Antibody pairs inducing selective clustering on cells co-expressing two antigens – a HexElect QSP model
Csaba Katai1, Raju Prasad Sharma2, Jeroen Elassaiss-Schaap1, Sieto Bosgra2
1PD-value B.V., 2Genmab B.V.
Introduction: Genmab has developed a range of Fc-mutation-based antibody platforms. The HexaBody® platform enables the creation of potent therapeutics that induce the formation of antibody clusters (hexamers) upon binding to the plasma membrane of target cells (De Jong et al., 2016). These clusters subsequently promote complement dependent cytotoxicity (CDC), cell-mediated effector functions, and/or initiate intracellular signaling. A detailed QSP model of oligomerization on the cell surface has helped understand the unique PK and PD properties of the HexaBody technology (Strasser et al. 2019; 2025). The HexElect® technology enhances the functional selectivity of therapeutic antibodies by making their activity dependent on co-expression of two antigens. The Fc domains of pairs of IgG antibodies are engineered to suppress their individual homo-oligomerization while promoting their pairwise hetero-oligomerization after binding co-expressed antigens on the cell surface (Oostindie et al., 2022). The combination of two antibodies with distinct targets and the alternating sequence of oligomerization of HexElect pose additional questions. How does the efficiency of hexamer formation depend on ratios of receptor expression, antibody exposure and receptor occupancy? What conditions may compromise the desired selectivity towards cells expressing both targets? Objectives: (1) Develop a QSP model, based on the HexaBody model, to describe the oligomerization dynamics of the HexElect platform on the cell surface. (2) Compare and calibrate the QSP model, in conjunction with a basic CDC model, to in vitro lysis measurements. (3) Use the developed HexElect model to perform a scan of the parameter space (target expression, binding affinity, antibody concentrations and their ratio) to identify effective combinations of targets, antibodies and dosing schemes. Methods: The HexaBody model was refined by including engagement of two receptors by distinct antibodies and the alternating sequence of oligomer expansion. A model branch was added allowing oligomerization through Fc-Fc interactions without Fab domain binding to one of the two antigens. In vitro CDC experiments with various combinations of targets with varying cell surface density across a range of total antibody concentrations and antibody ratios were used for model calibration and validation. Experimental conditions included a HexaBody variant to independently estimate cell sensitivity to hexamer induced lysis, HexElect variants to calibrate the HexElect model, and a HexElect pair with one non-binding antibody to inform an oligomerization pathway involving Fab binding to only one antigen. Results: The model was successfully calibrated to the available in vitro data. Simulations confirmed effective oligomerization and downstream activity on cells sufficiently expressing both antigens. However, highly skewed relative occupancy of either target may lead to stable formation of lower order oligomers, hampering completion of hexameric rings. Furthermore, hexamers only partially bound to antigen may form primarily at high density and occupancy of the first target in combination with high concentrations of the antibody against the second target. Such conditions may lead to undesired activity towards cells expressing only one of the antigens. These results identify absolute and relative receptor expression on target cells as an important target identification parameter, and target binding affinities and dosing ratio as key parameters in candidate selection and therapeutic application. Conclusions: A HexElect oligomerization scheme was developed guided by lysis data. A parameter study was performed to infer optimal dosing levels and ratios, and the characteristics of ideal target and antibody pairs that maximize cytotoxicity on target cells while retaining low activity towards non-target cells. The QSP model has identified key factors determining functional activity and selectivity of the platform and may support discovery and development of therapeutic applications.
De Jong, R.N., Beurskens, F.J., Verploegen, S., Strumane, K., van Kampen, M.D., Voorhorst, M., Horstman, W., Engelberts, P.J., Oostindie, S.C., Wang, G. and Heck, A.J., 2016. A novel platform for the potentiation of therapeutic antibodies based on antigen-dependent formation of IgG hexamers at the cell surface. PLoS biology, 14(1), p.e1002344. Oostindie, S.C., Rinaldi, D.A., Zom, G.G., Wester, M.J., Paulet, D., Al-Tamimi, K., van der Meijden, E., Scheick, J.R., Wilpshaar, T., de Jong, B. and Hoff-van den Broek, M., 2022. Logic-gated antibody pairs that selectively act on cells co-expressing two antigens. Nature Biotechnology, 40(10), pp.1509-1519. Strasser, J., de Jong, R.N., Beurskens, F.J., Schuurman, J., Parren, P.W., Hinterdorfer, P. and Preiner, J., 2019. Weak fragment crystallizable (Fc) domain interactions drive the dynamic assembly of IgG oligomers upon antigen recognition. ACS nano, 14(3), pp.2739-2750. Strasser, J., Frischauf, N., Schustereder, L., Karner, A., Bosgra, S., Labrijn, A.F., Beurskens, F.J. and Preiner, J., 2025. A kinetic model of antigen-dependent IgG oligomerization and complement binding. Submitted.