Pharmacokinetics, Tissue Distribution, and Efficacy of VIO-001 (Meropenem/Piperacillin/Tazobactam) for Treatment of Methicillin-Resistant Staphylococcus aureus Bacteremia in Immunocompetent Rabbits with Chronic Indwelling Vascular Catheters

ABSTRACT

Methicillin-resistant Staphylococcus aureus (MRSA) infections of surgically implanted subcutaneous vascular catheters (SISVCs) cause serious morbidity in patients with chronic illnesses. Previous in vitro and murine models demonstrated the synergistic interaction of equimolar concentrations of meropenem/piperacillin/tazobactam (MPT) (VIO-001) against MRSA infection. We investigated the pharmacokinetics (PK) and efficacy of VIO-001 for the treatment of MRSA bacteremia in immunocompetent rabbits with SISVCs. In PK studies, we determined that optimal dosing to achieve a time above 4× MIC (T_>4×MIC) of a duration of 3 to 3.30 h required a 1-h infusion with every-4-h (Q4h) dosing. Study groups in efficacy experiments consisted of MPT combinations of 100/150/100 mg/kg of body weight (MPT100), 200/300/200 mg/kg (MPT200), and 400/600/400 mg/kg (MPT400); vancomycin (VAN) at 15 mg/kg; and untreated controls (UC). The inoculum of MRSA isolate USA300-TCH1516 (1 × 10³ organisms) was administered via the SISCV on day 1 and locked for 24 h. The 8-day therapy started at 24 h postinoculation. There was a significant reduction of MRSA in blood cultures from the SISVCs in all treatment groups, with full clearance on day 4, versus UCs (P < 0.05). Consistent with the clearance of SISVC-related infection, full eradication of MRSA was achieved in lungs, heart, liver, spleen, and kidneys at the end of the study versus UC (P < 0.01). These results strongly correlated with time-kill data, where MPT in the range of 4/6/4 μg/ml to 32/48/32 μg/ml demonstrated a significant 6-log decrease in the bacterial burden versus UC (P < 0.01). In summary, VIO-001 demonstrated a favorable PK/pharmacodynamic (PD) profile and activity against SISCV MRSA infection, bacteremia, and disseminated infection. This rabbit model provides a new system for understanding new antimicrobial agents against MRSA SISVC-related infection, and these data provide a basis for future clinical investigation.

INTRODUCTION

Infections of surgically implanted subcutaneous vascular catheters (SISVCs), including Hickman-, Broviac-, and Porta-Cath-type vascular catheters, cause serious morbidity in patients with chronic illnesses (1–5). Such patients include those receiving cancer chemotherapy, hematopoietic stem cell transplantation (HSCT), and solid-organ transplantation. Methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible Staphylococcus aureus (MSSA) infections of SISVCs typically require the removal of these devices for optimal cure outcomes (6). Unfortunately, the removal of such devices usually occurs in patients with limited venous access, and replacement will warrant another costly operative procedure with concomitant risks of anesthesia, hemorrhage, and pneumothorax. Furthermore, MSSA and MRSA infections of intraluminal surfaces of SISVCs are also associated with bacteremia and hematogenous dissemination of bacteria to distant organ sites to cause endocarditis, pneumonia, osteomyelitis, and abscesses in the liver, spleen, kidneys, and central nervous system (7, 8).

The development of strategies for the prevention and treatment of SISVC infections caused by MRSA may greatly improve quality of life, reduce costs, and decrease morbidity. The use of “lock therapy” for the treatment of MRSA SISVCs has been described using ethanol, EDTA, and proprietary antiseptics (9–13). However, these agents could not be reliably used for the treatment of infections in all distant organ sites. While vancomycin (VAN) can also serve as a viable agent for lock therapy, it has the added utility of systemic administration for MRSA infections, albeit this route is encumbered by the need for therapeutic drug monitoring and the risk of nephrotoxicity. Although daptomycin and oxazolidinones are also effective in the treatment of MRSA bacteremia, there are limited data on their management of vascular catheters that cannot be readily removed.

As a novel approach to the treatment of MRSA infections, Rosenblatt et al. hypothesized that the combination of meropenem (MER), piperacillin (PIP), and tazobactam (TAZ) operates through meropenem inhibition of penicillin binding protein 1 (PBP1), piperacillin binding to PBP2, and tazobactam’s protection of piperacillin from class A β-lactamases as well as by meropenem’s allosteric opening of the active site of PBP2a for inhibition by other antimicrobial molecules in the combination, resulting in synergistic inhibition of multiple components of MRSA cell wall synthesis (13). VIO-001 is a formulation developed by Viosera Therapeutics as a combination of MER, PIP, and TAZ at a ratio that covers clinical MRSA isolates with a variety of resistance profiles while suppressing the development of drug resistance due to its internal collateral sensitivity (14, 15). Composed of safe, well-studied components, VIO-001 is intended to be used as an extended infusion every 8 h in patients suffering from life-threatening systemic infections. Antimicrobial strategies that are developed for the treatment of SISVC infections should be able to treat both vascular catheter and concomitant systemic staphylococcal infections in deep tissues.

