Development of a pharmacokinetic/pharmacodynamic evaluation model for osteomyelitis and usefulness of tedizolid as an alternative to vancomycin against MRSA osteomyelitis

Abstract

Objectives

This study aimed to develop a suitable osteomyelitis model for pharmacokinetic/pharmacodynamic (PK/PD) evaluation and to investigate the target PK/PD values of vancomycin and tedizolid against methicillin-resistant Staphylococcus aureus (MRSA) osteomyelitis.

Methods

An osteomyelitis model was established by implanting an MRSA-exposed sterilized suture in the tibia of normal mice and mice with cyclophosphamide-induced neutropenia. The suitability of the osteomyelitis mouse model for PK/PD evaluation was assessed using vancomycin as an indicator. The target PK/PD values for tedizolid were determined using this model.

Key findings

In neutropenic mice, to achieve a static effect and 1 log₁₀ kill against MRSA, the ratios of the area under the free drug concentration–time curve for 24 h to the minimum inhibitory concentration (fAUC₂₄/MIC) of vancomycin were 91.29 and 430.03, respectively, confirming the validity of the osteomyelitis model for PK/PD evaluation. In immunocompetent mice, the target fAUC₂₄/MIC values of tedizolid for achieving a static effect and 1 log₁₀ kill against MRSA were 2.40 and 49.20, respectively. Additionally, only a 0.28 log₁₀ kill was achieved in neutropenic mice with 20 times the human equivalent dose of tedizolid.

Conclusions

In patients with restored immunity, tedizolid can potentially be used as an alternative to intravenous vancomycin therapy.

tedizolid, vancomycin, methicillin-resistant Staphylococcus aureus, osteomyelitis, pharmacokinetics/pharmacodynamics

Introduction

Osteomyelitis is an inflammatory process caused by the invasion of bacterial pathogens into the bone and is typically categorized into three types: haematogenous osteomyelitis caused by haematogenous spread, osteomyelitis resulting from adjacent tissue infection, and osteomyelitis caused by vascular insufficiency. Staphylococcus aureus is the most common microorganism in all types of osteomyelitis [1]. The most significant epidemiological change in osteomyelitis is the ongoing increase in methicillin-resistant S. aureus (MRSA) [2]. The recent increase in the number of patients with sarcopenia, frailty, and locomotive syndrome due to an ageing society is a risk factor for bone fractures in older individuals, and the prevention of MRSA infection is an important issue during surgery for fractures and artificial joint replacements.

Vancomycin is the antimicrobial of choice for the treatment of MRSA infections. Based on a clinical study in patients with lower respiratory tract infections, a ratio of area under the curve over 24 h to minimum inhibitory concentration (AUC₂₄/MIC) ≥ 400 is the current acceptable critical pharmacokinetic/pharmacodynamic (PK/PD) target value for almost all MRSA infections [3, 4], including osteomyelitis [5]. However, this target value was determined based on clinical studies in patients with pneumonia or bacteraemia [3, 6] and differs from the results of clinical studies in patients with osteomyelitis alone [7]. Thus, the PK/PD target value of vancomycin for osteomyelitis remains unclear. We established an osteomyelitis mouse model for evaluating PK/PD to determine the optimal PK/PD index value for MRSA osteomyelitis.

Linezolid, the first available oxazolidinone agent, is used for the oral treatment of MRSA osteomyelitis [8, 9]. However, despite its efficacy and convenience, the incidence of adverse effects increases with prolonged treatment for more than 2 weeks [10]. Tedizolid is a new oxazolidinone with equivalent effectiveness and fewer adverse effects than linezolid [11, 12] and can be administered orally [13]; therefore, it could be an alternative to linezolid for the treatment of osteomyelitis. As osteomyelitis treatment is long (usually 4–6 weeks) [13, 14], if oral administration is possible, it can significantly improve the quality of life of patients who are unable to tolerate long-term intravenous therapy or frequently visit the hospital.

In this study, we calculated the optimal target value of the PK/PD index using a vancomycin-treated mouse model of osteomyelitis to provide data on tedizolid therapeutics for MRSA osteomyelitis.

