Immune escape mechanisms in Hodgkin’s disease

Annals of Oncology 9 (Suppl. 5): S21-S24. 1998. 1998 Khmer Academic Publishers. Printed in the Netherlands.

Review Immune escape mechanisms in Hodgkin's disease S. Poppema, M. Potters, L. Visser & A. M. van den Berg Department of Pathology & Laboratory Medicine, University Hospital Groningen. The Netherlands

Results and discussion: The T cells surrounding RS cells have an immuno-phenotype and cytokine production capabilBackground: The nodular sclerosis and mixed cellularity sub- ity consistent with a Th2-type response. RS cells express several types of Hodgkin's disease are histologically characterised by a members of the TNF receptor family such as the FAS ligand small population of neoplastic cells, the so-called Reed-Stern- (CD95L) that may induce apoptosis of activated, FAS expressberg cells and their mononuclear variants (RS cells) and an ing, CD8+ T cells and NK cells. The RS cells also produce extensive admixture of other cell types including lymphocytes, TGFP and interleukin-10 that may downmodulate the Thl plasma cells, eosinophils, and histiocytes. The nature of this response. In addition, the Reed-Sternberg cells produce the infiltrate is largely known, but the mechanisms and functional chemokineTARC that could lead to the specific attraction of a effects are not. The small lymphocytes immediately surround- Th2 T-cell subset. ing the RS cells are mostly CD4+ T cells that express early Conclusion: RS cells have several mechanisms that may activation markers. The absence of prominent specific cyto- allow it to escape an effective immune response. The relative toxic T cell or natural killer (NK) cell populations seems to contributions of each of these and other potential mechanisms argue against a Thl-type response, whereas the sometimes are not yet known. prominent admixture of plasma cells and eosinophils is suggestive of a Th2-type response. Enrichment of the CD4 T-cell population may result from selective influx of CD4 T cells or Key words: Hodgkin's disease, Th2, chemokine, TARC, FAS, from selective depletion of CD8 cells and NK cells. FASL,TGFy, IL-10 Summary

The lymphocytes in close vicinity of HRS cells are almost exclusively positive for CD4, while essentially no CD8+ or NK-cells are present in this area [1]. These T cells are not clonal and are not part of the neoplastic population of RS cells which have no B- or T-cell markers but frequently do have immunoglobulin gene rearrangements [2]. One possible reason for the presence of an extensive inflammatory infiltrate would be an immune response against the neoplastic cells. In light of the presence of the EB virus [3, 4] and at least some EBV-induced membrane antigens (LMP1, 2) in the RS cells of many cases of Hodgkin's disease, this would be a possibility [5]. The increase of CD4+ cells in Hodgkin nodes is associated with a decrease in CD4+ cells in the circulation and predominantly reflects increased influx of mature CD4+ T lymphocytes into the involved tissues [6]. Results of imaging with indium-labelled lymphocytes showing positivity in involved nodes support this concept [7]. Activated CD8+ T cells, as determined by granzyme positivity, have been described in some cases but not immediately surrounding the HRS cells. Surprisingly, the presence of these granzyme positive lymphocytes was associated with an unfavourable prognosis [8].

The CD4+ T cells that surround the RS cells are in a state of activation as indicated by the presence of several activation associated surface markers, including CD38, CD69, CD71, and HLA class II [9 10]. A considerable number of these lymphocytes haveTIA-1 positive granules, but the significance of this finding is not clear. There are no indications that the lymphocytes are of clonal origin or have a restricted TCRP-chain gene variable region repertoire [11, 12]. The lymphocytes lack expression of another activation marker, CD26 (dipeptidylpeptidase IV) [13], a surface molecule involved in co-stimulation of T lymphocytes [14]. Further characterisation of CD4+ T cells derived by flow cytometry from lymph nodes of classical as well as NLP Hodgkin's disease, revealed that they are predominantly CD45RO+/ CD45RBdim, suggesting an activated/memory Th2 phenotype [15]. When the total lymph node cell population is optimally stimulated in vitro with phorbolester and ionomycin IL-2, IFNy as well as IL-4 are produced. When the CD4+/CD26- lymphocytes that immediately surround the HRS cells are sorted from Hodgkin lymph nodes, these cells do not produce IL-2 but do produce IFNy and increased amounts of IL-4 when optimally stimulated [15, 16]. In vitro, naive CD4+ T cells can be primed for a Th2 response by anti-CD28 without anti-CD3 in the pres-

Downloaded from http://annonc.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 13, 2015

Characterisation of the cells surrounding the RS cells

22 ence of IL-2. Approximately 30% of T cells primed in this way express CD57 [17]. CD57+ Th2-primed clones produced three times as much IL-4 and IL-5 than the CD57- Th2-primed clones and also produced IFNy. Interestingly, the T cells in the nodular lymphocyte predominance subtype have a CD57+CD4+ phenotype [10, 18]. Improper stimulation of the T cells by co-stimulatory molecules such as B7.1 and B7.2 on the RS cells [19] could lead to this unusual Th2-like cytokine production. This may explain the observed Th2-immunophenotype of the T cells in Hodgkin's disease (CD45RBdim/CD26-), but also the absence of a cytotoxic response against RS cells.

