Two studies have illuminated some of the molecular underpinnings of T-cell exhaustion. The first pinpoints the subset of exhausted T cells that revive upon PD-1 blockade. The second describes key metabolic deficiencies—restricted glucose uptake and mitochondrial dysfunction—that drive T cells to exhaustion.

Recent research has illuminated some of the molecular underpinnings of T-cell exhaustion. One study from Emory University in Atlanta, GA, pinpoints the subset of exhausted T cells that are revived upon PD-1 blockade. A second study, from the University of Pennsylvania in Philadelphia, describes key metabolic deficiencies—namely restricted glucose uptake and mitochondrial dysfunction—that drive T cells to exhaustion.

Both research groups studied mice infected with a chronic strain of lymphocytic choriomeningitis virus, a well-established model in which T-cell exhaustion was first documented nearly two decades ago. In that work, led by Rafi Ahmed, PhD, director of the Emory Vaccine Center, “we observed that cytotoxic T cells were present in the mice, but they couldn't proliferate robustly or kill well,” Ahmed explains. By showing that exhausted T cells could be reenergized with antibodies blocking PD-1 or PD-L1, his group helped lay the foundation for the development of immune checkpoint inhibitors.

Not all exhausted T cells respond to anti–PD-1 therapy, however, so in the new Emory study, Ahmed's team set out to identify molecular features distinguishing the T-cell subset that undergoes a proliferative burst upon checkpoint blockade. They uncovered a vital role for TCF1 in generating these cells. Besides expressing this transcription factor, the cells also had high levels of the chemokine receptor CXCR5 and various costimulatory molecules, including ICOS and OX40, which are essential for an effective immune response. This T-cell subset was located predominantly in the lymph nodes and spleen, and had a self-renewal capacity akin to that of stem cells. By contrast, exhausted T cells that failed to proliferate after PD-1 blockade lacked CXCR5 or costimulatory molecule expression. Instead of having stem cell–like characteristics, they were more terminally differentiated.

“We've provided one of the first clear definitions, at least in this model, of what it is that responds to anti–PD-1 therapy,” Ahmed says. “The proliferative burst from this T-cell subset with clearly defined markers is key to the therapeutic efficacy of checkpoint blockade.”

Meanwhile, John Wherry, PhD, director of UPenn's Institute for Immunology, had observed metabolic dysregulation in exhausted T cells years ago, “but at the time, we didn't have the tools to carefully dissect that finding,” he says. The present study's fresh exploration indicated that metabolic deficiencies were an early driver rather than a consequence of T-cell exhaustion, which Wherry's group had previously assumed.

In chronically infected mice, “we found that the T cells were metabolically confused,” Wherry says. Although the cells' metabolism was ramped up due to viral antigen persistence, their ability to import glucose had been downregulated through PD-1. “There was insufficient fuel to sustain such high metabolic activity, and this helped drive the cells to early exhaustion,” he explains.

Wherry's team also showed that the mitochondria in these T cells were unable to efficiently generate energy because PD-1 signaling was suppressing PGC1α, a coordinator of the pathway that helps maintain mitochondrial health. PD-L1 blockade reversed the exhaustion induced by restricted glucose uptake and mitochondrial dysfunction, improving the T cells' overall metabolic fitness, he adds.

Ahmed and Wherry want to see if exhausted T cells in human cancer share any of the markers and metabolic characteristics they've found in mice. If so, “we should be able to design immunotherapy combinations more rationally by pairing a PD-1 inhibitor with, say, an OX40 agonist,” Ahmed says. If TCF1 and CXCR5 mark exhausted-but-revivable T cells in human cancer, these cells should be easier to identify and track during clinical studies of checkpoint-blocking agents, Wherry observes. He's also interested in exploring pharmacologic manipulation of PGC1α or related pathways, to improve mitochondrial health and thereby enhance the revival of exhausted T cells.

“If you think of PD-1 blockade as a lever, we want to find ways to complement the effects it [the blockade] initiates and strengthen the torque on that lever,” Wherry says. –Alissa Poh

©2016 American Association for Cancer Research.
2016
American Association for Cancer Research.