Use of quantitative pharmacology in the development of HAE1, a high-affinity anti-IgE monoclonal antibody

Abstract

HAE1, a high-affinity anti-IgE monoclonal antibody, is discussed here as a case study in the use of quantitative pharmacology in the development of a second-generation molecule. In vitro, preclinical, and clinical data from the first-generation molecule, omalizumab, were heavily leveraged in the HAE1 program. A preliminary mechanism-based pharmacokinetic/pharmacodynamic (PK/PD) model for HAE1 was developed using an existing model for omalizumab, together with in vitro binding data for HAE1 and omalizumab. When phase I data were available, the model was refined by simultaneously modeling PK/PD data from omalizumab studies with the available HAE1 phase I data. The HAE1 clinical program was based on knowledge of the quantitative relationship between a pharmacodynamic biomarker, suppression of free IgE, and clinical response (e.g., lower exacerbation rates) obtained in pivotal studies with omalizumab. A clinical trial simulation platform was developed to predict free IgE levels and clinical responses following attainment of a target free IgE level (</=10 IU/ml). The simulation platform enabled selection of four doses for the phase II dose-ranging trial by two independent methods: dose-response non-linear fitting and linear mixed modeling. Agreement between the two methods provided confidence in the doses selected. Modeling and simulation played a large role in supporting acceleration of the HAE1 program by enabling data-driven decision-making, often based on confirmation of projections and/or learning from incoming new data.

Fig. 1

Inhibition of histamine release from ragweed-specific IgE-loaded RBL-48 cells by HAE1 and omalizumab

Fig. 2

Structure of mechanism-based PK/PD model for an anti-IgE monoclonal antibody

Fig. 3

HAE1, total IgE, and free IgE concentration-time profiles for five HAE1 subjects who received a single subcutaneous dose of 7.5, 30, 90, 180 or 360 mg

Fig. 4

Simulated probability of HAE1-treated subjects achieving target free IgE level of ≤10 IU/ml as a function of dose

Fig. 5

Simulated probability of attaining clinical response (Δ total symptom score) as a function of dose for three cases: an optimistic response rate (70%), a response rate similar to that expected for omalizumab (64%), and a low response rate (50%)