We hypothesized that the VIO-001 formulation might possess these properties for simultaneous lock therapy and systemic treatment. The proposed strategy would use a solution of meropenem/piperacillin/tazobactam (MPT) and heparin in SISVCs of HSCT recipients and patients receiving therapy for hematological malignancies. In order to evaluate the optimal dosing regimen and prevention and treatment durations of the VIO-001 formulation in the treatment of infections caused by MRSA of surgically implanted SISVCs, we studied the combination of MPT in an experimental model of MRSA chronic indwelling vascular catheter infection in immunocompetent rabbits.

The VIO-001 combination of meropenem/piperacillin/tazobactam may have particular application in the empirical treatment of suspected sepsis where vancomycin may be contraindicated. Such conditions would include preexisting hypersensitivity to vancomycin or renal impairment. High dosages of meropenem, piperacillin, and tazobactam may also be especially effective in the empirical management of septic patients where suspected Gram-negative bacillary infections with elevated MICs would require plasma exposures to maintain the time above the MIC (T_>MIC).

RESULTS

In vitro efficacy studies.

The MICs of the three compounds MER, PIP, and TAZ as single agents and in the VIO-001 formulation (MPT) against MRSA are depicted in Table 1. Studies were conducted in three replicate experiments with the final MIC of MER being 4 μg/ml, that of PIP being >128 μg/ml, that of TAZ being 128 μg/ml, and that of VAN being 1 μg/ml. The combination of the formulation gave final concentrations of MPT at a ratio of 2:3:2 equal to 2/3/2 μg/ml and at a ratio of 1:1:1 equal to 2/2/2 μg/ml.

TABLE 1.

MICs of meropenem, piperacillin, and tazobactam alone or VIO-001 (meropenem/piperacillin/tazobactam [MPT]) against MRSA isolates used in the study

Drug	Final MIC(s) (μg/ml)
Meropenem	4
Piperacillin	>128
Tazobactam	>128
Vancomycin	1
MPT at 2:3:2	2/3/2
MPT at 1:1:1	2/2/2

When examined in time-kill assays, the VAN control demonstrated growth at 0.5 μg/ml, an ∼90% reduction at 1 to 2 μg/ml, and nearly 99% reduction at 4 μg/ml. In comparison, the VIO-001 formulation produced a concentration-dependent killing effect of >99.9% kill at the two highest concentrations (Fig. 1). Of note, meropenem degrades rapidly in aqueous solutions in a temperature- and concentration-dependent manner (16), which makes the 24-h time-kill assay unsuitable for this compound.

FIG 1.

Time-kill assay of vancomycin (VAN) (A) and VIO-001 (meropenem/piperacillin/tazobactam [MPT]) (B) against methicillin-resistant Staphylococcus aureus (MRSA). The effects of treatment with concentrations of VAN at 0.5, 1, 2, and 4 μg/ml and MPT at 1/1.5/1, 2/3/2, 4/6/4, and 8/12/8 μg/ml were studied in relation to no treatment (UC). Data are plotted as the means ± SEM from three separate experiments for each growth curve. As the SEM was small for several time points, the error bars may not always be apparent in the time-kill curves.

Pharmacokinetics.

In order to determine the plasma profile of the VIO-001 formulation, simultaneous plasma pharmacokinetics (PK) of MER, PIP, and TAZ were performed with single-dose pharmacokinetics using 10-min, 30-min, and 1-h infusions. A single-dose plasma kinetic profile of the 10-min infusion demonstrated that an area under the plasma drug concentration-time curve extrapolated to 4 h (AUC_0–4) depicted dose proportionality for MER and TAZ and a somewhat wider range of AUCs with piperacillin over three dosage regimens from 100 mg/kg of body weight to 200 mg/kg depending on the compound.

The plasma concentration-time curves of the 10-min infusion demonstrated that there was no detectable MER, PIP, or TAZ by the end of 2 h in virtually all the dosing regimens. However, at a dosage of 200 mg/kg in the 10-min infusion cohort, detectable drug was observed after 3 h. Tissue distribution studies after the 10-min infusion demonstrated substantial concentrations of PIP and TAZ in lung, liver, spleen, and kidney, with only PIP achieving any notable concentrations in the cerebral tissue (Fig. 2 and 3).

FIG 2.

Plasma pharmacokinetics expressed as concentration-time curves of meropenem, piperacillin, and tazobactam after a single-dose intravenous 10-min infusion of VIO-001 (meropenem/piperacillin/tazobactam) at 100/150/100 mg/kg (n = 4), 150/225/150 mg/kg (n = 4), and 200/300/200 mg/kg (n = 4) to healthy New Zealand White rabbits.

FIG 3.

Concentrations of the VIO-001 formulation (meropenem/piperacillin/tazobactam [MPT]) in cerebrum, heart, lung, spleen, liver, and kidney after single-dose intravenous 10-min infusions of 100/150/100 mg/kg (MPT100) (n = 4), 150/225/150 mg/kg (MPT150) (n = 4), and 200/300/200 mg/kg (MPT200) (n = 4) to healthy New Zealand White rabbits. Tissues were obtained after rabbits were euthanized 10 min after the 10-min infusion.

