ABSTRACT
The rising rates of antibiotic resistance accentuate the critical need for new antibiotics. Daptomycin is a new antibiotic with a unique mode of action and a rapid in vitro bactericidal effect against gram-positive organisms. This study examined the kinetics of daptomycin's bactericidal action against peritonitis caused by methicillin-susceptible Staphylococcus aureus (MSSA) and methicillin-resistant S. aureus (MRSA) in healthy and neutropenic mice and compared this activity with those of other commonly used antibiotics. CD-1 mice were inoculated intraperitoneally with lethal doses of MSSA (Xen-29) or MRSA (Xen-1), laboratory strains transformed with a plasmid containing the lux operon, which confers bioluminescence. One hour later, the animals were given a single dose of daptomycin at 50 mg/kg of body weight subcutaneously (s.c.), nafcillin at 100 mg/kg s.c., vancomycin at 100 mg/kg s.c., linezolid at 100 mg/kg via gavage (orally), or saline (10 ml/kg s.c.). The mice were anesthetized hourly, and photon emissions from living bioluminescent bacteria were imaged and quantified. The luminescence in saline-treated control mice either increased (neutropenic mice) or remained relatively unchanged (healthy mice). In contrast, by 2 to 3 h postdosing, daptomycin effected a 90% reduction of luminescence of MSSA or MRSA in both healthy and neutropenic mice. The activity of daptomycin against both MSSA and MRSA strains was superior to those of nafcillin, vancomycin, and linezolid. Against MSSA peritonitis, daptomycin showed greater and more rapid bactericidal activity than nafcillin or linezolid. Against MRSA peritonitis, daptomycin showed greater and more rapid bactericidal activity than vancomycin or linezolid. The rapid decrease in the luminescent signal in the daptomycin-treated neutropenic mice underscores the potency of this antibiotic against S. aureus in the immune-suppressed host.
Daptomycin is a lipopeptide antibiotic with rapid bactericidal activity against a broad range of aerobic and anaerobic gram-positive bacteria, including methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) (48). As such, daptomycin is of particular interest given the increasing prevalence of gram-positive bacteria in hospital-based and community-acquired serious infections and the critical need for drugs effective against organisms resistant to other agents (5, 8, 23, 35, 43, 50).
New antibiotics are specifically needed to combat infections in patients with cancer and neutropenia, which are increasingly caused by antibiotic-resistant, gram-positive pathogens (21, 51) and which may lead to serious complications or death if they are not treated promptly and appropriately (21). It is proposed that antibiotics with rapid bactericidal action may be particularly beneficial to immunocompromised patients, although data are lacking (13). Bactericidal antibiotics are specifically recommended for the treatment of endocarditis, another serious infection frequently recalcitrant to therapy, due to a significant correlation with improved clinical outcomes (13, 34). Daptomycin was recently investigated for use in patients with Staphylococcus aureus bacteremia with known or suspected infective endocarditis (10). On the basis of that study, daptomycin was approved by the FDA for use in patients with S. aureus bacteremia with known or suspected infective endocarditis (10).
Previously, we demonstrated the rapid bactericidal activity of daptomycin against systemic S. aureus infections in healthy mice. In a model of murine septicemia, daptomycin produced dose-dependent bactericidal activity (>99% reduction) against MSSA and MRSA at 5 h postdosing and was more rapidly active and more potent than vancomycin (28). Prior studies have also demonstrated that daptomycin is bactericidal against MSSA (31) and MRSA (11) in neutropenic mouse models of thigh infection.
In contrast to the techniques used in the studies cited above, which required tissue specimen extraction and in vitro quantification of the bacterial CFU, the recently developed bioluminescence methodology allows the real-time, noninvasive tracking of infections in live laboratory animals. Bioluminescent bacteria were engineered by genetic transformation with a lux operon, a series of genes encoding enzymes responsible for the luminescent reaction in naturally light-emitting organisms (e.g., fireflies and bacteria such as Photorhabdus luminescens [previously Xenorhabdus luminescens] and Vibrio harveyi) (32). Prior studies have shown that light emitted from viable, bioluminescent bacteria can be imaged in vivo and quantified, and measurements of bioluminescent flux correlate well with viable CFU counts (14, 15, 20, 24-27, 33, 40). For example, the P. luminescens lux operon has been expressed successfully in numerous bacteria, including S. pneumoniae (15), Escherichia coli (40), Pseudomonas aeruginosa (20, 25), and S. aureus (14, 25-27, 33, 39, 49), which were monitored for light emission in mouse models of infection. It was found that the light emission had a high correlation with the viable CFU counts.
