Regulatory Forum Opinion Piece*

Abstract

The immune system has been recognized for decades as a potential “target organ” of toxicity. Immune system activation can result in cytokine release resulting in severe systemic toxicity. Immunosuppression can result in impaired host defense and an increase in opportunistic infection, reemergence of latent infection, poor responses to vaccination, or increased risk of certain cancers. Several regulatory documents have addressed various aspects of immunotoxicity assessments. Nonhuman primates (NHPs) and in particular macaques are often the only relevant species for biotechnology-derived investigational new drugs based on cross-reactivity with human and NHP targets. This article reviews the challenges and opportunities associated with monitoring immune function in NHPs in the context of regulatory expectations. The article emphasizes how a comprehensive assessment of immunotoxicity remains a challenge due to interanimal variability associated with certain parameters (e.g., T-dependent antibody response)and it identifies gaps, such as the stage of development of certain assays (e.g., cytotoxic T-cell function). Despite these challenges, a thorough assessment of target biology-driven theoretical risks, in combination with proper integration of all information from the standard toxicology studies, and the refinement of certain assays should enable proper risk assessment. To this effect, emphasis should be placed on leveraging predictive in vitro assays using human cells.

The immune system is a complex, highly regulated network encompassing multiple compartments, cell types, and mediators, and has been recognized for decades as a potential “target organ” of toxicity (Vos 1977). Immune system activation can result in a range of effects, including the potential for cascading (uncompensated) cytokine release resulting in severe systemic toxicity (i.e., cytokine storm). Immunosuppression or immunodepletion can result in impaired host defense and an increase in opportunistic infection, reemergence of latent infection, poor responses to vaccination, or increased risk of certain cancers. For some drug candidates, there may be elements of both immune system activation and immunosuppression (e.g., antibodies depleting immune cells and associated with infusion reactions). While some immunotoxicities that arise in humans are poorly predicted in animal models (e.g., allergic reactions), multiple endpoints can be considered to evaluate the risks associated with immunomodulation in nonclinical studies.

Several regulatory documents have addressed various aspects of immunotoxicity assessments. ICH S6 (Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals 1997) indicated that an important aspect of the nonclinical evaluation for biotechnology-derived pharmaceuticals is the assessment of potential immunogenicity issues to aid in the interpretation of toxicity studies. ICH S6 acknowledged that induction of antibody in animals is not predictive of a potential for antibody formation in humans (this point was further emphasized in ICH S6(R1) 2011). However, this guidance document emphasized that many biologicals are targeting the immune system and suggested that strategies to assess potential effects on the immune system should be developed. ICH S6 proposed strategies tailored to each molecule, and it advised that routine tiered testing approaches or standard testing batteries were not recommended. The Committee for Medicinal Products for Human Use (CHMP) of EMA issued a note for guidance on repeated dose toxicity studies, with a fairly prescriptive appendix dedicated to immunotoxicity, stating that signs of immunotoxicity in animal studies may trigger the incorporation of immunotoxicological endpoints into the safety monitoring of clinical trials for conventional medicinal products (EMEA 2000). This note for guidance proposed conducting nonclinical immunotoxicology testing using a tiered approach consisting of an initial screening phase followed, if necessary, by extended studies. The FDA (CDER) published guidance on the immunotoxicology evaluation of new investigational drugs, which described the type of assays that can be used to assess immunosuppression, immunogenicity, hypersensitivity, autoimmunity, and adverse immunostimulation (FDA 2002). The FDA guidance was not intended to apply to biologics and proposed an approach similar to what was later proposed in ICH S8: a case-by-case tiered approach. ICH S8 (Immunotoxicity Studies for Human Pharmaceuticals) provides (1) recommendations on nonclinical testing approaches to identify new chemical entities with potential immunotoxic liabilities and (2) guidance on a weight-of-evidence decision-making approach (ICH 2005). ICH S8 does not address hypersensitivity and autoimmunity issues specifically, but focuses on unintended immunosuppression or enhancement. The EMEA published in 2007 a Guideline on strategies to identify and mitigate risks for first-in-human (FIH) clinical trials with investigational medicinal products (EMEA 2007). The EMA guideline cited molecules targeted toward the immune system, notably CD3 or CD28 superagonists, as molecules requiring special attention and proposed that a risk evaluation should be conducted based on the mechanism of action of the investigational drug, its target, and the relevance of the animal species available. The guidance document states that in addition to standard core battery safety pharmacology studies, additional studies need to be considered on a case-by-case basis, and that when targeting the immune system, potential unintended effects should be investigated, for instance using in vitro studies, including studies with human cell systems.

