Physiologically Based Modeling of the Pharmacokinetics of "Catch-and-Release" Anti-Carcinoembryonic Antigen Monoclonal Antibodies in Colorectal Cancer Xenograft Mouse Models.

Engineered monoclonal antibodies (mAbs) with pH-sensitive target release, or "catch-and-release" (CAR) binding, have shown promise in decreasing the extent of target-mediated mAb elimination, increasing mAb exposure, and increasing dose potency. This study developed a mechanistic physiologically based pharmacokinetic (PBPK) model to evaluate the effects of pH-sensitive CAR target binding on the disposition of anti-carcinoembryonic antigen (CEA) mAbs in mouse models of colorectal cancer. The PBPK model was qualified by comparing model-predicted plasma concentration-time data with data observed in tumor-bearing mice following the administration of T84.66, a "standard" anti-CEA mAb that demonstrates strong binding at pH 7.4 and 5.5. Further simulations evaluated the effects CAR pH-dependent binding, with decreasing CEA affinity with decreasing pH, on anti-CEA mAb plasma pharmacokinetics. Simulated data were compared with data observed for a novel CAR mAb, 10H6. The PBPK model provided precise parameter estimates, and excellent data characterization (median prediction error 18.4%) following fitting to T84.66 data. Simulations well predicted 10H6 data (median prediction error 21.4%). Sensitivity analyses demonstrated that key determinants of the disposition of CAR mAbs include the following: antigen binding affinity, the rate constant of mAb-CEA dissociation in acidified endosomes, antigen concentration, and the tumor vasculature reflection coefficient.

"Catch-and-Release" Anti-Carcinoembryonic Antigen Monoclonal Antibody Leads to Greater Plasma and Tumor Exposure in a Mouse Model of Colorectal Cancer.

In this study, we examined the effects of target expression, neonatal Fc receptor (FcRn) expression in tumors, and pH-dependent target binding on the disposition of monoclonal antibodies (mAbs) in murine models of colorectal cancer. A panel of anti-carcinoembryonic antigen (CEA) mAbs was developed via standard hybridoma technology and then evaluated for pH-dependent CEA binding. Binding was assessed via immunoassay and radioligand binding assays. 10H6, a murine IgG1 mAb with high affinity for CEA at pH = 7.4 (KD = 12.6 ± 1.7 nM) and reduced affinity at pH = 6.0 (KD = 144.6 ± 21.8 nM), and T84.66, which exhibits pH-independent CEA binding (KD = 1.1 ± 0.11 and 1.4 ± 0.16 nM at pH 7.4 and 6.0), were selected for pharmacokinetic investigations. We evaluated pharmacokinetics after intravenous administration to control mice and to mice bearing tumors with (MC38CEA+, LS174T) and without (MC38CEA-) CEA expression and with or without expression of murine FcRn, at doses of 0.1, 1, and 10 mg/kg. 10H6 displayed linear pharmacokinetics in mice bearing MC38CEA+ or MC38CEA- tumors. T84.66 displayed linear pharmacokinetics in mice with MC38CEA- tumors but dose-dependent nonlinear pharmacokinetics in mice bearing MC38CEA+ In addition to the improved plasma pharmacokinetic profile (i.e., linear pharmacokinetics, longer terminal half-life), 10H6 exhibited improved exposure in MC38CEA+ tumors relative to T84.66. In mice bearing tumors with CEA expression, but lacking expression of murine FcRn (LS174T), 10H6 demonstrated nonlinear pharmacokinetics, with rapid clearance at low dose. These data are consistent with the hypothesis that pH-dependent CEA binding allows mAb dissociation from target in acidified endosomes, enabling FcRn-mediated protection from target-mediated elimination in mice bearing MC38CEA+ tumors.

Development and validation of an enzyme-linked immunosorbent assay for the quantification of gelonin in mouse plasma.

This article details the development and validation of an enzyme-linked immunosorbent assay (ELISA) for the quantification of gelonin in mouse plasma. The ELISA was validated for intra- and inter-day variability and for accuracy over a standard curve range of 7.5-100 ng/mL. The assay was then applied to assess gelonin pharmacokinetics in mice. Results from the ELISA were compared to data obtained from a parallel study conducted with (125)Iodine-labeled gelonin, with quantification via gamma counting. The ELISA demonstrated good precision, as the percent coefficient of variation of quality control samples in intra-day and inter-day validation ranged from 5.4-9.3% and 2.9-7.3%, respectively. Sample recoveries ranged from 98.3-105% of nominal values. The ELISA method yielded lower plasma concentrations of gelonin than found from the less-specific gamma counting method. Consequently, pharmacokinetic analyses yielded significantly higher estimates for volume of distribution (106 ± 31 vs. 55.8 ± 13 mL/kg) and plasma clearance (34.7 ± 6.6 vs. 10.9 ± 2.1 mL/min/kg) for data determined by ELISA vs. by gamma counting.

Investigation of the influence of nephropathy on monoclonal antibody disposition: a pharmacokinetic study in a mouse model of diabetic nephropathy.

PURPOSE: This study employed a mouse model to evaluate the effects of diabetic nephropathy on the pharmacokinetics of 8C2, a murine monoclonal antibody (mAb). METHODS: Streptozotocin (STZ) was administered to mice to induce diabetic nephropathy (125 mg/kg/day × 2). Mice were grouped (n = 8-10) based on time after STZ-treatment (control, 1, 2, 3, 4, or 6 weeks), and injected intravenously with 10 mg/kg 8C2. Blood samples were collected up to 7 days, and 8C2 plasma concentrations were determined via immunoassay. Inulin clearance and urinary albumin excretion rate (UAE) were determined to assess renal function. RESULTS: UAE, inulin clearance, and 8C2 clearance increased significantly following STZ. Comparing control and 6 week STZ-treatment groups, UAE and inulin clearance increased from 25.7 ± 3.3 to 99.3 ± 13.7 µg/day, and from 421 ± 31 to 584 ± 78 µl/min. 8C2 clearance increased from 121 ± 12.5 to 228 ± 61 µl/hr/kg (p < 0.01). 8C2 clearance was highly correlated with UAE (r(2): 0.731). Inclusion of UAE as a covariate in population modeling explained significant residual variability in 8C2 clearance. CONCLUSIONS: The clearance of 8C2 increased significantly in STZ-treated mice. Population pharmacokinetic modeling suggests that UAE has potential for use in predicting mAb clearance in subjects with diabetic nephropathy, possibly assisting in the individualization of mAb dosing.

Stabilization of a pH-sensitive apoptosis-linked coiled coil through single point mutations.

The apoptosis-associated Par-4 protein has been implicated in cancers of the prostate, colon, and kidney, and in Alzheimer's and Huntington's diseases, among other neurodegenerative disorders. Previously, we have shown that a peptide from the Par-4 C-terminus, which is responsible for Par-4 self-association as well as interaction with all currently identified effector molecules, is natively unfolded at neutral pH, but forms a tightly associated coiled coil at acidic pH and low temperature. Here, we have alternately mutated the two acidic residues predicted to participate in repulsive electrostatic interactions at the coiled coil interhelical interface. Analysis of circular dichroism spectra reveals that a dramatic alteration of the folding/unfolding equilibrium of this peptide can be effected through directed-point mutagenesis, confirming that the two acidic residues are indeed key to the pH-dependent folding behavior of the Par-4 coiled coil, and further suggesting that alleviation of charge repulsion through exposure to either a low pH microenvironment or an electrostatically complementary environment may be necessary for efficient folding of the Par-4 C-terminus.