Materials and methods

Ethics

The Keio University Institutional Animal Care and Use Committee approved all animal experiments in this study (approval number: #A2022-059, approval date: 7 July 2019) and complied with the standards outlined in the National Institute of Health Guidelines for the Care and Use of Laboratory Animals and Guidelines for the Care and Use of Laboratory Animals at Keio University.

Materials

MRSA (ATCC33591; ATCC, Manassas, VA, USA) was stored in a Microbank storage system (IWAKI Co., Ltd., Tokyo, Japan) at −80°C and subcultured on mannitol salt agar (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) for 24 h at 37°C before experiments. Vancomycin (Cayman Chemical, Michigan, USA) and tedizolid (ChemScene, Monmouth Junction, NJ, USA) were stored at −20°C. Tedizolid phosphate (SIVEXTRO^®; Merck & Co., Inc, Kenilworth, NJ, USA) and vancomycin hydrochloride (Sawai Pharmaceutical Co., Ltd., Osaka, Japan) were stored in the dark at room temperature (20–25°C) and −4°C, respectively, until use.

Susceptibility tests

The broth microdilution (BMD) method was used to determine MRSA susceptibility to vancomycin and tedizolid according to the Clinical & Laboratory Standards Institute guidelines. Briefly, MRSA was incubated at 37°C overnight. The next day, using cation-adjusted Mueller–Hinton broth as the solvent, a bacterial suspension equivalent to 1 × 10⁶ colony-forming units (CFU)/ml was mixed with two-fold serial dilutions of tedizolid and incubated overnight at 37°C in 96-well plates. The final inoculum concentration was approximately 5 × 10⁵ CFU/ml. Minimum inhibitory concentration (MIC) was determined using a concentration range of 0.25 to 128 µg/ml.

Animals

Five-week-old female ddY mice (weighing 22–24 g; Sankyo Labo Service Corporation, Inc., Tokyo, Japan) were used in all animal experiments. The mice were housed under a 12-h light/dark cycle at 25°C and fed ad libitum. All experiments were performed over a 1-week acclimatization period.

Test for plasma protein binding of vancomycin

Plasma protein binding of vancomycin was assessed using ultrafiltration. Briefly, different concentrations of vancomycin were added to mouse plasma to achieve final concentrations of 1, 10, and 200 μg/ml and then incubated at 37°C for 30 min to achieve equilibrium. Subsequently, samples were placed in a centrifugal filter unit (Merck Co., Ltd., Tokyo, Japan) and centrifuged at 950×g and 25°C for 10 min to separate the free drug from the bound drug. Vancomycin concentration was quantified using high-performance liquid chromatography (HPLC). The following formula was used to calculate plasma protein binding:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{\%}\mathrm{P}\mathrm{B}=([V-\mathrm{V}\mathrm{U}\mathrm{F}]/V)\times 100}\end{array}}$$

where V is the initial vancomycin concentration in the serum and VUF is the vancomycin concentration in the ultrafiltrate.

PK experiment of vancomycin

Vancomycin was administered subcutaneously at 100, 200, 400, and 800 mg/kg. Subsequently, 1 ml of blood was collected via intracardiac puncture under anaesthesia at 0.1, 0.25, 1, 2, and 4 h (n = 3 at each time point). Following centrifugation at 6000×g for 10 min, the plasma was collected and stored at −80°C until use. Plasma vancomycin concentrations were determined by HPLC analysis.

Quantitation of vancomycin and tedizolid by HPLC

Vancomycin was measured using our previously described protocol [15] with an HPLC system comprising a HITACHI Chromaster 5160 Pump, HITACHI Chromaster 5430 Diode Array Detector, and a HITACHI Chromaster 5310 column oven (Hitachi, Ltd., Tokyo, Japan). The HPLC conditions were as follows: column (100 × 4.6 mm) and guard column (both LaChrom LM Type A; Hitachi, Ltd.); column temperature, 40°C; ultraviolet detection, 235 nm; flow rate, 2 ml/min; and mobile phase, acetic acid aqueous buffer (10 mM, pH 4.7). A linear gradient of acetonitrile was used for determination: at the start, linear gradient from 5% to 30% acetonitrile was applied from 0 min to 3 min; then, at 3.1 min, the acetonitrile composition was set to 60% and was kept constant until 4.5 min; and finally, at 4.6 min, the acetonitrile composition was changed back to 5% and maintained until the end of the determination.

