T cells modified to express tumour-directed chimeric antigen receptors (CARs) have shown anti-tumor efficacy in early phase clinical trials. There is no consensus yet about the optimal CAR T- cell dose needed for achieve optimal anti-leukemic activity. Early phase clinical trials have shown that the level of in vivo expansion of CAR T-cells rather than CAR T-cell dose determines the efficacy of CAR-T cell therapy. Here we identify factors that modify CAR-T cell expansion and anti-tumour activity in vitro and in vivo in a ‘humanised’ mouse model using CARs targeting CD19+ B-cell leukaemia (CAR19 T-cells). Human peripheral and cord blood T-cells were genetically modified using second generation of CAR19-T cells engineered to express CD3zeta and co-stimulatory CD28 domains cloned into PiggyBac-transposon vector. Antigen-specific stimulation using autologous mononuclear cells was used to enrich and expand CAR19 T-cells. CAR19 T-cells induced cytolysis of leukaemia cells in vitro and exhibited anti-tumour activity in vivo in xenograft mouse model of leukaemia. Antigen-specific stimulation induced rapid activation and proliferation of CAR19 T-cells, however, terminal effector differentiation and activated T cell death limited CAR19-T cell expansion. We have analysed CAR19 T-cell expansion using the panel of primary CD19+leukaemia cells expressing different levels of CD19 antigen. CAR19 T-cells expansion and cytolytic function correlated with the level of CD19 expression on target cells. In addition, increasing CD19 expression in leukaemia cells using hypomethylating agent 5-aza-2'-deoxycytidine (5-AZA) promoted antigen-specific recognition of target cells by CAR19-T cells and increased their expansion. Treatment with 5-AZA also increased CAR expression on the CAR19 T-cells, however, cytotoxicity towards CAR19 T cells mediated by the agent delayed CAR19T-cell expansion. Interestingly, allogeneic stimulation of CAR19 T-cells using CD3/CD28 activation of TCR signalling or co-culture with allogeneic bone marrow stroma cells promoted CAR19 T-cell expansion, however, addition of antigen-specific target cells (CD19+leukaemia cells) attenuated allogeneic expansion indicating the antagonism between antigen-specific and allogeneic CAR19 T- cell activation.

Effector to target (E:T) cell ratio appears to be a strong determinate of CAR19 T-cell expansion and cytolytic activity towards cancer cells. CAR19 T-cells used at high E:T ratios exhibit higher cytolytic activity compared to effector cells used at a low E:T. Cell division analysis demonstrats lower proliferation of CAR19 T-cells when used at a high E:T compared to those used at low E:T ratio. The repetitive antigen-specific activation of CAR19 T-cells occurring at a high density of target cells (low E:T ratio) promotes proliferation, however, also acts to induce terminal effector differentiation resulting in rapid T cell extinction due to activated T cell death. Thus the excessive proliferation and rapid extinction of CAR19 T-cells may account for the low efficacy of CAR19 T- cell therapy when given at low E:T ratios. Similar effects were observed in a stem cell reconstituted mouse model where stem cell-derived CD19+ B cells were targeted by CAR19 T-cells. A single infusion of CAR19-T cells infused at a high E:T ratio in mice with low B cell engraftment resulted in efficient B cell depletion while the same dose of CAR19 T-cells infused to mice with high B cell engraftment level(low E:T ratio) was less efficient. These findings show that rapid exhaustion of effector cells used at low E:T ratio prevents long lasting anti-tumor effects justifying the need of pre-conditioning in patients with advanced tumors.

.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution