In vitro potency of amikacin and comparators against E. coli,K. pneumoniae and P. aeruginosa respiratory and blood isolates

The purpose of this study was to define the potency of amikacin and comparator agents against a collection of blood and respiratory nosocomial isolates implicated in ICU based pulmonary infections gathered from US hospitals. Minimum inhibitory concentrations of amikacin, aztreonam, cefepime, ceftazidime, ceftolozane/tazobactam, ceftriaxone, ciprofloxacin, imipenem, meropenem, piperacillin/tazobactam and tobramycin were tested against 2460 Gram-negative isolates. Amikacin had 96 % susceptibility against the combined E. coli and K. pneumoniae isolates and 95 % susceptibility against P. aeruginosa. Ninety-six percent of all of isolates tested were susceptible (i.e., MICs ≤16 mg/L) to amikacin by current laboratory standards which demonstrates a high level of activity to combat infections caused by these organisms including ESBL, MDR, β-lactam and fluoroquinolone resistant strains. Moreover, 99 % of all organisms had amikacin MICs ≤64 mg/L. Overall, these data highlight the continued potency of amikacin and suggest that the achievable lung concentrations of approximately 5000 mg/L with the administration of the amikacin by inhalation (Amikacin Inhale, BAY41-6551) will exceed the MICs typically observed for P. aeruginosa, E. coli and K. pneumoniae in the hospital setting.


Background
The management of hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP) has been made increasingly difficult due to the emergence of resistance and the potential for reduced antibiotic lung penetration in the intubated patient. HAP continues to be the second most common cause of nosocomial infections in the United States and is associated with increases in hospital length of stay, healthcare costs and represents a major cause of mortality especially in critically ill patients [1,2]. VAP, a subset of HAP, that occurs in mechanically ventilated patients more than 48 h after tracheal intubation and occurs in 9-40 % of mechanically ventilated patient's making it among the most frequent infections in the ICU [1,3].
Gram-negative bacteria are responsible for a substantial proportion of HAP and VAP infections. Pseudomonas aeruginosa, along with the Enterobacteriacae, Escherichia coli and Klebsiella pneumoniae are amongst the most common etiological organisms representing approximately two-thirds of causative agents [4,5]. Nosocomial pneumonia caused by P. aeruginosa, E. coli and K. pneumoniae continues to pose significant challenges in US hospitals due to their prevalence and the acquisition of numerous antimicrobial resistance mechanisms. As a result, the selection of empirical antibiotic therapy in patients with nosocomial respiratory tract infections has become increasingly challenging as the number of potentially effective agents has been reduced due to evolving resistance. Yet more challenging is delivering sufficiently Amikacin Inhale (BAY41-6551) is a reformulated solution of amikacin (AMK) combined with a drug-delivery module that is currently under phase III study as an adjunctive therapy for the treatment of Gram-negative pneumonia in intubated and mechanical ventilated patients. In vitro pharmacodynamic models evaluating the achievable epithelial lining fluid (ELF) concentrations after the administration of AMK inhalation against Gram-negative organisms demonstrated rapid and sustainable bactericidal killing of AMK both alone and in combination with systemic exposures of meropenem when AMK MICs were ≤256 mg/L [6]. Our objective was to define the potency and MIC distribution of AMK against a US collection of P. aeruginosa, E. coli, K. pneumoniae nosocomial isolates and relate these data to achievable lung concentrations of the compound when delivered via the aerosol route.

Methods
Fifty US hospitals, 41 teaching and 9 community provided non-duplicate nosocomial blood and respiratory isolates of E. coli, K. pneumoniae and P. aeruginosa from adult inpatients. Additionally, five of these hospitals also provided S. maltophilia respiratory isolates. Organisms were identified at each participating site using the standard methods. The isolates were transferred to trypticase soy agar slants for shipping to the Center for Anti-Infective Research and Development, Hartford Hospital, Hartford, CT, USA. Collection occurred from 2013 into 2014.
Clinical Laboratory Standards Institute (CLSI) defined broth microdilution methods were employed to determine minimum inhibitory concentration (MICs) for AMK, aztreonam (ATM), cefepime (FEP), ceftazidime (CAZ), ceftolozane/tazobactam (C/T), ceftriaxone (CRO), ciprofloxacin (CIP), imipenem (IPM), meropenem (MEM), piperacillin/tazobactam (TZP) and tobramycin (TOB) [7]. Antibiotics were purchased from Sigma (St. Louis, MO) except for C/T which was provided by Cubist Pharmaceuticals. Quality control was performed on each batch of MIC testing using E. coli 25922 and P. aeruginosa 27853 as defined by CLSI. All transfer and colony counts were performed on trypticase soy agar plates containing 5 % blood. CLSI and FDA breakpoints were used to define susceptibility. For C/T the FDA breakpoints of 2 mg/L for Enterobacteriaceae and 4 mg/L for P. aeruginosa were utilized [8]. Isolates that were non-susceptible to AMK (i.e., ≥32 mg/L) by current laboratory definitions were repeated and confirmed.
Escherichia coli and K. pneumoniae were tested for extended spectrum β-lactamases (ESBL) production if they had an MIC of ≥1 to 2 mg/L of the following: ATM, CRO or CAZ. CLSI defined ESBL confirmation studies were then undertaken using additional MIC testing with CAZ, CAZ with clavulanate, cefotaxime and cefotaxime with clavulanate [7].
Isolates testing non-susceptible to ertapenem, imipenem, or meropenem were evaluated for carbapenemase production using the CarbaNP test [10].
To determine if the susceptibility of these organisms are influenced by the source of infection and location of the patient, we evaluated the % susceptibility of ICU versus Non-ICU patients. The Enterobacteriaceae had no more than a 2 % difference in % susceptibility between the ICU (n = 539) and Non-ICU (n = 1106) patients. For the P. aeruginosa, we noted differing susceptibility profiles derived from the ICU and non-ICU setting. The susceptibility profile of ATM, FEP, CAZ, IPM, MEM and TZP was 5-11 % lower from patients in the ICU. In contrast, ≤3 % difference in was observed between the potency of ICU and non-ICU P. aeruginosa for AMK, C/T, CIP and TOB.