A subsequent study was then conducted on single-dose plasma pharmacokinetics with a 30-min infusion, which demonstrated an AUC_0–4 value for MER and PIP similar to that of a 10-min infusion but with a somewhat greater AUC_0–4 of TAZ (data not shown). The plasma concentration-time curves of these compounds demonstrated a longer exposure time above the MIC. However, since the target for the time above the MIC had not been fully met, a 1-h infusion was then attempted at higher dosages that explored three times the dosage level of up to 600 mg/kg of MER, 900 mg/kg of PIP, and 600 mg/kg of TAZ (Table 2). These dosage regimens resulted in substantially higher AUCs and more sustained times above the MIC over the 4-h interval (Table 3 and Fig. 4).

TABLE 2.

Single-dose plasma total drug pharmacokinetic parameters of meropenem/piperacillin/tazobactam after intravenous 1-h infusion administration at 200/300/200 mg/kg, 400/600/400 mg/kg, and 600/900/600 mg/kg to healthy New Zealand White rabbits

Dose (mg/kg Q4h)	Mean AUC0–4 (μg · h/ml) ± SEM	Mean Cmax (μg/ml) ± SEM	Mean CL (ml/h/kg) ± SEM	Mean V (liters/kg) ± SEM
Meropenem
200	284.8 ± 22.8	229.9 ± 20.0	715.7 ± 59.2	281.3 ± 27.2
400	746.4 ± 25.9	716.9 ± 91.5	536.8 ± 17.8	455.4 ± 271.2
600	782.6 ± 78.5	690.1 ± 93.3	787.0 ± 69.3	332.1 ± 41.1
Piperacillin
300	456.9 ± 81.7	483.5 ± 46.3	734.2 ± 147.8	230.1 ± 38.5
600	1,550.3 ± 113.3	1,245.8 ± 130.7	392.1 ± 31.3	183.8 ± 12.5
900	1,554.3 ± 104.1	1,096.9 ± 120.1	585.4 ± 41.8	464.51 ± 73.7
Tazobactam
200	483.0 ± 78.7	425.6 ± 62.3	450.8 ± 75.8	194.3 ± 21.8
400	1,235.4 ± 84.13	990.2 ± 53.1	327.7 ± 23.6	178.1 ± 14.9
600	1,084.8 ± 103.9	722.4 ± 96.6	569.3 ± 59.6	558.2 ± 116.8

TABLE 3.

Percentage of time above the predicted MIC after a single dose of meropenem/piperacillin/tazobactam after intravenous 1-h infusion administration at 200/300/200 mg/kg to healthy New Zealand White rabbits over a 4-h interval

Drug (dosage [mg/kg])	%T>MIC over a 4-h interval at MIC (μg/ml) of:
	1	2	4	8
Meropenem (200)	94	75	73	69
Piperacillin (300)	83	74	71	68
Tazobactam (200)	94	86	75	74

FIG 4.

Plasma pharmacokinetics expressed as concentration-time curves of meropenem, piperacillin, and tazobactam after a single-dose intravenous 1-h infusion of VIO-001 (meropenem/piperacillin/tazobactam) at 200/300/200 mg/kg (n = 4), 400/600/400 mg/kg (n = 4), and 600/900/600 mg/kg (n = 4) to healthy New Zealand White rabbits.

Tissue distribution studies after the 1-h infusion again demonstrated the highest tissue concentrations for piperacillin compared to those of tazobactam and meropenem for most sites, including the cerebrum, heart, lung, spleen, and kidney. Reflecting the renal concentrations of these agents, the highest tissue levels were observed in the kidneys (Fig. 5 and 6).

FIG 5.

Concentrations of the VIO-001 formulation (meropenem/piperacillin/tazobactam [MPT]) in cerebrum, heart, lung, spleen, liver, and kidney after single-dose intravenous 1-h infusions of 200/300/200 mg/kg (MPT200) (n = 4), 400/600/400 mg/kg (MPT400) (n = 4), and 600/900/600 mg/kg (MPT600) (n = 4) to healthy New Zealand White rabbits. Tissues were obtained after rabbits were euthanized 10 min after the 1-h infusion.

FIG 6.

Concentrations of the VIO-001 formulation (meropenem/piperacillin/tazobactam [MPT]) in skin, muscle, and bone after single-dose intravenous 1-h infusions of 200/300/200 mg/kg (MPT200) (n = 4), 400/600/400 mg/kg (MPT400) (n = 4), and 600/900/600 mg/kg (MPT600) (n = 4) to healthy New Zealand White rabbits. Tissues were obtained after rabbits were euthanized 10 min after the 1-h infusion.

In vivo efficacy studies.