Tedizolid was quantified according to our previously reported protocol [16] using an HPLC system comprising an LC-10ATvp pump, a CTO-10Avp oven, and an SPD-10Avp UV detector (Shimadzu Co., Kyoto, Japan). The HPLC conditions were as follows: column, Kinetex^® 5-µm EVO C18 100 Å (150 × 4.6 mm; Phenomenex, Torrance, CA, USA); guard column, TSKgel^® ODS-80Ts (Tosoh Co., Tokyo, Japan); column temperature, 40°C; mobile phase, acetic acid aqueous buffer (1 mM, pH 3.5)/acetonitrile (81:19); flow rate, 1 ml/min; and ultraviolet detection, 300 nm. The validation data for vancomycin and tedizolid are provided in Supplementary Table S1.

PK analysis of vancomycin

The PK parameters of vancomycin were determined using Phoenix^TM WinNonlin^TM software v.6.4 (Certara, Princeton, NJ, USA) based on vancomycin plasma concentrations in mice. The concentration–time data were fitted to a one-compartment model with no lag time or first-order elimination to calculate the PK parameters.

PD experiment based on an implanted mouse osteomyelitis model

We established an implanted mouse osteomyelitis model using both neutropenic and immunocompetent mice to evaluate PK/PD based on a previous study [17]. This implanted mouse osteomyelitis model induces chronic infection for more than one month, as observed in humans. Briefly, 5-mm pieces of sterile sutures were cut and cultivated in MRSA suspensions (5 × 10⁶ CFU/ml for the neutropenic model and 1.5 × 10⁸ CFU/ml for the immunocompetent model) on a bioshaker (90 rpm, 45 min at 37°C). The implant was then inserted into the left tibia of the mice after surgical exposure by drilling a hole with a 21G needle (Terumo Corporation, Tokyo, Japan) under anaesthesia with 0.3 mg/kg medetomidine, 4.0 mg/kg midazolam, and 5.0 mg/kg butorphanol. The wound was closed with Aron Alpha and sterilized using an iodine solution. Sustained neutropenia in a neutropenic mouse model was achieved before implant surgery according to a previously described method [18]. In brief, neutropenia was induced via an intraperitoneal injection of cyclophosphamide (Shionogi & Co., Ltd., Osaka, Japan) at 150 mg/kg (4 days before inoculation) and 100 mg/kg (the day before inoculation).

Following surgery, the establishment of infection in the neutropenic and immunocompetent models was confirmed on Days 1–2 and 6–7 (n = 3), respectively. In the neutropenic model, untreated controls (n = 3) were euthanized 24 h after infection. The infected tibia was homogenized in sterile saline and serially diluted 10-fold. Serial dilutions were inoculated into mannitol salt agar at 37°C for 24 h to determine the bacterial burden of each tibia before treatment. Vancomycin was subcutaneously administered in 15 dosing patterns (Supplementary Table S2, n = 3) at 100–800 mg/kg over 24 h after infection. Additionally, a single dose of tedizolid was administered intraperitoneally at human equivalent clinical (HEC) dosage, 10-fold HEC dosage, and 20-fold HEC dosage (n = 3). The value of the area under the free drug concentration–time curve (fAUC) of tedizolid in mice was calculated by the AUC of healthy male subjects after a single 200 mg/day tedizolid administration (approximately 30 µg·h/ml) [19, 20] and the humans’ protein binding rate of tedizolid (86.6% [21]), to simulate HEC dosage. In the immunocompetent model, controls were euthanized on Day 6 after surgery and 1–30 mg/kg tedizolid was intraperitoneally administered in 13 dosing patterns (Supplementary Table S3, n =3). At 24 h after the initial treatment, all mice were euthanized, including the untreated controls (n = 3), and the infected tibias were aseptically collected. Tibial sample homogenates were prepared by mixing tibias with sterile saline. The bacterial counts were determined by 10-fold step dilution of homogenates spread on cation-adjusted Mueller–Hinton agar plates. All the PD experiments were conducted in a P2-level laboratory to avoid contamination.

