Frederic Lusoli1, Dr. Celine Pitou2, Prof. Leon Aarons1, Dr Kayode Ogungbenro1
1Centre for Applied Pharmacokinetic Research, University of Manchester, 2Global PK/PD & Pharmacometrics, Eli Lilly and Company
Introduction: Abemaciclib is an orally administered drug targeting the cyclin-dependent kinase 4 and 6 (CDK4/6), responsible for tumour growth. It showed a complex absorption, which has been described by a number of complex absorption models in the literature, with mixed results [1,2]. The aim of this analysis is therefore to further explore different models of absorption for abemaciclib, with the objective of providing a better description of the observed plasma concentration data at different doses, and to explore the impact of the complex absorption on overall exposure. Objectives: •Use a deconvolution method to transition from an individual to a population absorption profile. •Develop a PopPK model for the observed plasma concentration of abemaciclib, focusing on the complex absorption kinetics. Methods: The dataset contains 43 individuals from two phase 1 studies with different formulations; an absolute-bioavailability (n = 10, formulation 1) and a dose-escalation (n = 33, formulation 2) study. Each individual received orally a single dose of abemaciclib ranging from 50 to 275mg. For the absolute-bioavailability study, abemaciclib was orally administered at time 0 followed by an intravenous injection of a ¹³C8 tracer 6 hours after the oral dose in the same individual. Different models were tested and developed in Monolix [3]: 1-, 2- or 3- compartment disposition model, normal or lognormal distribution of the error model. The best model was selected based on the parameter estimates, the goodness-of-fits and the visual predictive check (VPC). A deconvolution method [4] was used to describe the individual input rate and percentage of dose absorbed at a time t of the absolute bioavailability study. A finite absorption time [5] of 50 hours was implemented, based on the assumption that 95% of the bioavailable-dose was absorbed by that time in every individual. A Weibull function was used to describe the population input rate for absorption, disposition was described by standard compartmental model and the parameters in the PopPK model were estimated. Results: The PopPK model was able to describe the observed plasma concentration data accurately with the implementation of a Weibull function. This was linked to a 3-compartment model for disposition. The parameters related to the Weibull function were split to characterise separately the absorption of the two formulations in the dataset. Therefore, for the formulation 1 (absolute-bioavailability study), the parameters ? (analogue to the absorption rate constant), s (shape parameter), F (Bioavailability) were estimated to be respectively 0.0071 h-s (% R.S.E = 21.1, BSV (CV %) = 93.28), 2.43 (% R.S.E = 3.83, BSV (CV %) = 16.45) and 0.34 (% R.S.E = 6.28, BSV (CV %) = 28.56). On the opposite, Weibull parameters for the formulation 2 were: ? = 0.036 h-s (% R.S.E = 18.4, BSV (CV %) = 157.55), s = 2.58 (% R.S.E = 4.84, BSV (CV %) = 31.07) and F = 0.47 (% R.S.E = 8.02, BSV (CV %) = 41.08). Other parameters of the model estimated were; volume of distribution of central compartment 145.91 L (% R.S.E = 23.4, BSV (CV %) = 245.59), volume of distribution of first peripheral compartment 184.73 L (% R.S.E = 12.7, BSV (CV %) = 40.81), volume of distribution of second peripheral compartment 298.63 L (% R.S.E = 10.7, BSV (CV %) = 41.03), first inter-compartmental clearance 57.08 L/h (% R.S.E = 37.7, BSV (CV %) = 605.48), second inter-compartmental clearance 122.45 L/h (% R.S.E = 25.4, BSV (CV %) = 188.33) and clearance 20.45 L/h (% R.S.E = 6.10, BSV (CV %) = 25.5). Conclusion: The PopPK model adequately described the observed plasma concentration of abemaciclib following a single dose across different dosage regimens. The model will be expanded with active metabolites (M2/M20), which are mostly metabolised by CYP3A4.
[1] Chigusta et al. CPT:PSP (2020) 9, 523-533. [2] Posada et al. J Clin Pharmacol. (2020) 60, 915-930. [3] Monolix 2023R1, Lixoft SAS, a Simulations Plus company [4] Veng-Pedersen P., J Pharmacokinet Biopharm. (1980) 8, 463-481. [5] Macheras, P., Tsekouras, A.A. J Pharmacokinet Pharmacodyn (2023) 50, 5–10
Reference: PAGE 33 (2025) Abstr 11551 [www.page-meeting.org/?abstract=11551]
Poster: Drug/Disease Modelling - Absorption & PBPK