Rabbits receiving VIO-001 had clearance of all blood cultures drawn through the SISVC on day 2 (1 day after the initiation of treatment), with sustained negative blood cultures measured each day from day 3 through day 10. VAN also showed similar activity, with clearance of blood cultures on day 2 and no subsequent positive blood cultures thereafter. In comparison, untreated control rabbits demonstrated progressively increased concentrations of Staphylococcus aureus drawn through the SISVC, reflecting an expanding population of biofilm from day 1 through day 8 (P < 0.05) (Fig. 7A and B). Consistent with these observations, the quantitative postmortem cultures of the SISVC wash specimens demonstrated approximately 10⁴ CFU/ml in the untreated controls, whereas the VIO-001-treated rabbits with MPT100, MPT200, and MPT400, as well as those treated with VAN, had no detectable organisms in the catheter wash specimens (P < 0.01) (n = 6 catheters per treatment group) (Fig. 7C). Also consistent with these data from quantitative blood cultures and quantitative catheter wash specimens, bacterial cultures of liver, spleen, lung, heart, and kidney demonstrated ≥10⁴ CFU/g in the untreated controls, while no organisms were detected in any of the tissues of the animals treated with VIO-001 or VAN (P < 0.01) (Fig. 8).

FIG 7.

Quantitative serial blood cultures of surgically implanted subcutaneous vascular catheters (SISVCs) (A), treatment with the VIO-001 formulation (meropenem/piperacillin/tazobactam [MPT]) only (B), and quantitative cultures of SISCV wash specimens (C) after treatment of methicillin-resistant Staphylococcus aureus (MRSA) infection of chronic indwelling vascular catheters in immunocompetent rabbits. Groups consisted of VIO-001 administered Q4h with a 1-h infusion at 100/150/100 mg/kg (MPT100), 200/300/200 mg/kg (MPT200), and 400/600/400 mg/kg (MPT400); vancomycin (VAN) Q12h at 15 mg/kg; and untreated control rabbits (Control). There were no detectable organisms in the catheter wash specimens from VIO-001- or VAN-treated rabbits (n = 6 catheters per treatment group). Data are plotted as the means ± SEM from three separate experiments. §, P < 0.01.

FIG 8.

Response of methicillin-resistant Staphylococcus aureus (MRSA) infection of chronic indwelling vascular catheters in immunocompetent rabbits after an 8-day treatment course with VIO-001 (meropenem/piperacillin/tazobactam [MPT]) administered Q4h with a 1-h infusion at 100/150/100 mg/kg (MPT100) (n = 6), 200/300/200 mg/kg (MPT200) (n = 6), and 400/600/400 mg/kg (MPT400) (n = 6); vancomycin (VAN) Q12h at 15 mg/kg; and untreated control rabbits (Control) (n = 6). The data demonstrated full eradication of residual bacterial burdens in liver, spleen, lung, heart, and kidney. Data are plotted as the means ± SEM from three separate experiments. §, P < 0.01.

DISCUSSION

Chronic indwelling central venous catheters are a critical component of supportive care for the administration of multiple chemotherapeutic agents, immunomodulators, and antimicrobial agents as well as for multiple blood draws in patients with chronic illnesses, including those with hematological malignancies, solid tumors, stem cell transplantation, solid-organ transplantation, and immune deficiencies. Infections caused by Staphylococcus aureus of SISVCs warrant the removal of the catheter device as the standard of care (6). However, the removal of such devices often poses considerable challenges for patients with chronic illness, who have limited venous access. The insertion of another catheter may also require another operative procedure, including general anesthesia.

Among the organisms causing SISVC infections, MRSA poses even more challenges in the limited therapeutic agents available to treat catheter-related infections or bacteremia. Although VAN is commonly used, patients with impaired renal function are at risk for dose-dependent cumulative nephrotoxicity as well as the challenges involved in therapeutic drug monitoring and optimized dosing in a setting of fluctuating renal function (17). Other alternatives, such as daptomycin and oxazolidinones, are effective in the treatment of MRSA bacteremias, but their spectrum is limited to Gram-positive pathogens for empirical therapy of catheter-related sepsis. Although trimethoprim/sulfamethoxazole and doxycycline may also be used to treat MRSA infections, they are seldom used in the management of MRSA vascular catheter-related bacteremias.

The near-equimolar concentrations of meropenem, piperacillin, and tazobactam when studied in vitro were found to be effective against MRSA as well as other Gram-positive organisms (14, 15). When studied in the immunosuppressed murine model of disseminated MRSA infection, this formulation was shown to clear MRSA within 11 h in treated mice, while untreated mice all succumbed to infection in 48 h (14). However, the short half-lives of VIO-001 components in mice (∼3 min) and rats (∼10 min) (18) in addition to the instability of tazobactam in rats made PK/pharmacodynamic (PD) studies extremely challenging. In order to further explore the potential for this formulation in the treatment of SISVC MRSA infections in a dosing regimen similar to the intended clinical dosing (extended intravenous [i.v.] infusions over >1 week), we investigated this formulation in immunocompetent rabbits with experimental SISVC infection, bacteremia, and deep-organ infection caused by MRSA.

In order to further understand the pharmacokinetics of the VIO-001 formulation, a series of single-dose studies was conducted with 10-min, 30-min, and 1-h infusions, with the ultimate objective to determine the infusion time for maintaining the time above 4× the MIC for at least 3 h above the MIC. A review of the PK data demonstrated that the infusion over 1 h in the rabbits would allow a plasma concentration in total serum of all three agents at 4× above the MIC for approximately 3.5 h. The parallel tissue distribution studies at the end of the infusion also demonstrated that there would be adequate concentrations in liver, spleen, kidney, heart, lung, and cerebrum. Substantiating these calculations, the dosage and infusion regimen eradicated bloodstream infection measured from the catheter within 2 days and completely cleared affected deep tissues to the lower limit of detection.