PK/PD analysis

The PK/PD indices (ratio of fAUC for 24 h to the MIC [fAUC₂₄/MIC], ratio of the maximum free drug concentration to the MIC [fC_max/MIC], and length of time at which the free drug concentration surpassed the MIC [fT>MIC]) of tedizolid were calculated based on the previously reported tedizolid-free plasma concentrations [16]. The PK/PD indices of vancomycin were estimated based on vancomycin-free plasma concentrations calculated using the protein binding rate identified in this study. The standard sigmoid I_max model in Phoenix^TM WinNonlin^TM software was used for all analyses.

$${\displaystyle \begin{array}{l}{\displaystyle E=E_0-(I_{max}\times C^{\gamma })/C^{\gamma }+IC{}_{50}^{\gamma })}\end{array}}$$

In the equation above, E is the vancomycin/tedizolid antimicrobial activity (change in log₁₀ CFU/tibia), E₀ is the baseline effect in the absence of vancomycin/tedizolid, I_max is the maximum kill, C is the PK/PD index, IC50 is the PK/PD index leading to 50% of the maximum kill, and γ is the Hill coefficient describing the steepness of the sigmoidal curve. The fit of the predicted curve to the data was based on visual assessment and the calculated coefficient of determination (R²).

Tissue-plasma partition coefficients of tedizolid

The tissue-plasma partition coefficient (K_p) of tedizolid was determined after a single intraperitoneal administration of tedizolid (20 mg/kg). After 4 h, the mice were euthanized, and the infected tibia, uninfected tibia, uninfected thigh, and plasma were collected (n = 3). The thigh and tibia were weighed, and 1–7 times 19% acetonitrile (v/v) was added for homogenization. The homogenate was vortexed for 30 s and incubated at 4°C for 24 h. Acetonitrile was added to the supernatant and blood for deproteinization, and the tedizolid concentration was determined using HPLC.

Statistical analysis

Differences between the tedizolid tissue penetration groups were compared using the Kruskal–Wallis test followed by the Dunn’s test in SPSS v. 28 software (IBM Corp., Armonk, NY, USA). Differences were considered statistically significant at P < .05.

Results and discussion

Generally, PK/PD parameters determined using the thigh infection model are utilized to predict the clinical target values of antimicrobial agents [22]. However, a suitable osteomyelitis model for evaluating PK/PD has not yet been established. Therefore, the effective PK/PD parameters for osteomyelitis have not been clarified for several antimicrobial agents. In this study, we established a PK/PD evaluation model for osteomyelitis and determined the optimal PK/PD target values for vancomycin and tedizolid against MRSA osteomyelitis.

The MICs of vancomycin and tedizolid were determined as 1.0 and 0.5 μg/ml, respectively. The validity of the osteomyelitis model for PK/PD evaluation was assessed using the PK/PD index of vancomycin, which is the standard treatment for MRSA osteomyelitis. First, we determined the plasma concentration–time profiles of vancomycin following subcutaneous administration in mice (100, 200, 400, and 800 mg/kg) (Fig. 1). The mean ± standard deviation (SD) of the main PK parameters of vancomycin were obtained using one-compartment analysis based on the concentration–time curves. The volume of distribution, absorption rate constant, clearance, and elimination rate constant of vancomycin were 4.29 ± 1.77 L/kg, 7.63 ± 3.68 h⁻¹, 2.14 ± 0.22 L/h/kg, and 0.64 ± 0.54 h⁻¹, respectively. The plasma protein binding of vancomycin was 23.95 ± 0.95%. Based on the plasma protein binding rate, single-dose administration of vancomycin at doses of 100, 200, 400, and 800 mg/kg resulted in fAUC of 41.1, 100.5, 203.5, and 370.4 mg·h/L, respectively, which were significantly correlated (R² = 0.998, Supplementary Fig. S1).

