Impact of MIC Range for Pseudomonas aeruginosa and Streptococcus pneumoniae on the Ceftolozane In Vivo Pharmacokinetic/Pharmacodynamic Target

Abstract

Ceftolozane is a novel cephalosporin with activity against drug-resistant pathogens, including Pseudomonas aeruginosa and Streptococcus pneumoniae. The in vivo investigation reported here tested the limits of this drug against 20 P. aeruginosa and S. pneumoniae isolates across a wide MIC range and defined resistance mechanisms. The times above the MIC (T>MIC) targets for stasis and 1- and 2-log reductions were 31%, 39%, and 42% for P. aeruginosa and 18%, 24%, and 27% for S. pneumoniae, respectively. The 1-log endpoint was achieved for strains with MICs as high as 16 μg/ml.

TEXT

Emerging drug resistance has compromised the utility of our current antimicrobial armamentarium (1–3). The development of the novel cephalosporin ceftolozane provides a solution for a subset of drug-resistant infections. The spectrum of activity includes two common multidrug-resistant respiratory pathogens, Pseudomonas aeruginosa and Streptococcus pneumoniae (4–9). Initial pharmacokinetic/pharmacodynamic studies with this drug demonstrate that the time above the MIC, as a percentage (%T>MIC), is the measure linked to efficacy, as with other cephalosporins (10). However, studies exploring the pharmacokinetic/pharmacodynamic (PK/PD) target associated with efficacy suggest that the %T>MIC for ceftolozane is lower than that for other drugs within the cephalosporin class (10, 11). Mechanistic investigations suggest that this may be related to the rate of organism killing and perhaps affinity for the penicillin-binding protein (PBP) (10, 12). The goal of the present study was to test the PK/PD limits of ceftolozane efficacy in vivo against P. aeruginosa and S. pneumoniae across a wide MIC range and with a diversity of resistance mechanisms.

Fourteen P. aeruginosa and 6 S. pneumoniae isolates were studied (Table 1). MICs were determined in triplicate according to CLSI guidelines (13). The ceftolozane MIC range for the 20 isolates was 0.125 to 16 μg/ml (i.e., it varied 128-fold). For P. aeruginosa, MICs ranged from 2 to 16 μg/ml, and for S. pneumoniae, they ranged from 0.125 to 16 μg/ml. Strain phenotypes and genotypes included ceftazidime, carbapenem, and ciprofloxacin resistance in P. aeruginosa due to AmpC overproduction and/or OprD mutations and penicillin resistance in four S. pneumoniae strains. The neutropenic-mouse thigh infection model was used for in vivo study of ceftolozane (14). Animals were maintained in accordance with American Association for Accreditation of Laboratory Animal Care (AAALAC) criteria. All animal studies were approved by the Animal Research Committee of the William S. Middleton Memorial VA Hospital and the University of Wisconsin. Mice were infected with 10⁵ to 10⁶ CFU of each strain/thigh.

TABLE 1.

Study organisms, phenotypes, genotypes, and ceftolozane MICs^a

Organism	MIC (μg/ml)	Phenotype	Genotype
S. pneumoniae
ATCC 10813	0.125	PSSP	NA
145	0.125	PRSP	NA
146	0.25	PRSP	NA
1329	8	PRSP	NA
1020	8	PRSP	NA
49619	16	PISP	NA
P. aeruginosa
3B	2	NA	NA
2638	2	DOR R, IPM R, MEM R, FEP R, CAZ R, TZP R, CIP R, TOB R	NA
4304A	2	NA	NA
3068	4	DOR R, IPM R, MEM R, FEP R, CAZ S, TZP R, CIP R	NA
3070	4	DOR R, IPM R, MEM R, FEP I, CAZ R, TZP S, CIP R	OprD mutant, AmpC hyperproducer
2757	4	NA	AmpC hyperproducer
2627	4	NA	NA
9139	4	DOR S, IPM S, MEM S, FEP R, CAZ S, TZP R, CIP S, TOB S	NA
3072	4	DOR R, IPM R, MEM R, FEP S, CAZ S, TZP S, CIP R, TOB R	OprD mutant, AmpC hyperproducer
3071	8	DOR R, IPM R, MEM R, FEP R, CAZ R, TZP R, CIP R	OprD mutant, AmpC hyperproducer
823	8	DOR S, IPM S, MEM S, FEP R, CAZ R, TZP R, CIP S, TOB S	NA
26975	8	DOR S, IPM S, MEM S, FEP R, CAZ R, TZP R, CIP S, TOB R	NA
3076	16	DOR R, IPM R, MEM R, FEP I, CAZ R, TZP R, CIP R, TOB S	OprD mutant, AmpC hyperproducer
24530	16	DOR R, IPM R, MEM R, FEP R, CAZ R, TZP R, CIP S, TOB S	NA

