The Race Is on: BiTE Vs CART As FLT3-Directed Immunotherapies in AML

Introduction Redirection of T cells has evolved as one of the most promising strategies for the treatment of hematologic diseases. The two main immunotherapeutic approaches consist of T-cell redirection via chimeric antigen receptor (CAR) T cells and T-cell engager (TCE) molecules.

A growing arsenal of anti-CD19 and -CD20 directed CAR T, as well as TCE constructs, have been FDA and EMA approved in different precursor and mature B-cell malignancies. Translation into myeloid malignancies is challenged by the choice of target antigen and T-cell fitness. Based on previous work, we identified FLT3 as a promising target antigen for the treatment of AML with limited on-target-off-leukemia toxicity (Brauchle et al. 2020). Hence, we compared FLT3-directed CAR T cells with a bispecific T cell engager (BiTE®) molecule in a preclinical AML model with a focus on T-cell function and evolving T-cell exhaustion.

Methods First, we implemented an AML cell line (MV-411, MOLM-13, and OCI-AML3) and primary AML in vitro model system to study CAR T- and BiTE®-mediated cytotoxicity using flow cytometry-based read-out assays. Impact of positive costimulation on T-cell function, in particular for the comparison of a CAR T containing a CD28 costimulatory domain vs a BiTE® molecule, were evaluated using our previously established Ba/F3 model system, which is devoid of any costimulatory molecules.

The impact on conjugate formation as surrogate for immunological synapse formation and consecutive T-cell degranulation were studied. To further enhance cytotoxicity, CAR-T cells and BiTE® molecules were combined with PD-1 blockade or tyrosine kinase inhibition. To test if CAR-T cells and BiTE® molecules are able to induce bystander killing of tumor cells not expressing the target antigen, we set up a mixing model with FLT3⁺ and FLT3^- cells including healthy bone marrow cells.

Finally, we compared CAR-T cells and BiTE® molecules in an in vivo xenograft model using the OCI-AML3-LUC-GFP cell line.

Results We observed that FLT3-targeted CAR-T cells and BiTE® molecules led to effector-to-target (E:T) ratio-dependent lysis of the FLT3⁺ cell lines and showed a similar cytotoxicity against all tested cell lines. Specific lysis of primary AML blasts was also similar in both conditions. However, using the Ba/F3 model system, we were able to show that the presence of the costimulatory ligand CD86 significantly increased the cytotoxic capacity of BiTE®-redirected T cells (p= 0.0312) while CAR-T cell-mediated lysis was unaltered.

The percentage of conjugates formed after 10 and 30 minutes of coculture was significantly higher in the CAR-T cell condition (p = 0.0298 and p = 0.0225). This was accompanied by a faster and stronger degranulation of the CAR-T cells. Combinatorial treatment with PD-1 blockade slightly enhanced the specific lysis of AML cell lines in both cocultures. The addition of the tyrosine kinase inhibitor Quizartinib significantly increased the CAR-T and BiTE®-mediated lysis of the MV-411^{(FLT3-ITD/FLT3-ITD)} cell line. Bystander killing of the target-antigen negative tumor cell line HEL92.1.7 was seen in both conditions, however low cytotoxicity against healthy primary bone marrow cells was observed. In contrast to our in vitro data, using an in vivoAML mouse model, the overall survival of mice treated with CAR-T cells was significantly prolonged when compared to the BiTE®-treated group (28 days vs 40 days, p = 0.0031). This was accompanied by enhanced proliferation and homing of the CAR-T cells as seen by significantly higher frequencies of T cells in the blood and spleen of these mice (p = 0.0019 and p = 0.0059).

Conclusion Together, our data underline the power of T-cell-based immunotherapy platforms to combat myeloid malignancies. FLT3-directed CAR-T cells and BiTE® molecules mediated cytotoxicity in the same range, and similar results were observed in various in vitro model systems. However, using an in vivo AML xenograft model, CAR-T cells performed better and led to significantly prolonged overall survival of the mice in this group. This might be related to an integrated positive costimulatory domain within the CAR-T construct, enabling prolonged T-cell fitness and enhanced proliferation. Future studies will further dissect the impact of CAR-T cells versus BiTE® molecule treatment on T-cell fitness and how this translates into response and outcome.

Brauchle:Adivo: Current Employment. Goldstein:Merck Research Laboratories: Current Employment; AMGEN Research, AMGEN Inc: Ended employment in the past 24 months. Kobold:Tabby Therapeutics: Research Funding; Arcus Biosciences: Research Funding; Carina Biotech: Other: IP License; TCR2 Inc: Honoraria, Other: IP License, Research Funding; Novartis: Honoraria; GSK: Honoraria; BMS: Honoraria. Buecklein:Miltenyi Biotech: Research Funding; Novartis: Honoraria, Research Funding, Speakers Bureau; Pfizer: Research Funding, Speakers Bureau; Amgen: Honoraria; Gilead: Honoraria. Chapman-Arvedson:Hexagon Bio: Current Employment; AMGEN Research, AMGEN Inc.: Ended employment in the past 24 months. Subklewe:Novartis: Consultancy, Speakers Bureau; Gilead: Consultancy, Research Funding, Speakers Bureau; Janssen: Consultancy, Speakers Bureau; Morphosys: Research Funding; Takeda: Other: Travel support; Roche: Consultancy, Research Funding; Seagen: Research Funding; Pfizer: Consultancy; Bristol-Myers Squibb: Research Funding; Seattle Genetics: Research Funding; Miltenyi Biotech: Research Funding; Celgene/BMS: Consultancy, Speakers Bureau; Amgen: Consultancy, Research Funding, Speakers Bureau.