Figure 1.

Plasma concentration of vancomycin after a single subcutaneous dose. Data are presented as means ± SD (n = 3).

In the untreated MRSA (ATCC 33591) neutropenic osteomyelitis model, the mean ± SD of bacterial burden was 7.37 ± 0.10 log₁₀ CFU/tibia on Day 1 after surgery. The bacterial burden in the control (2 days after surgery) mice increased by 0.77 ± 0.19 log₁₀ CFU/tibia by 24 h after no treatment. In contrast, within 24 h of vancomycin administration, the bacterial burden decreased to 1.37 ± 0.12 log₁₀ CFU/tibia. The PK/PD indices (fAUC₂₄/MIC, fC_max/MIC, and fT>MIC) of vancomycin were extrapolated from the calculated PK, PD, and MIC values. The sigmoid I_max model parameters used in the PK/PD study of vancomycin are listed in Table 1. The relationships between vancomycin PK/PD indices and antibacterial activity against MRSA osteomyelitis are presented in Fig. 2. The most closely correlated PK/PD index for vancomycin was fAUC₂₄/MIC (R² = 0.89). We calculated the fAUC₂₄/MIC values required for a static effect and 1 log₁₀ kill reduction in the bacterial load for vancomycin against MRSA (Table 2). To achieve a static effect and 1 log₁₀ kill against MRSA in neutropenic mice, the fAUC₂₄/MIC values for vancomycin were 91.29 and 430.03, respectively. We estimated the AUC/MIC of vancomycin corresponding to the target values in a certain range because of the high variability in the mean rates of human protein binding of vancomycin [32.8 ± 7.5% (range 7.9–52.8%) [23], 36.4 ± 25.8% (range 0–88%) [24], and 27.1 ± 12.2 (range 0–50%) [25]]. Considering that the vancomycin protein binding rate ranged from 10% to 40%, the bactericidal effect could be achieved at an AUC₂₄/MIC_BMD of 477.8–716.7, when the MIC was measured using the BMD method. A significant difference in vancomycin AUC/MIC was observed between BMD and Etest MIC (AUC₂₄/MIC_BMD:AUC₂₄/MIC_Etest = 400:226) [26]. The target PK/PD value recommended in the therapeutic drug monitoring (TDM) guidelines for vancomycin (AUC/MIC ≥ 400) is also based on BMD [11, 27]. Gawronski et al. [7] reported that the target PK/PD value for MRSA osteomyelitis was AUC₂₄/MIC_Etest > 293, which was converted to an AUC₂₄/MIC_BMD value of 533. This value is similar to the PK/PD target value obtained in this study using the osteomyelitis mouse model (AUC₂₄/MIC_BMD ≥ 477.8–716.7). Therefore, the target values obtained using the mouse model of osteomyelitis are suitable for PK/PD evaluation and can be extrapolated to humans.

Table 1.

Sigmoid I_max model parameters calculated by PK/PD analysis for fAUC₂₄/MIC, fC_max/MIC, and fT>MIC of vancomycin and tedizolid.

Drugs	PK/PD parameters	E0	Imax	γ	IC50
Vancomycin	fAUC24/MIC	1.52	7.51	0.45	2004.07
	fCmax/MIC	1.44	3.34	0.73	28.63
	fT>MIC	0.56	4.13	0.96	161.48
Tedizolid	fAUC24/MIC	2.92	4.75	0.36	0.66
	fCmax/MIC	1.70	2.84	0.78	0.17
	fT>MIC	0.01	0.91	5.92	7.58

E₀, baseline effect in the vancomycin/tedizolid absence; fAUC₂₄/MIC, the ratio of the area under the free drug concentration–time curve for 24 h to the minimum inhibitory concentration; fC_max/MIC, the ratio of the maximum free drug concentration to the MIC; fT > MIC, the length of time at which the free drug concentration surpassed the MIC; I_max, maximum kill; IC50, PK/PD index leading to 50% of the maximum kill; γ, Hill coefficient.

Table 2.

Target values of vancomycin and tedizolid for a static effect and a 1 log₁₀ Kill reduction against MRSA in the osteomyelitis model.

