Expression of the immune checkpoint receptor TIGIT in Hodgkin’s lymphoma

Hodgkin’s lymphoma (HL) is a malignant transformation of B cell origin that accounts for about 20–30% of lymphomas in Western societies [1‐3]. Although the majority of Hodgkin’s lymphoma patients are curable with a multi-agent chemotherapy and/ or radiotherapy protocol, about 10–20% of patients develop therapy refractory disease [4]. A characteristic feature of HL is the presence of few malignant cells, including Hodgkin cells, Reed-Sternberg cells and histiocytic cells, in a high background of inflammatory cells. Current clinical trials provide first evidence that immune checkpoint inhibitors targeting the PD-1/PD-L1 axis may hold promise in these refractory or relapsed Hodgkin’s lymphomas [5‐7].

T cell immunoglobulin and ITIM domain (TIGIT), a co-inhibitory transmembrane glycoprotein of the poliovirus receptor (PVR)/−nectin superfamily, is another interesting candidate for novel checkpoint therapies [8, 9]. Using multiplex fluorescence immunohistochemistry, we have recently shown that TIGIT typically co-localizes with PD-1 on CD8+ cytotoxic T cells, CD4+ T helper cells and FOXP3 regulatory T cells [10]. Tumor associated lymphocytes expressing TIGIT have so far been demonstrated in acute myeloid leukemia, non-small cell lung cancer, colorectal carcinoma and melanoma [11‐13]. Although the downstream signaling cascade of TIGIT has not been clarified, there is evidence that TIGIT negatively regulates T cell activity through downregulation of T cell receptor expression [8, 14, 15]. In mouse models and ongoing clinical studies, blockade or ablation of TIGIT, alone or in combination with blockade of programmed cell death protein (PD-1), can restore tumor suppressive effects [11, 12, 16‐18]. These findings indicate that TIGIT, similar to PD-1, has a crucial role in inhibiting the tumor-directed immune response and, thus, might be a suitable and relevant target for novel immune-modulating therapies. Several drugs targeting TIGIT are currently under development [19].

As to yet, data on the possible role of TIGIT in Hodgkin’s lymphoma are lacking. Here, we made use of a microenvironment (ME) tissue microarray (TMA) that was constructed from 2 mm tissue punches each from 40 Hodgkin’s lymphomas and studied patterns of TIGIT and PD-1 expression by means of conventional bright field and multiplex fluorescence immunohistochemistry.

The data from this study demonstrate that TIGIT is frequently expressed in cells of the lymphocytic background accompanying HL and that the frequency and intensity of TIGIT expression varies markedly between individual patients.

All 40 HL analyzed in our study showed detectable TIGIT expression in all types of CD3 positive T cells including CD8 cytotoxic T cells, CD4 positive T helper cells, and FOXP3 positive regulatory T cells. This corresponds to the situation in other normal and inflamed tissues, where we also found that TIGIT expression was largely limited to these three major types of immune cells. In this earlier study, we also found that TIGIT expression often paralleled PD-1 expression in these cells. It was therefore expected that a large fraction (more than 80%) of TIGIT positive T cells also expressed PD-1 in our study. In normal lymph nodes and tonsils, TIGIT and PD-1 co-expression was found in 50–99% of T cells depending on the analyzed T cell subtype [10]. These findings are in line with a study describing that TIGIT (and PD-1) are constitutively expressed on about one third of circulating FOXP3 positive regulatory T cells and highly upregulated on regulatory T cells in an inflamed microenvironment [20, 21]. This suggests that the lymphocytic background of HL is affected by the same immunoregulatory mechanisms that also apply in inflammatory and cancerous tissues.

In contrast to many other studies that reported PD-1 staining in 20–60% of classical HL [22‐27], we detected PD-1 positive lymphocytes in all analyzed cases. Some of the discrepant findings may be related to different definitions of PD-1 positivity in earlier work, for example studies where PD-1 positivity was assumed only when more than 20% of cells [27] or more than 23 cells/mm² [26] had detectable PD-1 staining. While it is obviously impossible to explain all discrepant reports, we previously found that certain levels of TIGIT and PD-1 expression are characteristic features of lymphatic infiltrations in a wide range of inflamed and cancerous tissue [10].