Discussion
Hospital-acquired pneumonia (HAP) and ventilatorassociated pneumonia (VAP) are among the most frequent nosocomial infections encountered in the ICU and are responsible for increases in length of stay, mortality and morbidity in critically ill patients. P. aeruginosa, along with the Enterobacteriacae, E. coli and K. pneumonia represent a substantial proportion of HAP and VAP infections. The use of empirical broad spectrum antibiotics for the treatment of pneumonia has led to an increase in antimicrobial resistance along with the development of multidrug resistant Gram-negative organisms.
Appropriate antibiotic exposure at the site of infection is an important component for therapy to be clinically effective. Epithelial lining fluid (ELF) is considered to be the site of infection for pneumonia and achieving sufficiently high drug concentrations in ELF are critical for treating pneumonia. The rationale for inhaling antibiotics is to maximize drug delivery to the target site of infection (ELF in the case of pneumonia) and limit the potential for systemic side effects. The uses of inhaled antimicrobials have a long history in the treatment of lower respiratory tract infections specifically in patients with cystic fibrosis [11]. Due to the increasing prevalence of resistant Gram-negative organisms in VAP causing difficult to treat pneumonia along with the limited lung concentrations that can be achieved with standard parentally given antibiotics there has been an increased interest in the use of inhaled antimicrobials such as inhaled amikacin for respiratory tract infections such as in HAP and VAP [12]. In a recently published study, So et al. [6] investigated the in vitro pharmacodynamics of human simulated ELF concentrations of inhaled amikacin against variety of phenotypically diverse susceptible and non-susceptible K. pneumoniae and P. aeruginosa isolates and showed rapid bactericidal activity against isolates with an amikacin MICs of ≤256 mg/L. The high overall susceptibility profile of amikacin against these Gram-negative organisms of interest in our current study appears to represent the sustained in vitro potency of the compound when compared to two previously reported surveillance programs [13,14]. The INFORM study collected 7062 clinical P. aeruginosa isolates from Europe, Asia/South Pacific, Latin America and Middle East/Africa [13]. Similar to our data the INFORM global surveillance program found amikacin to have a high susceptibility (89.4 %) against the P. aeruginosa isolates tested. Moreover, the authors also reported similar (±6 %) susceptibility profiles for ceftazidime, cefepime, piperacillintazobactam and meropenem to that in our current study. The SMART surveillance program collected intraabdominal isolates of E. coli, K. pneumoniae, K. oxytoca and P. mirabilis [14]. E. coli (n = 434) maintained high susceptibilities to amikacin of >99 %, while K. pneumoniae (n = 231) had 93.5 %. Both organisms had ≥92 % susceptibility to imipenem and ertapenem. In the SMART program amikacin retained the highest susceptibility to MDR Enterobacteriaceae of all the agents tested.

Conclusions
In this study we defined the phenotypic profile of amikacin against 2460 blood and respiratory nosocomial isolates implicated in ICU based pulmonary infections collected from 50 US hospitals. AMK demonstrated a high level of activity to combat infections caused by the Enterobacteriacae, E. coli and K. pneumoniae including those strains producing ESBLs. When considering E. coli and K. pneumoniae our study found amikacin to have an MIC 50 and MIC 90 of 4 and 16 mg/L, respectively. Moreover, 96 % of organisms had a MIC of ≤16 mg/L and nearly all (99 %) organisms had MICs ≤64 mg/L despite concomitant β-lactam or fluoroquinolone resistance. In a study conducted by Sader et al. [15] analyzing the in vitro activity of amikacin against isolates gathered from patients hospitalized with pneumonia including VAP, across 62 US hospitals, the MIC 50 and MIC 90 was reported to be 1 and 32 mg/L, respectively, against K. pneumoniae. In a study investigating E. coli isolates across 66 Canadian medical centers the reported amikacin MIC 50 and MIC 90 were ≤2 and 4 mg/L, respectively [16].
For P. aeruginosa, our study found amikacin to have an MIC 50 and MIC 90 of 8 and 16 mg/L, respectively. Against these isolates 95 % of organisms had amikacin MICs of ≤16 mg/L, moreover nearly all (97 %) organisms have MICs ≤32 mg/L. Similarly when compared to the Sader et al. [15] study, P. aeruginosa isolates from 62 US hospitals, amikacin was reported to have an MIC 50 and MIC 90 of 4 and 8 mg/L, respectively. Another study analyzing similar data from 127 isolates from Canadian hospitals, reported MIC 50 and MIC 90 of 4 and 16 mg/L, respectively [17].
With respect to resistant Gram-negative bacteria, our study demonstrated that amikacin maintained potent activity against P. aeruginosa MDR organisms. In regards to ESBLs this study showed that AMK was a potent antimicrobial with a high level of activity against these MDR isolates as 87 % had an MIC of ≤16 mg/L. Against MDR P. aeruginosa isolates, amikacin was the most potent antibiotic tested as 87 % of resistant isolates were considered susceptible (i.e., MIC of ≤16 mg/L).
The current surveillance study which incorporated a large number of US hospitals demonstrated the high potency of amikacin against contemporary isolates of E. coli, K. pneumoniae, P. aeruginosa originating from a