	Vancomycin (neutropenic mice)	Tedizolid (immunocompetent mice)
Static effect	91.29	2.40
1 log10 Kill	430.03	49.20

1 log₁₀ Kill, 1 log₁₀ CFU/tibia reduction in viable count from initial value.

Figure 2.

PK/PD analysis of vancomycin in the neutropenic MRSA osteomyelitis mouse model. Relationship between log₁₀ CFUs/tibia at 24 h and PK/PD indices in a neutropenic MRSA osteomyelitis model. (a) Ratio of the area under the free drug concentration–time curve for 24 h to the minimum inhibitory concentration (fAUC₂₄/MIC), (b) ratio of the maximum free drug concentration to the MIC (fCmax/MIC), and (c) length of time at which the free drug concentration surpassed the MIC (fT>MIC). Symbols represent the means ± SD of viable counts in the tibiae (n = 3). The dashed line represents the initial bacterial load at the start of vancomycin administration and R² is the coefficient of determination.

Furthermore, the PK/PD target value for maximizing the efficacy of vancomycin in patients with MRSA osteomyelitis varied depending on the protein binding rate, which was estimated to be AUC₂₄/MIC_BMD ≥ 477.8–716.7. Except for prophylactic medication, TDM of vancomycin is recommended to avoid the major adverse effect of renal impairment [28]. Exposure to vancomycin with an AUC above 600 μg·h/ml significantly increases the risk of nephrotoxicity [29]. Therefore, when vancomycin is used for MRSA osteomyelitis, the upper limit of AUC should be 600 μg·h/ml and if it is not effective, another drug should be considered.

It is difficult to achieve sufficient bactericidal effect of tedizolid against MRSA in a neutropenic mouse model [16, 30]. We evaluated the PK/PD of tedizolid in neutropenic and immunocompetent mice. In neutropenic mice, even with 20 times the HEC dose of tedizolid, only a reduction of 0.28 ± 0.19 log₁₀ CFU/tibia was achieved (Fig. 3). A bacteriostatic effect was achieved at a dose 10 times of HEC (0.00 ± 0.38 log₁₀ CFU/tibia). Thus, it is difficult to achieve an antibacterial effect in the absence of neutrophils in any type of MRSA infection, within the clinical dosage range of tedizolid.

Figure 3.

Pharmacodynamics of tedizolid in the neutropenic osteomyelitis mouse model. The dashed line represents the initial bacterial load at the start of tedizolid treatment. Data are presented as means ± SD (n = 3).

Next, we evaluated tedizolid in an immunocompetent mouse model of MRSA osteomyelitis, in which the mean ± SD of the initial bacterial burden was 6.20 ± 0.72 log₁₀ CFU/tibia on Day 6 after infection, with an increase of 0.84 ± 0.90 log₁₀ CFU/tibia on Day 7 (24 h after no treatment). The sigmoid I_max model parameters used in the PK/PD study of tedizolid are listed in Table 1. PK parameters of tedizolid were obtained from our previous report [16]. The relationship between the PK/PD indices of tedizolid and its antibacterial activity against MRSA osteomyelitis is shown in Fig. 4. The most correlated PK/PD index for tedizolid was fAUC₂₄/MIC (R² = 0.95). In immunocompetent mice, the target fAUC₂₄/MIC values of tedizolid for achieving a static effect and 1 log10 kill against MRSA were 2.40 and 49.20, respectively (Table 2). The corresponding AUC₂₄ based on MIC (MIC₉₀, 0.25 μg/ml [21]) and human protein binding of tedizolid (86.6% [31]) were 4.5 and 91.8 µg·h/ml, respectively. Given that the AUC provided by the clinical dose of tedizolid will be approximately 30 µg·h/ml [19, 20], tedizolid can effectively treat MRSA osteomyelitis, although it cannot achieve 1 log ₁₀ kill. According to the PK/PD analysis, the human AUC of 30 µg·h/ml produces a bactericidal effect of approximately 0.7 log₁₀ kill. Accordingly, we hypothesized that tedizolid could be a potential therapeutic agent for immunocompetent patients with MRSA osteomyelitis if the treatment duration is extended. More detailed studies are required to determine the exact duration of treatment. Owing to these properties, tedizolid is rarely used in immunocompromised patients with severe neutropenia. Initially, intravenous vancomycin should be administered until the immune system recovers. Generally, antibiotic treatment for osteomyelitis requires 4–6 weeks; however, the safety of long-term vancomycin treatment should be carefully considered [32, 33]. However, the cutoff value of immune status that allows the administration of tedizolid has not yet been clarified, and there are no reports of such information in prior clinical or animal experiments. This is a limitation of this study. Therefore, it is not yet clear whether the index of immune status can be used as a reference. In addition, we considered that tedizolid can be given from the initial stage of treatment if a patient’s immune status is normal. However, although the risk of haematological toxicity (e.g. thrombocytopenia) of tedizolid is lower than that of linezolid, its administration should be carefully considered in patients with a high risk of thrombocytopenia when the drug is administered for a longer period of 4–6 weeks, as in osteomyelitis.