The effect of VIO-001 was comparable to that of VAN in clearing the bloodstream as well as in eradicating the organism burden in deep tissues. The extent of infusion in the rabbits was well tolerated, and all rabbits survived deep-seated MRSA infection. Although the standard of care in patients with MRSA bacteremia includes the removal of the source of infection, including SISVCs, these findings demonstrate that both VIO-001 and VAN over the course of time eradicated organisms from vascular catheters, the bloodstream, and deep tissues. Further study is warranted to assess the potential role of these agents in eradicating infection, particularly from vascular catheters in patients with SISVCs in whom removal is feasible and in whom the potential for clearance of MRSA bacteremia is possible. As β-lactams are considered to be superior to VAN in treating severe MSSA infections (19–21), VIO-001 is a β-lactam combination therapy that warrants further study for the treatment of MRSA diseases due to its unique resistance-suppressing property.

The broad spectrum of VIO-001 has also been demonstrated in vitro to be active against MSSA, VAN-resistant Enterococcus faecium (VRE), and other drug-resistant staphylococci, including Staphylococcus pseudintermedius (15). The very broad combination of meropenem and piperacillin/tazobactam, while at a different ratio than that of commercially available products, also introduces the concept of this potential combination being used in complex life-threatening polymicrobial infections that may include MRSA and Enterobacterales, Pseudomonas aeruginosa, and anaerobes, for example, polymicrobial pneumonia in immunocompromised patients, complicated polymicrobial skin and soft tissue wounds, and/or osteomyelitis, as may occur in diabetes mellitus or in traumatic injuries and burns.

The levels of exposure, as measured by the AUC and the maximum observed plasma concentration (C_max), in previous animal model systems evaluating meropenem and piperacillin/tazobactam, as well as typical human exposures to these antimicrobial agents, are substantially higher in our rabbit model in order to achieve an effective level of MRSA activity. The safety and tolerability of these levels of meropenem, piperacillin, and tazobactam warrant preclinical toxicological studies and phase 1 clinical trials in healthy adult volunteers. Such exposures, however, may also be beneficial in the treatment of Gram-negative bacillary infections in order to maintain the time above the MIC for organisms that may already have elevated MICs.

A rabbit model of catheter-related MRSA bacteremia and disseminated infection provides a potentially powerful system by which to assess new antimicrobial agents for the treatment of MRSA central line-associated bloodstream infection (CLABSI). The system presented here allows the assessment of the direct impact of antimicrobial agents on established MRSA biofilms as well as infection in a wide range of organ and tissue sites. Moreover, this rabbit model permits the evaluation of the efficacy of the impact of the removal of the vascular catheter as well as the potential for therapeutic agents to eradicate infection on the central venous line. The experiments conducted in this study specifically evaluated the impact of the VIO-001 combination of meropenem, piperacillin, and tazobactam versus vancomycin in the eradication of MRSA infection in a CLABSI rabbit model. Study of the efficacy of the impact of the removal of vascular catheters was beyond the scope and objectives of this particular study. However, one should note that the given regimens used were able to fully eradicate MRSA from the vascular catheter, prompting the question of novel strategies that may be developed for patients whose vascular catheters may not be removable for the management of MRSA CLABSI.

MATERIALS AND METHODS

Study drugs.

The VIO-001 formulation (meropenem/piperacillin/tazobactam) was provided by Viosera Therapeutics (Brentwood, MO) as the separate compounds MER, PIP, and TAZ: TAZ (sodium salt) as 1-g vials, PIP (sodium salt) as 1-g vials, and MER for injection as 1-g vials (meropenem plus sodium carbonate) (ACS Dobfar). A 0.9% sodium chloride solution in 250-ml i.v. bags was used as the vehicle to reconstitute VIO-001.

Bacterial isolates.

USA300-TCH1516, a Panton-Valentine leukocidin (PVL)-positive clinical MRSA isolate (ATCC BAA-1717; American Type Culture Collection [ATCC]), was used in the study. The reference (quality control [QC]) isolate ATCC 29213 was acquired from the ATCC and used as the internal control for the standardization of in vitro susceptibility testing of antimicrobial compounds. The isolates were maintained on CryoBeads at −80°C in the specimen archives of the Transplantation-Oncology Infectious Disease Program at Weill Cornell Medicine. The MRSA isolate USA300-TCH1516 and QC isolate ATCC 29213 were subcultured onto tryptic soy agar (TSA) plates and incubated at 37°C for 24 h prior to susceptibility testing.

In vitro studies. (i) In vitro antimicrobial susceptibility.

The MIC assays for inhibition of growth were performed according to the recommendations of the Clinical and Laboratory Standards Institute (CLSI). The MICs of the VIO-001 formulation (MPT) at different ratios and of MER, PIP, and TAZ as single agents were determined according to the testing standard broth microdilution methods described in CLSI document M07-A8 against the MRSA USA300-TCH1516 isolate (22). The VIO-001 formulations and MER, PIP, and TAZ were dissolved in sterile water and serially diluted with cation-adjusted Mueller-Hinton broth (CAMHB) to the appropriate concentrations for in vitro susceptibility testing.

