Bispecific antibodies (BsAb) are a class of monoclonal antibodies that can target two different antigens/epitopes simultaneously. It has been more than a decade since the first BsAb was approved by the European Medicines Agency (EMA) in 2009, and during this time, emerging data have demonstrated the huge advantage of BsAb over conventional monospecific antibodies. As a result, BsAbs have become a popular medication class for treating cancer, and their potential for therapeutic application has grown in recent years. The number of marketed BsAbs in the year 2022 has already exceeded the number of approved BsAbs ever before, thus BsAbs have finally reached their golden age.
BsAbs can be roughly divided into two classes: IgG-like BsAbs and fragment-based BsAbs. Fragment-based BsAbs have a much shorter serum half-life due to the absence of the Fc fragment, which has limited their application in the clinic. In contrast, IgG-like BsAbs have relatively long serum half-lives, however, generating pure IgG-like BsAbs and improving their yield is still challenging due to chain association issues. The production of a BsAb in one expression cell line is very difficult and unfavourable due to challenges in extracting the desired BsAb from the lysates and the inherently low yield (Chapter 1, Figure 2). Thus, the aim of the research described in chapter 3 was to design a novel format to improve the purity and yield of BsAb during production. We have combined a conventional antigen-binding fragment with a single-domain antibody and generated BsAbs in a Fab x sdAb-Fc format. Our data show that this format can avoid potential heavy-light chain mis-pairing during the production of BsAbs, and increases the purity of BsAbs to above 95%. Further characterization assays showed that the BsAbs in this configuration nevertheless maintained the ability of binding to two distinct antigens concurrently.
During the design of this novel BsAb format, we found that the hinge region can be adapted for different application scenarios. There are already several studies indicating that enhanced T-cell mediated tumor cell elimination can be achieved by decreasing the distance between T cells and tumor cells. In our case, the distance between tumor cells and T cells can be modulated by different hinge designs. Thus, we modulated the hinge region in T-cell redirecting bispecific antibodies (TRBAs) and studied their anti-tumor activity in in vitro assays. Our data show that with less space separating the two arms of the TRBA, tumor cells and effector T cells can bridge more tightly, which strengthens T cell activation and, in turn, increases tumor cell death. Therefore, our data indicate that the modulation of the ‘hinge region length’ parameter can possibly contribute to future design of similar molecules (chapter 4).
Since the antibody Fc-tail activates specific immune effector mechanisms, antibody isotype plays an important role in cancer therapy. Previous studies in a mouse tumor model showed the efficacy of prophylactic application of mIgG2a isotype antibodies. However, human cancer patients usually receive antibodies in a therapeutic setting. Thus, in the last study, we developed a panel of anti-Thy1.1 antibodies with various isotypes (mIgG1, mIgG2a, mIgE and Fc-silenced) and evaluated their effectiveness in a therapeutic setting in mouse tumor models. Our data demonstrated that mIgG2a is the most effective isotype in treating cancer in a therapeutic setting in mice. Therefore, isotype selection is a critical parameter determining the efficacy of tumor-targeting antibody therapy. We believe future research in tumor immunotherapy may benefit from the knowledge we gained in designing tumor antigen targeting antibodies for cancer therapy (chapter 5).