The blood endothelial cell chamber - An innovative system to study immune responses in drug development.

The risk for adverse immune-mediated reactions, associated with the administration of certain immunotherapeutic agents, should be mitigated early. Infusion reactions to monoclonal antibodies and other biopharmaceuticals, known as cytokine release syndrome, can arise from the release of cytokines via the drug target cell, as well as the recruitment of immune effector cells. While several in vitro cytokine release assays have been proposed up to date, many of them lack important blood components, required for this response to occur. The blood endothelial cell chamber model is an in vitro assay, composed of freshly drawn human whole blood and cultured human primary endothelial cells. Herein, its potential to study the compatibility of immunotherapeutics with the human immune system was studied by evaluating three commercially available monoclonal antibodies and bacterial endotoxin lipopolysaccharide. We demonstrate that the anti-CD28 antibody TGN1412 displayed an adaptive cytokine release profile and a distinct IL-2 response, accompanied with increased CD3+ cell recruitment. Alemtuzumab exhibited a clear cytokine response with a mixed adaptive/innate source (IFNγ, TNFα and IL-6). Its immunosuppressive nature is observed in depleted CD3+ cells. Cetuximab, associated with low infusion reactions, showed a very low or absent stimulatory effect on proinflammatory cytokines. In contrast, bacterial endotoxin demonstrated a clear innate cytokine response, defined by TNFα, IL-6 and IL-1ß release, accompanied with a strong recruitment of CD14+CD16+ cells. Therefore, the blood endothelial cell chamber model is presented as a valuable in vitro tool to investigate therapeutic monoclonal antibodies with respect to cytokine release and vascular immune cell recruitment.

Exploratory toxicology as an integrated part of drug discovery. Part II: Screening strategies.

In an effort to reduce toxicity-related attrition, different strategies have been implemented throughout the pharmaceutical industry. Previously (in Part I), we have outlined our 'integrated toxicology' strategy, which aims to provide timely go/no-go decisions (fail early) but also to show a direction to the drug discovery teams (showing what will not fail). In this review (Part II of the series) we describe our compound testing strategies with respect to cardiovascular safety, hepatotoxicity, genotoxicity, immunotoxicity and exploratory in vivo toxicity. We discuss the in vitro, ex vivo and in vivo assays and models we employ to assess safety risks and optimize compound series during the drug discovery process, including their predictivity and the decisions they generate.

Exploratory toxicology as an integrated part of drug discovery. Part I: Why and how.

Toxicity and clinical safety have major impact on drug development success. Moving toxicological studies into earlier phases of the R&D chain prevents drug candidates with a safety risk from entering clinical development. However, to identify candidates without such risk, safety has to be designed actively. Therefore, we argue that toxicology should be fully integrated into the discovery process. We describe our strategy, including safety assessment of novel targets, selection of chemical series without inherent liabilities, designing out risk factors and profiling of candidates, and we discuss considerations regarding what to screen for. We aim to provide timely go/no-go decisions (fail early) and direction to the discovery teams, by steering away from safety risk (showing what will not fail).

A triple-transgenic immunotolerant mouse model.

Avoiding unwanted immunogenicity is of key importance in the development of therapeutic drug proteins. Animal models are of less predictive value because most of the drug proteins are recognized as foreign proteins. However, different methods have been developed to obtain immunotolerant animal models. So far, the immunotolerant animal models have been developed to assess one protein at a time and are not suitable for the assessment of combination products. Our aim was to develop an animal model for evaluating the impact of manufacturing and formulation changes on immunogenicity, suitable for both single protein and combination products. We constructed two lines of transgenic mice expressing the three human coagulation factors, II, VII, and X, by inserting a single vector containing the three coagulation factors encoding sequences separated by insulator sequences derived from the chicken beta-globin locus into the mouse genome. Immunization of transgenic mice from the two lines and their wild-type littermates showed that transgenic mice from both lines were immunotolerant to the expressed human coagulation factors. We conclude that transgenic mice immunotolerant to multiple proteins can be obtained, and that these mice are potentially useful as animal models in the assessment of immunogenicity in response to manufacturing changes.

Refined analysis of antigen-specific antibody responses--a new one-step tool in immunogenicity studies.

Immunogenicity is a continuous efficacy and safety issue of biopharmaceuticals. Pre-clinical models for prediction of immunogenicity itself as well as biomarkers to reveal potential mechanisms behind an already existing antibody response are still needed. A sensitive, robust and specific immunogenicity assay has therefore been developed that can detect and measure antibodies of five classes against an administered recombinant human protein drug. Additionally, a validation was performed to evaluate the reproducibility and specificity of this newly developed assay. The production of drug-induced antibodies in mice injected with a recombinant human protein drug has been measured by using a modified version of a multi-parametric bead analysis technique. Competitive binding was used to verify drug-specificity of the antibodies. Results showed that the mouse response against the recombinant human protein was IgG1- and IgG2b-specific, suggesting that the drug-induced response was driven by both Th1/Th2 cells; a finding confirmed by measurement of the cytokine profile. With this assay, anti-drug antibody class and subclass screening may be executed in one step.

TNF-alpha impairs peripheral tolerance towards beta-cells, and local costimulation by B7.1 enhances the effector function of diabetogenic T cells.

Maintenance of peripheral tolerance and inactivation of autoreactive T cells is based on a delicate balance between pro-inflammatory and protective cytokines that is poorly understood. We have here addressed how the local expression of the inflammatory cytokine TNF-alpha can impair peripheral tolerance and lead to autoreactivity. After transplantation of pancreata that are immunogenic due to beta-cell expression of B7.1 and TNF-alpha, into thymectomized and euthymic syngeneic mice, we found that only euthymic mice rejected the grafts. This result suggests that under normal circumstances autoreactive T cells are functionally inactivated, and initiation of an autoreactive response requires de-novo generation of T cells. By contrast, thymectomized mice expressing TNF-alpha on the endogenous islets rejected the grafts, showing that expression of TNF-alpha prevents functional silencing of the autoreactive T cells. Thus, this study provides a mechanism by which TNF-alpha and possibly chronic inflammatory responses may promote autoimmune diseases. Furthermore, we have investigated whether B7.1 can enhance T cell responses of already activated T cells leading to islet rejection. By transplantation of wild-type and B7.1-expressing islets into overtly diabetic mice we found that only the wild-type islets could restore normoglycemia, suggesting that costimulation by B7.1 is required in the expansion or effector phase of the response.