oxazolidinones

Linezolid is an oxazolidinone with bacteriostatic activity against Gram-positive bacteria, including E. faecium and E. faecalis. Oxazolidinone antimicrobials (linezolid, tedizolid, and contezolid) bind to the 50S ribosomal subunit and disrupt protein synthesis by altering the position of the aminoacyl moiety of the aa-tRNA. Linezolid was first approved by the FDA in 2000 (600 mg every 12 hours) and remains the only approved antimicrobial for VRE BSIs. Resistance to linezolid is reported globally in enterococci and is often mediated through mutations in genes encoding the 23S rRNA involving all alleles (four alleles in E. faecalis, six alleles in E. faecium) (54). Of note, recombination between alleles can rapidly lead to the replacement of linezolid-susceptible 23S rRNA genes with the resistant allele (55). Linezolid resistance in VRE is also associated with the presence of several transferable resistance genes, including cfr, poxtA, and optrA, encoded within mobile transposons or plasmids (56). The Cfr enzyme methylates the C8 of adenosine at position 2,503 of the 23S rRNA, and poxtA and optrA are thought to mediate oxazolidinone resistance by protecting the ribosome, preventing binding of the antibiotic (57, 58).

The clinical breakpoint for linezolid susceptibility in Enterococcus spp. differs between CLSI and EUCAST guidelines, with CLSI reporting isolates with a MIC ≤2 mg/L susceptible and a MIC ≤4 mg/L intermediate. There is no intermediate classification in EUCAST, with all isolates with a MIC ≤4 mg/L deemed susceptible to the antibiotic. However, both CLSI and EUCAST consider linezolid resistance in Enterococcus spp. isolates with an MIC ≥8 mg/L. High levels of variation in linezolid plasma concentrations are reported in patients with normal to augmented renal function, the elderly, and those with critical illnesses (59). Clinical data from the compassionate use of linezolid suggested that a free drug time above the area under the curve to MIC ratio (fAUC24/MIC) of >100 correlated with successful treatment outcomes (59). Using this fAUC24/MIC target, one study concluded that the standard linezolid dose (600 mg every 12 hours) may result in insufficient exposure to effectively treat VRE isolates with higher, but susceptible, MICs (50% probability of target attainment for MIC 2 mg/L) (60, 61). In a separate multicenter observational study, there was a significant association with linezolid dose per body weight and body height and patient mortality (adjusted OR = 0.02; 95% CI = 0.002–0.23; P = 0.002) (62). However, the mean calculated fAUC24~/MIC was no different between the groups (213 versus 233, P = 0.42), and linezolid MIC up to 4 mg/L was not associated with patient mortality (P = 0.95) (62). Increased linezolid dosing also enhances the risk of hematologic toxicity (thrombocytopenia). Thus, additional studies are needed to clearly define the therapeutic target and role of linezolid dose adjustment in serious VRE infections.

Several studies have examined risk factors for clinical failure or the emergence of resistance to linezolid. In an 11-year retrospective cohort study (2007–2018) of patients who failed linezolid therapy for VRE BSI, early initiation of linezolid treatment within the first 48 hours of bacteremia (as compared to no VRE active therapy) was a protective factor from mortality in the univariate analysis (P < 0.001), but not in multivariate analyses (P = 0.088) (63). However, these results were not substantiated in a prospective 3-year, case-control study (2019–2022) of linezolid-resistant VRE infections in a region where linezolid use is high (64). Acquisition of a linezolid-resistant isolate was significantly associated with prior use of linezolid (OR = 10.13; 95% CI = 4.13–24.82; P ≤ 0.001) and carbapenems (OR = 2.85; 95% CI = 1.62–5.02; P ≤ 0.001), with linezolid-resistant isolates significantly associated with increased patient mortality (P = 0.003). Linezolid-resistant isolates were more likely to be resistant to several antibiotic classes, such as vancomycin, with some strains being only susceptible to DAP. The different outcomes between these two case series could be related to the time at which the isolates were collected (2007–2018 versus 2019–2022) since resistance to linezolid is reported to be increasing in VRE as linezolid consumption has increased (65). In a large retrospective cohort of patients with enterococcal bacteremia in the Veterans Affairs system in the US, linezolid appeared to perform worse than DAP (increased 30-day mortality; RR, 1.13; 95% CI = 1.02 to 1.26; P = 0.015), even in patients who were started on linezolid and later switched to DAP (66, 67). Thus, these data suggest that linezolid may not be the most effective antibiotic in DTR infections due to enterococci.