Distinct clinical concerns exist for immunosuppressive and immunostimulating agents. The nonclinical assessment should be tailored to address these specific concerns. The specific endpoints and assays used in a given study need to be tailored to evaluate the nature of the immunomodulation of concern in that species.

Why Focus on NHP and Immunotoxicology?

Despite a constant focus on reducing the use of NHPs, macaques are often the only relevant species for biotechnology-derived investigational new drugs based on cross-reactivity of these molecules with human and NHP target proteins. Monitoring immune function in NHPs has been and still is the subject of extensive research, thus enabling the development of reagents and assays applicable to immunotoxicology assessments.

Similar to what has been developed for studies in rodents, the evaluation of potential for immunosuppression in NHPs should include standard toxicology parameters (e.g., clinical observations, hematology, macroscopic observations, spleen and thymus weight, histopathology of the bone marrow, spleen, thymus, lymph nodes, tonsils, and gut-associated lymphoid tissues; STP Immunotoxicology Working Group 2005). Based upon cause for concern, these standard endpoints can be combined with immunophenotyping of peripheral blood cells and evaluation of the functional response of the immune system (e.g., T-dependent antibody response [TDAR], natural killer [NK] cell activity, cytotoxic T-cell [CTL] activity).

Immunophenotyping of peripheral blood cells in NHPs is readily available for a variety of populations of lymphocytes where a standard panel can include total T lymphocytes (CD45⁺/CD3⁺), helper T lymphocytes (CD45⁺/CD3⁺/CD4⁺), cytotoxic T lymphocytes (CD45⁺/CD3⁺/CD8⁺), B lymphocytes (CD45⁺/CD20⁺), and NK lymphocytes (CD45⁺/CD3⁻/CD16⁺ or CD45⁺CD3-CD159a⁺). Because some populations reflect only a fraction of total lymphocytes, conducting immunophenotyping routinely for immunomodulatory compounds may be the only way to detect significant changes. It has been the case in our experience during the development of a monoclonal antibody directed against a target found to be critical for NK homeostasis in certain species, including NHPs (ongoing publication). Decreases in NK counts could not be anticipated simply by monitoring standard hematology endpoints. Immunophenotyping peripheral blood lymphocytes for immunomodulatory compounds at an early stage of development is important. Populations typically not evaluated unless warranted to measure a specific pharmacodynamic response may include regulatory T helper lymphocytes (CD3⁺/CD4⁺/CD25^hi/FoxP3⁺), activated T helper lymphocytes (CD45⁺/CD4⁺/CD25⁺), activated cytotoxic T lymphocytes (CD45⁺/CD8⁺/CD25⁺), effector memory T helper lymphocytes (CD3⁺/CD4⁺/CD95⁺/CD28⁻), effector memory cytotoxic T lymphocytes (CD3⁺/CD8⁺/CD95⁺/CD28⁻), central memory T helper lymphocytes (CD3⁺/CD4⁺/CD95⁺/CD28⁺), central memory cytotoxic T lymphocytes (CD3⁺/CD8⁺/CD95⁺/CD28⁺), naïve T-helper lymphocytes (CD3⁺/CD4⁺/CD95⁻/CD28⁺), and naïve cytotoxic T lymphocytes (CD3⁺/CD8⁺/CD95⁻/CD28⁺). The selection of such assessments should be driven by the biology of the drug target and is typically not driven by safety considerations. Listings of reagents cross-reacting with NHP CDs are available at http://nhpreagents.bidmc.harvard.edu/NHP/literature.aspx.

