Optimizing Clinical Use of Polymyxin B

Multidrug-resistant (MDR) Gram-negative 'superbugs' are rapidly spreading around the world,
and polymyxin B and colistin (polymyxin E) are often the only effective antibiotics. Since
polymyxin B was released in the 1950s, its pharmacokinetics, pharmacodynamics, toxicodynamics
(PK/PD/TD) have never been defined. Recent pharmacological research on polymyxins has
predominantly focused on colistin methanesulfonate (CMS, an inactive prodrug of colistin) and
demonstrates that CMS has significant limitations. Thus, polymyxin B is increasingly being
viewed as the preferred polymyxin. Unfortunately, recently developed scientifically-based
dosing recommendations for CMS cannot and should not be applied to polymyxin B, as the latter
is administered as its active entity. Therefore, it is essential to determine the PK/PD/TD of
polymyxin B in critically-ill patients, refine optimal dosage regimens, and develop the
user-friendly adaptive feedback control (AFC) clinical tool.

The Specific Aims are:

1. To develop a population PK model for polymyxin B;

2. To investigate relationships between the PK of polymyxin B, duration of therapy and
patient characteristics, with the development and timing of nephrotoxicity; and to use
next-generation proteomics to identify the most predictive biomarker(s) of polymyxin B
associated nephrotoxicity; and to develop the population PK/TD model;

3. To establish the relationships between polymyxin B PK, bacterial susceptibility and
patient characteristics, with the probability of attaining and time to achieving
clinical and bacteriological outcomes; and

4. To employ the models from Aims 1-3 and Monte Carlo simulation to develop
scientifically-based dosage regimens of polymyxin B and to develop an AFC algorithm for
future individual patients.

Research Design: Patients being treated with intravenous polymyxin B will be identified at
three clinical sites in the USA and one in Singapore. Patients (n = 250) will have blood
collected at various times surrounding a dose of polymyxin B between days 1 and 5 of therapy.
Development of nephrotoxicity, clinical response, and bacteriological response will be
examined. Total and free plasma concentrations of polymyxin B will be determined. Bacterial
isolates will be examined for the emergence of polymyxin resistance. The relationships
between polymyxin B PK, PD and TD end-points (e.g. clinical and bacteriological responses,
development of toxicity and resistance) will be assessed using pharmacometric analyses.
Finally, the obtained information will be used to apply Monte Carlo simulation to examine the
impact of various patient characteristics and other factors on polymyxin B PK, PD and TD, in
order to establish optimal dosage regimens and AFC algorithms for individual critically-ill
patients.

Significance: No new antibiotics will be available for Gram-negative 'superbugs' for many
years. This landmark multicenter study will provide essential information for optimizing
polymyxin B use in critically-ill patients, while minimizing resistance and toxicity. This
proposal aligns perfectly with the NIAID priority "To teach old drugs new tricks" and the
recent Executive Order of the White House to combat antibiotic resistance.

Inclusion Criteria:

1. Patient of 18 years of age or older

2. Expectation of hospitalization and receipt of polymyxin B of ≥ 48 hours

3. Receipt of intravenous polymyxin B for treatment of a bloodstream infection (according
to CDC criteria 55) and/or pneumonia

4. Provision of written informed consent by the patient or by the patient's health care
proxy if the patient cannot give consent

5. Adequate venous access to enable collection of blood for determination of
concentrations of polymyxin B and co-administered antibiotics

Exclusion Criteria:

1. Age <18 years

2. Currently incarcerated

3. Concomitant use of polymyxin B delivered directly into the respiratory tract

4. Cystic fibrosis

5. Known allergy to CMS/colistin or polymyxin B

6. Anticipated death within 48 h of commencing polymyxin B therapy