2016 - Lisboa - Portugal

PAGE 2016: Methodology - New Modelling Approaches
Martin Fink

Improving priors for human mAb linear PK parameters by using half-lives from pre-clinical studies

Martin Fink (1), Philip J Lowe (1), Vittal Shivva (2)

(1) Novartis Pharma AG, Basel, Switzerland; (2) University of Otago, Dunedin, New Zealand

Objectives: Obtaining a good prior for the linear PK part of new monoclonal antibodies (mAb) is essential for designing first-in-man (FIM) studies but also for fitting possible non-linear target-mediated disposition observed in these studies early-on.

Methods: Non-human primate (NHP) studies and FIM studies for six mAbs were fitted with 2cmt PK models, first separately, later together using a simple pool, a model with a 3rd hierarchical random effects ($LEVEL, NM7.3+) and a model including covariates for between mAb differences. Two other mAbs with slightly nonlinear PK were included for comparison.

Results: There was good agreement between compounds for the central volume (reflecting the rapidly accessible plasma volume Vp of 2.9L for a 70kg man), but the tissue volume (Vti) and clearance (CL) differed substantially – leading to terminal half-lives ranging from 15 to 28 days (plausibly due to differences in FcRn binding, charge distribution, glucolysation, etc. – see e.g., [1,2]). Inter-compartmental flow estimates were variable but mostly with poor precision.
One of the two nonlinear compounds showed, despite similar typical parameter values, greater inter-individual variability (IIV) in Vti (47 %CV) whereas the other showed much larger CL, Vp and Vti, perhaps due to rapid binding to readily available membrane-bound receptors.
The simple pool of human studies (similar to [3]) gave larger IIV estimates (CL 32 %CV, Vti 35 %CV) than the separate fits (CL 13-26 %CV, Vti 10-36 %CV); as inter-mAb-variability was being added to the IIV. The pool using a 3rd hierarchical random effect and adding drug specific covariates gave IIV estimates close to those of the separate fits (CL 23 %CV, Vti 23 %CV).
The between-mAb differences were predictable using allometric scaling and terminal half-life estimates from NHP, which showed less than 13% difference from their respective human estimates (consistent with [4]).

Conclusions: Ignoring inter-mAb variation leads to inflated estimates of IIV. However, by using just the terminal half-life estimates from NHP data one can account for between-mAb-differences in human and thus provide non-inflated priors for the linear PK of new mAbs.

[1] Suzuki et al., Importance of Neonatal FcR in Regulating the Serum Half-Life of Therapeutic Proteins Containing the Fc Domain of Human IgG1: A Comparative Study of the Affinity of Monoclonal Antibodies and Fc-Fusion Proteins to Human Neonatal FcR, J Immunol 2010.
[2] Datta-Mannan et al., The interplay of non-specific binding, target-mediated clearance and FcRn interactions on the pharmacokinetics of humanized antibodies, mAbs 2015.
[3] Davda et al., A model-based meta-analysis of monoclonal antibody pharmacokinetics to guide optimal first-in-human study design, mAbs 2014.
[4] Zhao et al., Across-Species Scaling of Monoclonal Antibody Pharmacokinetics Using a Minimal PBPK Model, Pharm Res 2015.

Reference: PAGE 25 (2016) Abstr 5736 [www.page-meeting.org/?abstract=5736]
Oral: Methodology - New Modelling Approaches
Click to open PDF poster/presentation (click to open)