Whole-body Bacteriophage Distribution Characterized by a Physiologically based Pharmacokinetic Model

In 2019 there were over 2.8 million cases of antibiotic-resistant bacterial infection in the US with gram negative organisms having up to a 6% rate of mortality. Bacteriophage (phage) therapy holds great promise to treat such infections. However, the biologic features which influence the pharmacokinetics (PK) of phage have been difficult to characterize due to a lack of standardized protocols of phage purification, tissue assay, and labeling. Here we present robust methods for ultrapure phage preparation as well as non-destructive highly stable attachment of radio-iodide to phage using the well described Sulfo-SHPP linker. We purified and radiolabeled the phage strains, PAML-31-1, OMKO1, and Luz24 lytic to drug-resistant Pseudomonas aeruginosa for biodistribution assay in normal young adult CD-1 mice injected via penile vein. Groups of 5 mice were euthanized and tissues/organs removed for weighing and scintillation well counting of I-125 activity at 30 min, 1h, 2h, 4h, 8h, and 24h. A physiologically based PK (PBPK) model was then constructed focusing on compartments describing blood, lung, muscle, bone, liver, stomach, spleen, small intestines, large intestines, and kidney. Model permeability coefficient (P_S) was estimated across all organs as being 0.0227. Tissue partition coefficients (K_P) were estimated for high perfusion organs (lung and kidney) as 0.000138, GI organs (liver, spleen, and stomach) as 0.627, and all other organs as 0.220. Elimination was governed by MPS-mediated elimination (T_MPS,deg) and active secretion at epithelial barriers (CL_Active), which were estimated as 0.00301 h and 0.0145 L/h/kg, respectively. Monte Caro simulations showed that the rapid elimination phage in humans is expected, resulting in phage blood concentrations being lower than 10² PFU/mL (limit of quantification by plaque assay) by 12 hours. As such, multi-dose regimens and continuous infusion regimens were the only strategies that allowed continuously detectible phage concentrations. Evaluation of different dose levels showed that at a maximum dose of 10¹² PFU, phage concentrations are expected to be approximately 10⁷ PFU/g. Our physiologically based PK model of phage represents the first rigorous pre-clinical assessment of phage PK utilizing contemporary pharmacometric approaches amenable to both pre-clinical and clinical study design.