The application of trifunctional, bispecific antibodies directed against CD38, a transmembrane glycoprotein, and CD3, a T-cell antigen, may be an effective therapeutic strategy for multiple myeloma. Hence, in this study we constructed and evaluated such antibody. The trifunctional antibody with an Fc region has the potential to engage natural killer cells and macrophages. This antibody efficiently recruited effector T cells to CD38-expressing myeloma cells and induced antibody-dependent cell-mediated cytotoxicity (ADCC).


The recombinant protein consisted of the extracellular domain of the human CD38 (aa 43-300), which was fused to the N-terminus of the Fc region of the human IgG1 and expressed in CHO-K1. These CD38-expressing cells were used to immunize mice. Monocytic cells from the spleen of the immunized mice were fused with a fusion partner, X63-Ag8.653 mouse myeloma cell line, using polyethylene glycol-mediated cell fusion. This was followed by the selection of a hybridoma using the method of Kinebuchi et al. One of the monoclonal antibodies (31C2, IgG1 κ) was confirmed to recognize CD38. The VH and VL genes of the hybridoma cells were cloned with the synthesized primers, using previously reported polymerase chain reaction (PCR) methods. The expression vectors of the chimeric and trifunctional antibodies, ch-31C2 and tri-31C2, respectively, were constructed from the cloned genes. Following this, we constructed the chimeric anti-CD38 monoclonal antibody (ch-31C2) as well as the trifunctional antibody (tri-31C2) which binds CD38 and CD3 and evaluated their anti-myeloma activities.


We assessed the potential of tri-31C2 to inhibit CD38-positive cell growth. CD38-positive RPMI 8226 and CD38-negative Colo 320 cells were treated with tri-31C2, ch-31C2, at concentrations of 0.1 and 1 µg/mL respectively; human polyclonal IgG was used as the control. These cells were subsequently co-cultured with PBMCs (E/T ratio 5). The results revealed that 1 but not 0.1 µg/mL of the ch-31C2 inhibited the cell viability by approximately 80-90% compared to the control. In contrast, both 0.1 and 1 µg/mL of the tri-31C2 inhibited myeloma cell growth to 30% viability compared to the control.

Co-culturing the CD38-positive RPMI 8226 myeloma cells and PBMCs with 0.1 and 1.0 µg/mL of tri-31C2 triggered significant secretion of TNFα, INFγ, and IL-2; however, the same concentrations of ch-31C2 did not increase cytokine secretion. Furthermore, there was an insignificant increase in secretion of these cytokines when CD38-negative Colo 320 cells were co-cultured with ch-31c2 or tri-31C2 Next, we examined the antitumor effects of the tri-31C2 and ch-31C2 antibodies in a mouse xenograft model of multiple myeloma. They were both found to significantly inhibit tumor formation compared to the control.


We have successfully generated the chimeric anti-CD38 monoclonal antibody (ch-31C2) as well as the trifunctional antibody (tri-31C2) which binds CD38 and CD3. Our results suggest that tri-31C2 had a stronger anti-myeloma effect compared to ch-31C2.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.