Figure 4.

PK/PD analysis of tedizolid in the immunocompetent MRSA osteomyelitis mouse model. Relationship between log₁₀ colony-forming units/tibia at 24 h and PK/PD indices in an immunocompetent MRSA osteomyelitis model. (a) Ratio of the area under the free drug concentration–time curve for 24 h to the minimum inhibitory concentration (fAUC₂₄/MIC) (b) ratio of the maximum free drug concentration to the MIC (fCmax/MIC), and (c) ratio of the maximum free drug concentration to the MIC (fCmax/MIC). Symbols represent means ± SD of viable counts in the tibiae (n = 3). The dashed line represents the initial bacterial load at the start of tedizolid administration and R² is the coefficient of determination.

The tedizolid dose that achieved a static effect against MRSA osteomyelitis was slightly different from that obtained in our previous study using an MRSA thigh infection model in neutropenic mice (static effect: 5.8-fold HEC) [16]. The K_P values of tedizolid in the normal tibia, infected tibia, and thigh were 0.28, 0.32, and 0.28, respectively, and no significant differences were observed among the tissues (P > .05, Fig. 5). Thus, the difference between the thigh model and the osteomyelitis model was not due to differences in tissue penetration but was more likely due to the altered antibiotic susceptibilities of S. aureus after exposure to the osteoblast intracellular environment [34].

Figure 5.

Tissue-plasma partition coefficients of tedizolid in the normal tibia, infected tibia, and thigh. Data are presented as means ± SD (n = 3).

Here, we suggest the following treatment plan for osteomyelitis: for immunocompromised patients, intravenous vancomycin is administered during the initial phase until the immune system recovers, followed by switching to oral tedizolid to avoid the adverse effects of long-term continuous vancomycin use. This allows patients to switch from inpatient to outpatient treatment, which not only improves their quality of life but also reduces medical costs. Our results provide new evidence for the optimal treatment of MRSA osteomyelitis with vancomycin and tedizolid depending on the immune status of the patient. However, these results need to be validated in future clinical studies.

Supplementary material

Supplementary data are available at Journal of Pharmacy and Pharmacology online.

Acknowledgement

We would like to thank Editage (www.editage.com) for English language editing services.

Author contributions

Conceptualization and design of study: X.L., Y.E., and K.M.; Acquisition of data: X.L. and Y.E.; Analysis of data: X.L., Y.E., and K.M.; Drafting of article and/or critical revision: X.L., Y.E., and K.T.; Final approval of manuscript: Y.E., K.T., and K.M.

Conflict of interest

K.M. received tebipenem and nacubactam, grant support, and payment for lectures from Meiji Seika Pharma Co., Ltd. (Tokyo, Japan) and grant support from Sumitomo Dainippon Pharma Co., Ltd. The other authors declare no conflicts of interest.

Funding

None declared.

Data availability

The datasets generated or analysed in the current study are available from the corresponding author upon reasonable request.

Author notes

Xiaoxi Liu and Yuki Enoki contributed equally to this work. The author order was determined by a discussion of all authors.