Experimental and emerging therapeutics modeling

Recent advances in medicine have led to modern therapeutics, such as immunotherapies, cell and gene therapies and RNA-based therapies that challenge the traditional pharmacological paradigm of a drug molecule binding to a specific target to effect its pharmacological activity. Quantitative modeling is an essential tool to understand the unconventional dose-activity relationships for these emerging and experimental therapies. Models that are informed by a clear understanding of disease biology and the mechanism of therapeutic action can simulate different dosing scenarios and efficacy metrics. Such simulations can help teams formulate effective study designs, interpret experimental data, and clearly communicate early results to executives and investors.

Gene therapies

Gene therapies using autologous stem cell transplantation have showed promising results for the treatment of inherited genetic blood disorders such as sickle cell disease (SCD) and beta thalassemias, that were until recently considered incurable. These therapies seek to replace a patient’s endogenous stem cells with edited versions containing a corrective gene. Engraftment of the edited stem cells into the bone marrow produces healthy red blood cells that mitigate effects of the disease. However, these therapies remain complex and expensive, and often require months to years before reaching maximal efficacy. Thus a model-informed approach is needed to understand and evaluate proposed therapies in these areas. In a published study, we developed a model of gene therapy for SCD informed by the known biology of erythropoiesis and SCD pathology. This model was validated with published clinical data, and can be used to simulate the effect of varying stem cell dose, degree of bone marrow preconditioning and the impact of vector copy number on long term outcome of the gene therapy.

Immunotherapies

Immunotherapies harness cells of the immune system to fight disease. Molecular immunotherapies use target receptors to activate specific immune cell types, such as effector T cells that kill cancer cells. Cell-based immunotherapies include chimeric antigen receptor (CAR) T cells and CAR-NK cells. These modified versions of immune cells express an engineered surface receptor, the CAR, that binds to a specific tumor antigen. CAR-expressing cells are expanded ex vivo and injected into a patient where they home into the tumor to target cancer cells. A modeling-driven approach is essential for development of immune therapies to understand and properly interpret their nonlinear dose response, and to account for secondary effects such as delayed expansion of targeted cells and elevated cytokine levels. Training a model on preclinical data can be used to predict the anticipated clinical response and degree of variability, and to inform clinical study design and patient selection criteria.

Bioconjugate therapies

Bioconjugate therapies refer to several emerging classes of therapeutics, such as antibody-drug conjugates (ADC) and radio-ligand therapies (RLT). These molecules consist of an antibody or peptide - that binds to a specific target receptor - conjugated via a biochemical linker to a cytotoxic payload, such as the radioactive isotope. Despite their molecular similarities, ADCs and RLTs require fundamentally different strategies to maximize efficacy and minimize systemics toxicity. An ADC benefits from a long residence time in the body which maximizes uptake into the target tissue, whereas with an RLT the goal is to minimize systemic exposure to ionizing radiation while maximizing the uptake into tumor. Therefore, a careful consideration of target expression levels on tumor versus healthy tissue, target turnover kinetics and the effect of soluble receptors is necessary to evaluate candidate molecules. A modeling approach can be used as an effective hypothesis generation and validation tool, and to optimize target selection and dosing strategy for bioconjugate therapies.