(T-076) Population Pharmacokinetic-Pharmacodynamic Modeling for Triapine to Optimize Dosing Regimen

Sarah Taylor, MD, PhD – Assistant Professor, University of Pittsburgh School of Medicine; Aman Chauhan, MD – Assistant Professor, University of Miami; S. Percy Ivy, MD – Program Director, National Cancer Institute; Jogarao Gobburu, PhD, MBA – Professor, University of Maryland School of Pharmacy; Jan Beumer, PharmD, PhD, DAB – Professor, UPMC Hillman Cancer Center; Allison Dunn, PharmD – Research Assistant Professor, University of Maryland School of Pharmacy

Objectives: Triapine is a ribonucleotide reductase inhibitor used in combination with radiation and/or chemotherapy for the treatment of various types of advanced stage solid cancers. While the clinical activity of intravenous (IV) triapine appears promising in Phase I and Phase 2 studies, oral dosing may be a more feasible option, and oral triapine in Phase I studies showed a comparable safety profile to that of IV triapine. The objective of this study was to develop a population pharmacokinetic-pharmacodynamic (PK/PD) model to characterize the disposition of triapine, the effect of smoking on exposure, and the association of methemoglobin levels with exposure for triapine.

Methods: A total of 36 patients with advanced stage cervical or neuroendocrine cancers from NCI9892 and NCI10388, two phase 1 studies (NCT02595879, NCT04234568) were included for population PK/PD model building. Triapine plasma concentrations and methemoglobin levels were sampled over a 24-hour period after the 1st, 6th, or 9th dose. The model qualification step was implemented using a visual predictive check. Using the parameters of the final model, plasma concentrations were simulated and the 5th, median, and 95th quantiles were calculated. Simulations will be performed to explore once versus twice daily dosing regimen for triapine to predict the percentage of patients above the methemoglobin level considered to be unsafe/symptomatic (~10%). Data were analyzed by a nonlinear mixed-effects modeling approach using Pumas software (www.pumas.ai).

Results: A two compartment PK model with two-compartment Erlang transit absorption and first-order elimination adequately described the concentration time-profile of triapine. The final PK model incorporated covariates of weight and smoking status in relation to their effect on both clearance and volume of distribution. The mean (%CV) estimates of apparent clearance and central volume of distribution were 55.15 L/hr (28.4%) and 59.37 L (26.9%), respectively. In smoking patients, a 37% increase in clearance was observed. An effect compartment PD model adequately described the methemoglobin levels.

Conclusions: Differences in PK behavior and methemoglobin levels with triapine exposure have been identified among smoking and non-smoking patients. The population PK/PD model can be effectively leveraged to inform optimal dosing regimen for triapine.

Citations: N/A