In Vivo Pharmacodynamic Target Assessment of Antofloxacin against Streptococcus pneumoniae and Staphylococcus aureus in a Neutropenic Murine Pneumonia Model

We determined in vivo efficacy and target PK/PD exposures of antofloxacin against Streptococcus pneumoniae and Staphylococcus aureus in the murine pneumonia model. The mean plasma free drug area under the concentration-time curve/MIC (fAUC/MIC) targets associated with stasis and 1-log₁₀ and 2-log₁₀ kill effects were 8.93, 19.2, and 48.1, respectively, for S. pneumoniae, whereas they were 30.

ABSTRACT

We determined in vivo efficacy and target PK/PD exposures of antofloxacin against Streptococcus pneumoniae and Staphylococcus aureus in the murine pneumonia model. The mean plasma free drug area under the concentration-time curve/MIC (fAUC/MIC) targets associated with stasis and 1-log₁₀ and 2-log₁₀ kill effects were 8.93, 19.2, and 48.1, respectively, for S. pneumoniae, whereas they were 30.5, 55.4, and 115.8, respectively, for S. aureus. The fAUC/MIC targets in murine lung epithelial lining fluids (ELF) for the same endpoints were nearly 2-fold higher than those in plasma.

TEXT

Community-acquired bacterial pneumonia (CABP) remains a significant health concern worldwide with substantial morbidity and mortality (1). The major CABP pathogens continue to be Streptococcus pneumoniae, regardless of age and geographical location, followed by Staphylococcus aureus and Gram-negative bacteria (2, 3). Albeit less frequently, S. aureus has been noted as a problematic pathogen due to increasing drug resistance (i.e., methicillin-resistant S. aureus [MRSA]) (4). Therefore, there is an urgent need to evaluate alternative antibiotics against CABP pathogens. Antofloxacin, a broad-spectrum fluoroquinolone antibiotic, has been approved in China for the treatment of acute bacterial exacerbation of chronic bronchitis (ABECB), wound infections, and multiple epifolliculitis (5). Previous clinical studies in ABECB patients have demonstrated comparable clinical and bacteriological outcomes compared with levofloxacin (6). Given its spectrum and potency in ABECB patients, a logical consideration is to what extent antofloxacin could be valuable for CABP, especially with respect to S. pneumoniae and MRSA. Hence, we determined the pharmacokinetic/pharmacodynamic (PK/PD) profiles and target exposures of antofloxacin associated with efficacy against S. pneumoniae and S. aureus in the neutropenic murine pneumonia model.

Four S. aureus strains, including the community-acquired MRSA strain MW2, and four S. pneumoniae strains were included (Table 1). MICs were determined using the broth microdilution method in accordance with CLSI guidelines (7). The MICs of antofloxacin varied from 0.03 to 0.5 mg/liter. The elevated MICs to antofloxacin were mirrored by a similar increase in MICs to levofloxacin (Table 1).

TABLE 1.

In vitro antimicrobial susceptibility results of S. aureus and S. pneumoniae strains selected for use in the murine lung infection studies^a

Organism	Source	Resistance	MIC (mg/liter) of:
			ATX	LEV	CTX	AZM
S. aureus
29213	ATCC	MSSA	0.125	0.25	2	1
43300	ATCC	MRSA	0.125	0.25	64	512
MW2	ATCC, septicemia	MRSA	0.063	0.25	128	1
SA 368	NPIN, sputum	MRSA	0.25	2	32	16
S. pneumoniae
49619	ATCC	PSSP	0.25	1	0.125	0.125
SP 18-1	NPIN, BAL	PSSP	0.063	0.125	0.063	0.25
SP 18-2	NPIN, BAL	PRSP	0.5	1	4	2
SP 18-3	NPIN, BAL	PSSP	0.031	0.25	0.03	0.063

^aNPIN, National Pathogen Identification Network of China (courtesy of Guangdong Provincial Center for Disease Control and Prevention of China); BAL, bronchoalveolar lavage fluids; MSSA, methicillin-susceptible S. aureus; MRSA, methicillin-resistant S. aureus; PSSP, penicillin-susceptible S. pneumoniae; PRSP, penicillin-resistant S. pneumoniae; ATX, antofloxacin; LEV, levofloxacin; CTX, cefotaxime; and AZM, azithromycin.

