LB-015 - POPULATION PHARMACOKINETICS OF THE PARP INHIBITOR VELIPARIB IN PATIENTS WITH ADVANCED CANCER.

N. Yahya¹, Y. Niyonzima², F. Boakye-Agyeman³, J. Reid¹; ¹Mayo Clinic, Rochester, MN, USA, ²Certara, Raleigh, NC, United States, ³Certara, Raleigh, NC, USA.

Background: Veliparib (VPB) is a potent oral small molecule inhibitor of poly (ADP-ribose) polymerase (PARP) 1 and 2. More than 70% of the drug is cleared by renal excretion as the parent compound and remaining is metabolized by CYP2D6 enzyme. Previous population (Pop) pharmacokinetic (PK) studies of VPB were based on doses below 100 mg/day and linear PK. The objective was to characterize VPB PK over a wider dose range (10-400 mg) and to identify potential factors contributing to VPB PK variability

Methods: VPB PK was determined in a Phase I clinical trial in patients with solid tumors treated with oral VPB (10-400 mg qd or bid) on days 1- 3, 8-10, 15-17 given with weekly IV infusions of topotecan (2 - 4 mg/m2) on days 2, 9 and 16 every 28 days in patients with advanced ovarian cancer. The full PopPK analysis dataset included a total of 823 concentration-time data from 49 subjects for model development. Base model development included evaluation of 1-, 2-, and 3- compartments with linear first-order absorption and elimination and nonlinear saturation based on Michaelis-Menten kinetics. Following structural model selection, potential covariate screening was performed using the stepwise method. Models were evaluated using objective function value, goodness-of-fit (GOF) plots, visual predictive checks and non-parametric bootstrap analysis.

Results: A 2-comparment model with first-order absorption, a lag-time and nonlinear saturable elimination based on Michaelis-Menten kinetics adequately described the VPB PK in this study. The residual unexplained variability was accurately characterized by a multiplicative error model. The parameter estimates Km, Vmax, Vc, Vp, Q, Tlag, Ka were 2.96 mg/L, 63.2 mg/L, 11.8 L, 96.5 L, 9.60 L/h, 0.112 h, and 0.220 1/h respectively. The GOF plots showed that the model-predicted concentrations were generally in agreement with the observed concentrations.

Conclusion: The PopPK model incorporating saturable Michaelis-Menten elimination adequately described PK data of VPB, and results were consistent with previous studies demonstrating linear PK with 100 mg/day doses and less. Saturable elimination at high doses, resulting in reduced VPB clearance may lead to accumulation.

This work is supported in part by T32 DK07013 and P30 CA15083.
Keywords: Population Pharmacokinetics (PopPK), Nonlinear models, Oncology