Forum focuses on dysfunctional, exhausted T Cells


Tumor-infiltrating T cells are impaired in a range of essential functions. Although development of therapeutic intervention to block the interaction of PD-1 with its ligand and release the brake on the T cells to attack cancer cells has been successful, there is a lack of knowledge about the basis for the T-cell functional impairments.

During the Monday, April 12, Forum Are Antitumor T Cells Exhausted or Dysfunctional? Does it Matter? W. Nicholas Haining, BM, BCh, Merck, Inc., moderated a discussion with several leaders in the field who presented key data that should help shape the understanding of T-cell function against cancer. Registrants can watch a replay of the session through June 21, 2021.

Road to dysfunction

Andrea Schietinger, PhD
Andrea Schietinger, PhD

Andrea Schietinger, PhD, Memorial Sloan Kettering Cancer Center, discussed what she called the road to T-cell dysfunction. Initially, tumor-specific T cells might not be aware of malignant lesions early during tumorigenesis and can be called “ignorant,” Schietinger said. This ignorance will allow the tumor to continue to grow. Eventually, a sufficient mass of cancer cells and tumor antigens are formed and presented in tumor-draining lymph nodes. Naïve tumor-specific T cells will eventually encounter antigen presenting cells.

Early activation drives T cells into anergy or “early dysfunction” in contrast to acute or chronic infection, where pathogenic antigens appear acutely and in high levels in a stimulatory context, which leads to optimal priming of naïve pathogen-specific CD8 T cells.  As cancer progresses and more tumor antigen is expressed, T cells within these progressing tumors receive continuous antigen and TCR stimulation, which drives the T cells into deep, profound late dysfunctional state.

Tumor-specific T-cell differentiation likely evolves in several phases. Early dysfunctional T cells could regain effector function. Schietinger referred to this as a plastic dysfunctional state in contrast to later dysfunctional T cells that are resistant to reprogramming, called fixed dysfunctional state. In their studies, Schietinger and colleagues were able to identify biomarkers and transcription factors associated with each of these states.

Functionally adapted

Daniela Stefanie Thommen, MD, PhD
Daniela Stefanie Thommen, MD, PhD

Daniela Stefanie Thommen, MD, PhD, The Netherlands Cancer Institute discussed the need for a better understanding of the heterogeneity of tumor-infiltrating T cells and the functional consequences of such heterogeneity in human cancers.

Thommen and colleagues found that PD-1T tumor infiltrating lymphocytes (TILs)—a transcriptionally distinct CD8 TIL pool with tumor-specific high PD-1 expression—acquire a new function in the tumor microenvironment: They express CXCL13 mRNA, which is usually not produced by CD8 T cells. This suggests that despite their impaired capacity to secrete classical effector cytokines, PD-1T TILs might display an altered function in the tumor microenvironment.

CXCL13 has been shown to be important for maintenance of lymph follicles. Researchers found that these cells predominantly localize in tertiary lymphoid structures (TLS) and hardly outside the rest of the tumors. This suggests a link between tumor reactive dysfunctional cells and the formation or maintenance of TLS.

Based on this high tumor activity, they looked into whether these cells could identify patients that benefit from PD-1 blockade. PD-1T TILs quantified in pretreatment biopsies of a pilot cohort of 21 patients with non-small cell lung cancer  showed that responding tumors had higher number of PD-1T TILs. This was associated with better survival.

The data does not show whether PD-1T TILs can be reactivated by PD-1 blockade, Thommen said, but it does suggest that they might be an indicator for the presence of a tumor-specific T-cell pool.

“A better understanding of the distinct role of T cells along the dysfunctional gradient should help us to better target these cells and develop treatments that can restore all components and that are crucial for an effective anti-tumor response,” Thommen said.

T-cell exhaustion

E. John Wherry, PhD
E. John Wherry, PhD

John Wherry, PhD, University of Pennsylvania, explained T-cell exhaustion. There are a lot of terms for T cells with altered function, Wherry said. A key question remains: Is everyone talking about the same thing?

Wherry discussed T-cell exhaustion as an adaptation to constant or frequent stressful stimulation and whether, if conditions are stressful enough, the population of exhausted T cells may erode over time.

“Exhausted T cells have adapted to withstand the strong or constant stimulation from frequent TCR stimulation from antigen,” Wherry said. “Transient loss of one negative regulatory pathway, PD-1, can drive temporary proliferative expansion, but eventually the cells burnout.”

That means that exhausted T cells probably engage multiple mechanisms to balance signal strength with survival and durability over time.

Exhausted CD8 T cells are an epigenetically distinct lineage of mature T cells, Wherry said, and the nuclear factor TOX has been identified as a key programmer of these epigenetic changes. This means other cells have had to solve for some of the things that exhausted T cells might face. For example, how do other tissues solve the problem of tissue renewal and formation of cells that have different roles in the tissues’ response?

Wherry noted that a framework to examine exhausted T cells could help identify context-specific biology, and it might be important to harness aspects of exhaustion biology instead of completely avoiding it.

CAR T-cell exhaustion

Rachel Lynn, PhD
Rachel C. Lynn, PhD

Rachel C. Lynn, PhD, of Lyell Immunopharma, discussed whether exhaustion matters in CAR-T cell dysfunction. When one thinks about exhaustion, Lynn said, they should think about it as a dysfunctional epigenetic state with chronic activation stimulation the root cause.

Lynn helped develop a model of T-cell exhaustion based on observation of GD2-directed CAR, which seems to spontaneously cluster on the cell surface. This manifests as a spontaneous, chronic T-cell activation signal driven through the CAR. Lynn tweaked the model and showed that chronic T-cell activation can be driven via a cross-linking CAR (HA-GD2), and this was sufficient to reproduce many functional and phenotypic characteristics of exhaustion.

This simple in vitro model can be used to rapidly and reproducibly examine biology behind T-cell exhaustion as well as rapid evaluation of novel engineering strategies that can potentially be used to overcome exhaustion induced dysfunction.