Antofloxacin concentration measurement methods and PK data in murine plasma and lung epithelial lining fluids (ELF) have been described in detail elsewhere (5). The half-lives of antofloxacin were 1.32 h in plasma and 1.05 h in ELF, respectively (5). The protein binding rates of antofloxacin in mice and humans are 20.3% and 17.5%, respectively, and the mean penetration ratio of free drug into ELF was 172% in mice (5, 8). Although pulmonary disposition studies of antofloxacin have not been carried out in humans, the good ELF penetration in mice is considered an acceptable model of penetration into human lung tissue (3, 5). In addition, compared to other respiratory fluoroquinolones such as moxifloxacin and gemifloxacin, antofloxacin was well tolerated and did not cause QTc (heart-rate-corrected QT interval) prolongation (6).

A well-characterized neutropenic murine lung infection model was utilized (3–5, 9). All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of the South China Agricultural University (no. 2019046). Bacterial suspensions were diluted in saline to final inocula of 10^8.0 to 10^8.3 and 10^7.1 to 10^7.5 CFU/ml for S. aureus and S. pneumoniae, respectively. Lung infections were produced by intratracheal injection of 0.05 ml of inoculum into the lungs of isoflurane-anesthetized neutropenic mice. Treatment was initiated 2 h after infection. Antofloxacin was administered by the subcutaneous route in doses of a 2-fold incremental scheme from 0.16 to 80 mg/kg of body weight twice daily. After 24 h, the bacterial burdens were enumerated from lung homogenates. The area under the concentration-time curve/MIC (AUC/MIC) was chosen as the predictive PK/PD index to describe antofloxacin efficacy in treating lung infections (5). The correlation between the AUC/MIC and efficacy was determined by the sigmoid maximum effect (E_max) model (8). The 24-h free drug AUC/MIC (fAUC_{24 h}/MIC) targets in plasma and ELF associated with the net static, 1-log₁₀, and 2-log₁₀ kill effects were calculated for each strain.

Bactericidal effects were noted at ≥10 mg/kg/12 h for S. aureus and ≥2.5 mg/kg/12 h for S. pneumoniae, respectively (Fig. 1a and b). The highest doses reduced the bacterial burden by >3.0 log₁₀ CFU/g compared with the initial burden. The relationship between the AUC/MIC and effectiveness was strong, with an R² of >0.95 (Fig. 1c and d). For S. aureus, the mean plasma fAUC/MIC targets for stasis and 1-log₁₀ and 2-log₁₀ kill effects were 30.5, 55.4, and 115.8, respectively. The corresponding ELF fAUC/MIC targets were 52.2, 95.0, and 196.4 (Table 2). Of note, the PD targets of antofloxacin in ELF were substantially higher than that in plasma (P < 0.01, paired Student's t test). This is consistent with other fluoroquinolone members that generally have good penetration into the ELF (3). However, the PD targets for S. pneumoniae were much lower than those for S. aureus (P < 0.005, unpaired Student's t test). Plasma fAUC/MIC targets for the same endpoints were 8.93, 19.2, and 48.1. The corresponding ELF fAUC/MIC targets were 14.7, 31.7, and 81.8 (Table 2).

FIG 1.

PK/PD profiles of antofloxacin against the study pathogens. (a and b) In vivo dose-response relationship of antofloxacin against S. aureus (a) and S. pneumoniae (b) strains in the murine lung infection model. Each symbol represents the mean and standard deviation from four mice (n = 4). Two-fold increasing dose levels were divided into a regimen of administration every 12 h. Bacterial burden was measured at the beginning and after 24 h of therapy. The horizontal dashed line represents the bacterial burden in murine lungs at the beginning of therapy. Data points below the line represent killing, and points above the line represent growth. (c and d) Relationship between in vivo treatment efficacies of antofloxacin and the PK/PD index of AUC/MIC ratios in plasma and ELF against S. aureus (c) and S. pneumoniae (d) strains in the murine lung infection model. Antofloxacin exposure is expressed as the free drug 24-h AUC/MIC. The 50% effective concentration (EC₅₀) represents the AUC/MIC ratios to achieve 50% of the maximal effect (E_max), and N is the Hill coefficient. The line drawn through the data points is the best-fitting line based on the sigmoid E_max model. R² represents the coefficient of determination.

