Date of Award

Fall 11-2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Molecular Science and Nanotechnology

First Advisor

Eric A. Sherer

Abstract

Physiology-based pharmacokinetic models are mathematical models that characterize the behavior of a drug and have compartmental equations that are representative of specific tissues and physiological processes.[1, 2] Doxorubicin is an anthracycline antibiotic that is effective and widely used in anticancer therapy due to its potent cytotoxicity. Unfortunately, with that potency comes cardiotoxic side effects related to cumulative lifetime dose.[3] Specifically, the toxicity is related to the accumulation of the primary metabolite doxorubicinol (DOXol) in the heart.[4] Since the toxicity is organ-specific, the best way to characterize the behavior is through PBPK modeling.[2] Since PBPK models tend to be large systems of ODEs, several numerical methods were attempted for solving the model before a matrix-based approach was chosen.[5, 6] The eigenvalue/eigenvector solution was evaluated at three time points which were then included in a Composite Simpson’s Rule numerical integration for the length of some time interval.[5, 7] The PBPK model, adapted from a pig model, was fit to mouse data and scaled to predict rat, rabbit, dog, pig, and human data sets using an allometric scaling equation on the blood:plasma partition coefficient B : P .[8, 9, 10]

Despite extensive investigation into dose adjustments for DOX, no covariates were consistently found to improve the efficacy and minimize toxicity except dosing schedule – infusion rate and duration.[11] The criterion for decreasing incidence of cardiotoxicity was maintaining a sub-toxic Cmax,heart,DOXol in the heart while maximizing exposure, represented by area under the concentration-time-curve (AUC). Thus, the original mouse data set was ideal since it included both DOX venous blood concentration and DOXol heart concentration.[12] The model was optimized at 10 time points between 1 minute and 72 hours with the goal of (AUC) maximization without exceeding Cmax,heart,DOXol. Using these predictions, therapeutic drug monitoring could be executed by taking the plasma concentration samples during a patient’s first DOX dose, PBPK model predictions could provide AUC and Cmax,heart,DOXol data, which could then inform the infusion parameters for the next dose. Clinical thresholds for Cmax,vb have been established for incidence of adverse effects, and in future work, perhaps a similar threshold for cardiotoxicity could also be established using tissue-specific measures.

Download

Share

COinS