S.Duwal, J. Zisowsky, A. Yassen and C. Niederalt
Bayer AG, Pharmacometrics/Modeling & Simulation, Leverkusen/Wuppertal, Germany
Objectives: Target-mediated drug disposition (TMDD) models play an important role in elucidating non-linear pharmacokinetics (PK) of monoclonal antibodies (mAbs), incorporating complexities arising from processes such as non-specific clearance, target turnover, target binding and target-drug internalization giving rise to non-linear and dose-dependent PK of mAbs [1,2,3]. When mAb target soluble proteins the formation of circulating Ab-target complexes introduces a challenge in characterizing kinetics and dynamics of both free Ab and target molecules. In some cases, no assays are available to measure free drug and target, further complicating the interpretation of the PK. This becomes even more relevant, when elimination characteristics of drug-target complexes masks the PK of free Ab, a phenomenon that have not been thoroughly discussed in literature context of TMDD to the authors´ best knowledge. We developed a physiologically based pharmacokinetic (PBPK) model for a plasma soluble target to perform a simulation study to characterize and understand the drivers of non-linear PK and pharmacodynamics (PD) behaviors. Learnings from a mechanistic and bottom-up approach like PBPK were used to generate structural models and parameter ranges for PopPK/PD models which in turn provide mechanistic insights.
Methods: Modeling and simulation for PBPK model were performed with PK-Sim, MoBi and ospsuite-R package belonging to Open Systems Pharmacology available as freeware under the GPLv2 License [4] and the adult model was built according to Kuepfer et al. [5]. A large molecule model implemented in PK-Sim were utilized representing 1) two-pore formalism for drug extravasation from blood plasma to interstitial space, 2) lymph flow, 3) endosomal clearance and 4) protection from endosomal clearance by neonatal Fc receptor (FcRn) mediated recycling as especially relevant for Abs [6]. The PBPK model included binding to the plasma soluble target with single binding site. Target turnover (target synthesis and degradation) was also considered. We investigated two scenarios: 1) mAb can bind to a single target generating 1-to-1 complex and 2) mAb can bind to two targets in a two-step fashion generating 1-to-1 and 1-to-2 complexes. The target, Ab and complexes are allowed to distribute in plasma and interstitial spaces with different volume of distribution due to different molecular weights. The Ab and complexes are eliminated via endosomal clearance and protected from endosomal clearance by FcRn mediated recycling. Clearance differences of complexes in comparison to Ab were considered. The PBPK model were used to develop structural PopPK/PD models in Monolix [7] mapping total Ab and target concentration to observed PK and PD data and R shiny app was used to visualize the behavior.
Results: In the investigated case, the measured Ab plasma concentration are the sum of free Ab and Ab-target complexes. In both 1-to-1 and 1-to-2 binding scenarios, PK of complexes can mask the underlying TMDD of free Ab. Thus, discerning the extent and duration of saturation of target from PK of total Ab can be misleading. PK of total Ab displayed a non-linear and dose dependent behavior, although different from usual TMDD profiles. E.g. in contrast to non-linear and dose over-proportional increase in AUClast for free mAbs, the total Ab showed dose under-proportional increase. This trend can be further strengthened due to clearance differences of complexes to free Ab. Similarly, complexes also interfere with target measurements confounding PD readouts.
Conclusions: PK of mAbs for membrane bound target can indicate the duration and extend of target saturation, however, PK of total mAbs for plasma soluble targets can lead to misinterpretation due to the potential masking as discussed above. Dynamic interactions of various species in plasma can give rise to a strong non-linear behavior of PK and PD effects. Identification of various parameters of TMDD model for PopPK/PD model from clinical PK data alone are already challenging due to over-parameterization [8]. Furthermore, in scenario where masking is relevant, the structure and parameter identification become even more difficult. Building PopPK/PD requires including comprehensive clinical PK and PD data and the learnings from a mechanistic and bottom-up approach like PBPK can be used to generate structural models and parameter ranges for PopPK/PD model which in turn provide mechanistic insights.
References:
[1] Levy G Pharmacologic target-mediated drug disposition. Clin Pharmacol Ther. 1994;56:248–52.
[2] Mager DE, Jusko WJ. General pharmacokinetic model for drugs exhibiting target-mediated drug disposition. J Pharmaco Phamacod. 2001;28(6):507–32.
[3] Gibiansky L, Gibiansky E, Kakkar T, Ma P. Approximations of the target-mediated drug disposition model and identifiability of model parameters. J Pharmacokinet Pharmacodyn. 2008;35(5):573–91.
[4] www.open-systems-pharmacology.org
[5] Kuepfer L, Niederalt C, Wendl T, Schlender JF, Willmann S, Lippert J, Block M, Eissing T, Teutonico D. Applied concepts in PBPK modeling: how to build a PBPK/PD model. CPT: Pharmacometrics Sys Pharmacol. 2016;5(10):516–531.
[6] Niederalt C, Kuepfer L, Solodenko J, Eissing T, Siegmund HU, Block M, Willmann S, Lippert J. A generic whole body physiologically based pharmacokinetic model for therapeutic proteins in PK-Sim. Journal of pharmacokinetics and pharmacodynamics 2018;45:235–257.
[7] Monolix 2023R1, Lixoft SAS, a Simulations Plus company
[8] Gibiansky L, Gibiansky E. Target-mediated drug disposition model: approximations, identifiability of model parameters and applications to the population pharmacokinetic–pharmacodynamic modeling of biologics. Expert opinion on drug metabolism & toxicology 2009;5(7):803–812.
Reference: PAGE 32 (2024) Abstr 11104 [www.page-meeting.org/?abstract=11104]
Poster: Methodology - Other topics