7.3.6 Cell-based neutralization assays

None of these assays can assess the ability of the antibodies to neutralize the biological activity of the therapeutic proteins that is an important element in the assessment of immunogenicity. For the evaluation of the neutralizing ability, the use of an appropriate noncell-based competitive ligand-binding assay or a cell-based neutralization assay is required. Testing for immunogenicity is performed during the preclinical and clinical phases. The U.S. FDA Guidance for Industry: Assay Development for Immunogenicity Testing of Therapeutic Proteins states that an immunogenicity assay should, in addition to being sensitive, also be able to detect all isotypes, in particular, IgM and all IgG isotypes. The recommended sensitivity is 250 to 500 ng/mL. Studies are performed in three steps: screening, confirmation, and characterization of positives. Initial screening can result in false positives, and, therefore, the initial screening assay is usually followed by a confirmatory assay. After identification and confirmation of positive samples, a full characterization of ADAs in terms of assessment of isotype (class or subclass), binding stability, epitope specificity, and neutralizing capacity gives valuable information of the nature of the studied immune response. The IgG4 is second to IgG1 as the major isotype in ADAs developed for therapeutic mAbs. IgG4 has been associated with immune responses generated under conditions of high doses and prolonged exposure to therapeutic proteins. IgG4 ADAs can be difficult to detect in traditional bridging or homogenous ELISA and ECL™ (enhanced chemiluminescent) assays due to their bispecific nature.

The development of improved assays, particularly cell-based assays for the detection of neutralizing antibodies that allow immunogenicity to be determined with precision and the comparison of immunogenicity data between biopharmaceuticals, are critical for the development of less immunogenic and safer biopharmaceuticals. The biological activity of a therapeutic is often evaluated using an in vitro cell-based assay based on a functional aspect of the protein or the MOA. These assays can be categorized into those that detect signaling responses soon after the protein-receptor interaction has occurred (early stage) or those that provide a measurable readout after the culmination of a cellular response (late stage). Since these assays assess the cellular response in vitro to a protein, they constitute an ideal and appropriate logical approach for the development of a cell-based neutralization assay. It should be realized that different types of bioassay procedures can be used as the basis of a neutralization assay for a biological. A cell-based neutralization assay can be defined as an in vitro assay utilizing cells that interact with or respond to the therapeutic either directly or indirectly in a measurable manner in the presence of a test sample for the detection of antiproduct neutralizing antibodies. The detection Safety similarity

279

Biosimilarity: The FDA Perspective

280 of neutralizing antibodies is based on the principle that any sample containing an antibody (of a neutralizing nature) would reduce or abolish the biological activity induced by a known concentration of the therapeutic in a cell-based assay. Most biological therapeutics can be broadly categorized into agonists or antagonists based on their desired effect in vivo. While agonists (e.g., cytokines, growth factors, hormones, agonistic mAbs) induce a response by directly binding to receptors on the target cell surface, therapeutics with antagonistic properties (e.g., soluble receptors, antagonistic mAbs) act by blocking the binding of a ligand to the target receptor expressed on the cell surface. As a result, assay formats and designs of neutralizing antibody assays can vary depending on the biological and the type of assay being used. Although regulatory authorities recommend the use of cellbased assays for the detection and the quantification of neutralizing antibodies, cellbased assays often give variable results and are difficult to standardize. Conventional cell-based assays for neutralizing antibodies are based upon the assessment of a drug-induced response in a drug-sensitive cell line, and the ability of an antibody to inhibit that response. The form of the assay can considerably vary, however, reflecting the diversity of the biopharmaceuticals currently employed in the clinic. Drug-induced responses vary from stimulation of cell proliferation in the presence of a growth factor, such as EPO or GMCSF, or induction of apoptosis in the presence of TNFα to inhibition of virus replication in interferontreated cells. Such druginduced responses are complex events involving the transcriptional activation or modulation of numerous genes. Druginduced biological responses also often take several days to develop and are influenced by a number of factors that are difficult to control. A series of engineered reporter cell lines has been developed for the quantification of the activity and the neutralizing antibody response to biopharmaceuticals that eliminate many of the limitations of conventional cellbased assays. In order to improve the performance of conventional cell-based assays and to develop assays that allow more direct comparisons of immunogenicity data, reporter cell lines were established based on transfection of cells with the firefly luciferase reporter gene regulated by a drugresponsive chimeric promoter. These assays allow the drug activity, and the neutralization of drug activity, to be determined selectively and with a high degree of precision within a few hours by measuring light emission. In addition, the use of a single common druginduced response for a variety of different drugs facilitates comparisons of immunogenicity data. Conventional cell-based assays are difficult to standardize due in part to assay variation resulting from changes in culture conditions and genetic and epigenetic changes that can occur, as cells are continuously cultivated in the laboratory. Assay variation can be minimized by the use of