Danice Wilkins, PhD
Predictive Immunotoxicology: The Cytokine Release Assay and Beyond
Gauging the risk that molecules may have for triggering unanticipated or exaggerated immune effects
Predictive immunotoxicology (ITOX) describes a battery of in vitro assays that can be used to identify the potential of large or small molecule therapeutics to elicit immunosuppression, enhanced on-target effects or risks of unanticipated pharmacology or off-target toxicities.
Predictive ITOX assays can measure the response of a broad array of immune cell targets and employ a variety of assay platforms ranging from multiplex cytokine analysis and flow cytometry, to ligand binding assays and cell-based assays. The common thread linking most predictive ITOX assays is the use of human immune cells, with human whole blood or peripheral blood mononuclear cells (PBMC) being the most commonly used matrices.
The selection of which predictive ITOX assays should be applied to the development of a molecule is largely driven by the target, structure, and mechanism of action of the molecule. While one can advocate for the inclusion of these assays early in the drug development process (fail early, fail cheaply), they can and are often applied at various stages in the drug development process. At times these assessments are performed in different matrices to support the selection of relevant toxicology species for preclinical development, as well as later in the development process to investigate unanticipated findings observed during nonclinical or clinical studies.
The Cytokine Release Assay (CRA)
The most notable predictive ITOX assay is the cytokine release assay (CRA), which has traditionally been used to assess the potential risk of immune modulating drugs to trigger exaggerated levels of cytokine release in vivo. Cytokines are small chemical mediators that allow the cells of the immune system to communicate with each other. When cytokines are produced in an uncontrolled fashion, such as in response to a strong immune agonist (stimulator), they trigger a condition known as cytokine release syndrome or “cytokine storm” and can have a devastating impact on the body. Cytokine release syndrome has long been investigated as a response to certain classes of therapeutics such as CAR-T cell therapies and immune-activating antibodies; however, it has also recently been in the spotlight due to its role in the pathology of severe SARS-CoV-2 infections. Cytokine release syndrome can impact a variety of systems including the heart and blood vessels, lungs, liver, kidneys, brain and nervous system, and can present with symptoms that range from vomiting and fever, to hypotension, confusion, reduced lung function, and multi-system organ failure.
Regulatory guidance suggests that because cytokines can create such a wide variety of detrimental effects, their induction in response to therapeutics should be tested during the drug development process. Because each assay is tailored to the individual characteristics of the molecule being tested, cytokine release assays come in a variety of formats. The basic assays require the culture of human blood or PBMCs with multiple combinations of plate-bound or soluble test molecule, as well as relevant positive and negative control conditions. The duration of culture can be varied, as well as the number and type of cytokines analyzed.
While the cytokine release assay was originally designed to assess molecules with known immunomodulatory function, the CRA has also been applied to characterize risks related to other components of a molecule, such as the Fc region of monoclonal antibodies. Monoclonal antibodies that have active or augmented Fc functions or antibodies that are of an isotype that is known to engage Fc gamma receptors expressed on immune cells may also be candidates for assessment via CRA, even if the primary binding region of the antibodies do not engage an immune-related target.
Beyond the Cytokine Release Assay
For therapeutics that engage immune cells, many of the traditional pharmacology assays used to characterize the function and potency of these molecules may provide insights into potential immunotoxicities and can double as predictive ITOX assays.
Beyond traditional receptor occupancy assessments, additional assessment examining the upregulation or downregulation of costimulatory molecules, activation markers, or cytotoxicity can be included as part of traditional in vitro pharmacology assessments and provide helpful insights, particularly in the case of T cell engaging molecules. Additionally, assays such as antibody dependent cellular cytotoxicity (ADCC), antibody dependent cellular phagocytosis (ADCP), and complement dependent cytotoxicity (CDC) are often employed when characterizing molecules with known or augmented Fc function and can provide early insights into the potential risk for exaggerated pharmacology related to complement release and innate immune cell effector functions.
Predictive ITOX assays may also be used to determine the risk of a molecule to impact other blood components in addition to immune cells. Flow cytometry assays are commonly used to assess the potential of peptide and oligonucleotide molecules to elicit platelet activation and aggregation, and in the case of large molecule therapeutics, it is becoming increasingly common to assess certain classes of monoclonal antibodies for their potential to bind to red blood cells, platelets, and neutrophils. These assays may be performed early in the drug development
process as predictive ITOX assays based on a known or anticipated class effect, or later in the development process as investigative tools in case unanticipated changes in these populations occur during a nonclinical or clinical study.
The predictive value of ITOX assays is often limited beyond therapeutics that directly target and bind to immune cells or blood components. Indeed, many of the matrices used for predictive ITOX assays are blood-derived; however, an expanding number of assays are being developed that utilize cells differentiated from sources rich in hematopoietic stem cells such as human bone marrow or umbilical cord blood. Macrophages, dendritic cells, and mast cells are examples of tissue‑resident cell types that require extended cell culture and differentiation in vitro. Predictive ITOX applications for these cell types may include assessments to determine the risk of a therapeutic to elicit myelosuppression by adversely impacting dendritic cell or macrophage differentiation, polarization, activation, or antigen presentation, as well as the potential for triggering mast cell degranulation resulting in histamine and cytokine release.
A growing toolkit for drug developers
In short, predictive immunotoxicology is a growing field comprised of a variety of assays that can be used to gauge the risk that molecules may have for triggering unanticipated or exaggerated immune effects. One of the key predictive ITOX assays continues to be the cytokine release assay, a versatile platform initially designed to detect the potential of large molecule immune agonists to elicit exaggerated pharmacology in the form of cytokine production. While the CRA continues to be the workhorse of predictive ITOX assays, there are several new and creative assays under development that can be used to derisk a variety of therapeutic types and platforms. As new molecular entities and drug targets are developed, the field of predictive ITOX will continue to evolve beyond the CRA to identify and characterize new risks for immunity and immune-related toxicities.
Danice Wilkins, PhD, is a Scientific Director--Immunology and Biomarkers, at Charles River Laboratories.