In order to optimize therapy, a full understanding of the pharmacokinetics of any systemic therapy is desired. We routinely model the pharmacokinetic (PK) data of agents being tested for antitumor activity and correlate that with activity and/or toxicity (pharmacodynamics modeling). We utilize compartmental and noncompartmental approaches to define the disposition of agents. Analysis of PK data (using concentration measurements provided by sample analysis using validated assays) allows for assessment of drug disposition, including the absorption, distribution, metabolism and excretion of a drug. Modeling this data, essentially describing these physiological processes as a mathematical equation, allows for optimization of drug administration (including dose and frequency of dosing,) in silico. Over the years, we have conducted population pharmacokinetic modeling of the following compounds: depsipeptide, romidepsin, sorafenib, olaparib, docetaxel in combination with the p-glycoprotein antagonist tariquidar, TRC105, and TRC102. Recent efforts have focused on building a population PK model to understand the disposition kinetics of mithramycin in the body to best optimize dose. We also performed population PK (PPK) modeling and simulation of belinostat, a second-generation histone deacetylase inhibitor (HDI) predominantly metabolized by UGT1A1-mediated glucuronidation. Two common polymorphisms (UGT1A1*28 and UGT1A1*60) were previously associated with impaired drug clearance and thrombocytopenia risk, likely from increased drug exposure. We conducted a PPK model to include a pharmacodynamic (PD) model describing the change in platelet levels in patients with cancer administered belinostat as a 48-h continuous intravenous infusion, along with cisplatin and etoposide. Several covariates were explored, including sex, body weight, UGT1A1 genotype status, liver, and kidney function, but none significantly improved the model. Platelet levels rebounded to baseline within 21 days, before the next cycle of therapy. Simulations predicted that higher belinostat drug exposure does cause lower thrombocyte nadirs compared to lower belinostat levels. However, platelet levels rebound by the start of the next belinostat cycle. This model suggests a q3week schedule allows for sufficient platelet recovery before the next belinostat infusion is optimal. We were also involved in developing a quantitative mathematical modeling of the dynamics and intracellular trafficking of far-red light-activatable prodrugs to describe the implications in stimuli-responsive drug delivery system.
Showing the most recent 10 out of 27 publications