The following concentration ranges of antimicrobial agents were tested for susceptibility against isolates of MRSA USA300-TCH1516: meropenem from 0.125 to 128 μg/ml, piperacillin from 0.125 to 128 μg/ml, tazobactam from 0.125 to 128 μg/ml, and the VIO-001 formulation at ratios of 2:3:2.

(ii) Preparation of the inoculum and inoculation of microplates.

A standardized inoculum was prepared using direct colony suspension. Within 15 min of preparation, the adjusted inoculum suspension was diluted in water to achieve final concentrations of approximately 5 × 10⁵ CFU/ml (range, 2 × 10⁵ to 8 × 10⁵ CFU/ml) in each well. The dilution procedure to obtain this final inoculum varied according to the method of delivery of the inoculum to the individual wells and was calculated for each situation. For microdilution tests, a 100-μl inoculum volume delivered to the wells was used. If the volume of broth in the well was 100 μl and the inoculum volume was 10 μl, a 0.5 McFarland suspension (1 × 10⁸ CFU/ml) had to be diluted 1:20 to yield 5 × 10⁶ CFU/ml. When 10 μl of this suspension was inoculated into the broth, the final test concentration of bacteria was approximately 5 × 10⁵ CFU/ml (or 5 × 10⁴ CFU/well in the microdilution method) (measured by a spectrophotometer [model no. UV-1600PC; VWR International, Radnor, PA, USA]). Within 15 min after the inoculum had been standardized as described above, each well was inoculated in a microdilution tray using an inoculator device that delivered a volume that did not exceed 10% of the volume in the well (e.g., ≤10 μl of the inoculum in 100 μl of the antimicrobial agent solution). A purity check was performed on the inoculum suspension by subculturing an aliquot onto a TSA plate for simultaneous incubation. Inoculated microdilution plates were incubated at 35°C for 16 to 20 h in an ambient-air incubator within 15 min of the addition of the inoculum.

(iii) Time-kill assays.

A microplate method was performed in sterile, flat-bottomed, 96-well microplates. The 5× drug solutions were diluted with CAMHB to obtain 2× the final concentrations. Next, 100 μl of the 2× drug solution was dispensed into the wells. The working suspension of the inoculum, prepared spectrophotometrically, was diluted in CAMHB to obtain a 2× final suspension. The bacterial suspension in CAMHB (100 μl) was added to each well, resulting in the desired final drug concentration and inoculum size.

The microplates were incubated for 24 h at 35°C in a Multiscan Sky microplate spectrophotometer (model no. 51119700D; Thermo Fisher Scientific Inc.) and read every hour with the microplate reader at 600 nm.

Bacterial inoculum and inoculation.

The MRSA USA300-TCH1516 isolate (independently determined to have an MPT MIC of 4/4/4 μg/ml by Viosera Therapeutics [Brentwood, MO] and JMI Laboratories [North Liberty, IA]) was used in the studies. The MIC for the MRSA USA300-TCH1516 isolate was determined according to CLSI standards.

The inoculum of MRSA USA300 was prepared using direct colony suspension. For the preparation of the inoculum, the isolate was subcultured from a frozen stock culture stored at −80°C on CryoBeads onto TSA plates and incubated at 37°C for 24 h. The concentration of the inoculum was adjusted in order to give each rabbit a predetermined inoculum of 1.0 × 10³ organisms suspended in a volume of 5 ml of 0.9% normal saline. The inoculum size was confirmed by serial dilution and culture on TSA plates.

Inoculation with MRSA was performed on day 1 of the study in order to establish colonization and biofilms in the catheter and establish subacute bloodstream infection. The MRSA inoculum of 1 × 10³ organisms (suspended in a 5-ml volume of 0.9% sterile normal saline) was slowly administered to each rabbit via the indwelling Silastic central venous catheter and left in the catheter for 24 h.

Animals.

Healthy female New Zealand White rabbits (Covance Research Products Inc., Denver, PA) were used in all studies. All rabbits were monitored under humane care and use standards, in the RARC facilities, accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International, according to the guidelines of the National Research Council for the care and use of laboratory animals (23), and under approval by the Institutional Animal Care and Use Committee (IACUC) of Weill Cornell Medicine of Cornell University, New York, NY.

New Zealand White rabbits weighing 2.5 to 3.5 kg and 6 months of age were individually housed in rabbit cages, and food (5326 Hi-Fiber Rabbit Feed LabDiet manufactured by PMI Nutrition International LLC) and water (supplied by New York City and meets human drinking standards) were provided ad libitum. Individual cage cards and ear tattoos were used for rabbit identification. The rabbits were used in the studies after acclimation for at least 72 h (3 days). The rabbits were housed in a temperature-controlled room at a temperature of 18°C ± 4°C and humidity of up to 70%. The lighting cycle in the room was 12 h of light and 12 h of darkness; the lighting cycle could be interrupted for the performance of protocol-defined activities. Atraumatic vascular access was established by the surgical placement of a Silastic tunneled central venous catheter as previously described by Walsh et al. (24).

