Single Cell Analysis for Mapping Heterogeneous Activation States of Chimeric Antigen Receptor T Cells.

9780438269552

Ann Arbor : ProQuest Dissertations & Theses, 2018

1 online resource (140 p.)

Access is available to the Yale community.

T cells collected from patients are engineered ex vivo to express chimeric antigen receptors (CARs) that recognize specific tumor cell surface antigens and actively kill tumor cells independent of major histocompatibility complex (MHC) recognition and signaling. Adoptive transfer of anti-CD19 CAR-T cells has shown remarkable efficacy in treating patients with B cell acute lymphoblastic leukemia (B-ALL), chronic lymphocytic leukemia (CLL), and other indolent lymphomas. Despite the demonstrated success of this "living" drug, its use can be associated with significant risks for the patients. First, there is substantial variability in therapy success and a patient's response. Second, this powerful therapy could be potentially life threatening due to severe adverse effects such as cytokine release syndrome (CRS) and neurotoxicity. Currently, only a limited number of tools are used to predict overall clinical efficacy and safety of CAR-T therapy but they fail to capture the full heterogeneous landscape of pre-infusion CAR-T product. In my thesis project, a panel of single-cell analysis technologies were combined to interrogate the behavior of anti-CD 19 CAR-T cells upon antigen-specific stimulation. These technologies include: live-cell imaging of CAR-T cell induced cytolysis, highly multiplexed measurement of cytokine secretion in single cells, and high-throughput transcriptome sequencing of single cells. This study provides the first comprehensive portrait of single-cell level transcriptional and secretory signatures of third-generation anti-CD19 4-1BB/CD28/CD3zeta CAR-T cells. First, it was shown that both CD4+ `helper' CAR-T cells and CD8+ cytotoxic CAR-T cells are equally effective in directly killing target tumor cells and their cytotoxic activity is associated with the elevation of a range of inflammatory and effector type cytokines. Second, both Th1 and Th2 responses were observed and confirmed in the CD4 subpopulation by the expression of master transcription factors TBX21 (T-bet) and GATA3 as well as signature secreted cytokines IFNgamma, TNFalpha, IL5, and IL13. Regulatory T cell (Treg) activity, although observed only in a small fraction of cells, was identified and associated with elevated TGFbeta and IL10 production. However, rather than conforming to stringent Th1, Th2, and Treg only responses, the same CAR-T cells showed multiple responses upon antigen stimulation. These results support the hypothesis that `polyfunctional' CAR-T cells that co-secrete multiple proteins (e.g. cytokines) correlate with stronger, more efficient objective response of patients in clinical trials. Third, the cytokine response is minimally dependent on differentiation status although all major differentiation states such as naive, central memory, effector memory, and effector were detected. Finally, the microchip platform uniquely enabled the tracking of CAR-T/tumor cell interaction in real time followed by multiplexed protein secretion assay or single-cell-level transcriptome sequencing. This potentially allows for the discovery of mechanisms associated with intrinsic resistance and the possibility to further improve the design of CAR-T cell in a patient-specific manner including the use of multiplexed genome editing such as CRISPR/Cas9. In summary, my thesis work demonstrated an integrated single-cell omics approach to characterize the heterogeneous activation states of CAR-T cells and to reveal new insights in CAR-T cell biology, which may facilitate the development of more accurate and informative assays for CAR-T product quality control and companion diagnostics of CAR-T therapies.

Books / Online / Dissertations & Theses

English

January 09, 2019

Thesis (Ph.D.)--Yale University, 2018.

Yale University.