Tedizolid is approved in the United States (2014) and European Union (2015) for the treatment of acute bacterial skin and skin structure infections (ABSSSIs) caused by susceptible Gram-positive bacteria. Tedizolid exhibits higher potency than linezolid and retains lower MICs against linezolid-resistant Enterococcus spp. and Staphylococcus spp. Moreover, tedizolid seems to be active in vivo against S. aureus strains carrying the cfr gene (68, 69). There is some concern as to whether the lower MIC in enterococci translates to efficacy, as tedizolid was found to be inferior to either DAP or linezolid against one cfr(B) positive isolate of E. faecium in a mouse peritonitis model (70). CLSI has established a clinical breakpoint for tedizolid only for E. faecalis, defining susceptibility at an MIC ≤0.5 µg/mL, while EUCAST has not endorsed a specific breakpoint for tedizolid (71, 72). It is worth noting that tedizolid and linezolid differ significantly in their dosing regimens and resulting drug exposures. Tedizolid, when administered orally as a once-daily 200 mg dose, results in an AUC0-24 of 25.6 ± 8.4 µg⋅h/mL (73). In comparison, oral linezolid is dosed at 600 mg twice daily, yielding an AUC0-τ of 138 ± 42.1 µg⋅h/mL (74, 75).

Contezolid is in late-stage clinical development and is only approved by the National Medical Products Administration in China for the treatment of complicated skin and soft tissue infections in susceptible bacteria (76). In the United States, contezolid is undergoing clinical development for the treatment of ABSSSIs and diabetic foot infections. Preliminary data from a Phase III trial in China suggest that contezolid shows a lower tendency to depress platelet counts than linezolid, indicating it could act as an alternative treatment for patients who are unable to tolerate linezolid. There are no breakpoints available for contezolid from either CLSI or EUCAST for Enterococcus spp. but preliminary epidemiological cut-off values have been proposed for E. faecalis and E. faecium at a MIC >4 mg/L (77).

Due to the difficulties in treating DTR enterococcal infections, there is a reasonable consideration to use linezolid as part of a combination regimen. However, most in vitro studies do not support an additional benefit of linezolid combined with other anti-enterococcal antibiotics (no evidence of synergism or antagonism). Synergism between linezolid and fosfomycin has been demonstrated in vitro, with the combination of fosfomycin and linezolid resulting in a synergistic or additive effect in a panel of clinical enterococcal isolates (68.4% or n = 13/19) (78). The synergism was present in fosfomycin-resistant strains in vitro, with high concentration fosfomycin (2,048 or 4,096 mg/L) resulting in a significant decrease (2 × log10 CFU/mL) during in vitro time kill experiments. However, there was limited data on the species of enterococci, the presence of vancomycin resistance, or the mechanism of fosfomycin resistance in the strain collection tested. The synergism between linezolid and fosfomycin was further tested against vancomycin-resistant E. faecium biofilms, with higher concentrations of linezolid (2 or 4× MIC) and fosfomycin (1× MIC) resulting in a significant decrease in biofilm biomass (79). These studies indicate that more experimental data are needed to investigate the potential synergistic effect between linezolid and other antimicrobials for VRE infections. Of note, linezolid has been added to other antibiotics to increase anti-enterococcal activity in certain circumstances of “rescue” of DTR enterococcal infections, independently of the synergistic effect. This approach has not been validated, and there is a paucity of clinical data to support this practice.

Newer tetracyclines (tigecycline, eravacycline, and omadacycline)

Tigecycline is a semisynthetic glycylcycline that has enhanced binding to the 16S rRNA, which can evade some tetracycline resistance mechanisms mediated through the tet resistance genes. Tigecycline is approved for the treatment of cSSTIs, complicated intra-abdominal infections, and community-acquired pneumonia caused by susceptible Gram-negative or Gram-positive bacteria and has emerged as a “last-resort” antibiotic for VRE infections due to activity in vitro against multidrug-resistant enterococci. Resistance to tigecycline is considered low worldwide in both E. faecium (1%) and E. faecalis (0.3%), with resistance occurring through mutations in the ribosomal protection protein S10 (above) and amplification of tetracycline resistance genes tet(L) and tet(M). Tigecycline resistance has been recently described to emerge after exposure to the drug from a patient with a BSI caused by a linezolid- and vancomycin-resistant E. faecium strain through two amino acid deletions (K57 and Y58) in the S10 protein (encoded by the rpsJ gene) and a deletion in the tet(M) leader peptide, suggesting resistance can emerge de novo (80).