The goal of the current study was to use bioluminescent bacteria to measure in real time the degree of bacterial cell killing following a single dose of daptomycin in live mice and to compare this bacterial killing with that of other commonly used antibiotics. We used strains of MSSA and MRSA genetically modified to express the P. luminescens lux genes in a model of peritonitis in healthy and neutropenic mice. The infected animals were either saline treated (control) or treated once with daptomycin or a comparator agent. Bioluminescence was monitored over time with an imaging system. The effectiveness of each antibiotic in the treated animals could be measured by the decrease in the emitted light signal. This allowed determination of the bactericidal time course of daptomycin against MRSA and MSSA in both healthy and neutropenic mice and a direct comparison of the activity of daptomycin with those of the other antibiotics used to treat MRSA and MSSA infections. This imaging system has previously been used to measure the time course of daptomycin killing of MRSA in a rat model of endocarditis (33) and to monitor S. aureus infections in murine models of thigh muscle infection (14, 27) and foreign body infection (25-27).
(Parts of this work were published previously in the abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy [32b]) and the abstracts of the 104th General Meeting of the American Society for Microbiology [32a]).
MATERIALS AND METHODS
Animals.CD-1 female mice (Charles River Laboratories) weighing 20 to 25 g were used in this study. Water and Agway rodent chow were provided ad libitum throughout the study. All studies involving the use of research animals at Cubist Pharmaceuticals are reviewed and approved by the Cubist Animal Care and Use Committee prior to initiation.
Bacteria.Two laboratory strains of S. aureus were obtained from Xenogen Corporation (Alameda, CA): Xen-1, an MRSA strain, and Xen-29, an MSSA strain, which contain copies of the lux genes as a stable plasmid and a chromosomal insert, respectively. The bacterial strains were cultured in Mueller-Hinton broth (Becton Dickinson, Baltimore, MD) at 37°C overnight. The infecting inoculum was prepared by suspending the overnight culture in 6% hog gastric mucin (M-2378; Sigma). Measurement of the MICs of all antibiotics was performed according to Clinical and Laboratory Standards Institute (CLSI; formerly NCCLS) guidelines, which for daptomycin involves microdilutions in broth supplemented with calcium chloride to 50 mg/liter.
Mouse peritonitis model.Groups of CD-1 female mice (n = 5/group) were inoculated intraperitoneally (i.p.) with a lethal dose of S. aureus (MSSA or MRSA, 2 × 107 to 5 × 108 CFU/mouse). The number of mice surviving each day after dosing was recorded for 7 days following inoculation. For the trials with immunocompromised mice, 150 mg of cyclophosphamide per kg of body weight was injected i.p. at 4 days, and 100 mg/kg of cyclophosphamide was injected i.p. at 1 day before infection. This regimen has been shown to maintain neutropenia in mice for at least 5 days (2, 37).
Antibiotic dosing.At 1 h after bacterial inoculation, the animals were given a single dose of daptomycin at 50 mg/kg subcutaneously (s.c.), nafcillin at 100 mg/kg s.c., vancomycin at 100 mg/kg s.c., linezolid at 100 mg/kg via gavage (orally [p.o.]); or saline at 10 ml/kg s.c. The daptomycin dose was chosen to closely mimic the exposure (the area under the concentration-time curve from time zero to 24 h) obtained with doses used in an infective endocarditis clinical trial of 6 mg/kg in humans. The doses for nafcillin, vancomycin, and linezolid were also chosen to reflect single doses used clinically (3, 12, 38). Note that daptomycin is approved for use once a day (10), while vancomycin and linezolid are dosed twice per day (12, 38), and nafcillin can be dosed between four and six times per day (3). This study examined the in vivo kinetics of bacterial killing following the administration of a single dose of each agent in mice.
