Model-Based Discovery and Development of Biopharmaceuticals: A Case Study of Mavrilimumab.

Drug development is a lengthy, costly process with low probability of success. Biopharmaceuticals are highly specific molecules, with efficacy and safety closely tied to target biology and pharmacology. The "learning-predicting-confirming" continuum by translational and clinical modeling and simulation (M&S) was implemented at every decision point for mavrilimumab, a human monoclonal antibody in development for rheumatoid arthritis (RA). This tutorial uses mavrilimumab as an example to demonstrate rational discovery, preclinical development, clinical study design, and dose selection of biotherapeutics by M&S.

Population pharmacokinetic analysis from phase I and phase II studies of the humanized monovalent antibody, onartuzumab (MetMAb), in patients with advanced solid tumors.

Onartuzumab is a unique, humanized, monovalent (one-armed) monoclonal antibody (mAb) against the MET receptor. The intravenous (IV) pharmacokinetics (PK) of onartuzumab were investigated in a phase I study and a phase II study in recurrent non-small cell lung cancer (NSCLC) patients. The potential for drug-drug interaction (DDI) was assessed during co-administration of IV onartuzumab with oral erlotinib, by measuring the PK of both drugs. The concentration-time profiles of onartuzumab were adequately described using a two-compartment model with linear clearance (CL) at doses between 4 and 30 mg/kg. The estimates for CL, central compartment volume (V1 ), and median terminal half-life were 0.439 L/day, 2.77 L, and 13.4 days, respectively. Statistically significant covariates included creatinine clearance (CrCL) on clearance, weight and gender on V1 , and weight on peripheral compartment volume (V2 ), but the clinical relevance of these covariates needs to be further evaluated. The current analysis did not indicate obvious DDI between onartuzumab and erlotinib. MET diagnostic status did not impact the exposure of either agent. Despite the slightly faster clearance compared with typical bivalent mAbs, the PK of onartuzumab support dosing regimens of 15 mg/kg every 3 weeks or doses equivalent to achieve the target minimum tumoristatic concentration in patients.

Characterization of the long-term pharmacokinetics of bevacizumab following last dose in patients with resected stage II and III carcinoma of the colon.

PURPOSE: The study characterizes the long-term pharmacokinetics (PK) following last dose of bevacizumab as adjuvant therapy in patients with resected stage II and III colon carcinoma in a Phase III clinical study (AVF3077s). METHODS: Patients in AVF3077s received bevacizumab (5 mg/kg every 2 weeks) as adjuvant therapy for 1 year. Following the last dose bevacizumab concentration, data at 3 and 6 months were used to characterize long-term bevacizumab PK based on the population-modeling approach. RESULTS: The long-term bevacizumab PK were consistent with previously reported results based on short-term bevacizumab PK. The clearance (CL), central volume of distribution (V(1)), intercompartmental clearance (Q), and the peripheral volume of distribution (V(2)) were 214 mL/day, 2,830 mL, 636 mL/day, and 2,490 mL, which correspond to a disposition and elimination half-life of 1.33 and 19.1 days, respectively. The empirical Bayes estimates of median post-treatment bevacizumab drug levels at 3 and 6 months were 6.14 and 0.23 µg/mL, respectively. For test covariates, the change in CL and V(1) of bevacizumab was less than 20% of the typical value. Body weight is the important covariate explaining the inter-individual variability on CL and V(1). CONCLUSIONS: Long-term bevacizumab PK in this study was predictable based on short-term PK data from metastatic settings in other tumor types. An exploratory analysis demonstrated no apparent association of the tested covariates with bevacizumab PK. Further, the extended serum persistence of bevacizumab following last dose should be considered in clinical study designs and post-treatment evaluations that may be affected by bevacizumab.

Anti-neuropilin-1 (MNRP1685A): unexpected pharmacokinetic differences across species, from preclinical models to humans.

PURPOSE: To compare the pharmacokinetics (PK) of MNRP1685A, a human monoclonal antibody (mAb) against neuropilin-1 (NRP1), in mice, rats, monkeys, and cancer patients from a Phase I study to model with parallel linear and nonlinear clearances. METHODS: Binding characteristics of MNRP1685A in different species were evaluated using surface plasmon resonance technology. PK profiles of MNRP1685A after single and/or multiple doses in different species were analyzed using population analysis. PK parameters were compared across species. RESULTS: MNRP1685A binds to NRP1 in all four species tested. Consistent with the wide expression of NRP1, MNRP1685A demonstrated pronounced non-linear PK over a wide dose range. PK profiles are best described by a two-compartment model with parallel linear and nonlinear clearances. Model-derived PK parameters suggest similar in-vivo target expression levels and binding affinity to target across all species tested. However, compared to typical human/humanized mAbs, non-specific clearance of MNRP1685A was faster in mice, rats, and humans (60.3, 19.4, and 8.5 ml/day/kg), but not in monkeys (3.22 ml/day/kg). CONCLUSIONS: Monkey PK properly predicted the target-mediated clearance of MNRP1685A but underestimated its non-specific clearance in humans. This unique PK property warrants further investigation of underlying mechanisms.

