Efficacy of Tedizolid against Enterococci and Staphylococci, Including cfr⁺ Strains, in a Mouse Peritonitis Model

In a mouse peritonitis model, tedizolid was comparable to linezolid and daptomycin against an Enterococcus faecium strain (VAN^r, AMP^r), an Enterococcus faecalis strain, and a methicillin-resistant Staphylococcus aureus (MRSA) strain with and without cfr. Against a cfr(B)⁺ E. faecium, tedizolid was inferior in vivo to linezolid and daptomycin, despite an ∼4-fold lower MIC.

ABSTRACT

In a mouse peritonitis model, tedizolid was comparable to linezolid and daptomycin against an Enterococcus faecium strain (VAN^r, AMP^r), an Enterococcus faecalis strain, and a methicillin-resistant Staphylococcus aureus (MRSA) strain with and without cfr. Against a cfr(B)⁺ E. faecium, tedizolid was inferior in vivo to linezolid and daptomycin, despite an ∼4-fold lower MIC.

INTRODUCTION

Tedizolid phosphate (TZD; Sivextro) is a next-generation oxazolidinone antibiotic approved for the treatment of acute bacterial skin and skin structure infections (ABSSSI) caused by susceptible pathogens in a number of countries, including the United States, Canada, and in the European Union (1, 2). The clinical approval encompasses Staphylococcus aureus, including methicillin-susceptible and -resistant strains (MSSA and MRSA, respectively), various Streptococcus spp., and Enterococcus faecalis. The MIC of TZD for 90% (MIC₉₀) of MSSA and MRSA strains has been shown to be 0.25 to 0.5 μg/ml (3–5). For vancomycin (VAN)-resistant E. faecalis and Enterococcus faecium (VRE), the MIC₉₀ was 0.5 μg/ml (3, 4). However, limited information is available regarding the in vitro activity of TZD against multidrug-resistant (MDR) strains of E. faecium resistant to linezolid (LZD), VAN, and daptomycin (DAP) (6–8).

Several studies have described TZD in vivo pharmacodynamics against S. aureus (MSSA and MRSA) and Streptococcus pneumoniae (9–14), but no data are available for the treatment of E. faecalis, E. faecium, and S. aureus (MRSA) and their cfr derivatives in a mouse peritonitis model. Here, we tested the in vivo activity of TZD, DAP, and LZD against several enterococci and S. aureus strains, including related strains with and without cfr.

Previously described E. faecalis OG1RF (15, 16), E. faecium TX82 (VAN^r, AMP^r), E. faecium TX4320 cfr(B) (17), S. aureus MRSA CM-05 cfr⁺, and MRSA CM-05 cfr-cured (18) strains were used in in vitro and in vivo experiments.

TZD for in vitro use (TR-700), TZD prodrug for in vivo use (TR-701), and DAP were provided by the study sponsor, Merck & Co., Inc. (Rahway, NJ). LZD was provided by Pfizer, Inc. (Peapack, NJ). These antibiotics were reconstituted as recommended by the manufacturers. MICs were determined against test bacteria by the broth microdilution method using cation-adjusted BBL Mueller-Hinton II (MH) broth (BD, Sparks, MD), according to Clinical and Laboratory Standards Institute (CLSI) guidelines (19). S. aureus ATCC 29213 and E. faecalis ATCC 29212 were used as controls. Experiments were repeated twice, and a range of MICs was reported where appropriate.

We used an established model of murine peritonitis, which has been previously used by our group in preclinical studies of other antibiotics (18, 20–22). In brief, female, 4- to 6-week-old outbred ICR mice (Envigo) with mean weights of 25 g were used in the study. The 50% lethal dose (LD₅₀) of test bacteria and therapy experiments in mice were performed as described in our previously published studies (23, 24), where LD₅₀s were determined by the method of Reed and Muench (25). Antibiotic dose and delivery route for TZD was based on the study sponsor’s recommendation; an earlier study showed high bioavailability and that a mouse dose of 8.42 mg/kg TZD phosphate given intraperitoneally was equivalent to a single human dose of 200 mg (12). For in vivo efficacy, TZD (10 mg/kg orally [p.o.] every 24 h [q24h]), DAP (50 mg/kg subcutaneously [s.c.] q24h) (26, 27), and LZD (80 mg/kg p.o. q12h) (18, 28) were administered 1 h following an intraperitoneal inoculation with ∼10 times the LD₅₀ of test bacteria. Bacterial identity was confirmed either by phenotypic antibiotic characteristics or by using pulsed-field gel electrophoresis.

After the animals were sacrificed, aliquots of spleen homogenates (18) were plated directly on MH agar plates supplemented with TZD or DAP, 4 μg/ml each, to detect possible in vivo mutants with decreased susceptibility. Samples showing any growth on these media were retested for growth on the same concentration of antibiotics according to CLSI methods (19).

For statistical evaluation, the numbers of mice that survived at the end of the study in treatment versus nontreated groups were compared with Fisher’s exact tests by using online GraphPad Prism 7 software. A P value of <0.05 was considered significant. Animal studies were performed according to the guidelines of UTHSC Animal Welfare Committee under an approved protocol (AWC-16-0083).