Eravacycline is a novel fluorocycline approved for treating complicated intra-abdominal infections caused by Gram-negative and Gram-positive bacteria. EUCAST breakpoints for eravacycline for both E. faecalis and E. faecium include susceptible (≤0.125 mg/L) and resistant (>0.125 mg/L) categories. However, the FDA breakpoint has a susceptible-only category of ≤0.06 mg/L for both species. An in vitro analysis of eravacycline activity against enterococci isolated from the Chinese CHINET study suggested robust activity against E. faecalis (97%–99.5% susceptible) and E. faecium (88.2%–94.4% susceptible) (81). Among the vancomycin-resistant E. faecium isolates, susceptibility rates were slightly lower (76.7%–90%) (81). A similar analysis using global enterococcal isolates collected from 2017 to 2020 indicated that eravacycline was highly active against E. faecalis (n = 1,876) and E. faecium (n = 1,724) (MIC90 0.06 mg/L for both), including vancomycin-resistant isolates (98.3% susceptible by EUCAST and 76.5% susceptible by CLSI) (82). Eravacycline is generally active against enterococcal isolates harboring the acquired tet [tet(M), tet(K), tet(L), tet(A), and tet(B)] resistance genes, although expansion in tet gene copy number in tandem with rpsJ mutations (encoding the ribosomal S10 protein) has been associated with resistance to both tigecycline and eravacycline (83).

Omadacycline is an aminomethylcycline antibiotic that has an FDA-approved indication for ABSSSI caused by susceptible bacteria (includes E. faecalis but not E. faecium) and also in community-associated bacterial pneumonia. This antibiotic is available in both oral and intravenous formulations. The FDA has established a susceptibility breakpoint for omadacycline at ≤0.25 µg/mL for E. faecalis, without differentiating based on vancomycin resistance status (84). Omadacycline has potent in vitro activity against E. faecium, with MIC50 and MIC90 of 0.06 and 0.12 µg/mL, respectively (>1,500 E. faecium strains, including 93.8% VRE and 93.7% harboring Tet determinants [efflux or ribosomal protection]) (85). In an in vivo murine model of peritonitis, omadacycline was highly active against multidrug-resistant E. faecium strains, including vancomycin and ampicillin-resistant and DAP-resistant, yielding significantly higher survival in animals compared to vancomycin and DAP (85).

One of the main limitations of the newer tetracyclines is that these compounds do not reach sufficient concentrations in blood to be considered agents of choice for enterococcal bacteremia. However, these compounds achieve high tissue concentrations, particularly in the biliary tract, making them useful in the treatment of intra-abdominal infections since their spectrum of activity also targets multidrug-resistant Enterobacterales, which are common pathogens in these infections. This pharmacological property and robust in vitro activity make them attractive as part of a combination therapy with agents that are optimal for bacteremia (e.g., DAP or the combination of DAP plus ampicillin) when the focus of infection involves the GI or biliary tract. Some case reports, including one with endocarditis treated with DAP plus tigecycline (86), seem to support this approach, but more robust clinical data are needed.

Oritavancin

Lipoglycopeptides (dalbavancin and oritavancin) are long-lasting, semisynthetic antimicrobials with potent activity against Gram-positive bacterial pathogens. Unlike dalbavancin, oritavancin has in vitro activity against both vanA and vanB VRE. Oritavancin inhibits peptidoglycan synthesis by binding to peptidoglycan precursors, and the mechanism of action involves additional points of contact on peptidoglycan stem peptide compared to vancomycin (87). Additionally, part of the oritavancin molecule also interacts with the membrane, likely increasing its affinity for peptidoglycan precursors and altering membrane homeostasis (88). Oritavancin has a half-life of ~393 hours, which allows for weekly administration in ABSSIs. Surveillance studies in the United States (89) and Europe (90) over 2010–2019 indicated that enterococcal isolates have remained susceptible (MIC < 0.5 mg/L) during the sampling timeframe, including vancomycin- and DAP-resistant strains. Although data are limited, preliminary studies evaluating oritavancin for the treatment of invasive enterococcal infections appear promising. A single case report suggested 1,200 mg of oritavancin for 6 weeks (with a more aggressive administration schedule) was successful for treating liver abscesses caused by linezolid- and vancomycin-resistant E. faecium (vanA, optrA, and cfrD genotype) (91). A retrospective cohort (92) and a case report (93) studied the potential for oritavancin use after BSIs or IEs treatment with standard-of-care antibiotics. Sequential therapy with oritavancin was associated with earlier discharge of patients (94%) and institutional cost avoidance (92), suggesting oritavancin could be effective as sequential antimicrobial treatment for BSIs caused by enterococci. Moreover, in vitro data suggest that oritavancin exposure can also lead to the see-saw effect and is likely to be synergistic with β-lactams (94), opening novel strategies for treatment. However, clinical data to support these assumptions are still lacking.

