Point-Of-Care CAR T-Cell Production (ARI-0001) Using a Closed Semi-automatic Bioreactor: Experience From an Academic Phase I Clinical Trial

Abstract

Development of semi-automated devices that can reduce the hands-on time and standardize the production of clinical-grade CAR T-cells, such as CliniMACS Prodigy from Miltenyi, is key to facilitate the development of CAR T-cell therapies, especially in academic institutions. However, the feasibility of manufacturing CAR T-cell products from heavily pre-treated patients with this system has not been demonstrated yet. Here we report and characterize the production of 28 CAR T-cell products in the context of a phase I clinical trial for CD19+ B-cell malignancies (NCT03144583). The system includes CD4-CD8 cell selection, lentiviral transduction and T-cell expansion using IL-7/IL-15. Twenty-seven out of 28 CAR T-cell products manufactured met the full list of specifications and were considered valid products. Ex vivo cell expansion lasted an average of 8.5 days and had a mean transduction rate of 30.6 ± 13.44%. All products obtained presented cytotoxic activity against CD19+ cells and were proficient in the secretion of pro-inflammatory cytokines. Expansion kinetics was slower in patient's cells compared to healthy donor's cells. However, product potency was comparable. CAR T-cell subset phenotype was highly variable among patients and largely determined by the initial product. T_CM and T_EM were the predominant T-cell phenotypes obtained. 38.7% of CAR T-cells obtained presented a T_N or T_CM phenotype, in average, which are the subsets capable of establishing a long-lasting T-cell memory in patients. An in-depth analysis to identify individual factors contributing to the optimal T-cell phenotype revealed that ex vivo cell expansion leads to reduced numbers of T_N, T_SCM, and T_EFF cells, while T_CM cells increase, both due to cell expansion and CAR-expression. Overall, our results show for the first time that clinical-grade production of CAR T-cells for heavily pre-treated patients using CliniMACS Prodigy system is feasible, and that the obtained products meet the current quality standards of the field. Reduced ex vivo expansion may yield CAR T-cell products with increased persistence in vivo.

Keywords: CAR T-cell production; CD19; CliniMACS Prodigy; chimeric antigen receptor; immunotherapy; leukemia; lymphoma.

Figure 1

ARI-0001-cell expansion in CliniMACS Prodigy. (A) Expansion kinetics of ARI-0001-cell products (Total cell number). Gray points indicate individual products. Black triangles indicate Mean ± SD and adjusting curve. (B) Expansion kinetics of CAR19+ cells (red) and total cell number (black). Mean ± SD is represented. (C) Expansion kinetics of ARI-0001 cells (Total cell number) comparing healthy controls and different types of disease. Mean ± SEM is represented. (D) Percentage of CD3 and CAR19 positive cells as determined by flow cytometry. Mean ± SD is also indicated. Panels on the right show flow cytometry representative image corresponding to CAR19 and CD3 staining in ARI-0001-cell final products and Control T cells (Untransduced).

Figure 2

ARI-0001-cell potency. (A) Cytotoxicity assay after 4 h of ARI-0001 co-culture with NALM6 cells, at the indicated ratios. Mean ± SD of all 27 CAR T cell products is indicated. (#) Dashed line indicates minimum of ARI-0001-cell cytotoxicity level for a product to be considered valid. (B) IFNγ, TNFα, and GranzymeB levels measured in the supernatants of the cytotoxicity assays. E:T ratio 0 indicates no target cells. (*) indicates statistical significance, p < 0.05. (C) Comparison of ARI-0001 cytotoxic potential after 4 h of co-culture with NALM6 cells, at the indicated ratios. Mean ± SD is shown. “n.s.” indicates not statistically significant (Non parametric test). (D) Comparison of IFNγ, TNFα, and GranzymeB levels measured in the supernatants of the cytotoxicity assay at E:T ratio 1:1. “HD” indicates Healthy donors. “n.s.” indicates not statistically significant (Parametric test applied to IFNγ and TNFα and non-parametric test applied to GranzymeB).

Figure 3

ARI-0001 cell subset characterization. (A) CD4/CD8 ratio of apheresis products, after CD4-CD8 cell selection and of the final product. (B) CD4/CD8 ratio variation during cell expansion. Left panel corresponds to products with an initial ratio < 1. Right panel corresponds to products with an initial ratio > 1. (C) CAR19 transduction efficiency in CD4 and CD8 cells. Mean ± SD is shown. (D) Percentage of T-cell subpopulations within initial (CD4-CD8 cell selection) and final products (CAR– and CAR+ cells). (E) Representative flow cytometry plots of three different patients showing T cell populations in initial and final products. (F) Differences in MFI for CD45RA and CCR7 in initial and final products. Lower panel shows paired analysis for CCR7 MFI. (*) indicates statistical significance, p < 0.05. n.s. indicates not statistically significant.