Population pharmacokinetic and pharmacodynamic modeling of MNRP1685A in cynomolgus monkeys using two-target quasi-steady-state (QSS) model.

MNRP1685A (anti-NRP1) is a fully human IgG1 monoclonal antibody against neuropilin-1 (NRP1), a protein necessary for blood vessel maturation. MNRP1685A binds to free membrane-bound NRP1 (mNRP1) and circulating NRP1 (cNRP1). Total cNRP1 increased in a dose-dependent manner following anti-NRP1 administration in mice, rats, and monkeys. The purpose of this study is to develop a mechanism-based model to simultaneously describe the kinetics of both unbound drug (MNRP1685A) and total cNRP1 in cynomolgus monkeys. Pharmacokinetic (PK) and pharmacodynamic (PD) profiles after single- or multiple-dose administrations were well described by the two-target quasi-steady-state (QSS) model. The estimated nonspecific clearance was 3.26 mL/day/kg and central compartment volume was 38.2 mL/kg. The maximum elimination rate for mNRP1-mediated disposition was 98.8 nM/day. The synthesis rate (3.8 nM/day), degradation rate constant (1.53 day(-1)), and complex elimination rate constant (0.260 day(-1)) for cNRP1 were also derived from the model. QSS constants were 6.94 nM for mNRP1 and 2.8 nM for cNRP1. The results suggest that cNRP1 has minimal effect on MNRP1685A PK while mNRP1 plays a major role in the target-mediated drug disposition. This finding is favorable as the desired pharmacological target is mNRP1, rather than cNRP1. The two-target QSS model provides mechanistic understanding of the nonlinear PK of MNRP1685A. Based on the model prediction, the free drug concentrations to maintain free mNRP1 and cNRP1 below 10% of baseline level are 10 and 20 µg/mL, respectively. This serves as a target concentration for clinical dose selection, assuming cynomolgus monkeys are predictive for humans.

A guide to rational dosing of monoclonal antibodies.

BACKGROUND AND OBJECTIVE: Dosing of therapeutic monoclonal antibodies (mAbs) is often based on body size, with the perception that body size-based dosing would reduce inter-subject variability in drug exposure. However, most mAbs are target specific with a relatively large therapeutic window and generally a small contribution of body size to pharmacokinetic variability. Therefore, the dosing paradigm for mAbs should be assessed in the context of these unique characteristics. The objective of this study was to review the current dosing strategy and to provide a scientific rationale for dosing of mAbs using a modelling and simulation approach. METHODS: In this analysis, the body weight-based or body weight-independent (fixed) dosing regimens for mAbs were systematically evaluated. A generic two-compartment first-order elimination model was developed. Individual or population pharmacokinetic profiles were simulated as a function of the body weight effects on clearance (θ(BW_CL)) and on the central volume of distribution (θ(BW_V1)). The variability in exposure (the area under the serum concentration-time curve [AUC], trough serum concentration [C(min)] and peak serum concentration [C(max)]) was compared between body weight-based dosing and fixed dosing in the entire population. The deviation of exposure for light and heavy subjects from median body weight subjects was also measured. The simulation results were then evaluated with clinical pharmacokinetic characteristics of various mAbs that were given either by body weight-based dosing or by fixed dosing in the case study. RESULTS: Results from this analysis demonstrated that exposure variability was dependent on the magnitude of the body weight effect on pharmacokinetics. In contrast to the conventional assumption, body weight-based dosing does not always offer advantages over fixed dosing in reducing exposure variability. In general, when the exponential functions of θ(BW_CL) and θ(BW_V1) in the population pharmacokinetic model are <0.5, fixed dosing results in less variability and less deviation than body weight-based dosing; when both θ(BW_CL) and θ(BW_V1) are >0.5, body weight-based dosing results in less variability and less deviation than fixed dosing. In the scenarios when either θ(BW_CL) or θ(BW_V1) is >0.5, the impact on exposure variability is different for each exposure measure. The case study demonstrated that most mAbs had little effect or a moderate body weight effect (θ(BW_CL) and θ(BW_V1) <0.5 or ∼0.5). The difference of variability in exposure between body weight-based and fixed dosing is generally less than 20% and the percentages of deviation for light and heavy subpopulations are less than 40%. CONCLUSIONS: The analysis provided insights into the conditions under which either fixed or body weight-based dosing would be superior in reducing pharmacokinetic variability and exposure differences between light and heavy subjects across the population. The pharmacokinetic variability introduced by either dosing regimen is moderate relative to the variability generally observed in pharmacodynamics, efficacy and safety. Therefore, mAb dosing can be flexible. Given many practical advantages, fixed dosing is recommended to be the first option in first-in-human studies with mAbs. The dosing strategy in later stages of clinical development could then be determined based on combined knowledge of the body weight effect on pharmacokinetics, safety and efficacy from the early clinical trials.