As mentioned earlier, the T cells surrounding the RS cells in the classical types of Hodgkin's disease do not express CD26, although more than 60% of normal peripheral blood and lymph node T cells are CD26+ [9]. CD26 physically interacts with adenosine deaminase (ADA) and with CD45R0, both of which are important in immune response [13, 14]. Normal CD26- T cells become CD26+ by stimulation with antigens/mitogens under physiological conditions [20, 21], but the CD26cells from Hodgkin lesions remain negative after stimulation [9]. This indicates that the absence of CD26 is potentially relevant with respect to the impaired immune response observed in Hodgkin's disease. Selected CD26- lymphocytes from cases of classical Hodgkin's disease can be stimulated in vitro to produce IFNy and IL-4, but not IL-2 [9]. The inability to produce IL-2 and reduced proliferation are the hallmarks of anergic T cells. Anergic T cells can be obtained by several mechanisms: lack of co-stimulation [22, 23], activation by superantigens [24], or the effect of certain cytokines (IL-10, TGFP) [25, 26]. Thus, a possible way for HRS cells to escape cytotoxic killing is by induction of anergy in T cells. The anergic state of the lymphocytes is probably not the result of a lack of co-stimulation by CD80 (B7.1), CD86 (B7.2), and various other adhesion/co-stimulatory molecules such as CD58 (LFA-3) and CD54 (ICAM-1), because these molecules are highly expressed on HRS cells [19, 27]. HRS cells are capable of producing a wide variety of cytokines including IL-10 and TGF(3 [27-29]. IL-10 and TGFp are known to be capable of anergy induction [3032]. Supernatant from Hodgkin cell line L428 inhibits CD25 expression and IL-2 production when peripheral blood mononuclear cells are stimulated with anti-CD3, while the cells still become CD69 positive. This indicates that a soluble factor is responsible for the improper activation [32]. L428 is known to produce TGF|3 and IL-10 [33], and depletion of TGF(3 from L428 supernatant completely prevents the inhibitory effect [32]. Adding the removed TGF(3 to RPMI gave a similar

Possible role for members of theTNFR/NGFR and TNFL superfamilies CD30, now identified as member of the TNFR family, wasfirstdiscovered with a monoclonal antibody prepared against the cell line L428 that showed a consistently high expression on HRS cells [36]. Further investigations revealed that CD30 expression is not restricted to HRS cells alone, but is also expressed on activated Tand B lymphocytes. The abundant presence of CD30 on HRS cells and the absence in most non-Hodgkin lymphomas suggests that it may play an important role in the development of Hodgkin's disease. The TNFR superfamily has gradually grown during recent years [37]. Their natural ligands form two superfamilies, the neutrophins (NGFL superfamily), and the TNFL superfamily. Several members of both the TNFR and TNFL superfamily have been observed in Hodgkin's disease. CD27L (CD70), CD40, and CD95 (Fas/Apo) are highly expressed on HRS cells, but the expression of CD120a and CD120b is also seen [16]. Some of their ligands, notably CD30L and CD40L, are expressed on the surrounding activated T cells [38, 39]. These T cells also express CD95 but only minimal amounts of CD95L. Somewhat surprisingly, CD40L is absent from the CD57+ lymphocytes in NLP-Hodgkin's disease (Poppema, unpublished). Several members of these superfami-

Downloaded from http://annonc.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 13, 2015

Anergy of surrounding T cells as a possible cause for impaired cytotoxicity

reduced CD25 and IL-2 expression pattern as with L428 supernatant [32]. IL-10 depletion did not affect the inhibitory effect of L428 supernatant. These findings indicate that TGFp is the T-cell inhibitory factor in the L428 cell line. TGFp is generally produced in a latent, inactive complex composed of the bio-active TGFP homodimer (25 kD) and a non-covalently bonded precursor protein [34]. Often a latent TGFp binding protein is associated with this complex. The bio-active TGFp homo-dimer can be released from the complex after very strong acidification. The TGFp produced by the L428 cell line is already active at physiological pH and has a much higher molecular weight than the usual active form [36]. By using SDS-PAGE the L428 TGFP was shown to contain a 25 kD molecule that crossreacted with antibodies against TGFp. High molecular weight TGFp was also observed in the urine of patients with Hodgkin's disease while it was absent from healthy controls, indicating that it is also produced in vivo by HRS cells [35]. Production of TGFp that is already active at physiological pH by HRS cells may enable them to (de)regulate the immune response in their favour. TGFP is also a potent growth factor for fibroblasts and is known to promote the formation of extracellular matrix and fibrosis [34]. The L428 cell line was derived from a patient with nodular sclerosis subtype of Hodgkin's disease. Therefore, TGFp may shape the environment for HRS cells by suppressing the T-cells response and inducing the collagen formation in the nodular sclerosis subtype.

