PI-085 - POPULATION PHARMACOKINETIC MODEL OF TOTAL AND UNBOUND PAMIPARIB, A PARP INHIBITOR, IN GLIOBLASTOMA PATIENTS

C. Wickramasinghe¹, S. Kim¹, y. Jiang², J. Jiang², Y. Yue³, X. Bao⁴, N. Kaibara⁵, A. Hong⁵, N. Sanai⁶, J. Li²; ¹Karmanos Cancer Institute, Detroit, United States, ²Karmanos Cancer Institute, Detroit, Michigan, United States, ³Karmanos Cancer Institute, Detroit, MI, United States, ⁴Karmanos Cancer Institute, Detroit MI, United States, ⁵Barrow Neurological Institute, Phoenix AZ, United States, ⁶Barrow Neurological Institute, Phoenix, Arizona, United States.

Background: Pamiparib is an orally bioavailable, small molecule inhibitor of Poly(ADP-ribose) polymerase-1 (PARP1) and PARP2. It has been evaluated as a single agent or combination with temozolomide or radiation in patients with newly diagnosed or recurrent gliomas. This study was to develop a population pharmacokinetic (PK) model that characterized the plasma concentration time profiles of total and unbound pamiparib in glioblastoma patients and identified patient factors influencing the PK.

Methods: Total and unbound pamiparib plasma concentration data were obtained from 41 glioblastoma patients treated with pamiparib 60 mg twice daily for 5 days. Plasma samples were collected on day 5 following a single oral dose at pre-dose, 0.5, 1, 2, 4, 8, and 24 h. Total and unbound drug plasma concentrations were determined by LC-MS/MS. Population PK analysis was performed using Monolix (2024R1). Total and unbound drug plasma concentration data were simultaneously fitted to one or two compartment model using stochastic approximation expectation maximization algorithm. Population mean PK parameters, interindividual variability (IIV), and residual error were assessed in the model. The covariate model was developed by covariate screening using generalized-additive modeling followed by stepwise model building.

Results: Total and unbound pamiparib plasma concentration – time profiles were well described by a one-compartment model with a first-order absorption and elimination. Creatinine clearance was a significant covariate on the apparent volume of distribution (V/F) of both total and unbound drug, explaining ~20% of IIV of V/F. Age was a significant covariate on the apparent clearance (CL/F) of the total (but not unbound) drug, explaining ~ 8% of IIV of CL/F. Population estimates of the absorption rate constant (Ka), V/F, CL/F, and unbound fraction for the total drug were 2.18 h-1, 64 L, 4.16 L/h, and 0.047, respectively; and the respective estimates for unbound drug were 2.29 h-1, 1375 L, and 110 L/h.

Conclusion: This is the first report on the population PK model of pamiparib in glioblastoma patients. Total and unbound pamiparib plasma pharmacokinetics are characterized by a linear one-compartment model. Creatinine clearance and age are identified as statistically significant covariates, and their clinical relevance remains to be determined.