Luminescent imaging.At hourly intervals following inoculation and dosing, groups of mice were anesthetized for bioluminescence measurements in the Xenogen Corporation imaging system (14, 26-28). The mice were anesthetized with an injection of 60 mg/kg of pentobarbital i.p. or, for repeated imaging of the same mice, via inhalation of aerosolized isoflurane mixed with oxygen. The anesthetized mice were transferred to the imaging chamber, ventral side up, and imaged for 1 to 5 min. The imaging system measures the number of photons reaching each detector of the charge-coupled device camera, and the IVIS Living Image software (Xenogen) translates these data into a false-color image that depicts areas of intense luminescence with red, moderate luminescence with yellow and green, and mild luminescence with blue. The images displayed in this paper are photographic images with an overlay image of bioluminescence that uses this computer-generated false-color scale.
Quantification of luminescence.Luminescent images were quantified with the imaging software. The total flux (number of photons/s/cm2) was calculated from a user-defined area (region of interest) covering the major portion of the infected site. The flux was averaged from each of five mice for each treatment at each time point. While the background luminescence varied slightly for each experiment, the limit of detection, determined by measuring the luminescence in background regions, was usually 2 × 105 to 3 × 105 photons/s/cm2. Bioluminescence has been shown to correlate directly with the number of CFU in several animal models (25, 27, 33). In this peritonitis model, declining luminescence correlates with bacterial cell death. The percent reductions in luminescence given in the Results refer to a comparison with the luminescence for the saline-treated controls at the same time point.
Correlation of luminescence with CFU.Separate studies were conducted to confirm the correlation of bioluminescence with these engineered bacterial strains and the number of CFU collected at harvest. The harvest of bacterial colonies and the CFU counts in models of peritonitis can be unreliable due to variability in the recovery of bacteria from the infection site (the peritoneal cavity). Additional mice were given an intramuscular injection of a known concentration of bacteria into the left thigh (n = 2/group). One hour later, the mouse thighs were imaged to measure bioluminescence and were then immediately harvested aseptically, homogenized and serially diluted, and plated to determine the bacterial CFU. The results are plotted (Fig. 1) as the means and standard deviations of the log mean flux (x axis) versus the log mean number of CFU harvested/mouse (y axis).
Correlation between in vivo flux and the numbers of CFU harvested from luminescent MRSA in the neutropenic mouse thigh. Mice were given an intramuscular injection of MRSA (Xen-1) at time zero. The luminescence in the anesthetized mice was measured 1 h later, and then the same animals were euthanized for measurement of the numbers of CFU/mouse thigh. sr, steradian.
Statistical analyses.The mean and standard deviation of the luminescence for five mice were calculated at every time point for all antibiotics. Differences in the bioluminescence obtained with the different antibiotics at each time point were analyzed by using a two-way analysis of variance, followed by the Bonferroni posttest (GraphPad Prism software, version 4.0). The correlation between bioluminescence and CFU was analyzed by linear regression (Microsoft Excel software; Microsoft Office, 2003).
RESULTS
Correlation of bioluminescence and CFU.Figure 1 shows the correlation of bioluminescent flux with bacterial CFU in mice infected intramuscularly with MRSA. There was a highly significant correlation (R2 = 0.98) between the measured luminescent flux and the numbers of CFU recovered from the mouse thigh. This study supports the correlation between the bioluminescent signal and the number of viable bacteria for this model system.
MICs for MSSA and MRSA.The MICs of daptomycin, nafcillin, vancomycin, and linezolid for these bioluminescent strains of MSSA and MRSA are shown in Table 1. Note the lack of activity for nafcillin against MRSA Xen-1. Both strains of S. aureus are susceptible to daptomycin, linezolid, and vancomycin, according to CLSI standards.
MICs for bioluminescent S. aureus strains Xen-29 and Xen-1
Survival.Daptomycin at 50 mg/kg s.c. resulted in a greater than 90% reduction in luminescence under all conditions tested by 2 to 3 h postdosing. This reduction in luminescence correlated with higher survival rates among the daptomycin-treated mice. In nonneutropenic mice, a single dose of daptomycin at 50 mg/kg s.c. resulted in 100% survival for 7 days for both MRSA- and MSSA-infected mice (data not shown). Nafcillin was the only other treatment that protected 100% of the healthy mice with MSSA infections. Most often, 100% of the MRSA- and MSSA-infected mice treated with saline died within 24 h. For the neutropenic mice, a single dose of daptomycin at 50 mg/kg s.c. resulted in 100% survival for 24 h but only 40% survival through 7 days. For the neutropenic mice infected with either MRSA or MSSA, single doses of nafcillin, vancomycin, or linezolid resulted in 60% or greater mortality within 24 h and 100% mortality by 4 days postinfection.