Teicoplanin for vanB-carrying enterococcal infections

Teicoplanin is a bactericidal glycopeptide antibiotic with a mechanism of action similar to vancomycin. Teicoplanin retains activity against VRE carrying the vanB operon but not isolates with the vanA gene cluster. However, mutations resulting in constitutive activation of the vanB operon have led to teicoplanin failures. Teicoplanin is not approved by the FDA but is used in both Australia (approved in 1994) and the European Union (approved in 1988) at a dose of 6 mg/kg/day for cSSTIs, pneumonia, and complicated UTIs and 12 mg/kg/day for bone and joint infections and infective endocarditis caused by Gram-positive bacteria. Few studies have been conducted on the efficacy of teicoplanin against enterococcal infections, despite teicoplanin being used routinely for vanB VRE BSIs in Australia. In a retrospective review at two Australian healthcare centers (2008–2014) where vanB VRE is endemic, the effectiveness of teicoplanin monotherapy was analyzed (95). Teicoplanin (6–12 mg/kg) administered 12-hourly for three doses was associated with lower rates of ICU admission at 48 hours of VRE BSI (OR = 4.16; 95% CI = 1.08–16.00; P = 0.038), but did not influence mortality at 30 days (OR = 0.57; 95% CI = 0.20–1.64; P = 0.299) compared to other VRE-active agents such as linezolid and daptomycin or sequential therapy with any of these agents (95). In a prospective multicohort study in South Korea, patients with vancomycin-susceptible enterococcal BSIs were treated with either teicoplanin or vancomycin for ≥48 hours (96). There was no significant difference in 30-day mortality (P = 0.358) or 7-day mortality (P = 0.212) in patients treated with teicoplanin or vancomycin for vancomycin-susceptible E. faecium bacteremia. Thus, more clinical data are needed to add confidence in the utilization of teicoplanin for severe infections caused by vanB-carrying VRE.

EMERGENCE OF THE PENICILLIN-RESISTANT, AMPICILLIN-SUSCEPTIBLE PHENOTYPE IN ENTEROCOCCUS FAECALIS AND IMPLICATIONS FOR THE EFFECTIVENESS OF THE AMPICILLIN-CEFTRIAXONE COMBINATION

The current standard of care for severe infections (including IE) caused by E. faecalis is the combination of ampicillin with ceftriaxone or ampicillin with gentamicin (or streptomycin). However, due to the toxicity of aminoglycosides, most clinicians are using the former combination as first-line therapy. The basis for this approach is the differential activity on PBPs of both drugs. Indeed, ampicillin has higher reactivity with PBP4 (a class B monofunctional transpeptidase), whereas ceftriaxone appears to have higher affinity for some bi-functional class A PBPs (PonA, PbpA, and PbpF) (97). This “dual” PBP effect is bactericidal and appears to be as efficacious as the combination of ampicillin plus gentamicin (98). However, in the last few years, several reports have described the emergence of reduced susceptibility of E. faecalis to penicillin, aminopenicillins, and piperacillin-tazobactam (99–101) associated with changes in the promoter of the gene encoding PBP4 or substitutions in the actual PBP that are likely to decrease the affinity of the PBP for β-lactam antibiotics. The emergence of the penicillin-resistant and ampicillin-susceptible phenotype in E. faecalis has been reported in clinical isolates from Brazil (102), South Korea (99), and Poland (103) and is associated with two major changes in PBP4. Some in vitro data suggest that these changes may diminish the activity of the ampicillin plus ceftriaxone combination, potentially leading to ineffective synergy in vivo (104). A prospective observational study suggested the 30-day mortality rates of patients with E. faecalis BSIs were twofold higher (26.9%; n = 18/67) in BSIs caused by penicillin-resistant strains than penicillin-susceptible isolates, likely due to treatment failures with piperacillin, where most MICs were ≥32 µg/mL (99). Moreover, an ampicillin non-susceptible strain of E. faecalis has emerged in Chile, harboring a deletion in the gene encoding PBP4, which may lead to a truncated PBP4 (105). However, western blot analysis showed no difference in PBP4 production compared to a wild-type laboratory strain. This finding is currently under active investigation (105). The role of the latest generation cephalosporins, such as ceftaroline and ceftobiprole, in the treatment of severe E. faecalis infections is unclear, although they are more potent than other cephalosporins and seem to have activity in combination with ampicillin in a retrospective case series (106, 107).

