Design and in Vitro Evaluation of a CAR-T Prototype (ARI-0003) Targeting CD123 for Acute Myeloid Leukemia

Introduction

The outcome of patients with Acute Myeloid Leukemia (AML) is still dismal, especially in patients with a relapsed or refractory (R/R) disease. In these patients, innovative treatment strategies must be considered in order to improve survival. Chimeric Antigen Receptor T-cell (CART) immunotherapy has demonstrated its efficacy and safety in diverse B-cell neoplasms and therefore is being investigated in other hematological malignancies. In AML, CART development is challenging due to the absence of a universal target antigen across AML subtypes, as well as on-target/off-tumor toxicity in healthy tissues. Among all explored AML antigens, CD123 seems to be a safe antigen to target, given its expresion in a high proportion of bulk and leukemic AML stem cells and with a higher level compared to normal hematopoietic progenitors cells. Herein we detail the generation and the in vitro assays of a CD123 CAR-T model (ARI-0003) for AML.

Methods

We developed an anti-CD123 antibody-secreting hybridoma and thereafter identified the variable domains of heavy and light chains of the immunoglobulin to create the scFv domain. The CAR (ARI-0003) genetic sequence was designed and cloned into a pCCL plasmid, and together with a VSV-G, RRE and REV plasmids we transfected HEK 293T cells to obtain 3 ^rd generation lentivirus (Figure 1A). T-lymphocytes were obtained from healthy volunteers and were transduced with lentiviral vectors to generate CAR-T cells. CAR expression in infected T-cells was monitored with flow cytometry using anti-F(ab) ₂ antibodies. To test in vitro activity against AML, non-transduced T-lymphocytes (NT) and CAR-T cells were incubated with AML cell lines and AML primary samples from diverse genetic subtypes, and cytotoxicity was evaluated by flow cytometry at 24 and 48 hours using CD123 and CD33 antibodies.

Results

At the time of co-incubation of target cells and lymphocytes, the mean scFv expression of ARI-0003 in CAR-T cells was 37% (range, 20-57). Cytotoxicity assays with AML cell lines (THP-1, Kasumi1) showed higher target cell mortality with CAR-T cells, compared to NT cells (mean percentage of alive cells at 48h, obtained with an effector-target cell ratio of 1:1, of 0 and 1% [CAR] vs 45 and 57 [NT] in THP-1 and Kasumi1, respectively; Figure 1B-C). This differential cytotoxic activity was maintained using distinct effector ratio, being statistically significant in higher ratios (namely, 1:1, 1:2 and 1:4). Moreover, cytotoxicity induced by CAR-T cells was higher at 48 hours compared to 24 hours. To demonstrate antigen specificity of ARI-0003 CAR-T cells, CD123 negative cell lines (Raji, 697) were used, and target cell mortality was not statistically significant between CAR-T and NT in CD123-negative cell lines.

Finally, ARI-0003 CAR-T efficacy was tested against 8 primary AML samples with a distinct genetic risk, including 5 unfavorable cases according to the European LeukemiaNet classification and 3 AML samples refractory to multiple high-intensity regimens (Figure 1F). Interestingly, a significant cytotoxic effect was observed against all 8 samples after incubation with ARI-0003 CAR-T cells at higher E:T ratio (p < 0.05 for ratios 1:1, 1:2 and 1:4; Figure D-E).

Conclusions

In vitro assays showed that immunotherapy with CAR-T cells ARI0003 against CD123 could be an effective approach for R/R AML. In vivo experiments are needed in order to confirm these results before translating this therapy to clinical phases. Moreover, on-target/off-tumor toxicity induced by ARI-0003 CAR-T needs to be explored, especially myelotoxicity due to target antigen expression in hematopoietic precursor cells.

Figure legend

A) Structure of the ARI0003 CAR. B) Cytotoxicity of CAR-T cells against cell lines (normalized to background toxicity observed with NT cells). C) Flow cytometry of THP1 cells against NT and CAR-T cells. D) Cytotoxicity of CAR-T cells against AML primary samples (normalized to NT percentage). E) Absolut number of remaining AML cells after 48h of coincubation with CAR or NT cells. F) Characteristics of AML primary samples used for in vitro assays.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal