(T-055) Population PK model on preclinical data for optimization of further clinical trial designs in oncology

Flavie Audy, NA – Specialist, Pharmacometrics & Data Sciences, Cencora Pharmalex; Elisabeth Rouits, NA – Senior Director, Global Head Clinical Pharmacology & Pharmacometrics, Cencora Pharmalex

Objectives: The purpose of the following analysis is to support the design of a Phase I study in cancer patients, based on preclinical pharmacokinetics (PK) and pharmacodynamics (PD) data. The overall pre-clinical and non-clinical PK/PD information available was summarized in order to either optimize the dosing regimen and evaluate the biomarker profile of clinical relevance.

Methods: Single-dose and multiple dose PK studies were conducted after oral administration of drug X in mice, rats, beagle dogs, and cynomolgus monkeys. In addition to plasma PK sampling, three PD biomarkers were collected in one monkey study. Clinical data from a first-in-human (FIH) clinical trial were available. PK and PD samples were available after single and multiple dose administration in cancer patients.
Based on all available preclinical PK data, a population PK analysis was conducted via nonlinear mixed effects modelling using NONMEM, V7.5. The First-Order Conditional Estimation with Interaction (FOCEI) method was used for the estimation process. Assessment of model adequacy and decisions about increasing model complexity were driven by the data and guided by goodness-of-fit criteria. Allometric scaling approach was used to predict the concentration-time profiles in human and to verify the adequacy of the approach with the available preliminary human data. Finally, simulations of PKPD and dosing regimen scenarios were considered to support further clinical trial designs.

Results: 1028 quantifiable concentrations from a total of 198 animals were used in the PopPK analysis. PK of drug X was adequately described by a 2-compartment model with linear elimination and first-order absorption. The estimated PK parameters (CV%) were absorption rate (ka) of 1.25 h-1 (69.4%), apparent central clearance (CL/F) of 0.707 L/h (72.7%), apparent central volume of distribution (Vc/F) of 0.498 L (32.7%), apparent peripheral clearance (Q/F) of 0.760 L/h (55.9%), apparent peripheral volume of distribution (Vp/F) of 4.07 L (89.9%). Inter-individual variability was estimated on all parameters. Body weight was included on absorption rate, clearances and volumes. Allometric scaling factors were estimated to be -0.223, 0.665 and 0.911 for ka, CL and V, respectively. This model allows to predict the mean (±SD) concentration-time profiles in human and to check adequation with available PK exposure from FIH study.

Conclusions: Thanks to the allometric scaling approach, the popPK model is able to characterize the PK of drug X by integrating the knowledge gathered from several preclinical studies in 4 different species. Adequate simulations of dosing regimens in further clinical trials can be supported by this model, by considering the uncertainty, in order to consider the optimization of PKPD relationship in human.

Citations: [1] Project Optimus, FDA
[2] Papachristos, A.; Patel, J.; Vasileiou, M.; Patrinos, G.P. Dose Optimization in Oncology Drug Development: The Emerging Role of Pharmacogenomics, Pharmacokinetics, and Pharmacodynamics. Cancers 2023, 15, 3233
[3] Jolling K., Perez-Ruixo J., Hemeryck A., Vermeulen A., Greway T., Mixed-effects modelling of the interspecies pharmacokinetic scaling of pegylated human erythropoietin, European Journal of Pharmaceutical Sciences, Volume 24, Issue 5, 2005