The fluorescence approach enabled us to quantify the expression level of TIGIT and PD-1. We have recently shown that highest physiological TIGIT and PD-1 expression was found in follicular T helper cells residing at the periphery of germinal centers in lymph nodes and tonsils [10]. Given this consistently high expression levels and because of the easiness to identify follicular T cells, we have used these cells as a reference to compare the variable TIGIT and PD-1 expression of T cells located in other tissue areas. Accordingly, all measured intensity levels that were lower than 100% do, however, not imply reduced TIGIT and PD-1 expression of the respective cells. We recorded a stunning degree of TIGIT and PD-1 expression level variability across the different HL specimens analyzed. The much less variable expression levels of CD3 in our cancers argue against relevant fixation related variations in tissue immunoreactivity as a possible reason for the observed interindividual TIGIT expression differences. Biological reasons for the variable expression might include modulation of the immune environment by the Hodgkin and Reed-Sternberg (HRS) cells or the lymphocytic background itself. For example, it has been suggested that the expression of PD-1 on T cells is likely driven by constitutive upregulation of its ligands, PD-L1 and PD-L2, on HRS cells [28] and that HRS cells as well as some leukocytes (e.g. helper and regulatory T cells or macrophages) [29] can selectively express immunomodulatory proteins connected to PD-1 signaling such as galectin 1 and IL 10 [30, 31]. Hence, it is tempting to speculate that also the level of TIGIT expression may be influenced by regulatory molecules in a similar way as known from PD-1.

Notably, our series of HL included one case of nodular lymphocyte predominant Hodgkin’s lymphoma (NLPHL). A unique feature of NLPHL is that the lymphocytic predominant (LP) cells (previously known as “popcorn” cells) arise in a follicular like microenvironment that shares characteristics of germinal centers in lymph nodes and tonsils, such as a high density of follicular T helper cells [32, 33]. That highest TIGIT and PD-1 levels in our current study were also found in a HL subtype with a characteristic follicular architecture suggests an important function of these receptors in dense meshworks of leucocytes. Highest levels of TIGIT and PD-1 in tightly packed lymphocytic rosettes surrounding the neoplastic cells fits well to the concept of compensatory downregulation of excessive inflammatory reactions through immune checkpoint upregulation [34, 35]. Exceedingly high PD-1 expression levels in NLPHL has earlier been described and PD-1 expression analysis was suggested as a diagnostic feature for this rare HL subtype [36]. Our fluorescent measurement revealed that the TIGIT and PD-1 expression in individual T cells could reach 2 to 4-fold higher levels than those found in normal human tonsils.

Another feature of HL is that the tumor cells have developed mechanisms to escape the excessive immune-cell infiltration of the host. For example, the cellular landscape of HL is characterized by a paucity of specific cytotoxic T cells and natural killer (NK) cells [37]. Much hope is, therefore, put on novel therapeutic strategies modulating the immune system and its response to the disease, particularly in heavily treated relapsed or refractory HL where effective therapies are lacking. After promising results were reported from a first clinical trial [5] using the PD-1 inhibitor nivolumab in chemotherapy refractory HL, a number of follow up studies have confirmed the potential of immune checkpoint therapy in this disease [38‐41]. Phase I/II studies using nivolumab as a combination or second line therapy in combination with or after anti-CD30 treatment in patients with relapsed and refractory Hodgkin’s lymphoma reported over all response rates of 66–85% [38, 39]. Combining immunotherapies is another option to increase response rates. For example, combination of anti-PD-1 and anti-CTLA4 drugs has demonstrated additive efficacy in melanomas, non-small-cell lung cancer and renal cell carcinoma [6, 42, 43]. At present, several drugs directed against TIGIT are in preclinical pipelines [17, 19]. In our study, TIGIT and PD-1 expression was found in all analyzed HL and was not limited to a specific subtype. These findings identify HL as a disease entity where it could be interesting to determine whether TIGIT inhibitors alone or in combination with other immunotherapies and anti-CD30 treatment might be effective. It might be speculated, that the variable levels of TIGIT expression could correspond to variable responses to putative therapies targeting TIGIT.

The large tissue size TMA (ME-TMAs) enable the analysis for large enough tissue areas to study a tumor’s microenvironment under perfectly standardized experimental conditions. Having all tissue samples of a study on one glass slide ensures not only fully identical staining procedures for every tissue specimen but also identical section thickness and slide age [44]. This is important for experimental standardization as automated immunostainers are not yet perfectly dealing with the requirements of multiplex fluorescence immunohistochemistry.

In summary, the results of our study identify HL as a tumor type with frequent but variable expression of immune checkpoint receptors such as TIGIT and PD-1. Patients with HL might benefit from potential future therapies targeting TIGIT alone or in combination with other drugs.