Abstract 3760: Preclinical optimization of a low cost PiggyBac transposase (PB) generated CD19-specific chimeric antigen receptor T cell (CART19) product for a first in man trial using local hospital cell manufacture

BACKGROUND: CAR19 T cells show remarkable efficacy against relapsed/refractory B cell malignancies. PiggyBac offers a less complex and more cost effective means for generating CAR19 T cells compared to viral vectors typically used, so we aimed to use PiggyBac to generate a product suitable for translation to the clinic. METHODS: CAR19 T cells utilizing the CAR1928z construct have been found to perform well in vitro but had poor in vivo efficacy and persistence, most likely due to deleterious FcγR interactions via the IgG1 Fc-containing CAR spacer domain. We designed two new CARs and replaced the IgG1 Fc component of the spacer with a (G4S)3 linker. We used either CD28 or 4.1BB co-stimulatory domains (CAR19h28z and CAR19h41BBz, respectively), and generated CAR19 T cells using electroporation of PiggyBac transposon and transposase plasmids. Cells were expanded over 22 days through CD19 stimulation with IL-15 support. The two new CARs were compared to CAR1928z in vitro and their in vivo activity was assessed over 12 weeks in NSG mice xenografted with B-ALL. RESULTS: CAR19 T cell products generated using PiggyBac showed vigorous expansion (minimum 120-fold) and robust CAR expression (>80% T cells in final product). T cells generated using CAR19h28z displayed CD4 predominance (mean 85%) while CAR19h41BBz showed greater CD8 predominance (mean 62%). Naive and central memory T cells constituted 30 to 45% of CAR19 T cells in all products. IFN-γ production was specific to CD19+ targets and was demonstrated in both CD4+ and CD8+ CAR19 T cells. CD19 specific dose-dependent cytotoxicity was seen with all products, although to a significantly lower degree with CAR19h28z. In vivo, a single dose of either CAR19h28z or CAR19h41BBz T cells 1 week after B-ALL injection significantly delayed leukemia progression compared to untreated mice (median OS: 42 days for untreated vs >82 days [median not reached] for each CAR19 T cell product, P CONCLUSIONS: CAR19 T cells generated with PiggyBac, utilizing optimized CAR constructs without IgG1 Fc-containing spacers and with either CD28 or 4.1BB co-stimulatory domains, had potent activity against B-ALL in preclinical studies. We selected CAR19h41BBz to proceed to a first in man clinical trial of PiggyBac generated CAR19 T cells. The low cost and simplicity of our manufacturing process utilizing electroporation with PiggyBac elements and patient-derived materials means that decentralized, local hospital production is feasible. We predict that CAR T cell production with PiggyBac will greatly improve accessibility to CAR19 T cell therapy. Citation Format: David Bishop, Ning Xu, Sylvie Shen, Tracey O’Brien, David Gottlieb, Alla Dolnikov, Kenneth Micklethwaite. Preclinical optimization of a low cost PiggyBac transposase (PB) generated CD19-specific chimeric antigen receptor T cell (CART19) product for a first in man trial using local hospital cell manufacture [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3760. doi:10.1158/1538-7445.AM2017-3760