Efficacy in immunocompetent mice.Among the immune-competent mice with MRSA peritonitis, daptomycin produced a greater and more rapid decline in luminescence compared with that produced by vancomycin (Fig. 2). In this model, 50 mg/kg daptomycin produced an 89% decrease in luminescence compared with that in saline-treated control mice only 2 h after s.c. dosing (Fig. 3). By 3 and 4 h after dosing, mice dosed with daptomycin had 97% and 98% decreases in luminescence, respectively, compared to the luminescence in the time-matched control mice. In contrast, 100 mg/kg vancomycin produced only a 76% reduction in luminescence at 2 h postdosing and produced 85% and 92% decreases in luminescence at 3 and 4 h after dosing, respectively.
Luminescent images of MRSA (Xen-1) peritonitis in healthy mice. Groups of mice (n = 5/group) were anesthetized with isoflurane and imaged for 3 min at 0 h (row 1), 2 h (row 2), and 4 h (row 3) after being dosed with 10 ml/kg saline (column 1), 50 mg/kg daptomycin (column 2), 100 mg/kg vancomycin (column 3), or 100 mg/kg nafcillin (column 4). All mice whose results are presented here were dosed via s.c. injections. The imaging system depicts false-color images representative of different levels of total flux (see Materials and Methods). False-color imaging represents intense luminescence in red, moderate luminescence in green, and low-level luminescence in blue/purple. Note the minimal luminescent signal measured for the daptomycin-treated mice at 4 h postdosing (row 3, column 2).
Mean flux levels (±standard deviation) from the experiment whose results are illustrated in Fig. 2. Groups of mice (n = 5/group) were anesthetized for measurement of luminescence (flux) every hour starting just prior to dosing (time zero) through 4 h postdosing. The mean flux for the daptomycin- and the vancomycin-treated mice from 2 to 4 h postdosing was significantly lower than that for the saline-treated control mice (P < 0.001 at 2 and 3 h for daptomycin and at 2 h for vancomycin; P < 0.01 at 4 h for daptomycin and at 3 h for vancomycin; P < 0.05 at 4 h for vancomycin).
Against MRSA, 100 mg/kg nafcillin was ineffective in altering the luminescence measured from peritoneal bioluminescent bacteria (Fig. 2 and 3). The MIC of nafcillin was 64 μg/ml for this strain of bacteria (Xen-1), whereas the MICs were 0.5 μg/ml for both daptomycin and linezolid and 2 μg/ml for vancomycin. The nafcillin-treated mice served as a negative control in this instance, demonstrating that ineffective antibiotic treatment results in luminescence measurements no different from those of the vehicle treatment.
At 100 mg/kg, nafcillin was nearly as effective as 50 mg/kg daptomycin in killing the MSSA bacteria in immune-competent mice (Fig. 4). At 2 h postdosing, mice dosed with 50 mg/kg daptomycin showed a 90% reduction in luminescence, whereas the reduction in luminescence was 82% for mice dosed with 100 mg/kg nafcillin s.c. (Fig. 4). By 3, 4, and 5 h after dosing, the bioluminescence in the daptomycin-treated mice had declined by 96%, 98%, and 99%, respectively, compared with the bioluminescence for the saline-treated controls at the same time points. Concurrently, the bioluminescence for the nafcillin-treated mice had declined by 93%, 96%, and 98%, respectively, compared with the bioluminescence for the controls. At a 99% reduction, almost no luminescence was seen in the false-color images of the daptomycin-treated mice 5 h after dosing (data not shown), as the bioluminescent signal was near the lower limit of detection for this model.
Mean flux levels (±standard deviation) measured during MSSA (Xen-29) peritonitis in healthy mice. Groups of mice (n = 5/group) were treated once with 10 ml/kg saline s.c., 50 mg/kg daptomycin s.c., 100 mg/kg nafcillin s.c., or 100 mg/kg linezolid via gavage (p.o.). The mice were anesthetized for the quantification of luminescence every hour starting just prior to dosing (time zero) through 5 h postdosing. The mean flux for the daptomycin- and the nafcillin-treated mice from 1 to 5 h postdosing was significantly lower than that for the saline-treated control mice (P < 0.01 at 1 h for nafcillin; P < 0.001 at 1 h for daptomcyin and for nafcillin and daptomycin 2 to 5 h postdosing). Mice dosed with linezolid showed a significant decline in mean flux relative to that for the saline-treated controls starting at 3 h postdosing (P < 0.001 at 3 to 5 h postdosing). The mean flux levels for daptomycin and nafcillin were significantly lower than that for linezolid at 2 h postdosing (P < 0.001). In the units of mean flux, p represents the number of photons.
