Improved Killing of AML Blasts By Dual-Targeting of CD123 and CD33 Via Unitarg a Novel Antibody-Based Modular T Cell Retargeting System

Acute myeloid leukemia (AML) is a hematologic malignancy of the myeloid line with high prevalence in older patients. As complete eradication of metastatic cancer cells is often not achieved by standard therapies, alternative treatment modalities are urgently needed.
In recent years, bispecific antibodies (bsAbs) and chimeric antigen receptors (CARs) emerged as promising candidates for an antigen-specific cancer immunotherapy. Both bsAbs and CARs are able to redirect T cells for efficient tumor cell lysis. Nevertheless, the development of a novel TAA specific bsAb or a CAR is a long lasting process. Therefore, we recently introduced a novel antibody-based modular platform (UniTARG) that can be rapidly and easily adapted for redirection of T cells to any TAA in both a bsAb or CAR related manner. The modular UniTARG system distributes the effector arm (the anti-CD3 domain or CAR) and the anti-TAA binding domain to two separate molecules: (I) an exchangeable target module (TM) comprising an anti-TAA binding moiety and a short peptide epitope (E5B9), and (II) a universal effector unit. The effector systems represent either a bsAb with specificity for CD3 and a peptide epitope (E5B9) termed UniMAB or a CAR directed to the E5B9 epitope (UniCAR). Thus, TMs can form a complex with the respective effector system that facilitates the cross-linkage of tumor and T cells similar to conventional bsAbs or CARs. For redirection of T cells to any kind of TAA only the binding moiety of the TM has to be adapted what saves costs and time. To increase tumor specificity and to reduce the risk of tumor escape variants, the modular UniTARG system further offers the possibility to apply simultaneously different monospecific or even bispecific TMs recognizing two TAAs.

For proof of concept of a dual targeting using the UniTARG system we selected as TAA on AML blasts the molecules CD33 and CD123. They represent promising target antigens as they are overexpressed on both rapidly proliferating terminal AML blasts and leukemic stem cells which might be responsible for disease relapse after initial chemotherapy. Thus, we generated an anti-CD123 TM and anti-CD33 TM that can be applied within the modular system for single-targeting or that can be combined for dual-targeting of AML blasts. By fusion of the anti-CD123 and anti-CD33 domains via the E5B9 epitope a bispecific TM was further constructed. As revealed by cytotoxicity assays with CD33+CD123+ AML cell lines, the novel mono- and bispecific TMs can be easily applied to the modular systems to trigger highly potent tumor cell lysis at low E:T ratios and picomolar Ab concentrations. By using the dual-targeting approach we can show that lysis of CD123+CD33+ AML blasts can be considerably improved in comparison to the mono-specific strategy.

Overall, due to the ease and cost-effectiveness of development the UniTARG platform technology represents a promising tool in the field of both bsAbs and CARs with the advantage of simultaneous or consecutive dual or even multispecific targeting. This approach might additionally improve anti-tumor activity by increasing tumor specificity and diminishing off-target effects.