David Janzén1, Erika Cavallin1, Xueqing Li1, Anders Björkbom1, Carina Ämmälä2, Shalini Andersson1, Rasmus Jansson-Löfmark1
1. DMPK, Cardiovascular, Renal and Metabolic diseases, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden 2. Bioscience Diabetes, Cardiovascular, Renal and Metabolic diseases, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden
Introduction:
Targeted drug delivery to a desired cell type is an emerging science which is conceptually straightforward but mechanistically complex. Tissue targeting can largely be divided into 1) chemical modifications of the drug 2) using a targeting ligand that is linked to the cargo (i.e. conjugation) 3) adapting the drug formulation or 4) changing the physical shape of the formulation.
Tissue targeting for oligonucleotides to hepatocytes by conjugation of drugs to the tirantennary N-acetyl galactosamine (GN3) enhances the uptake into hepatocytes through internalization of the oligonucleotide by the cell-surface lectin receptor ASGR, improving apparent potency of the ASO by 10-60-fold [1,2]. Currently there are several oligonucleotides in clinical trials also demonstrating the advantage of this targeting approach for hepatocyte targets. Tissue targeting to other cell types is less well established and challenging [3]. Recent data reported by Ämmälä et al showed thatan anti-sense oligonucleotide (ASO) conjugated to an engineered GLP1 peptide resulted in a >75% knockdown of the target gene in mouse beta cells in pancreas [4,5] at low doses. These data were of particular interest since an untargeted ASO will not reach these cell types. Although these data were encouraging, little information is known about the pharmacokinetics of the ASO conjugated to a GLP1 agonist.
Objectives:
- Assess the plasma and pancreas pharmacokinetics of the GLP1-linker-ASO conjugate and its components after single dose administration to mice
- Build a population pharmacokinetic plasma-pancreas model of the GLP1-linker-ASO conjugate to further guide chemical optimization of GLP1-linker-ASO conjugates
Methods:
The in vivo study was performed in two parts. In the first part C57B6 mice (n=63) were divided up into three dose groups. Group 1 (n=15) were dosed with the GLP1 targeting ligand alone, group 2 (n=24) was dosed with the targeting conjugate and group 3 (n=24) was dosed with the ASO alone. All animals received subcutaneous administration at equimolar dosing close to the expected maximum tolerable dose of the conjugate. Based on prior knowledge the sampling time points for group 1 were 10 min, 30 min, 1, 4, and 8h. For group 2 and 3 the sampling points were 0.5 h, 1, 2, 4, 8, 24, 72, and 120 h. In the second part of the study only the targeting conjugate was dosed to C57B5 mice (n=13) but at two lower doses in order to asses nonlinearity in PK. The samplings points here for this part were 4, 8 and 24 h. All sampling were terminal readouts with three animals per time point in the first part and 2 or 3 animals per time point in the second part. At each readout plasma, kidney, liver and pancreas were collected for bioanalysis of free targeting ligand, total targeting ligand, free ASO and the sum of ASO still connected to the linker and ASO in intact conjugate form.
Since all sampling were only terminal readouts and no individual time-profiles were generated the data were in this sense sparse. However, because a substantial number of animals were used a population modelling approach was ideal to handle this sparsity. The model considered the pharmacokinetics of GLP1-linker-ASO conjugate, its formation and clearance of the GLP1 agonist, the ASO with the linker and the unconjugated ASO both in plasma and pancreas for a total of five different observation compartments.
Results:
The model could adequately describe the pharmacokinetics of the GLP1-linker-ASO conjugate, its formation and clearance of the GLP1 agonist, the ASO with the linker and the unconjugated ASO both in plasma and pancreas. In the current studied dose range, pharmacokinetics was linear for all components in all matrixes.
The degradation rate of the intact GLP1-ASO targeting conjugate and the GLP1 agonist was rapid with a half-life of less than one hour in plasma. In pancreas, the intact GLP1-ASO targeting conjugate and the GLP1 agonist was also rapidly degraded with a half-life of less than one hour whereas the formed unconjugated ASO had a pancreas tissue half-life of approximately 70 hrs.
Conclusion:
The current presented pharmacokinetic model characterizes the pharmacokinetics of the GLP1-linker-ASO after single dose administration to mice. The model can be used as a platform to further gain insights in characteristics for a successful targeting conjugate to beta-cells and to further guide chemical optimization of GLP1-linker-ASO conjugates.
References:
[1]. T.P. Prakash et al Nucleic Acids Res 42, 8796-8807 (2014))
[2]. Nair et al 2014, J Am Chem Soc 136 (49), 16958-16961)
[3]. Dowdy SF., Nat Biotechnol. 2017 Mar; 35(3):222-229.
[4]. Ämmälä et al., American Diabetes Association Scientific Session 77, San Diego, June 9-13, 2017, http://www.abstractsonline.com/pp8/#!/4297/presentation/20558
[5]. Ämmälä et al., American Diabetes Association Scientific Session 77, San Diego, June 9-13, 2017, http://www.abstractsonline.com/pp8/#!/4297/presentation/20560
Reference: PAGE 27 (2018) Abstr 8585 [www.page-meeting.org/?abstract=8585]
Poster: Drug/Disease Modelling - Other Topics