The TDAR assay is referred to in several regulatory documents as a functional assay of choice as it encompasses multiple aspects of the immune response (i.e., antigen capture and presentation, T cell help, and antibody formation [including class switching]) in one endpoint: antigen-specific antibody titers. This assay can be included in NHP toxicology studies (to allow for primary and secondary immunizations to be administered, each followed by the collection of several serum samples for determination of antigen-specific antibody titers). It is worth noting the limitations associated with the TDAR assay—in particular the interanimal variability limiting the sensitivity of the assay in studies powered for general toxicology evaluations. Inhibition of the TDAR can be clearly evidenced with immunosuppressive agents in NHPs (Haggerty 2007), but demonstrating the lack of activity in the assay has been problematic while the definition of biologically irrelevant variations is not established. It was demonstrated that TDAR magnitude (in response to different antigens) and interanimal variability are similar in male and female cynomolgus macaques (Lebrec et al. 2011). Antibody titers from males and females should therefore be combined in NHP studies where both the biology of the drug target and the drug toxicokinetics support no gender differences, in order to increase power and minimize the impact of interanimal variability. Combining results from males and females is not standard practice for toxicology studies and can be controversial, but our experience shows that it is a necessary trade-off to avoid increasing animal numbers while keeping power and being able to inform hazard identification with confidence. Keyhole-limpet hemocyanin is extensively used in NHPs as an immunogen of choice, similar to the trend observed for rodent TDAR assays conducted for pharmaceuticals; however, it should be noted that several antigens may be used and are studied (e.g., tetanus toxoid, bacteriophage). Assay optimization and harmonization of the TDAR in the NHP remains a priority.

A functional ex vivo assay for NK cell-mediated lysis is available for NHPs (using the K562 human myelogenous leukemia target cell line), but the development of a functional assay for CTL activity remains a need and is subject to significant efforts. Approaches leveraging naturally occurring CTL activity against NHP viruses such as cytomegalovirus (CMV) and lymphocryptovirus (LCV) are being explored. Peripheral blood mononuclear cells from whole blood of NHPs can be cultured in the presence of lysates of CMV-infected fibroblasts or LCV-infected lymphoblastic cell lines to measure T-cell mediated interferon-gamma (IFN-γ) secreting cells by ELISPOT or IFN-γ and tumor necrosis factor-alpha (TNF-α) by intracellular cytokine staining (Kamperschroer, Kaur, and Lebrec 2012). Other endpoints, including release of perforin, expression analysis of CD8⁺ T cell CD107a (lysosomal-associated membrane protein-1) and intracellular granzyme B content in immunized animals, may help evaluate NHP CTL function (Morrow et al. 2010). It should be noted that the optimization and validation of such exploratory approaches for implementation of a functional CTL assay in NHP nonclinical safety studies remain to be conducted.

A range of potential liabilities associated with immune agonism may be explored in NHPs (Gribble et al. 2007). The cynomolgus monkey has been used as a relevant toxicology species for immunostimulatory agents such as toll-like receptor agonists; however, while a special recognition is warranted regarding the clinical risk of systemic cytokine release for certain targets, the nonhuman primate is not always predictive of this potential toxicity in humans. The lack of good predictivity of the NHP model for liabilities associated with immune agonism was illustrated by the case of the CD28-specific antibody TGN1412. While the target CD28 is expressed on human regulatory T cells, nonclinical safety studies using TGN1412 in cynomolgus macaques did not predict the life-threatening “cytokine storm” observed in all six healthy volunteers in the phase I clinical trial of this superagonist, partly because CD28 is not expressed on effector memory T cells in macaques (Stebbings et al. 2007; Eastwood et al. 2010). In vitro assessments of cytokine release using human peripheral blood cells should be considered for such agonistic molecules (Vidal et al. 2010) and should complement in vivo NHP studies to strengthen hazard identification and risk assessment.

Immunotoxicology assays are often conducted in the context of hazard identification: to screen out molecules associated with unexpected effects and to identify parameters to monitor during clinical development. Ultimately, immunotoxicology assessments should be refined and further leveraged for risk assessment purposes. For instance, the increasing knowledge of key components of the immune system involved in the control of different types of tumors should be utilized to inform and improve “carcinogenicity risk assessments” conducted for immunomodulators and the consequent labeling process and product.

In conclusion, it should be recognized that NHPs are often the only relevant species for the development of biopharmaceuticals but that a comprehensive assessment of immunotoxicity remains a challenge due to the interanimal variability associated with certain parameters (e.g., TDAR). There are also identified gaps, such as the stage of development of certain assays (e.g., CTL function). A greater understanding of background incidence of certain pathogens in NHPs is also needed for proper interpretation of infectious agent–related findings that may be misinterpreted as treatment-related findings as a result of immunomodulation. These challenges and gaps can be addressed within different organizations and through interlaboratory efforts, and emphasis should be placed on leveraging predictive in vitro assays using human cells.