Bach Tran Nguyen 1,2, Aurélie Barrail-Tran 3,4, Moreno Ursino 5, Nicolas Tournier 6, Emmanuelle Comets 1,2
1 Univ Rennes, Inserm, EHESP, Irset (Institut de recherche en santé, environnement et travail) - UMR_S 1085 (Rennes, France), 2 Université Paris Cité and Université Sorbonne Paris Nord, Inserm, IAME (Paris, France), 3 AP-HP. Université Paris Saclay, Hôpital Bicêtre, Pharmacie (Le Kremlin-Bicêtre, France), 4 UMR 1184, Center for Immunology of Viral, Auto-immune, Hematological and Bacterial diseases, INSERM, Université Paris Saclay, CEA (Paris, France), 5 Inserm, Université Paris Cité, Inria, HeKA, F-75015 (Paris, France), 6 Université Paris-Saclay, Inserm, CNRS, CEA, Laboratoire d’Imagerie Biomédicale Multimodale Paris-Saclay (BioMaps) (Orsay, France)
Introduction: First-in-human (FiH) studies require a range of doses extrapolated from pharmacokinetic (PK) animal studies combined with safety assessments. Subtherapeutic doses of radiolabelled drugs can be administered in preclinical and early clinical development to gain a dynamic understanding of PK [1]. In the present study, we investigated their potential to build Pharmacologically-Based PK (PBPK) models through whole body PET imaging data, and how they can be used for human extrapolation to provide prior information for FiH studies.
Aims: to develop a PET-informed PBPK workflow for preclinical-to-human extrapolation, with dolutegravir (DTG) as a case study.
Methods: We developed a structured workflow integrating micro- and therapeutic-dose data for extrapolation of PK across species and dose ranges. The full workflow revolves around PBPK models and extrapolation from micro- to therapeutic doses, to examine non-linearity across dose ranges [2], and across animals to human, to take into account interspecies differences.
We applied this approach to dolutegravir, a widely used HIV integrase inhibitor, for which we collected data in non-human primates (NHP), first in 4 animals after a radiolabelled microdose of DTG and then in 6 animals given a therapeutic dose. We first built a whole-body PBPK model for NHP in PK-Sim/MoBi (v12.1) [3] incorporating DTG physicochemical properties, protein binding, UGT1A1-mediated metabolism, and P-gp/BCRP efflux. We then estimated a first set of parameters θPET using PET-data collected over 3h in the 4 animals receiving a micro-dose: we used the data from 5 key organs, converted from radioactivity to concentrations, as well as the fraction of parent DTG over total DTG and its main metabolites in arterial blood, measured by LCMSMS. To determine the parameters to be estimated, we first performed sensitivity analyses to assess their identifiability and determine which parameters could be assumed to vary across subjects. We used the algorithms from PKSim to compute the sensitivity functions and to estimate parametersvia the Monte-Carlo algorithm . Starting from θNHP,PET, we refined the estimates using DTG concentrations measured up to 24h after dose in fluids and tissues following a 2.5 mg/kg IV injection in 6 additional NHPs. A sensitivity analysis was again performed to identify the parameters which could be estimated, and we fixed the remaining parameters to the geometric means from the first set of parameters θNHP,PET. This resulted in a second set of parameters θNHP,ther. The two sets of parameters θNHP,PET and θNHP,ther were extrapolated to humans using the built-in extrapolation formulas in PKSim. Finally, 100 virtual healthy adults (50% male, 20–80 years) receiving 50 mg qd po after high-fat meals were simulated using these two sets of extrapolated parameters over the distributions of physiological parameters in Caucasian subjects provided by PKSim. We predicted steady-state DTG concentrations and compared the range of simulated concentrations and AUC to reported phase I PK profiles in the literature.
Results: DTG data were well described in NHPs over the PET acquisition window. Parameter adjustment to fit the therapeutic-dosing data was able to handle the changes in dose range. While parameters extrapolated from microdose estimates systematically underpredicted exposure, combining NHP micro and macrodose data predicted steady-state geometric mean AUC0-24 and Cmax closely matched observed profiles in humans, although Ctrough was slightly underpredicted. The resulting profile range therefore provided an adequate prior for PK in humans that could be used to select an appropriate dose range for dose-finding methods [4].
Conclusions: Combining micro- and therapeutic doses in NHP through PET-PBPK modelling successfully predicted DTG concentrations in healthy volunteers. Whole-body imaging also allowed access to tissue distribution. This proof of concept study supports early acquisition of PET data to build robust priors for FIH studies,. We aim to complete the framework by acquiring PET data in human after microdose administration to strengthen the prediction of human PK and using more advanced estimation approaches [5].
Acknowledgments: This work has been supported by the SMATCH project (ANR-22-PESN-0003) funded under the Programmes et Equipements Prioritaires de Recherche – Santé Numérique of the Stratégie d’Accélération France 2030.
References:
[1] Burt T, Young G, Lee W, et al. Phase 0/microdosing approaches: time for mainstream application in drug development? Nat Rev Drug Discov 2020; 19:801–818.
[2] Schou M, Varnäs K, Lundquist S, et al. Large Variation in Brain Exposure of Reference CNS Drugs: a PET Study in Nonhuman Primates. Int J Neuropsychopharmacol 2015; 18:pyv036.
[3] Willmann S, Lippert J, Sevestre M, Solodenko J, Fois F, Schmitt W. PK-Sim®: A physiologically based pharmacokinetic ‘whole-body’ model. Biosilico. 1. 121-124. 2003.
[4] O’Quigley, J., and Shen, L. Z. (1996), Continual Reassessment Method: A Likelihood Approach, Biometrics, 52, 673–684.
[5] Teutonico D, Marchionni D, Lavielle M, Nguyen L, Integrating Population Approaches With Physiologically Based Pharmacokinetic Models: A Novel Framework for Parameter Estimation, CPT: Pharmacometrics & Systems Pharmacology 15, no. 1 (2026): e70186
Reference: PAGE 34 (2026) Abstr 12146 [www.page-meeting.org/?abstract=12146]
Poster: Methodology - New Modelling Approaches