Charlotte Bon (1), Thomas Hofer (2), Mark R. Davies (3), Ben-Fillippo Krippendorff (1)
(1) Roche Pharmaceutical Research and Early Development, Roche Innovation Center Basel, Switzerland, (2) Roche Pharmaceutical Research and Early Development, Roche Innovation Center Zurich, Switzerland, (3) QT-Informatics Limited, Macclesfield, UK
Objectives:
Efficient delivery of therapeutic molecules into the targeted tissues and cells remains a limiting factor for antibody based therapies and the wider use of new therapeutic technologies such as nucleic acid therapeutics and CRISPR mediated gene editing [1, 2]. The abundant cell surface asialoglycoprotein receptor (ASGPR) is a highly selective receptor found on hepatocytes with a high potential for being exploited as a selective shuttle for delivery. Various nucleic acid therapeutics that bind to this receptor are already in clinical development. Despite its high capacity, this receptor mediated delivery mechanism can be saturated resulting in a reduced selectivity for the liver and therefore increase the likelihood for systemic adverse effects. Therefore, it is important to optimize both molecular properties and the administration protocol. To investigate the in-vivo drug delivery capacity of the ASGPR we here aim to exploit the Target Mediated Drug Disposition (TMDD) of a newly developed anti-ASGPR antibody (ASGPR Ab) in mice.
Methods:
The developed anti-ASGPR antibody was generated and first characterized in-vitro. Pharmacokinetic and biodistribution studies were performed in mice and analyzed by modeling. The ASGPR antibody was administered via intravenous administration or subcutaneous dosing at doses ranging from 1 to 30 mg/kg. A mechanistically-based mathematical modeling approach was used for PK data analysis. The model was based on a generalized pharmacokinetic model for drugs exhibiting target-mediated drug disposition (TMDD) as originally described by Mager and Jusko in 2001 [3]. Parameters have been estimated using a non-linear mixed-effect approach in MONOLIX 4.3.3. A biodistribution study was performed using 111-indium radiolabeled ASGPR antibodies to assess the accuracy of the TMDD model in predicting the liver uptake. It is of note however that the liver has been found to be a major elimination organ for monoclonal antibodies. Therefore, a radiolabeled non-targeting antibody (IL17 Ab) was used for comparison. To be successful as a targeted therapy, the protocol of administration should be selected such as to maximize internalization into hepatocytes via the ASGPR route and minimize saturation that would lead to off-target distribution and unspecific clearance. Therefore, simulations were performed to show the influence of receptor saturation and frequency of administration on delivery efficiency (percentage of the dose distributed to the liver) and on the delivered quantity.
Results:
After IV and SC dosing, the PK profiles show non-linear concentration time profiles that are consistent with TMDD. Using pharmacokinetic data and an in-silico TMDD model we estimate an ASGPR expression level of 1.8 million molecules per hepatocyte in mice. The half-life of the degradation of the receptor was found to be equal to 15 hours and the formed ligand-receptor complex is internalized with a half-life of 5 days. The biodistribution study confirmed the specific uptake in liver in comparison with a non-targeting antibody. The kinetics of the ASGPR shows that a saturation of the shuttle at therapeutic concentrations is possible. Indeed, it was shown that at 1mg/kg 96% of the total dose is predicted to be cleared via the target pathway while at 30 mg/kg this percentage drops to 76% due to receptor saturation. In addition, the rate at which free receptors become available again is critical to optimize a protocol of administration in the case of multiple dosing regimens. Therefore, simulations were performed to investigate how quickly the ASGPR system recovers from a dose challenge. The free receptors return to 90% of the baseline level in 3 days, while at 30 mg/kg it takes more than a week. Simulations were then performed to show the influence of receptor saturation, varying the administrated dose, and frequency of administration on delivery efficiency (percentage of the dose distributed to the liver) and on the delivered quantity. This optimization can also be extended to different modalities, to incorporate the different pharmacokinetic properties (non ASGPR related). For example, we examined what would be the impact of a faster clearance rate on the delivery efficiency.
Conclusions:
We proposed using the developed TMDD to support the development of therapies that use the ASGPR as a shuttle into hepatocytes.
References:
[1] Liu, J. and S.L. Shui, Delivery methods for site-specific nucleases: Achieving the full potential of therapeutic gene editing. J Control Release, 2016. 244(Pt A): p. 83-97.
[2] Juliano, R.L., The delivery of therapeutic oligonucleotides. Nucleic Acids Res, 2016.
[3] Mager, D.E. and W.J. Jusko, General pharmacokinetic model for drugs exhibiting target-mediated drug disposition. J Pharmacokinet Pharmacodyn, 2001. 28(6): p. 507-32.
Reference: PAGE 27 (2018) Abstr 8643 [www.page-meeting.org/?abstract=8643]
Poster: Drug/Disease Modelling - Other Topics