^aPSSP, penicillin-susceptible Streptococcus pneumoniae; PRSP, penicillin-resistant Streptococcus pneumoniae; PISP, penicillin-intermediate Streptococcus pneumoniae; DOR, doripenem; IPM, imipenem; MEM, meropenem; FEP, cefepime; CAZ, ceftazidime; TZP, piperacillin-tazobactam; CIP, ciprofloxacin; TOB, tobramycin; S, susceptible; I, intermediate; R, resistant; NA, not available.

The in vivo fitness of the strains was relatively similar in untreated control mice based upon a similar increase in thigh burden over the 24-h period, 2.94 ± 0.49 log₁₀ CFU/thigh for all isolates. Two hours after thigh infection, ceftolozane was administered subcutaneously for 24 h using 6 or 7 dosing regimens that ranged from 0.39 mg/kg to 800 mg/kg every 6 h. A 1- or 2-log kill was achieved for most strains (Fig. 1A and C). In general, the dose response for each strain was linked to the MIC. More specifically, the drug dose required for efficacy, on a mg/kg basis, increased proportionally as the MIC increased. Efficacy was observed across the resistance genotypes and other drug resistance phenotypes. Plasma pharmacokinetics from our recent study with this infection model were used for %T>MIC determinations (10). Total drug concentrations were used, given the low degree of binding in this animal model (<10%). The sigmoid E_max model was used to analyze the exposure response data. The ceftolozane dose and associated %T>MIC needed to achieve net stasis and 1-log and 2-log kills were calculated and are shown in Table 2. Although the dose level (on a mg/kg basis) associated with these endpoints varied 62- to 138-fold, the %T>MIC varied only 1.3- to 2.5-fold. Thus, the %T>MIC needed for efficacy was relatively similar across the wide range of MICs studied. There was no difference in the PK/PD target T>MIC based on MIC across each species group (P = 0.568). As shown in Fig. 1B and D, regression of the treatment effect data with the %T>MIC measure resulted in a strong relationship, demonstrated by the relatively high coefficients of determination (R² = 0.80 for P. aeruginosa and 0.85 for S. pneumoniae).

FIG 1.

(A) Dose-response relationship for the 6-hourly dosing of ceftolozane against 6 S. pneumoniae isolates. The dose is expressed as mg/kg/24 h. Each symbol represents the mean log₁₀ CFU/thigh for 2 mice (4 thighs), and the error bars represent standard deviations. Black symbols represent an MIC of 0.125 mg/liter; green, 0.25 mg/liter; blue, 8 mg/liter; and red, 16 mg/liter. The dashed horizontal line represents the burden of organisms in the thighs at the start of therapy. (B) Dose-response relationship for the 6-hourly dosing of ceftolozane against 14 P. aeruginosa isolates. The dose is expressed as mg/kg/24 h. Each symbol represents the mean log₁₀ CFU/thigh for 2 mice (4 thighs), and error bars represent standard deviations. Black symbols represent an MIC of 2 mg/liter; green, 4 mg/liter; blue, 8 mg/liter; and red, 16 mg/liter. The dashed horizontal line represents the burden of organisms in the thighs at the start of therapy. (C) Relationship between the percent time above MIC and change in CFU/thigh over 24 h of treatment with ceftolozane against 6 S. pneumoniae isolates. Each symbol represents the mean log₁₀ CFU/thigh for 2 mice (4 thighs). Black symbols represent an MIC of 0.125 mg/liter; green, 0.25 mg/liter; blue, 8 mg/liter; and red, 16 mg/liter. R², coefficient of determination. E_max, ED₅₀, and N represent the maximal effect, 50% effect, and slope of the relationship resulting from the sigmoid E_max model, respectively. The dashed horizontal line represents the burden of organisms in thighs at the start of therapy. The solid sigmoid line represents the best fit using the sigmoid E_max model. (D) Relationship between the percent time above MIC and change in CFU/thigh over 24 h of treatment with ceftolozane against 14 P. aeruginosa isolates. Each symbol represents the mean log₁₀ CFU/thigh for 2 mice (4 thighs). Black symbols represent an MIC of 0.125 mg/liter; green, 0.25 mg/liter; blue, 8 mg/liter; and red, 16 mg/liter. R² represents the coefficient of determination. E_max, ED₅₀, and N represent the maximal effect, 50% effect, and slope of the relationship resulting from the sigmoid E_max model, respectively. The dashed horizontal line represents the burden of organisms in the thighs at the start of therapy. The solid sigmoid line represents the best fit line using the sigmoid E_max model. Black symbols represent an MIC of 2 mg/liter; green, 4 mg/liter; blue, 8 mg/liter; and red, 16 mg/liter.

