Stephan Schaller (1), Klaus Lindauer (2), Oliver Plettenburg (3), Thomas Klabunde (2), and Thomas Eissing (1)
(1) Bayer Technology Services GmbH, Computational Systems Biology, Leverkusen, Germany, (2) Sanofi-Aventis Deutschland GmbH, R&D LGCR/Struct.,Design & Informatics FF, Frankfurt, Germany, (3) Sanofi-Aventis Deutschland GmbH, R&D Diabetes Devision, Frankfurt, Germany
Objectives: Realistic in-silico models of the glucose metabolism are valuable tools in diabetes research, drug development and development of treatment strategies. Existing models [1, 2] provide a good basis, but do not yet provide detail at molecular and organ level to drive fundamental research in diabetes as well as to address specific questions in drug development. We here present a framework to close this gap and specifically applied it to investigate organ-selectivity and binding properties of insulin-derivatives.
Methods: We developed a coupled physiologically-based pharmacokinetic and pharmacodynamic (PBPK/PD) model of glucose metabolism [3], which provides mechanistic detail on multiple scales from whole-body level down to tissue specific insulin receptor dynamics and subcutaneous insulin absorption. The model was parameterized using literature, e.g. [4], and in-house data and has been used within clinical glucose-control trials. The variability in insulin action on a molecular level was analyzed in-silico w.r.t. molecular (i.e. receptor binding) properties of insulin analogs and organ selectivity and quantified to assess the hypoglycemia risk profile.
Results: Using this model framework the PK and PD profile of new “virtual” short-acting insulin analogues with different insulin receptor binding kinetics have been evaluated. Following a subcutaneous injection of an insulin analogue, the time profile of blood insulin plasma levels, the phosphorylation of the insulin receptor, peripheral glucose uptake, the net hepatic glucose exchange and the resulting effect on blood plasma glucose levels have been simulated and analyzed. The simulation studies provide detailed and mechanistic insights how changes of the receptor binding kinetics affect the PK and PD profile and support to shape the target product profile of new short-acting insulin analogues.
Conclusions: The modeling framework allows generating virtual diabetic populations or individualized models to support pharma diabetes R&D in defining a target product profile and in assisting lead optimization of new insulin analogues. The simulation studies in virtual patients provided novel insight, highlighted optimization potential, and supported decision making during the drug development of new insulin analogues. Overall, the diabetes PBPK/PD model provides a powerful tool to the pharmaceutical industry as decision support during the evaluation and development of novel diabetes targets, drugs, and treatment strategies on the basis of virtual diabetes populations.
References:
[1] Wilinska, M.E. and R. Hovorka, Simulation models for in silico testing of closed-loop glucose controllers in type 1 diabetes. Drug Discovery Today: Disease Models, 2008. 5(4): p. 289-298.
[2] Kovatchev, B.P., et al., In Silico Preclinical Trials: A Proof of Concept in Closed-Loop Control of Type 1 Diabetes. J Diabetes Sci Technol, 2009. 3(1): p. 44-55.
[3] Schaller, S., et al., A generic integrated physiologically-based whole-body model of the glucose-insulin-glucagon regulatory system. CPT: PSP, 2013. 2:e65.
[4] El-Khatib, F.H., et al., A bihormonal closed-loop artificial pancreas for type 1 diabetes. Sci Transl Med, 2010. 2(27): p. 27ra27.
Reference: PAGE 23 (2014) Abstr 3196 [www.page-meeting.org/?abstract=3196]
Poster: Drug/Disease modeling - Endocrine