FcRn-IgG Interaction: Development of FcRn Inhibitors and PBPK Modeling
MetadataShow full item record
The neonatal Fc receptor (FcRn) is known to play an essential role in extending IgG systemic persistence and in maintaining IgG homeostasis by protecting IgG from lysosomal catabolism. Inhibiting FcRn-IgG interaction with anti-FcRn monoclonal antibodies (mAbs) may have potential for treating antibody mediated autoimmune diseases. In addition, there has been growing interest in enhancing FcRn-IgG interaction as a means of improving the pharmacokinetics of therapeutic IgG antibodies. This dissertation has pursued the development of new anti-human FcRn mAbs, investigated the pharmacokinetics of anti-FcRn mAbs in rats, and has led to the development of new mathematical models to characterize and predict the effects of FcRn binding on the plasma and tissue disposition of monoclonal antibodies. We have studied the pharmacokinetics and tissue distribution of two anti-FcRn mAbs, 1G3 (anti-rat FcRn heavy chain) and 4C9 (anti-rat β 2 microglobulin), in rats at different dose levels. Both antibodies exhibited rapid clearance from plasma compared to normal murine IgGs. Based on the assessment from non-compartmental analyses, the clearance and distribution volume of each mAb decreased with increasing dose. CL and V ss values for 4C9 were larger than the values for 1G3 at each dose level. Pharmacokinetic data of 1G3 and 4C9 were fitted to target-mediated drug disposition (TMDD) models. The estimated total target receptor concentration (R tot ) for 4C9 was 5412 nM, which was 3-fold of the value estimated from 1G3 data (1660 nM). The tissue distribution of the antibodies was assessed by whole body autoradiography and by individual counting of excised tissues. The two anti-FcRn mAbs displayed similar patterns of tissue distribution, but 4C9 showed much higher absolute values of the tissue: blood concentration ratios (0.15∼5.16 vs. 0.03∼0.91 for 1G3). In comparison to 4C9, 1G3 demonstrated slower clearance and less tissue distribution, which suggests that anti-FcRn heavy chain antibodies may be more specific for FcRn binding and inhibition compared to anti-FcRn light chain mAbs. As such, mAbs directed against human FcRn (hFcRn) heavy chain were developed using the hybridoma technology. Hybridoma m1d5 was identified through screening using surface plasmon resonance (SPR) assays. Binding of mAb m1d5 to human FcRn and to human FcRn peptide, which was used for immunization, was characterized using SPR assays. The estimated binding affinity of m1d5 for the hFcRn peptide was 1.18 nM, and binding affinity for recombinant human FcRn was 58.7 nM. Pre-incubation of antibody with the hFcRn peptide in solution led to complete inhibition of m1d5 binding to human FcRn, demonstrating that the antibody binds specifically to the peptide epitopes. m1d5 may have utility for studying human FcRn, and may have utility as a hFcRn antagonist in vivo, as a potential therapy for autoimmune diseases. To study the effect of FcRn binding on IgG pharmacokinetics, a new catenary PBPK model of IgG disposition has been developed. The PBPK model was able to well characterize the pharmacokinetics of a murine IgG (7E3) in both wild type and FcRn deficient mice. When compared to predictions made by prior PBPK models, which assume equilibrium binding of FcRn and IgG within the endosomes of vascular endothelial cells, the catenary PBPK model provides improved predictions of the effects of altered FcRn binding on mAb disposition. To facilitate the evaluation of the mathematical model, a new chimeric anti-CEA mAb (cT84.66) was developed, and point mutations were performed to create variants with altered FcRn binding (N434A: increased FcRn binding compared to the wild type; I253A: decreased FcRn binding compared to the wild type). The catenary PBPK model was successfully scaled up and applied to characterize the pharmacokinetic data of mAbs in monkeys. Lastly, the model was employed to identify, by simulation, optimal FcRn binding properties to enhance the systemic persistence of mAb. The simulation results suggest that optimal increases in mAb persistence may be achieved by combining increased rates of association with slow rates of dissociation throughout the pH range of 7.0 – 6.0. Those results may provide guidance for antibody engineering for pharmacokinetics improvement of therapeutic mAbs.