The presence of the central Silastic venous catheter in our rabbit experimental models also closely recapitulates the clinical picture of critically ill patients. The vascular catheter permits atraumatic venous access for detailed sampling of concentrations of antimicrobial agents, blood cultures, biochemical parameters, cytokines, and biomarkers as well as the administration of multiple medications, including chemotherapeutic agents and immunosuppressive agents, similar to those of patients. Peripheral blood sampling was performed from the auricular vena under intravenous anesthesia with 0.3 ml to 0.5 ml of a 2:1 (vol/vol) mixture of ketamine at 100 mg/ml and xylazine at 20 mg/ml. Rabbits were euthanized according to Animal Care and Use Committee-approved prespecified humane endpoints or at the end of the study by the intravenous administration of pentobarbital (120 mg/kg of body weight of pentobarbital sodium in the form of Euthasol euthanasia solution [Virbac AH Inc., Forth Worth, TX]).

Study design. (i) Single-dose plasma pharmacokinetics: 10-min infusion.

Single-dose plasma pharmacokinetics (PK) of the VIO-001 formulation were performed in three dosage cohorts with 4 rabbits per group (n = 12 rabbits). Healthy rabbits with central Silastic venous catheters received the VIO-001 formulation at a ratio of 2:3:2 in three dosage cohorts, 100/150/100 mg/kg, 150/225/150 mg/kg, and 200/300/200 mg/kg, as a single dose intravenously over 10 min.

Blood samples were obtained from each rabbit at the following time points: baseline, 10 min, 15 min, 20 min, 30 min, 45 min, 1 h, 2 h, and 4 h. Rabbits received the next i.v. dose of the VIO-001 formulation after a 24-h washout. A blood sample was obtained before euthanizing the rabbit at 10 min postinfusion at the baseline and 10-min time points.

Blood samples were collected into 3-ml heparinized syringes, transferred into 15-ml polypropylene conical Falcon tubes (Becton, Dickinson Labware, Franklin Lakes, NJ), and separated via centrifugation at 400 × g for 10 min at 4°C, and plasma was stored at approximately −80°C in 2-ml Sarstedt microtubes until analysis.

(ii) Tissue distribution studies (10-min infusion).

For the assessment of tissue concentrations of the VIO-001 formulation, rabbits were euthanized 10 min after infusion of the last dose by i.v. administration. Cerebrum, heart, lung, liver, spleen, and kidney tissues were harvested for tissue concentration determinations.

(iii) Single-dose plasma pharmacokinetics: 1-h infusion.

Single-dose plasma PK of the VIO-001 formulation were performed in three dosage cohorts with 4 rabbits per group (n = 12 rabbits). Healthy rabbits with central Silastic venous catheters received the formulation at a ratio of 2:3:2 in three dosage cohorts, 200/300/200 mg/kg, 400/600/400 mg/kg, and 600/900/600 mg/kg, as a single dose. The VIO-001 formulation was administered i.v. over a 1-h infusion.

Blood samples were obtained from each rabbit at the following time points: baseline, 30 min, 1 h, 2 h, 3 h, 4 h, and 6 h. Rabbits received the next i.v. dose of the VIO-001 formulation after a 24-h washout. Blood samples were obtained before administration of the VIO-001 formulations and after the 1-h infusion before euthanizing the rabbits at the baseline and 1-h time points.

Blood samples were collected into 3-ml heparinized syringes, transferred into 15-ml polypropylene conical Falcon tubes (Becton, Dickinson Labware, Franklin Lakes, NJ), and separated via centrifugation at 400 × g for 10 min at 4°C, and plasma was stored at approximately −80°C in 2-ml Sarstedt microtubes until analysis.

(iv) Tissue distribution studies (1-h infusion).

For the assessment of tissue concentrations of MPT, rabbits were euthanized 10 min after the 1-h infusion. Cerebrum, heart, lung, liver, spleen, kidney, skin, muscle, and bone tissues were harvested for tissue drug concentration determinations.

In vivo studies.

In vivo studies were performed to determine the efficacy of VIO-001 formulations versus VAN for the treatment of MRSA-infected SISVCs in immunocompetent rabbits (n = 30 rabbits).

Antimicrobial therapy.

The objective of the study was to determine the efficacy of the three VIO-001 formulations versus VAN for the treatment of MRSA-infected SISVCs in immunocompetent rabbits. Rabbits were divided into 5 cohort groups: three treatment groups with the VIO-001 formulation at 100/150/100 mg/kg (MPT100) (n = 6), 200/300/200 mg/kg (MPT200) (n = 6), and 400/600/400 mg/kg (MPT400) (n = 6); VAN at 15 mg/kg; and untreated control rabbits (UC) (n = 6). Antimicrobial therapy with VIO-001 formulations began 24 h after intravenous inoculation every 4 h (Q4h) with a 1-h infusion and continued for 8 days. The last dose of the VIO-001 formulation was administered in the morning on day 8. The untreated control rabbit group received intravenous administration of 0.9% normal saline 24 h after inoculation for 8 days of the study. VAN powder was dissolved in 0.9% normal saline and administered intravenously 24 h after inoculation Q12h. Quantitative blood cultures of SISVCs and quantitative peripheral blood cultures were performed every day throughout the study.