In contrast, mice dosed via gavage with 100 mg/kg linezolid had no significant decrease in MSSA luminescence compared with that for the saline-treated control mice 2 h after dosing (Fig. 4). At 3, 4, and 5 h after dosing, the linezolid-treated mice had mean decreases in luminescence of 59%, 85%, and 90%, respectively, compared with the luminescence for the control mice. Linezolid, dosed via oral gavage in this model, was significantly less effective than daptomycin or nafcillin in decreasing the MSSA bacterial luminescence in immune-competent mice.
Efficacy in neutropenic mice.Against MRSA in neutropenic mice, 50 mg/kg daptomycin maintained mean luminescent signals near the starting values at 2 h postdosing, but the saline-treated control mice showed a significant increase in bioluminescence that correlated with the uncontrolled bacterial cell growth in these immune-compromised mice (Fig. 5 and 6). The daptomycin-treated mice produced 93% less luminescence compared with that produced by the time-matched controls at 2 h postdosing, a number that increased to 99% by 3 h and to 99.9% by 5 h. Thus, daptomycin at 50 mg/kg in mice resulted in a nearly 3-log10 reduction in bacterial luminescence, corresponding to a 3-log10 killing of luminescent bacteria in this peritonitis model.
Luminescent images of MRSA (Xen-1) peritonitis in neutropenic mice. Groups of mice (n = 5/group) were rendered neutropenic via two i.p. injections of cyclophosphamide 1 and 4 days before inoculation (see Materials and Methods). Luminescent images are shown just prior to dosing (0 h; row 1) and 2 and 4 h (rows 2 and 3, respectively) after dosing for mice treated with 10 ml/kg saline s.c. (column 1), 50 mg/kg daptomycin s.c. (column 2), 100 mg/kg vancomycin s.c. (column 3), or 100 mg/kg linezolid p.o. (column 4). False-color images are presented by using the same scale used in Fig. 2 (the range of color for all images is 8,000 to 100,000, with the exception of saline at 0 h and 4 h, for which the range is 10,000 to 100,000). Note the dramatic increase in red (high intensity) luminescence, which represents the highest levels of flux, in the saline-treated mice at 2 and 4 h postdosing.
Mean flux levels (±standard deviation) measured from the experiment whose results are illustrated in Fig. 5. Groups of mice (n = 5/group) were anesthetized for quantification of the luminescence every hour starting just prior to dosing (time zero) through 5 h postdosing. Note the rapid decline in mean flux for the daptomycin-treated mice starting at 3 h postdosing. The vancomycin- and the linezolid-treated mice maintain a slower decline in flux starting at 3 h postdosing. The means for daptomycin, vancomycin, and linezolid were significantly lower than that for the saline-treated controls from 2 to 5 h postdosing (P < 0.001). In the units of mean flux, p represents the number of photons.
Mice dosed via gavage with 100 mg/kg linezolid and those dosed s.c. with 100 mg/kg vancomycin produced slowly declining luminescent signals, on average, while the saline-treated controls and the nafcillin-treated mice displayed dramatic increases in bioluminescence over 5 h (Fig. 5 and 6). Mice dosed with linezolid had declines in luminescence, on average, of 91%, 98%, and 99.2% at 2, 3, and 5 h postdosing, respectively, compared with the luminescence for the saline-treated controls. Mice dosed with vancomycin had declines in luminescence, on average, of 84%, 95%, and 98.8% at 2, 3, and 5 h postdosing, respectively. By 3 h postdosing, daptomycin-treated mice had distinctly less luminescence than either the vancomycin- or the linezolid-treated mice.
