PROJECT SUMMARY/ABSTRACT
The goal of this F31 application is to investigate intracellular signaling events that promote death of tumor-specific
T cells in vivo. To achieve appropriate immune homeostasis, populations of effector T cells and regulatory T
cells must be tightly controlled. Cancer is one disease state where regulatory T cells can suppress effector T
cells and prevent elimination of tumors. To overcome immune suppression and tolerance that occurs during
cancer, adoptive T cell therapies have been introduced clinically, including adoptive transfer of CAR (chimeric
antigen receptor) T cells. CAR T cells contain TCRs with tumor antigen specificity, a signaling component of the
TCR (CD3 zeta chain) as well as one or more co-stimulatory domains (such as CD28). CAR T cells are efficiently
activated and have been very effective for the treatment of some cancers, especially hematologic cancers. One
problem, however, is that CAR T cells often do not persist in vivo, especially for solid tumors, and have been
shown to undergo activation induced cell death (AICD). Currently, knowledge of the mechanisms of CAR T cell
death and strategies to promote CAR T cell survival are limited. We discovered a novel form of AICD that occurs
when effector T cells are stimulated on anti-CD3 and anti-CD28 coated plates. This process is dependent on
p53 and is termed PICA (p53-induced CD28 dependent T cell apoptosis). We found that that effector T cells die
during PICA, but that Foxp3+ regulatory T cells are resistant to PICA through TGF-ß signaling. Our data also
show that effector CAR T cells stimulated with large numbers of antigen-expressing cells (>1:20) undergo cell
death. The CAR T cells that survive this high dose antigen stimulation are Foxp3+. We therefore hypothesize
that CAR T cells undergo PICA when stimulated with high dose antigen. Elucidating this mechanism of CAR T
cell death can help us better design CAR T cells that persist and kill more efficiently in vivo. While investigating
PICA in effector T cells, we discovered that the protein RasGRP1, a Ras activator, is highly elevated in effector
T cells after repeated stimulation. Tregs maintain low levels of RasGRP1, and this is dependent on TGF-ß
signaling. We currently do not know the function and phenotype of CAR T cells that survive high dose antigen
stimulation from solid tumors in vivo, and if RasGRP1 promotes CAR T cell death in vitro and in vivo. Aim 1 will
determine if CAR T cells that survive stimulation by solid tumors in vivo are Foxp3+ and functionally suppressive,
due to TGF-ß signaling. Aim 2 will determine if RasGRP1 promotes death in CAR T cells, and if engineering a
CAR with dominant negative RasGRP1 can rescue CAR T cell death and promote tumor killing in vivo. Overall,
we expect this study to illuminate mechanisms of CAR T cell death, which can aid in promoting CAR T cell
survival in solid tumors. Additionally, this study will test the use of a novel CAR construct, which we predict will
enhance CAR T cell survival and could inform future design of CAR T cells that are used clinically.