Surviving rabbits were euthanized on day 10 of the study 24 h after the last administration of antimicrobial drugs.

Outcome variables. (i) Pharmacokinetic analysis.

(a) Analytical assay. Concentrations of the VIO-001 formulation in plasma and tissues were determined using validated high-performance liquid chromatography assays at the Center for Anti-Infective Research and Development (CAIRD), Hartford Hospital. The MER assay was linear over a range of 0.25 μg/ml to 40 μg/ml (R² ≥ 0.995). Intraday and interday coefficients of variation for the low (0.5 μg/ml)- and high (30 μg/ml)-quality control samples were <8% (25). The PIP/TAZ assay was linear over a range of 2 to 100 μg per ml/1 to 50 μg per ml (R² ≥ 0.995). Intraday and interday coefficients of variation for the low (6/3 μg/ml)- and high (80/40 μg/ml)-quality control samples were <6% (26). Concentration results were considered acceptable if mean sample concentration results are within or equal to ±10% of the theoretical concentrations. Each sample concentration result was considered acceptable if it was within or equal to ±15%. After acceptance of the analytical results, backup samples were discarded.

(b) Pharmacokinetic analysis. Pharmacokinetic parameters for the VIO-011 formulation were determined from plasma concentration data using noncompartmental methods (WinNonlin Professional version 4.1; Pharsight Corp., Mountain View, CA). Pharmacokinetic measures are the maximum observed plasma concentration (C_max), the area under the plasma concentration-time curve (AUC) through 1 h after the first dose (AUC_0–1) calculated by the linear trapezoidal rule, clearance (CL) calculated by dividing the dose by the AUC, the volume of distribution (V) at steady state (V_ss) calculated by multiplying the dose by the ratio of the area under the first-moment curve to the square of the AUC, and the terminal elimination half-life (t_1/2) calculated from a linear regression of the log-linear portion of the log concentration-time curve.

(ii) Antimicrobial efficacy.

The antimicrobial efficacy of the VIO-001 formulation in the model of methicillin-resistant Staphylococcus aureus infection of chronic indwelling vascular catheters in immunocompetent rabbits was determined by quantitative clearance of MRSA USA300 from the intravenous Silastic catheter and multiple deep tissues.

(a) Quantitative blood cultures of SISVCs. Quantitative blood cultures of SISVCs were performed 24 h after inoculation with MRSA, and later quantitative blood cultures of SISVCs were performed every day throughout the study. An aliquot of blood (100 μl to 1,000 μl) was drawn from the intravenous Silastic catheter and cultured onto TSA plates and mannitol-salt agar plates directly or serially diluted for quantitative blood cultures. Plates were incubated at 37°C for 24 h, and the number of CFU of MRSA was counted and recorded as log CFU per milliliter.

(b) Quantitative cultures of vascular catheters. Representative sections of vascular catheters (proximal, middle, and distal) were quantitatively cultured for assessment of the eradication of infection in postmortem studies. Catheters were removed carefully with aseptic techniques.

For luminal lavage of proximal, middle, and distal vascular catheter segments, the lumen of each catheter was washed with a known volume of sterile 0.9% normal saline (100 μl to 1,000 μl), and the washout was aseptically collected in a sterile flask.

Samples were cultured on TSA plates and mannitol-salt agar plates. Plates were incubated at 37°C for 24 h, and the number of CFU of MRSA was counted and recorded as log CFU per milliliter. Data were graphed as the mean log CFU per gram ± standard error of the mean (SEM).

(c) Quantitative cultures of tissues. Representative sections of tissues (anterior vena cava or fibrovascular sheath [vascular tissues surrounding the catheter], cerebrum, heart, lung, liver, spleen, and kidney) were sampled and quantitatively cultured. Each tissue sample was weighed, placed in a sterile reinforced polyethylene bag (Tekmar Corp., Cincinnati, OH), and then homogenized with sterile 0.9% normal saline for 30 s (Stomacher 80; Tekmar Corp., Cincinnati, OH).

Aliquots (100 μl) from homogenates and homogenate dilutions were plated onto TSA and mannitol-salt agar plates and incubated at 37°C for the first 24 h. Carryover of the drug was controlled by serial dilution and by the streaking of a small aliquot (100 μl) onto a large volume of agar (one full agar plate per 100-μl aliquot). The number of CFU of MRSA was counted and recorded for each tissue, and the CFU per gram was calculated. A finding of one colony of MRSA was considered positive. Data were graphed as the mean log CFU per gram ± SEM.

Statistical analysis.

All bacterial counts are presented as log CFU per gram (means ± SEM). The endpoint analysis for eradication of MRSA from tissue includes analysis of proportions by Fisher’s exact test of the antimicrobial response (≥90% reduction of the residual bacterial burden measured in log CFU per gram of tissue). Differences in bacterial counts in blood (CFU per milliliter) and tissue (CFU per gram) for treated and untreated animals were evaluated for statistical significance by a Mann-Whitney U test. Differences in survival were plotted by Kaplan-Meier methods and analyzed by a log rank test. For all tests, differences are considered to be statistically significant when the P values are ≤0.05.