Against MSSA peritonitis in neutropenic mice, daptomycin-treated mice once again showed the fastest and largest declines in bioluminescence, correlating with in vivo killing that was faster and more powerful than that achieved with nafcillin or linezolid (Fig. 7). By 2 h postdosing, 50 mg/kg daptomycin s.c. had reduced the luminescence signals by 92% compared with those for the saline-dosed controls. This decline in bioluminescent signal increased to 98%, 99.4%, and greater than 99.7% by 3, 4, and 5 h postdosing, respectively, for daptomycin-treated mice. The luminescence in the mice dosed with 100 mg/kg nafcillin decreased by only 84% compared with that in the controls at 2 h postdosing, but reductions of 95.6%, 98.3%, and 99.2% were achieved by 3, 4, and 5 h postdosing, respectively. Linezolid dosed at 100 mg/kg p.o. was significantly less effective than nafcillin or daptomycin, producing no significant decline in luminescence at 2 h postdosing and producing reductions of 68%, 89%, and then 96.6%, at 3, 4, and 5 h postdosing, respectively, compared with those in the time-matched controls.
Mean flux levels (±standard deviation) measured during MSSA (Xen-29) peritonitis in neutropenic mice. Groups of mice were treated once with saline, daptomycin, or nafcillin, all s.c., or with linezolid p.o. Mice were anesthetized for the measurement of luminescence every hour starting just prior to dosing (time zero) through 5 h postdosing. The means for daptomycin and nafcillin were significantly lower than that for the saline-treated controls from 2 to 5 h postdosing (P < 0.001); the means for linezolid were significantly lower than that for the saline-treated controls from 3 to 5 h postdosing (P < 0.001). The means for daptomycin and nafcillin were significantly lower than that for linezolid at 1 and 2 h postdosing (P < 0.05, except P < 0.01 for daptomycin at 2 h postdosing). In the units of mean flux, p represents the number of photons.
DISCUSSION
Daptomycin is rapidly bactericidal in vivo in a peritonitis infection model induced by MSSA or MRSA in both healthy and neutropenic mice (Fig. 2 to 7). The bactericidal activity of daptomycin was greater and faster than those of its comparator drugs (vancomycin for MRSA, nafcillin for MSSA, and linezolid via gavage for both MRSA and MSSA) in both healthy and neutropenic mice. The activity of linezolid was comparable to that of vancomycin against MRSA in neutropenic mice and was inferior to that of nafcillin against MSSA in healthy and neutropenic mice. Nafcillin was ineffective against MRSA, as expected. The superior in vivo killing kinetics of daptomycin, vividly illustrated by bioluminescence imaging in this model, closely matched its rapid bactericidal activity in vitro (1, 18, 42).
Daptomycin is among the most rapidly bactericidal agents available, typically achieving a greater than 99.9% bacterial reduction in vitro by 8 h or less (1, 7, 18, 42). In in vitro time-kill studies against staphylococcal pathogens, including MRSA and methicillin-resistant Staphylococcus epidermidis, daptomycin has shown greater activity than vancomycin, quinupristin-dalfopristin, and linezolid. Against vancomycin-resistant enterococci and other enterococci, daptomycin has the greatest bactericidal activity of any antibiotic used as monotherapy (18, 42). Furthermore, when daptomycin was tested in an in vitro model of simulated endocardial vegetations, which incorporates protein and high bacterial inoculum (9 log10 CFU/g), daptomycin was shown to be significantly more active than vancomycin, linezolid, or nafcillin against MSSA and MRSA in both growing and stationary-phase bacteria (47). By using the same simulated endocardial vegetation model against the first clinical isolate of vancomycin-resistant S. aureus, daptomycin was rapidly bactericidal, again achieving a greater than 99.9% bacterial reduction by 8 h, which was three times faster than that achieved by linezolid and matched only by that achieved with quinupristin-dalfopristin (7).
In vitro and in vivo, daptomycin is equally potent against MSSA and MRSA (10), making it a logical choice for empirical treatment of staphylococcal infections of undetermined antibiotic sensitivity. Nafcillin and other β-lactam antibiotics are of limited use for empirical treatment of serious staphylococcal infections due to the high rates of methicillin resistance among these organisms (5, 8, 23, 35, 36, 43) and the negative consequences of delayed treatment with effective coverage. Two recent studies demonstrated that delayed effective antimicrobial treatment for nosocomial S. aureus bacteremia was a statistically significant, independent predictor of infection-related mortality (22, 30).