PHAGE AS AN EMERGING TREATMENT OPTION

The dissemination of front- and last-line antimicrobial resistance in E. faecium and E. faecalis increases the need for developing alternative therapies, particularly for long-term DTR infections where effective treatment is vital. Bacteriophage (phage) therapy is an alternative to antibiotics and a newer therapeutic option for infections caused by multidrug-resistant bacteria, including VRE, with a Phase I/Phase IIa clinical trial to evaluate the safety and efficacy of phage therapy (VRElysin) ongoing in the United States (Intralytix). Several studies have isolated phages active against VRE, including phages that can infect strains resistant to linezolid and DAP (108, 109). The development of phage cocktails, where multiple phages are given to a patient, is of particular interest since in vitro experimental evidence suggests multiple phages are more effective at killing than a single phage, potentially due to the narrow host range. Phage cocktails are thought to decrease the chance that phage-resistant mutants will emerge, as such a development may require the simultaneous evolution of multiple resistance mechanisms. A mixture of 19 enterococcal phages was combined into different cocktails to test this hypothesis with VRE, where combinations of two or three phages were able to prevent the growth of phage-resistant VRE mutants that emerged against single phages (110). Strains of VRE that were phage-resistant consistently gained mutations in the exopolysaccharide synthesis genes (epaE, epaR, or epaW) present in the Enterococcus core genome. However, further genomic studies using a different panel of phages are needed to elucidate the full suite of resistance mechanisms and the baseline prevalence of these mutations in the VRE population.

Phages also offer a potential adjunctive therapy for patients who do not respond to antibiotic treatment alone. Pre-clinical studies have demonstrated a synergistic effect from combining phages with systemic antibiotics, with certain phages able to “rescue” DAP susceptibility in resistant E. faecium strains in vitro. In a patient with recurrent vancomycin-resistant E. faecium bacteremia, antibiotic combination therapy (DAP plus β-lactam or DAP plus tigecycline) was ineffective at long-term treatment since the patient was colonized with vancomycin-resistant E. faecium, and subsequent infections were emerging from the GIT (51). The addition of phage (Φ9184 and ΦHi3) to the patient’s antimicrobial therapy was able to successfully manage the BSI and reduce the abundance of VRE in the GIT. However, an anti-phage antibody response emerged (day 337) that neutralized the phage’s activity. The rational design of phage therapy with knowledge of phage-host interactions may advance phages as antibiotic alternatives or antibiotic adjuvants for the treatment of Enterococcus spp. Despite this initial promising data, clinical data to support phage use in DTR enterococcal infection are scarce.

SUGGESTED APPROACH FOR DTR ENTEROCOCCAL INFECTIONS

Difficult-to-treat enterococcal infections usually occur in debilitated patients with multiple co-morbidities, particularly in those with hematological malignancies or solid organ transplants. Infective enterococcal endocarditis (mostly caused by E. faecalis) is also challenging to treat, even though patients may not be as immunocompromised. In any case, these infections are considered high-risk for poor outcomes and should be aggressively treated (Fig. 1). In case of bacteremia and IE, prompt microbiological eradication seems of paramount importance to improve patients’ outcomes. Additionally, source control plays a major role in treating enterococcal infections and may prevent the development of resistance during therapy.