TABLE 2.

Dose and %T>MICs for stasis, 1-log kill, and 2-log kill for ceftolozane against six S. pneumoniae and 14 P. aeruginosa strains

Organism	Static dose (mg/kg/24h)	Stasis %T>MIC	1-log kill (mg/kg/24h)	1-log kill %T>MIC	2-log kill (mg/kg/24 h)	2-log kill %T>MICa
S. pneumoniae
10813	10.3	20.9	18.9	23.8	44.4	27.7
145	4.44	17.0	8.38	20.0	30.2	25.9
146	6.02	15.2	20.0	20.8	66.0	26.4
1329	642	25.4	1,101	30.0	>1,600	NA
1020	271	17.1	696	26.2	>1,600	NA
49619	231	12.6	677	22.0	>1,600	NA
Mean	194	18.1	420	23.8	46.8	26.7
Median	121	17.1	349	22.9	44.4	26.4
SD	250	4.52	468	3.75	18.0	0.94
P. aeruginosa
3B	55.7	15.9	140	20.1	389	27.6
2638	1,473	40.1	2,538	43.2	4,501	46.4
4304A	907	36.2	5,029	47.0	>12,800	NA
3068	717	30.3	1,958	37.8	6,942	44.9
3070	1,814	37.3	10,228	47.1	>12,800	NA
2757	1,848	37.5	4,672	42.7	>12,800	NA
2627	2,040	38.0	7,558	45.4	>12,800	NA
9139	456	25.6	1,597	36.6	9,242	46.5
3072	608	28.7	>12,800		>12,800	NA
3071	780	27.2	11,226	43.7	>12,800	NA
823	444	21.5	1,847	33.5	>12,800	NA
26975	974	29.9	3,021	36.3	12,800	44.5
3076	5,602	35.9	>12,800		>12,800	NA
24530	3,141	32.6	10,658	39.5	>12,800	NA
Mean	1,490	31.2	5,039	39.4	6,775	42.0
Median	940	31.4	3,847	41.1	6,942	44.9
SD	1,438	6.99	3,921	7.53	4,700	8.11

^aNA, not achieved.

These data both confirm and extend the PK/PD information for ceftolozane. First, the results affirm the importance of the %T>MIC PK/PD measure for ceftolozane (15–18). Second, the PD targets identified in the current studies are similar to those noted in a previous in vivo assessment of ceftolozane (10). Specifically, the stasis and killing targets for the Gram-negative group were T>MIC values near 30% and 40%, respectively. These values are considerably lower than for other cephalosporins and may be due to more rapid killing kinetics (10, 19). For example the %T>MIC stasis target for ceftazidime against P. aeruginosa was found to be 40 to 45% in this animal infection model, which is closer to the 1-log kill target for ceftolozane and has also been suggested from clinical PK/PD analyses (17, 20, 21). The inclusion of a wider MIC range, higher MICs, and defined resistance mechanisms in the present studies provides an opportunity to understand the target pathogen “ceiling” for this new compound. The current results suggest that ceftolozane may be a useful treatment option for infections with few alternatives, including those due to strains resistant to other cephalosporins, quinolones, and even carbapenems. Efficacy was observed against organisms with MICs as high as 16 μg/ml. Importantly, human kinetic studies demonstrate that a dosing regimen of 1 g every 8 h produces serum concentrations near this MIC for nearly 50% of the dosing interval (22, 23). Additionally, these data represent the first PK/PD exploration for ceftolozane against pneumococci. Interestingly, comparison of these data with those from studies of Gram-negative organisms show that the %T>MIC target for similar efficacy in S. pneumoniae is nearly half of that for P. aeruginosa (P < 0.001). This PK/PD target difference has been described for other drug-bug combinations, but the mechanistic explanation in this case is unknown (24). The current results verify the promising utility of this new cephalosporin against multidrug-resistant pathogens and across a wide range of ceftolozane MICs. The data should be useful in guiding clinical use and the development of susceptibility breakpoints.