Table 1 shows the results of in vitro broth microdilution activity of TZD, DAP, and LZD against the test bacteria. MIC ranges of TZD, DAP, and LZD were 0.5 to 2 μg/ml, 2 to 4 μg/ml, and 1 to 16 μg/ml, respectively, against all strains tested. TZD MIC breakpoints for S. aureus (including MRSA) are ≤0.5 μg/ml (susceptible), 1 μg/ml (intermediate), and ≥2 μg/ml (resistant), while for E. faecalis, the breakpoint for susceptibility is ≤0.5 μg/ml (19). For E. faecium, there are currently no approved TZD CLSI breakpoints available.

TABLE 1.

MICs and in vivo efficacy

Strain	MIC (μg/ml) for:a			% survival (no. of animals) and P value for:b
	DAP	LZD	TZD	Nontreated	TZD	DAP	LZD
E. faecium TX82 (VANr, AMPr)	2	2	0.5	0 (8)	100 (6), 0.002	100 (7), 0.002	100 (7), 0.002
E. faecium TX4320 cfr(B)	4	8–16	2	0 (14)	14 (13)c, 0.352	86 (7), 0.0004	86 (14), 0.0004
E. faecalis OG1RF	2	1–2	0.5–1	0 (5)	100 (6), 0.002	100 (6), 0.002	100 (7), 0.002
S. aureus MRSA CM-05 cfr+	2	16	0.5	0 (9)	100 (7), 0.0001	100 (6), 0.0002	100 (7), 0.0001
S. aureus MRSA CM-05-cfr cured	2	2–4	0.5–1	0 (5)	100 (6), 0.002	100 (6), 0.002	NDd

^aCLSI breakpoints (19): DAP, ≤4 μg/ml (susceptible) for Enterococcus spp. and ≤1 μg/ml (susceptible) for S. aureus; LZD, ≤2 μg/ml (susceptible), 4 μg/ml (intermediate), and ≥8 μg/ml (resistant) for Enterococcus spp. and ≤4 μg/ml (susceptible) and ≥8 μg/ml (resistant) for S. aureus; TZD, ≤0.5 μg/ml (susceptible), 1 μg/ml (intermediate), and ≥2 μg/ml (resistant) for S. aureus and ≤0.5 μg/ml (susceptible) for E. faecalis; there are currently no CLSI breakpoints for E. faecium.

^bNumber of survivors at the end of study comparing nontreated versus antibiotic-treated animals using a mouse peritonitis model; P values determined by Fisher’s exact tests.

^cTested twice, with almost identical results each time.

^dND, not done because of high activity against the cfr⁺ strain.

The LD₅₀s of E. faecium TX82, E. faecium TX4320 cfr(B), E. faecalis OG1RF, and MRSA CM-05-cfr⁺ strains were 5.3 × 10⁸, 6.9 × 10⁸, 1.4 × 10⁸, and 5.2 × 10⁶ CFU/ml, respectively.

TZD, DAP, and LZD treatments against E. faecium TX82- and E. faecalis OG1RF-infected mice showed highly significant protection versus nontreated animals (Table 1). Both DAP and LZD also showed significant protection against TX4320 cfr(B) infection in mice (Table 1). However, TZD showed less in vivo efficacy compared to that of LZD against this strain, despite showing an ∼4-fold lower in vitro MIC than LZD (Table 1). Both MIC and in vivo efficacy (∼6 mice in each group) against this strain were retested to validate the results and showed the same result. Whether the greater efficacy of LZD, despite the strain being LZD resistant in vitro, reflects differing blood levels achieved by the different doses of TZD and LZD or a difference in the response in vivo to Cfr(B) is not known. A previous report found that TX4320 cfr(B) also possesses one mutated allele (G2576T) out of six 23S rRNA alleles and has wild-type L3, L4, and L22 sequences (17). We also showed that E. faecium TX4320 cfr(B) does not have optrA via PCR (data not shown), which has been implicated earlier as a cause of increased TZD MICs in S. aureus and E. faecalis hosts when present with cfr (14). It is worth noting that the TZD MIC against this strain was 2 μg/ml versus 0.5 μg/ml in the case of E. faecium TX82. While a breakpoint for E. faecium has not been established, this MIC is above the CLSI susceptibility breakpoint for E. faecalis as well as for S. aureus (19), and our study indicates that caution should be used if considering the use of TZD for E. faecium when the TZD MIC is >1 μg/ml.

TZD and DAP treatments also showed significant protection in mice inoculated with S. aureus MRSA CM-05-cfr⁺ and MRSA CM-05-cfr cured of its cfr gene versus nontreated animals. LZD was not tested against the cfr-cured derivative strain, since it showed protection similar to that seen with TZD and DAP against MRSA CM-05-cfr⁺. LZD protection in MRSA CM-05-cfr⁺-inoculated mice corroborated our previously reported results against this strain in a peritonitis model (18), repeated here to ensure that there was no change in bacterial infectivity or therapy results over time. Additionally, none of the test bacteria produced in vivo-generated TZD- and DAP-resistant colonies in therapy groups during the course of this study.

In conclusion, our study revealed comparable in vivo treatment outcomes against several strains in a murine peritonitis model using single doses of TZD, DAP, and LZD that correlate with human dosing. However, with one strain, E. faecium TX4320 cfr(B) (TZD MIC, 2 μg/ml), TZD showed less efficacy in mice than LZD despite lower MICs; the reason for this difference in efficacy remains to be determined.