Quantifying renal injury among the most commonly used clinical antibiotic combinations - Project Abstract: There is a fundamental need to improve pre-clinical models used to predict Acute Kidney
Injury (AKI) in the setting of polypharmacy (such as with nephrotoxic antibiotics). Until a translational model is
developed, risks can only be predicted individually and semi-quantitatively for groups. The long term goal is to
improve the kidney safety profile for critically necessary antibiotic therapies. Vancomycin (V) and piperacillin-
tazobactam (PT) are exemplars for study as they are the most commonly used antibiotics in the hospital setting
and have been widely suggested in clinical literature to synergistically cause AKI. Despite the clinical data, bio-
logic plausibility has not been established; it is unclear if elevated serum creatinine (SCr) is a real or a false
positive finding for AKI. SCr is neither highly sensitive nor specific for AKI, PT shares transporters with creatinine
that could affect clearance independent of AKI, and preliminary data from experimental models strongly suggests
that PT does not worsen V induced AKI (and may be protective). The overall objective is to leverage rat models
to quantify individual contributions of multiple drugs on AKI. The central hypothesis of this research is that SCr
from clinical studies misclassifies AKI, and alternate models more appropriately define AKI. Rigor of prior re-
search with urinary biomarkers suggests that non-creatinine biomarkers can improve predictions of AKI in the
setting of polypharmacy. Thus, we will create a novel framework to assess AKI in a pre-clinical manner that has
distinct advantages over the status quo by pursuing the following AIMs: 1. Define glomerular filtration rates
(GFR) for V, PT, and V+PT in rats as calculated by A) serum creatinine clearance (GFRCRCL) and B) inulin
(GFRinulin). 2. Create a Physiologic-Based Pharmacokinetic (PBPK) model that describes mass transit and
clearance of V, PT, and V+PT between blood, kidneys, and urine AND quantify real-time PK/Toxicodynamic (TD)
relationships for A) serum creatinine, B) urinary biomarkers, and C) terminal histopathology. 3. Develop a hu-
manized-rat, PK/TD model for simultaneous administration of two nephrotoxins. The proposed studies will ac-
complish 3 major goals. First, defining actual GFR in V, PT, and V+PT will provide mechanistic insight on the
role of SCr in V+PT therapy. Second, the PBPK model further informs mechanism by understanding the indi-
vidual components that increase SCr in V+PT. Finally, this project is innovative because there is not currently
a humanized animal model that can separate the AKI contribution from each drug in the setting of polypharmacy.
This model has potential benefits compared to allometrically scaled animal models (which result in high peak
concentrations for animals) or cellular models (which use static and supraphysiological concentrations). This
contribution will be significant because A) results from AIM 1 and 2 will clarify the toxicity profile for the two
most commonly used antibiotics and B) successful development of our novel humanized model can vertically
advance the field’s understanding of individual drug contribution to AKI in the setting of polypharmacy. This
proposal directly addresses the NIH mission to create ‘innovative research strategies’ for ‘improving health.’
