Overcoming the challenges associated with CD3+ T-cell redirection in cancer

Abstract

The development of bispecific antibodies that redirect the cytotoxic activity of CD3+ T cells to tumours is a promising immunotherapeutic strategy for the treatment of haematological malignancies and solid cancers. Since the landmark FDA approval at the end of 2014 of the anti-CD3 × anti-CD19 bispecific antibody blinatumomab (Blincyto^®) for the treatment of relapsed/refractory B-cell acute lymphoblastic leukaemia, ~100 clinical trials investigating the safety and efficacy of CD3+ bispecific T-cell redirectors for cancer have been initiated. However, despite early success, numerous challenges pertaining to CD3+ T-cell redirection in the context of cancer exist, including the recruitment of counterproductive CD3+ T-cell subsets, the release of systemic cytokines, the expansion of immune checkpoint molecules, the presence of an immunosuppressive tumour microenvironment, tumour antigen loss/escape, on-target off-tumour toxicity and suboptimal potency. The aim of the present review is to discuss novel approaches to overcome the key challenges associated with CD3+ bispecific T-cell redirection in order to achieve an optimal balance of anti-tumour activity and safety.

Fig. 1. Mechanism of action of CD3+ bispecific T-cell redirection in cancer.

The schematic depicts an IgG-like bispecific antibody simultaneously binding a tumour-associated antigen (TAA) on a cancer cell and CD3 epsilon on a T cell to redirect the cytotoxic activity of T cells to these tumour cells.

Fig. 2. Bispecific antibody constructs for CD3+ T-cell redirection.

A typical IgG-like bispecific antibody (shown) consists of two heavy chains and two light chains, subdivided into variable domains and constant domains. The Fab region (antigen-binding domain) and Fc region (effector-function domain) are indicated. V_H variable heavy, V_L variable light, C_H constant heavy, C_L constant light. Various bispecific antibody formats exist for CD3+ T-cell redirection: bispecific T-cell engager (BiTE), BiTE-Fc; dual-affinity re-targeting (DART), DART-Fc; tetravalent DART; bispecific killer cell engager (BiKE); tri-specific killer cell engager (TriKE); single-chain variable fragment (scFv)-scFv-scFv; diabody; tandem diabody (TandAb); knobs-in-holes cognate light chains; knobs-in-holes common light chains; DuoBody; TrioMab; CrossMab Fab; CrossMab VH-VL; 2:1 CrossMab; 2:2 CrossMab; scFv-IgG; DVD-Ig; IgG-IgG; and Fab-scFv-Fc. DuoBody bispecific antibodies are formatted on an IgG1 backbone and are generated by controlled Fab arm exchange of complementary C_H3 domain mutations that promote heterodimerisation (K409R refers to a lysine (K) to arginine (R) amino acid substitution; F405L refers to phenylalanine (F) to leucine (L) amino acid substitution).

Fig. 3. Next-generation anti-CD3 × anti-CD28 × anti-CD38 tri-specific T-cell-engaging antibody.

The diagram depicts a tri-specific antibody targeting anti-CD3ε and anti-CD28 on T cells as well as anti-CD38 on multiple myeloma cells. Co-engagement of CD3, CD28 and CD38 results in efficient T-cell stimulation and activation, cytokine release, and redirection of T-cell-mediated cytotoxicity to myeloma cells.