TABLE 2.

Recommendations for murine neutropenic thigh and lung infection models to determine nonclinical in vivo PK/PD targets^a

Study component	Recommendationb	Comments
Mouse strain	Outbred (e.g., CD-1, ICR or Swiss Webster)	Historically female; studies in both sexes have been strongly encouraged recently and, if feasible, should be considered
Induction of neutropenia	Cyclophosphamide i.p. or s.c. at 150 mg/kg of body weight at 4 days prior to infection and 100 mg/kg at 1 day prior to infection	Results in <100 neutrophils/mm3 for at least 2 days
Inoculum preparation	Culture should be in log-growth phase	Subculture aliquot from an overnight broth culture in fresh medium for several hours prior to study start
Mouse inoculation	Infect thigh via i.m. injection of 100 μl and lung via intranasal inhalation of 50 μl (i.e., 25 μl per naris)c	Culture for inoculation should be 106 to 107 CFU/ml
Baseline bacterial burden	106 to 107 CFU/tissue (may differ by pathogen and strain)	Note that this represents the burden at the time therapy begins
Start of therapy	2 h postinfection	Delay may be necessary for baseline tissue burden to reach 106 to 107
Study duration	24 h (sometimes 48 h)	After start of antibacterial dosing
Bacterial growth over study period	Tissue burden should increase by 2–3 log10 CFU in untreated mice compared to baseline at initiation of therapy	Note that this assumes that the initial inoculum is sufficiently below the plateau for a given strain; the use of less-virulent strains may result in underestimation of the PK/PD target
No. of strains	At least 4 strains of each target pathogen (including a reference strain), if possible, with relevant resistance profiles and mechanisms	Include enough strains to assess strain-to-strain variability; mean and median PK/PD target values should converge
Bacterial phenotypes	Cover MIC range of compound, include clinically relevant resistant phenotypes	Consider in vivo virulence when choosing strains
Control therapies	Inclusion of active comparator control (e.g., standard of care) may be beneficial; dosage regimen (with/without humanizing) should be considered	Especially important for evaluation of combination therapies against multidrug-resistant strains; dosing algorithm should be supported by PK/PD considerations

^aData are from Andes and Lepak (101). CD-1, outbred strain of albino mice; ICR, outbred strain of albino mice; i.p., intraperitoneal; s.c., subcutaneous; i.m., intramuscular.

^bThese specific recommendations are for “routine” establishment of PK/PD targets. Study design elements may need to be modified to achieve different experimental goals. Examples include the use of other bacterial phenotypes (including growth stages), use of immunocompetent mice (which can inform how targets may differ in the presence of white blood cells and/or support longer treatment durations), and use of a different bacterial burden (such as using a higher burden to study resistance).

^cThe maximum volume of the bacterial suspension which can be given per naris will depend on the mouse weight. This volume may affect the regional deposition of bacteria in the lung.