As Staphylococcus aureus bacteremia (SAB) is commonly encountered and carries a mortality rate of 15-50%1,2, early implementation of effective therapy is essential3. While vancomycin (VAN) remains a mainstay in MRSA bacteremia treatment, a number of factors limit its use4.
The lipopeptide daptomycin (DAP) remains an attractive alternative given its ease of dosing, bactericidal nature, and limited adverse effects with long-term use relative to other MRSA active agents4. Of concern is the recent observation of rising DAP MICs with long-term therapy largely mediated via gain-of-function mutations in the mprF gene, which is responsible for synthesis of a positively-charged outer cell membrane lipid. Increased positive surface charge reduces susceptibility to cationic antibiotics such as DAP5. While other agents exist for SAB, they may be less desirable related to pharmacodynamics, adverse effects, or stewardship concerns, making preservation of DAP highly desirable.
An alternative approach involves combination therapy with either VAN or DAP and an anti-staphylococcal β-lactam. The recent CAMERA2 trial, a multicenter randomized controlled trial of VAN or DAP with or without seven days of flucloxacillin, cloxacillin, or cefazolin, did not demonstrate improvement in a composite endpoint of mortality, persistent bacteremia at day 5, or microbiological treatment failure, and was terminated early due to increased acute kidney injury incidence in the combination therapy arm6. While the benefit of early combination therapy for MRSA bacteremia remains uncertain, current IDSA guidelines recommend high-dose DAP combined with either gentamicin, rifampin, linezolid, trimethoprim-sulfamethoxazole, or a β-lactam for refractory bacteremia7. Proposed mechanisms underpinning DAP + β-lactam therapy invoke synergy and the so-called “seesaw effect.”
The seesaw effect describes an inverse relationship between MICs to either glycopeptides or lipopeptides and β-lactams following simultaneous or consecutive exposure, generally observed in Staphylococcus aureus. The phenomenon was first observed in 1997 when a VAN resistant SA isolate (MIC = 100µg/mL) exposed to VAN developed increased methicillin susceptibility, and was later also described with DAP8,9. The effect has been demonstrated in-vitro and in-vivo, whereby β-lactams apply a selective pressure on VAN nonsusceptible or DAP nonsusceptible (DAP-NS) isolates leading to alteration in the expression of specific penicillin-binding proteins (PBPs). The magnitude of DAP potentiation has varied by PBP target, but seems to be most pronounced with β-lactams with high affinity for PBP1 (eg, cefazolin, nafcillin, meropenem)10.
Jenson et al in their recent manuscript, “Prolonged exposure to β-lactam antibiotics susceptibility of daptomycin-nonsusceptible Staphylococcus aureus to daptomycin,” examined this phenomenon in 50 MSRA isolates (25 DAP-S and 25 DAP-NS, MIC≥2µg/mL) derived from patients with SAB11. They hypothesized that exposure to β-lactams may resensitize DAP-NS S. aureus to DAP. First, they determined MICs to DAP, nafcillin (NAF), cloxacillin (LOX), ceftriaxone (CFO), and cefoxitin (FOX). Then they selected DAP-NS isolates for 28-day serial subculturing onto media with subinhibitory concentrations of these antibiotics. Finally, DAP MICs and time-kill assays, host defense peptide (HDP) susceptibilities, and whole genome sequencing were performed to examine the relationship of genetic changes to observed phenotypes.
The investigators found an inverse relationship in baseline MICs between β-lactams and DAP in DAP-NS isolates compared to DAP-S isolates. DAP MICs among DAP-S isolates ranged from 0.19—0.75 mg/L, and in DAP-NS isolates from 2-4 mg/L. A fourfold or greater decrease in MICs for at least one β-lactam was observed in 48% of pairs of overall isolates. The effect was most pronounced with NAF and LOX which have high affinity for PBP1, and less pronounced with CFO and FOX, which preferentially inhibit PBP2 and PBP4, respectively. The investigators found a statistically significant inverse correlation between DAP-NS and β-lactam MICs.
Four DAP-NS isolates either containing preexisting mprF mutations or no mutations were selected for serial passage, whereby colonies were subcultured daily for 28 days with or without low concentrations of above mentioned β-lactams. Resensitization to DAP occurred as early as day 7. LOX was most effective across all isolates, with a 6-fold DAP MIC decrease in one isolate. No change in DAP MIC was observed in the negative control
Following serial passage experiments, the authors sought to determine if MIC reductions translated to enhanced DAP killing. DAP-NS isolates that had been passaged or not with β-lactam for 28 days were evaluated in time-kill curves. The β-lactam-passaged DAP-NS isolates were all significantly killed with DAP exposure, similar to DAP-S isolates. Among the β-lactam-passaged isolates, the LOX-passaged strains exhibited the lowest survival profiles.
Finally, whole genome sequencing was performed on serially passaged isolates on days 0, 7, 14, 21, and 28. The investigators noted that DAP-NS isolates maintained pre-existing mprF mutations and gained additional mprF mutations by day 7. Impressively, LOX-passaged isolates that acquired additional mprF mutations demonstrated up to 32-fold decrease in MIC to DAP (3-4 mg/L to 0.125 mg/L).
This study demonstrates proof-of-principle of the seesaw effect and the ability of β-lactams reverse DAP-NS by inducing additional mutations in mprF. Importantly, this phenomenon is likely PBP specific, with PBP-1 specific β-lactams showing the greatest effect. Limitations of the study include use of only four isolates for serial passage, limiting generalizability, in-vitro only experiments, and the more general observation that outcomes in MRSA bacteremia are not purely driven by MICs.
While the study does not clarify when or in whom to implement combination therapy, it suggests that this need not only be done with MRSA-active β-lactams such as ceftaroline. Further research must clarify if manipulating the seesaw effect (particularly with PBP-1 selective β-lactams) is clinically relevant, or only adds drug toxicity.
Sean Bullis, MD is an infectious diseases fellow at the University of Vermont Medical Center, Burlington, VT.Andrew J. Hale, MD is an infectious diseases physician at the University of Vermont Medical Center and Assistant Professor of Medicine at Larner College of Medicine at the University of Vermont, Burlington, VT.
References
DOI: 10.1128/AAC.00890-20.