Awakening exhausted NK cells in lymphomas

In this issue of Blood, Vari et al describe a novel immune evasion strategy in classical Hodgkin lymphoma (cHL) and diffuse large B-cell lymphoma (DLBCL) mediated by the expansion of an exhausted programmed cell death protein 1 (PD-1) CD3^-veCD56^brightCD16^-ve natural killer (NK) cell population.¹ 

PD-1 is the most clinically relevant immune checkpoint molecule to date.²  Although it is physiologically expressed on activated T cells to prevent autoimmunity and maintain peripheral tolerance in healthy individuals, engagement of PD-1 with its ligand PD-L1 expressed on cancer cells leads to T-cell exhaustion and tumor immune evasion.²  Monoclonal antibodies targeting PD-1 have shown significant clinical activity in solid tumors, cHL, and, to a lesser extent, in DLBCL by enhancing immune-mediated antitumor response.^3-5  Lymph nodes from cHL patients are characterized by a small percentage of malignant Reed-Sternberg (RS) cells within an extensive but ineffective inflammatory and immune-cell infiltrate. RS cells express high levels of PD-L1 because of genetic alterations in most cases.⁴  Unlike cHL, lymph nodes from DLBCL patients demonstrate a diffuse proliferation of large lymphocytes with minimal immune infiltrate. PD-L1 expression is reported in ∼10% to 30% of DLBCL cases, possibly explaining the differences in treatment response to PD-1 targeting antibodies between cHL and DLBCL.⁶  In addition, PD-L1 expression appears to be higher in activated B-cell DLBCL and is associated with inferior overall survival.^5,6  In both cHL and DLBCL, PD-L1 is also expressed by nonmalignant cells in the tumor microenvironment.

Most of the published work thus far focuses on the role of immune checkpoint molecules in modulating T-cell expansion and function in B-cell lymphoma, with significantly less information available on the relevance of the PD-1/PD-L1 network in other immune cell subsets such as NK cells. NK cells play a major role in cancer immune surveillance and are conventionally classified in 2 main subsets identified as CD56^brightCD16^-ve and CD56^dimCD16⁺.⁷  Whereas CD56^brightCD16^-ve NK cells represents a less mature population, CD56^dimCD16⁺ NK cells, the major circulating subset in the peripheral blood of healthy individuals, are able to spontaneously kill tumor cells and mediate antibody-dependent cell-mediated cytotoxicity (ADCC).⁷ 

Vari et al showed through flow cytometry analysis that CD56⁺ NK cells were reduced in the peripheral blood of treatment-naïve cHL and DLBCL patients compared with healthy individuals, with no significant differences between the 2 lymphoma subtypes. In addition, the proportion of circulating CD56^brightCD16^-ve NK cells from these patients was significantly elevated compared with controls, and no differences were noted in the lymph nodes. Phenotypic profiling of circulating NK cells showed that C-C chemokine receptor type 7 (CCR7) expression was reduced on CD56⁺ NK cells whereas PD-1 was expressed at a significantly higher level on CD56^brightCD16^-ve compared with controls, with this effect being more pronounced in cHL than DLBCL. CD16, also known as FcϒRIII, binds to the fragment crystalizable (Fc) region of immunoglobulin G antibodies. Lack of this receptor limits the ability of NK cells to activate ADCC, whereas reduction in CCR7 expression affects NK cell homing and trafficking within lymph nodes.⁷  These results identify a novel immune evasion strategy in cHL and DLBCL mediated by the expansion of an exhausted PD-1–enriched CD56^brightCD16^-ve NK cell population. In addition, multispectral immunofluorescent imaging confirmed previous findings that cHL/DLBCL biopsy specimens are enriched in tumor-associated CD14⁺CD68⁺CD163⁺ monocytes/macrophages that express high levels of PD-L1 and PD-L2.^8,9  Interestingly, monocyte depletion of peripheral blood mononuclear cells from healthy donors did not affect NK cell activation in coculture experiments, whereas the same experiment resulted in marked activation of CD16^-ve NK cells from both cHL and DLBCL patients, suggesting a monocyte-mediated suppressive role on NK cells that can be overcome by PD-1 blockade (see figure).

This study represents a step forward in our understanding of the complexity of PD-1/PD-L1 dynamics in lymphoma patients and highlights a potential novel immune evasion mechanism mediated by exhausted PD-1⁺CD56^brightCD16^-ve NK cells. However, several questions arise that require further investigation. This work suggests that B-cell lymphoma cells mediate the changes in the circulating NK population, and authors speculate that tryptophan depletion caused by expansion of monocytic myeloid–derived suppressor cells in these patients may affect NK cell proliferation and function. Without mechanistic studies to support this idea, it is intriguing to hypothesize that a defect in NK cell immune surveillance could be the primary defect that ultimately leads to lymphoma development. Although cHL and DLBCL are very distinct entities, biologically and clinically, major histocompatibility complex (MHC) class I expression is frequently lost in both because of mutations in the β2 microglobulin gene (B2M) in most cases.¹⁰  Loss of MHC class I provides an immune escape mechanism to CD8⁺ cytotoxic T cells. However, lymphoma cells lacking MHC class I are normally targeted by NK cells for cytolysis, suggesting a potential role for PD-1 blockade in B-cell lymphoma with loss of MHC class I through immune restoration of exhausted NK cells.

The data presented by Vari et al will require further validation on a larger number of patients with more homogenous characteristics. Although their findings support the role of exhausted NK cells in establishing a tumor friendly environment in cHL and DLBCL, they also support NK cells as just one of the “exhausted” immune players given the different response to microenvironment manipulation in cHL and DLBCL. In addition, it will be important to continue the functional characterization of novel and existing checkpoint molecules for the development of rational combination therapeutic strategies.