The rise in the rate of methicillin resistance has led to an increased reliance on vancomycin for the treatment of documented MRSA infections, as well as for empirical therapy of infections in populations where the prevalence of MRSA is high (6). However, gram-positive pathogens with reduced susceptibilities to vancomycin are increasingly isolated from patients, especially those with frequent antibiotic exposure, such as immunocompromised patients (51). While they are still rare worldwide, S. aureus strains with reduced susceptibilities to vancomycin (MIC ≥ 4 μg/ml) have emerged in patients with a history of MRSA infection and antecedent vancomycin exposure and have responded poorly to vancomycin treatment (17). Thus, for both clinical and public health reasons, the indiscriminate use of vancomycin is no longer prudent and alternatives need to be identified (17, 44). Daptomycin has proven in vitro activity against vancomycin-intermediate S. aureus and vancomycin-resistant S. aureus (7, 9) and, as demonstrated in the current study, is a more rapidly bactericidal alternative to vancomycin against MRSA and MSSA in vivo.
Like daptomycin, linezolid is a first-in-class antibiotic with activity against most methicillin- and vancomycin-resistant pathogens (29). However, linezolid is inhibitory rather than bactericidal (29), with slower and less potent in vitro activity than daptomycin against S. aureus (1, 7, 18, 42, 47). The current study supports prior in vitro data by demonstrating that daptomycin decreases the bacterial load in vivo faster than comparator agents, including linezolid. Resistance to linezolid and persistent MRSA bacteremia during linezolid therapy have been reported (41, 45). Furthermore, linezolid has been associated with myelosuppression, including neutropenia (16, 19), which could exacerbate preexisting immunosuppression in special populations, including cancer patients.
The rapid bactericidal action of daptomycin against S. aureus in neutropenic mice demonstrated in the present study suggests its potential clinical applicability in immunocompromised patients. Among a small contingent of immunocompromised patients in two large, phase III clinical trials of patients with complicated skin and skin structure infections, higher clinical success rates were observed in patients treated with daptomycin than in those treated with comparator agents (nafcillin and vancomycin): 77% (10/13) of patients treated with daptomycin were clinical successes, while only 60% (9/15) of comparator agent-treated patients were considered clinical successes. Due to the small number of patients, this difference did not achieve statistical significance (95% confidence interval for daptomycin minus comparator, −16.8% and 50.7%). Furthermore, within this immunocompromised subpopulation, those treated with daptomycin required shorter courses of therapy than those treated with comparator (4). Of those patients who received only intravenous therapy, 62.5% (five of seven) and 37.5% (three of eight) of daptomycin- and comparator agent-treated patients, respectively, received 7 days or less of intravenous therapy. Again, due to the small numbers, these differences were not statistically significant (95% confidence interval for daptomycin minus comparator, −13.5% and 81.3%).
As mentioned in the introduction, bactericidal action may be a relevant property for effective therapy in the immunocompromised population, because bacteriostatic antibiotics depend on host immunity for bacterial eradication. In neutropenic mice in the current study, daptomycin remained rapidly bactericidal against MSSA and MRSA, suggesting that daptomycin activity does not rely on intact host immunity. Clinical research directly addressing this theory of bactericidal advantage in patients with neutropenia is currently lacking (13). Another important potential benefit of bactericidal antibiotics is that they may limit the development of resistance, whereas bacteriostatic agents may not (46). Rapid bactericidal activity could be important in the treatment of patients with neutropenia and serious infections with gram-positive bacteria, such as bacteremia.
ACKNOWLEDGMENTS
We thank the many employees of Cubist Pharmaceuticals, Inc., who helped edit the manuscript and, specifically, Richard Baltz, Bill Martone, Anton Ehrhardt, and Barry Eisenstein. Thanks go to Nicole Cotroneo, Jared Silverman, and Valerie Laganas for determining the MICs. Thanks go to Conni Otradovec for assistance with statistical analyses. We also thank Bill Anderson of Xenogen Corporation and Carol Bellinger-Kawahara, formerly of Xenogen, for their extensive help with the use of the imaging system.
This study was supported by Cubist Pharmaceuticals, Lexington, MA.
L. I. Mortin, T. Li, A. D. G. Van Praagh, S. Zhang, X. Zhang, and J. D. Alder are employees of Cubist.
FOOTNOTES
- Received 16 June 2006.
- Returned for modification 6 August 2006.
- Accepted 5 February 2007.
- Copyright © 2007 American Society for Microbiology
