A simulation platform to predict dose and therapeutic window of immunocytokines

点击翻译按钮获取中文摘要

Project Summary/Abstract Immunocytokines (ICs) are fusion proteins of an engineered cytokine conjugated to an antibody. These novel molecules are the next generation of cytokine-based immunotherapies with potential applications in a diverse range of diseases such as rheumatoid arthritis (RA), psoriasis, and cancer. ICs are designed to selectively target diseased tissue or specific immune cells with minimal systemic immune activation that typically leads to dose-limiting toxicity in recombinant cytokine therapy. However, it is challenging to design a molecule with high target specificity, predict its pharmacokinetics and identify doses that achieve high efficacy but low toxicity, i.e. the therapeutic window. We are proposing to develop a simulation platform for IC screening that will computationally predict dose and therapeutic window of novel ICs under development. The platform will implement a quantitative systems pharmacology (QSP) model that mechanistically describes the binding of an IC to target and off-target cells and links cytokine receptor occupancy to cellular activation and expansion dynamics. The model will predict in vivo pharmacokinetics (PK) and pharmacodynamics (PD) for an input dose and dosing regimen of a proposed IC. Simulations will report readouts such as cell counts and soluble cytokine levels that are clinically observable biomarkers of efficacy and toxicity. The model will be general enough to simulate pro- and anti-inflammatory ICs. A modular design will allow us to add new cell types and cytokines/receptors as needed to adequately model the crosstalk between the inflammatory and regulatory arms of the immune response. In Phase I of this Fast Track proposal, we will demonstrate the technical feasibility of developing a single mechanistic QSP model structure that captures drug dose- dependent expansion and contraction of four unique IC molecules. By fitting preclinical and clinical data for each molecule, we will establish a robust translational strategy for human dose prediction. In Phase II the platform model will be integrated with and made accessible through Applied BioMath’s Assess™ browser-based interface. With this setup, users can interactively explore the IC design space and use simulations to understand the impact of varying dose, dosing interval, target affinities, cytokine potency and drug half-life on clinical PK/PD. We expect that this interactive tool will foster effective communication within multidisciplinary drug development teams, and help them rationally identify optimal molecular characteristics and dosing strategies for novel ICs. The platform will also allow virtual patient cohort simulations to guide selection criteria for clinical trials. There are currently no effective tools to screen candidate molecules in the IC space. Our proposed computational platform to predict the optimal dose and therapeutic window of novel ICs will accelerate the lengthy and expensive lead candidate selection process, and thus lower the cost of IC development, facilitate clinical trial design, reduce late stage attrition and bring new drugs to the market faster to benefit patient healthcare.