By James DeGiulio --
On November 23, the European Medicines Agency (EMA) published guidelines for biosimilar monoclonal antibodies (mAb). Earlier this week, we reported on the clinical and non-clinical study portion of the guidelines (see "EMA Publishes Guidelines for Biosimilar Antibodies - Part I"). The EMA released a second set of guidelines along that directly deals with one of the premier safety challenges in working with mAbs -- immunogenicity. The second set of guidelines seeks to address the problems with detection of and risk related to the development of the mAb immune response. The guidelines are primarily aimed at products at the final marketing authorization stage of development, and unfortunately, the vast majority of the guidelines are spent describing all of the complications involved in controlling mAb immunogenicity, rather than offering much real guidance. Providing a general framework is a near-impossible task in this subject area due to the many differences from one mAb to the next, but the common thread throughout all of the EMA guidelines appears to be comprehensive risk planning at the early stages of development and commitment to diligent monitoring post-approval.
As was suggested in the first set of guidelines pertaining to the studies required to establish biosimilarity of mAbs, these guidelines support a subjective risk-based approach to reducing unwanted immunogenicity. Several sources of the immune response are illustrated, such as rejection of non-human or chimeric mAbs, response to unique CDRs, and the unknown nature of emerging next generation mAbs, such as bivalent mAbs. Different glycosylation patterns can also affect immunogenicity as they can shield the protein backbone or trigger toll-like receptor responses. Even the formulation, container system, or storage conditions can alter the mAb by interaction with the storage materials. Patient-related factors also present a substantial challenge over classic small-molecule drugs, as the MHC and human leukocyte antigen alleles of each patient are different, as well as the physiological rate of innate antibody production.
The guidelines suggest a limited number of approaches that may help in predicting and reducing unwanted immunogenicity of mAb, but first and foremost place the responsibility of design and selection of the mAb studies squarely on the biosimilar applicant. After clarifying the applicant's burden, in silico modeling is suggested to identify T-cell epitopes, but the guidelines warn that this is not a complete predictor of immunogenicity. The guidelines note that potential deletion of T-cell epitopes in the mAb may also result in reduced immunogenicity.
The guidelines particularly focus on the problems manufacturers may experience with current assays used to assess mAb immunogenicity. Detection of antibodies against mAb using ELISAs or radio-immunoprecipitation are difficult because of nonspecific binding, thus requiring a new generation of assays to be developed which are not yet sensitive enough to rely on in the clinical setting. Both the newer "bridging" ELISA and Surface Plasmon Resonance (SPR) techniques are introduced, but they both have limitations which will need to be resolved prior to their acceptance as standard assays. Also, patient samples (serum or plasma) typically contain substances that are known to interfere with these assays and skew results, which will in turn require substantial and costly customization for each mAb. Positive control sera for these assays is also challenging to produce, particularly at the early stages of development.
The guidelines advocate a risk-based approach, but acknowledge that it is a mere starting point. Due to the diversity of complications involved, standards for immunogenicity of mAbs simply cannot be generalized. Nonetheless, three subgroups of immunological "risk" are defined: product, process, and patient-related risk factors. Each factor within each group must be ranked and justified early in product development, with higher-ranked risk factors requiring more stringent clinical trials. Product factors include choice of cell line, potential impurities, and product isoforms and degradation products. Route of administration is an example of a process risk that must be considered, with intravenous mAb ranked as lowest risk due to short duration and physician observation, and subcutaneous mAb as highest risk due to patient home administration. Patient risk factors such as age, genetic background, and underlying disease must also be taken into account.
Most importantly, the guidelines stress the evaluation of the clinical consequences of unwanted immune response for each mAb. The mode of action of the mAb is critical in this assessment, as many mAb function to lyse or induce apoptosis in cells. Further, the nature of the target molecule as immunosilencing or immunostimulating must be comprehensively investigated, which can also be difficult to isolate and quantify, particularly using in vivo studies. Also important is whether adverse events such as infusion reactions can be properly handled by neutralizing the mAb with other antibodies or with medication. Early detection of these adverse events must occur in first rounds of clinical trials.
Finally, one standard requirement provided by the EMA is that all mAb will require a validated screening and confirmatory assay followed by a validated neutralizing assay. All antibodies must be classified as neutralizing or non-neutralizing, regardless of their risk level. Patient samples should be undertaken as a routine basis, tested in real-time, and banked during the course of development.
The EMA is requesting comments on either set of guidelines until May 31, 2011. Comments must be submitted using a template provided by the EMA and sent to BMWP.Secretariat@ema.europa.eu.