Preclinical pharmacokinetics of MEHD7945A, a novel EGFR/HER3 dual-action antibody, and prediction of its human pharmacokinetics and efficacious clinical dose.

PURPOSE: MEHD7945A is a novel dual-action monoclonal antibody in which each of the two antigen-binding fragments is capable of binding to EGFR and HER3 with high affinity. It is being evaluated as a potential therapy for human cancer. The purpose of these studies was to characterize the pharmacokinetics (PK) of MEHD7945A in mouse and monkey and predict its human PK and efficacious dose. METHODS: PK of MEHD7945A was determined in SCID beige mice and cynomolgus monkeys after administration of single intravenous doses. Human PK profiles were projected from monkey PK profiles using a species-invariant time method, and human population PK parameters were estimated using a nonlinear, two-compartment model comprising specific (target-mediated) and nonspecific clearance pathways. The antitumor efficacy in mice bearing human tumor xenografts was used in conjunction with human PK projections to estimate human efficacious doses. RESULTS: The total clearance of MEHD7945A decreased with increase in dose in both mouse and monkey. The nonspecific clearance in monkey was estimated to be 14 mL/day/kg. The predicted nonspecific clearance range in humans was 6-10 mL/day/kg. Doses of 8-12 mg/kg administered every 2 weeks in humans were predicted to achieve exposure of 300 day µg/mL per week to match the efficacious exposure observed in xenograft models. CONCLUSIONS: The PK of MEHD7945A was nonlinear in mouse and monkey in the dose range tested. The nonspecific clearance in monkey was approximately twofold higher than typical humanized IgG1 antibodies. The projected human efficacious dose and dose regimen appear to be achievable in patients.

Preclinical pharmacokinetics of MFGR1877A, a human monoclonal antibody to FGFR3, and prediction of its efficacious clinical dose for the treatment of t(4;14)-positive multiple myeloma.

PURPOSE: MFGR1877A is a human IgG1 monoclonal antibody that binds to fibroblast growth factor receptor 3 (FGFR3) and is being investigated as a potential therapy for relapsed/refractory FGFR3+ multiple myeloma. The purpose of these studies was to characterize the pharmacokinetics (PK) of MFGR1877A in mouse, rat, and monkey and predict its human PK and efficacious dose. METHODS: PK of MFGR1877A was determined in athymic nude mice, Sprague-Dawley rats and cynomolgus monkeys after administration of single intravenous doses. Human PK profiles were projected from monkey PK profiles using a species-invariant time method, and human population PK parameters were estimated using a non-linear, two-compartment model comprising specific (target-mediated) and non-specific clearance pathways. The anti-tumor efficacy in mice bearing human tumor xenografts was used in conjunction with inhibitory activity in cell proliferation assays and human PK projections to estimate clinical efficacious dose. RESULTS: The PK of MFGR1877A in mice was non-linear in the dose range of 1-50 mg/kg, while in rats and monkeys, PK was non-linear in the dose range of 1-10 mg/kg and linear at doses ≥ 10 mg/kg. The predicted non-specific clearance range in humans was 2.6-4.4 mL/day/kg. Doses ranging from 2 to 3 mg/kg weekly to 6-10 mg/kg every 4 weeks were predicted to achieve the target exposure in ≥ 90% of multiple myeloma patients. CONCLUSIONS: The predicted non-specific clearance of MFGR1877A in humans is similar to typical human IgG1 antibodies and will be verified in a Phase 1 study. The projected human efficacious dose and regimen appear to be achievable in patients.