Next-Generation CAR T Cells Counter Exhaustion

T-cell exhaustion continues to limit the long-term durability of cellular immunotherapies. In a new model for studying this mechanism of immune failure in human cells, scientists have discovered a transcription factor that, when overexpressed, helps make chimeric antigen receptor (CAR) T cells resistant to exhaustion (Nature 2019;576:293–300).

The study authors—through their company Lyell Immunopharma—hope the findings will help inspire a new generation of exhaustion-proof adoptive cell therapies for solid tumors. “There certainly are multiple barriers to overcome to get to curative cell therapies for solid tumors, and we believe exhaustion is one of those major barriers,” says Rachel Lynn, PhD, associate director of research at Lyell, who conducted the research at Stanford University School of Medicine in California.

Her team's new CAR T-cell model “captures some of the downstream biology that's active in these cells and offers a great way to test new hypotheses so we can eventually make better exhaustion-resistant cells,” she adds.

Working with Crystal Mackall, MD, Lynn and her colleagues had previously engineered a CAR with antigen-independent signaling, but they wanted to boost its activity levels further. So they swapped in a different single-chain variable fragment, while keeping the same hinge and costimulatory domains. In this way, they obtained a receptor that, when transduced into healthy T cells, displayed more severe signs of exhaustion, including increased expression of inhibitory receptors and diminished capacity for cytokine signaling, thereby hastening T-cell dysfunction and impairing antitumor effects.

“There haven't been good models of human T-cell exhaustion,” notes Mackall, a cofounder of Lyell. “So, we created this reductionist model of exhaustion—which is, by design, all about excess signaling.”

In these cells, the researchers also noticed widespread epigenetic changes, including less densely packed chromatin near exhaustion-associated genes and a surfeit of one particular type of regulatory binding site in those regions. The binding sites corresponded to those for transcription factors of the AP1 family. One, known as c-Jun, helps drive expression of IL2, a cytokine involved in promoting T-cell effector function and keeping exhaustion at bay.

The researchers overexpressed c-Jun in their cells and found it could prevent exhaustion. They tweaked the CAR construct to boost c-Jun activity and observed enhanced expansion potential, improved functional capacity, diminished terminal differentiation, and increased potency against tumor cells with low antigen density. What's more, the cells displayed greater tumor control in mouse models of leukemia and osteosarcoma, demonstrating that “c-Jun overexpression can enhance the antitumor activity of CAR T cells in the solid-tumor setting,” says Tony Tiganis, PhD, of Monash University and the Peter MacCallum Cancer Centre in Melbourne, Australia.

In the osteosarcoma xenograft model, Lynn says the cancers appeared to be “on their way to full regression,” but her team never tracked the tumors long enough to demonstrate complete clearance, and a propensity for GVHD made it impossible to gather long-term survival data. “By contrast,” Tiganis says, “our studies have shown that deleting PTPN2 in CAR T cells allows for the clearance of antigen-bearing tumors and the marked extension of life span.”

In his latest paper, Tiganis demonstrated that the PTPN2 deficiency enhanced immunosurveillance and CAR T-cell function, allowing mice with implanted breast tumors to live 6 months or longer, instead of dying within 50 days (EMBO J 2019;e103637). He and his colleagues additionally showed that the genetic manipulation promoted recruitment of the CAR T cells to tumors expressing particular chemokines.

“In this way,” Tiganis says, “targeting PTPN2 can overcome key limitations for CAR T-cell therapy, ensuring both homing and sufficient activation in the immunosuppressive tumor microenvironment.” Simultaneous overexpression of c-Jun and deletion of PTPN2 could have even more pronounced effects, he adds. –Elie Dolgin

Expression of IL2 (above) is driven in part by c-Jun. IL2 helps promote T-cell effector function and limit T-cell exhaustion.