Targeting the FMS-like Tyrosin Kinase 3 with the Unicar System: Preclinical Comparison of Murine and Humanized Single-Chain Variable Fragment-Based Targeting Modules

Clinical translation of chimeric antigen receptor (CAR) T cell therapy in myeloid malignancies is progressing slowly compared to its success in treatment of B cell malignancies. Clinical experiences with CAR T cell therapies against the currently investigated tumor-associated antigens (TAA) (e.g. CD33, CD123 and FMS-like tyrosine kinase 3 (FLT3)) were discouraging and severe side effects occurred (cytokine release syndrome, neurotoxicity and myeloid aplasia) (Hoffmann et al. Journal of Clinical Medicine 2019). Probably targeting a single TAA is insufficient to treat high risk myeloid malignancies with CAR T cell therapies. Therefore, combined targeting of two or even more TAAs seems to be a promising approach. In order to implement such a multiple tumor targeting strategy, we developed a modular CAR T cell system termed UniCAR. The system consists of a universal CAR (UniCAR) directed against the La peptide epitope E5B9 combined with single-chain variable fragment (scFv) -based target modules (TM). In contrast to conventional CARs, anti-tumor activity of UniCAR T cells is only turned on in the presence of the TMs. Thus, this approach will allow UniCAR T cell control due to the short half-life of the TM and therefore has a favorable safety profile. Furthermore, different TMs against several TAAs can be administered both sequentially or in parallel to increase the anti-tumor efficacy or face disease relapse due to antigen escape mechanisms. In the field of myeloid malignancies our group developed retargeting strategies against the TAAs CD33 and CD123 (Cartellieri et al. Blood Cancer Journal 2016). In addition, we have developed a new TM for the UniCAR system that is directed against the TAA FLT3. FLT3 is highly expressed on acute myeloid leukemia (AML) cells and also present on CD123^low AML samples (Riccioni et al. British Journal of Haematology 2011).

The novel FLT3 TM was constructed by fusion of the variable domain of the heavy and the light chain of the murine anti-FLT3 monoclonal antibody (4G8) to the E5B9 UniCAR epitope. In light of a potential clinical application, we in parallel generated a humanized FLT3 TM to further decrease its immunogenicity. Both FLT3 TMs were tested in vitro against different AML cell lines, by using flow cytometry based killing assays as described elsewhere (Fasslrinner et al. British Journal of Haematology 2019). The functionality of the FLT3 TMs in vitro was highly effective. Both FLT3 TMs were able to redirect UniCAR T cells for AML cell lysis already in the picomolar range and were moreover comparable effective than the previously developed CD123 TM. Thus, humanization of the FLT3 TM did not lead to a decrease in anti-tumor efficacy.

In summary, we could show that both the novel murine FLT3 TM and the humanized counterpart redirected UniCAR T cells and induced highly effective elimination of AML cells in vitro. Thus, the flexible application of the FLT3-based UniCAR system seems to be a promising tool for cell-based AML therapy alone or even in combination with other AML-specific TMs (e.g. CD33, CD123).