TABLE 2.

PK/PD target values of antofloxacin in plasma (fAUC_plasma/MIC) and ELF (fAUC_ELF/MIC) necessary to achieve stasis and 1- and 2-log₁₀ kill effects for the study organisms in the neutropenic murine lung infection model^a

Organism	Antofloxacin MIC (mg/liter)	Burden at start of therapy (log10 CFU/g)	24-h control growth (log10 CFU/g)	Target value of antofloxacin AUC/MIC ratio in plasma and ELF
				fAUCplasma/MICb			fAUCELF/MIC
				Stasis	1-log10 kill	2-log10 kill	Stasis	1-log10 kill	2-log10 kill
S. aureus
29213	0.125	6.30	2.01	42.0	69.7	132.1	72.1	120.3	223.2
43300	0.125	6.25	1.97	32.4	60.1	124.8	56.2	103.1	210.9
MW2	0.063	6.43	1.59	22.8	46.8	112.9	38.0	80.4	198.5
368	0.25	6.35	1.86	24.6	45.0	93.4	42.5	76.2	153.2
Mean	NA	6.33	1.86	30.5	55.4	115.8	52.2	95.0	196.4
SD	NA	0.08	0.19	8.76	11.6	16.9	15.3	20.6	30.6
S. pneumoniae
49619	0.25	5.82	2.12	6.31	15.6	53.2	10.5	26.1	89.1
18-1	0.063	5.83	2.24	9.61	21.7	54.1	16.0	36.0	90.3
18-2	0.5	5.69	1.91	7.29	17.1	45.3	12.3	28.8	75.6
18-3	0.031	5.99	2.04	12.5	22.5	40.0	20.0	36.0	72.1
Mean	NA	5.83	2.08	8.93	19.2	48.1	14.7	31.7	81.8
SD	NA	0.13	0.14	2.76	3.38	6.68	4.22	5.11	9.28

^aMean values and standard deviations (SD) are shown in boldface. NA, not applicable.

^bfAUC/MIC, the free drug AUC/MIC. P < 0.01 for fAUC/MIC targets in plasma versus ELF (paired Student's t test). P < 0.005 for fAUC/MIC targets between S. aureus and S. pneumoniae (unpaired Student's t test).

Clinical studies have demonstrated that dose optimization based on the murine-derived PK/PD target assessments can help to predict the likelihood of clinical success. For example, high clinical and microbiological responses in CABP patients treated with levofloxacin were associated with an fAUC/MIC of >33.7 in plasma (10). This clinical PK/PD endpoint correlates well with the stasis target exposures (AUC/MIC of 28.2 to 31.6) in the murine pneumonia model due to S. pneumoniae (11). Thus, we herein explored antofloxacin PK/PD targets obtained in this study in the context of PK data in humans and surveillance MIC data. Human plasma PKs have been evaluated for single and multiple oral administrations of antofloxacin (12). Plasma protein binding is 17.5% in humans and results in a free steady-state plasma AUC_{0–24 h} of 43.5 mg · h/liter based on the currently approved human dose regimens, which consist of an oral 400-mg loading dose followed by 200 mg daily (5, 12). Previous surveillance studies for antofloxacin susceptibility demonstrated MIC₉₀s of 0.5 and 2.0 mg/liter for S. aureus and S. pneumoniae, respectively (6, 13). Therefore, the clinical dose of antofloxacin would produce the corresponding plasma fAUC_{24 h}/MIC exposures of 87 and 22 at the MIC₉₀ values, respectively. These parameters exceed the 1-log₁₀ kill targets in plasma for both S. aureus and S. pneumoniae found in this study. Given the previous levofloxacin PK/PD target correlation between animal model and clinical outcomes in CABP patients (10, 11), we reasoned that antofloxacin, a levofloxacin derivative, may be an attractive candidate for treatment of CABP infections involving S. aureus and S. pneumoniae.

Data availability.

The raw data supporting the conclusion of this article will be made available by the authors without undue reservation to any qualified researcher.