Diffuse Large B-Cell Lymphoma Is Infiltrated with Functional CD8⁺ T-Cells Lacking the Hallmarks of Exhaustion

Background. B-cell non-Hodgkin lymphomas (NHL) are hematologic malignancies that arise in the lymph node but are not cleared by the immune cells present. The failure of anti-tumor immunity may be due to immune checkpoints such as the PD-1/PD-L1 axis, which can cause T-cell exhaustion. In contrast to Hodgkin lymphoma, checkpoint blockade in NHL has showed limited efficacy. Here we demonstrate that T-cells in DLBCL do not exhibit an exhausted phenotype which may explain the poor response to immune checkpoint inhibitors.

Results.In order to better understand how the tumor microenvironment may impact NHL, we performed an extensive characterization of malignant and non-malignant human lymph nodes using high dimensional flow cytometry. We compared follicular lymphoma (FL), diffuse large B-cell lymphoma (DLBCL), and non-malignant reactive hyperplasia (RH). Using the unsupervised clustering algorithm Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP), we identified several T-cell populations that were altered in NHL compared to RH. These included follicular helper T-cells, regulatory T-cells, CD4⁺ PD-1⁺ T-cells, and CD8⁺ PD-1⁺ T-cells. Notably, DLBCL was highly enriched with CD8⁺ PD-1⁺ T-cells. Given the important role of PD-1 in regulating T-cell exhaustion, we thus hypothesized that DLBCL is infiltrated with exhausted T-cells. Consistent with this, we demonstrated that the CD8⁺ T-cells also expressed other exhaustion markers and could be divided into single positive for PD-1; double positive for PD-1 and TIM-3; and triple positive for PD-1, TIM-3, and CTLA-4.

Given the large expansion of CD8⁺ PD-1⁺ T-cells in DLBCL, we restricted further analysis to this histology. Since CD8⁺ T-cells are an important part of anti-tumor immunity, we further analyzed these cells to determine if PD-1 could serve as a potential therapeutic target. We first performed in vitro stimulation with PMA/ionomycin to determine the cytokine production capacity of CD8⁺ cells. Compared to PD-1^- cells, CD8⁺ PD-1⁺ cells retain their production of IFNg but have decreased capacity to produce IL-2. They also express lower levels of IL-7R (CD127) and lack CD45RA consistent with an effector phenotype. Additionally, as they acquire TIM-3 and CTLA-4, they make increasing amounts of granzyme B and perforin and exhibit greater degranulation capacity. Compared to PD-1^- cells, PD-1⁺ cells also have no baseline defects in apoptosis or proliferation.

Since the cohort appeared to be infiltrated with highly activated CD8⁺ T-cells, we next examined whether this could be suppressed by PD-1 signaling. We identified a single lymphoma in our cohort that highly expressed PD-L1. Despite this, the CD8⁺ T-cells retained their ability to produce cytokines. Together, these data suggesting that CD8⁺ T-cells in DLBCL lack many hallmarks of exhaustion. Additionally, it suggests that PD-L1 expression by a lymphoma is insufficient by itself to cause T-cell exhaustion.

Conclusion. Our work may explain the failure of single-agent immune checkpoint inhibitors in the treatment of DLBCL. Accordingly, functional differences of CD8⁺ T-cells in DLBCL may inform different therapeutic targeting strategies.

Disclosures. A.K.G. reports grants and nonfinancial support from Teva, Bristol-Myers Squibb, Merck, Takeda, TG Therapeutics, and Effector; grants, personal fees, and nonfinancial support from Seattle Genetics, Pfizer, Janssen, Gilead, Spectrum, Amgen and Incyte; personal fees from Aptevo, BRIM Bio, Seattle Genetics, and Sanofi.