Abstract

T cells may be genetically modified ex vivo to target specific antigens by retroviral transduction of genes encoding chimeric antigen receptors (CARs). We have previously constructed a CAR, termed 19z1, specific for the CD19 antigen expressed on most B cell malignancies. Human T cells modified to express the 19z1 CAR specifically eradicate systemic human CD19+ tumors in SCID-Beige mice. However, these models are limited by the xenogeneic nature of the human T cells and tumor cells and the immune compromised state of the host. Here, we studied the biology of adoptively transferred 19z1+ T cells in a syngeneic immune competent murine model designed to better mimic the clinical setting of patients with B cell malignancies. We utilized transgenic C57BL6 mice which lack expression of mouse CD19 (mCD19−/−) and have a single copy of the human CD19 (hCD19+/−) gene (C57BL6(mCD19−/− hCD19+/−)) kindly provided by Dr. T. Tedder, Duke University. These mice are functionally immune-competent with hCD19 expression restricted to the B cell population. To assess whether syngeneic 19z1+ T cells were capable of eradicating normal hCD19+ B cells, we infused C57BL6(mCD19−/− hCD19+/−) mice with either 19z1+ or control prostate specific membrane antigen-targeted (Pz1+) T cells. As assessed by flow cytometric analysis of peripheral blood, we neither found evidence of hCD19+ B cell aplasias in 19z1+ T cell treated mice nor were able to demonstrate the persistence of infused CAR+ T cells. To investigate whether the lack of 19z1+ T cell efficacy and persistence was due to an absence of homeostatic drive, we next lymphodepleted C57BL6(mCD19−/− hCD19+/−) mice with cyclophosphamide prior to T cell infusion. Mice lymphodepleted prior to 19z1+ T cell infusion demonstrated marked and sustained B cell aplasias when compared to lymphodepleted Pz1+ T cell and non-lymphodepleted T cell treated controls. Furthermore, while no CAR+ T cells were identifiable in the Pz1 and non-lymphodepleted control groups, 19z1+ T cells were consistently present in the peripheral blood of the cyclophosphamide pre-treated, 19z1+ T cell treated mice (3–5% of white blood cells). To assess the anti-tumor efficacy of the 19z1+ T cells, we next established a systemic tumor model utilizing mouse EL4 thymoma cells retrovirally modified to express hCD19 (EL4(hCD19)). C57BL6(mCD19−/− hCD19+/−) mice pre-treated with cyclophosphamide, subsequently infused systemically with EL4(hCD19) tumor, followed by systemic 19z1+ T cell infusion, had a significant survival advantage (80% survival at >120 days) over untreated controls or controls treated with Pz1+ T cells or 19z1+ T cells in the absence of lymphodepletion (0% survival). In conclusion, we have developed a syngeneic immune competent tumor model of hCD19 disease that is highly relevant to the clinical setting. Using this model, we demonstrate the significance of lymphodepletion on the prolonged in vivo persistence and anti-tumor efficacy of 19z1+ T cells. Data derived from this model will be correlated to findings obtained from a recently initiated clinical trial for patients with chronic lymphocytic leukemia, and will significantly impact the design of subsequent trials in the future.

Author notes

Disclosure: No relevant conflicts of interest to declare.