The standard of care for severe infections caused by E. faecalis continues to be the combination of ampicillin plus ceftriaxone (see rationale above). This combination is preferred by many since it is less prone to toxicity with apparent similar efficacy. However, the emergence of the penicillin-resistant, ampicillin-susceptible phenotype may jeopardize the effectiveness of the ampicillin-ceftriaxone combination (104). A good approach to identify this phenotype is to perform penicillin susceptibilities on isolates from deep-seated infections such as IE. If the organism is found to be resistant to penicillin (likely mediated by changes in or expression of PBP4/pbp4, see above), the ampicillin-ceftriaxone combination should be cautiously used, and consideration should be given to alternative treatments. The combination of DAP plus ampicillin is likely to be highly efficacious in these circumstances (Fig. 1). Indeed, we favor the combination over DAP monotherapy due to the ability of these organisms to develop DAP resistance during therapy, particularly since DAP resistance pathways overlap with those triggered by antimicrobial peptides produced by the immune system. DAP exposure is likely to select for resistant subpopulations that have adapted the cell envelopes to the antibiotic attack. The addition of ampicillin is likely to not only prevent the emergence of resistance but also increase the therapeutic efficacy of DAP via the seesaw effect. The caveat to this approach is the limited supporting clinical data and potential toxicities. Moreover, it is unclear if a combination would be required for the entire treatment period or can be “de-escalated” (e.g., DAP monotherapy), once the patient is stable and blood cultures have cleared.

For multidrug-resistant E. faecium, the most DTR of the enterococcal species, we favor initial empiric DAP combination therapy with a β-lactam (specifically ampicillin if possible despite phenotypic resistance to this antibiotic), while awaiting susceptibility testing and guided by local antimicrobial resistance data. This preference is based on the retrospective data suggesting better outcomes with high-dose DAP as compared to linezolid, and the in vitro data suggesting ampicillin both potentiates DAP activity and may prevent the emergence of resistance. Definitive therapy should be adjusted based on susceptibility data. Again, it is unclear whether the combination would be needed for the entire course of treatment, but it seems reasonable to maintain combination therapy if source control is not fully achieved. In case of intra-abdominal infections, particularly affecting the biliary tract, or in cases of persistent bacteremia in neutropenic patients where the source is thought to be related to disruptions in the GIT microbiome, the addition of eravacycline (or tigecycline) to the DAP plus ampicillin combination might be reasonable to achieve clearance of the blood stream (Fig. 1), although only anecdotal clinical data are available for this approach.

Linezolid monotherapy or in combination with other agents is also a possibility. However, as discussed above, linezolid does not show robust evidence of synergism with any particular anti-enterococcal agent, resistance can emerge with single mutations in the 23S rRNA genes, and the toxicity profile may limit its administration for prolonged periods of time. The rationale for adding linezolid or other agents is based on increasing activity against the infecting organisms and, possibly, impairing the ability to develop resistance.

An interesting phenomenon is the emergence of a “triple threat” in enterococci that includes resistance to ampicillin, vancomycin, and DAP. The antimicrobial choices are limited in this scenario. However, due to the see-saw effect, it seems reasonable to include the combination of DAP plus ampicillin, plus the addition of another active agent (e.g., linezolid, eravacycline/tigecycline, Fig. 1) as an option in high-risk patients. A caveat of this approach is that resistance mediated by pathways other than LiaFSR may not exhibit the see-saw effect, particularly in the setting of high MICs. An interesting alternative to treat this difficult multidrug-resistant infection is the use of oritavancin (see above). This lipoglycopeptide is active in vitro against DAP-resistant and vancomycin-resistant E. faecium and seems to also trigger the see-saw effect, suggesting that combination with ampicillin (94) may add therapeutic benefit. Oritavancin can also be combined with other active agents (Fig. 1). The major issue of the use of oritavancin is that the appropriate dosing scheme is not known when used in patient populations with severe VRE infections. In addition, clinical data to support oritavancin efficacy are also lacking.

CONCLUDING REMARKS AND FUTURE DIRECTIONS

Multidrug-resistant enterococcal infections continue to be a major clinical challenge. The situation is compounded by the fact that clinical data are lacking to inform the best therapeutic approaches for these DTR infections. In the majority of cases, mechanistic insights supported by in vitro and in vivo data are the guiding principles to the therapeutic approach. Novel therapies that include phages, microbiome manipulation, and possibly, novel compounds with in vitro activity against these organisms are being tested, but most of these approaches are still far from becoming available for routine use. As such, there is an urgent need to gather clinical data to guide the best available therapies for these infections.