Development of a mechanistic based model for neonatal Fc receptor recycling to design human serum albumin mutants with extended half-lives
K. Bergmann (1), T. van Steeg (1), J. Smeets (1), C. Chaudhury (2) and B. Agoram (2)
(1) LAP&P Consultants, Leiden, Netherlands, (2) MedImmune, Cambridge UK and Gaithersburg MD, USA.
Objectives: Similar to immunoglobulins, increasing the pH dependent affinity of binding to neonatal Fc receptor (FcRn) is a possible strategy for engineering human serum albumin (HSA) mutants with longer half-life than natural HSA [1]. The purpose of this study was to develop a mechanism-based model to help in the identification of HSA mutants with appropriate in vivo half-life and to evaluate the possible translation of half-life between animal models and human for HSA mutants.
Methods: The final model contained distribution of HSA, degradation of free HSA in the endosome and recycling of HSA bound to the FcRn receptors. The model further included competition between endogenous and exogenous albumin and explicitly accounted for their different affinities to FcRn. The mechanism-based model was originally validated against literature albumin profiles of wild-type albumin from mouse, monkey and human. Plasma concentration profiles from HSA wild-type and three mutants following intravenous administration to huFcRn transgenic mice were compared to model predictions. Due to the model complexity compared to the simplicity of the observed PK profiles, it was neither possible nor desirable to fit all model parameters to data. For mouse and human all but two model parameters could be obtained from literature, and the unknown parameters were derived from steady state conditions. Monkey parameters were scaled from human.
Results: For human and monkey the model predictions were in good agreement with literature-reported data, and the parameters were found to be mutually scalable. Distribution parameters were fitted to the HSA profile of huFcRn transgenic mouse PK profile, in order to obtain a good prediction of the mouse PK. The model was applied to predict the half-life of HSA mutants as a function of FcRn affinity. Half-life changes caused by improved affinity were found to be similar for monkey and human but up to 2 orders of magnitude smaller for mouse. Simulations were performed which showed that increased affinity has a small effect on the terminal half-life for HSA in the huFcRn transgenic mouse due to a high affinity of endogenous MSA for the huFcRn receptor.
Conclusions: The mechanism-based model provides a convenient tool to help identify HSA mutants with optimal in vivo PK. Furthermore, the model suggests that in huFcRn transgenic mice a large change in FcRn affinity results in a relatively small change in HSA half-life compared to monkey and human.
References:
[1] Chaudhury et al (2003) J. Exp. Med, 197 (3), 315-322.