23

Conclusion

ing effect. IL-10 that is also produced by some HRS cells also may modulate the immune response towards a Th2 type. TheTNF/TNFR superfamilies probably play an important role in Hodgkin's disease, as several members of these families, notably CD30, CD40, CD70, CD95 and also CD95L are highly expressed on the HRS cells. Moreover, their natural ligands are expressed on the surrounding lymphocytes. A general feature of members of the TNF/TNFR family is their involvement in cell activation and/or apoptosis, and therefore they may play a crucial role in improper activation of the lymphocytes. Unequal susceptibility of lymphocyte subpopulations to Fas-mediated cell death could result in the absence of appropriate effector cells such as CD8+ T cells and NK cells, or appropriate helper cells like Thl CD4+ cells. Finally, a novel explanation for the specific phenotype of Hodgkin's disease is offered by our recent finding that the tumor cells of the classical subtypes of Hodgkin's disease produce high quantities of the CC chemokine TARC (reported at the fourth International Conference on Hodgkin's Disease, Cologne, April 1998). This chemokine is normally produced in much smaller amounts by a subset of antigen-presenting cells and strongly attracts T lymphocytes expressing the CCR4 receptor. Activated Th2 lymphocytes indeed express the CCR4 receptor, whereas Thl lymphocytes do not and express the CCR5 receptor. The production of TARC by the RS cells may therefore result in the specific attraction of Th2-typeT lymphocytes. In conclusion, there is a range of factors present in Hodgkin's disease that may contribute to the paradox of an extensive inflammatory infiltrate and concomittant ineffectiveness of the host anti-tumor response. The increased knowledge of these factors may contribute to the design of biologicals that can interrupt the cycle of tumor cell proliferation and lymphocyte influx of Hodgkin's disease in an effective and non-toxic manner.

Hodgkin's disease is characterised by HRS cells surrounded by CD4+ T lymphocytes. These lymphocytes express a variety of activation markers, but are incapable of mounting an effective immune response against the HRS cells. The lymphocytes typically are CD45RO+/ CD45RBdim, lack the expression of CD26, and in vitro can be stimulated to produce IFNy and IL-4, but not IL-2, but do not spontaneously produce cytokines. The References CD45 isotype expression, the absence of CD26, and the potential for IL-4 production suggest that these lym1 Poppema S, Bhan AK. Reinherz EL et al. In situ immunologic characterization of cellular constituents in lymph nodes and phocytes have aTh2 phenotype. Activation of T cells via spleens involved by Hodgkin"s disease. Blood 1982; 59: 226-32. the CD28 co-receptor in a cytokine-rich environment 2. Kanzler H, Kuppers R, Hansmann ML. Rajewski K. Hodgkin without appropriate stimulation via the TCR may lead and Reed-Sternberg cells in Hodgkin's disease represent the to the IL-4 and IFNy producing cytokine profile found outgrowth of a dominant tumor clone derived from (crippled) germinal center B cells. J Exp Med 1996; 184: 1495-505. in the T cells in Hodgkin tissues. Absence of IL-2 pro3. PoppemaS,Van ImhoflTG,Torensma R, Smit J. Lymphadenopathy duction is not only a characteristic of Th2 cells but also morphologically consistent with Hodgkin's disease associated of anergic cells, indicating that the lymphocytes could with Epstein-Barr virus infection. Am J Clin Pathol 1985: 84: also be in a state of anergy. Lack of IL-2 production, the 385-90. induction of a predominant Th2 response, or the induc4. Weiss, LM, Mohaved, LA, Warnke RA. Sklar J. Detection of tion of anergy, would all contribute to an ineffective Epstein-Barr viral genomes in Reed-Sternberg cells of Hodgkin's disease. N Engl J Med 1989; 320: 502-6. immune response against the HRS cells. Another poten5. Pallesen G, Hamilton-Dutoit SJ, Rowe M, Young LS. Expression tial factor responsible for the lack of an effective immune of Epstein-Barr virus latent gene products in tumour cells in response is the cytokine TGF|3 that is produced in an Hodgkin's disease. Lancet 1991; 337: 320-2. active form by HRS cells and has potent immuno6. Romagnani S. Del Prete GF, Maggi E et al. Displacement of T suppressive effects on T cells as well as afibrosispromotlymphocytes with the 'helper/inducer phenotype from peripheral

Downloaded from http://annonc.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 13, 2015

lies have an important role in the regulation of proliferation and apoptosis [40, 41]. CD40/CD40L interaction is extremely important in B-cell activation. CD95/ CD95L interaction is capable of apoptosis induction. CD95/CD95L-induced apoptosis is involved in the maintenance of immune privilege and peripheral tolerance [41]. It is also one of the mechanisms used in the cytotoxic response by T cells. It is therefore interesting that there is no adequate cytotoxic response towards HRS cells, although they express substantial amounts ofCD95. Possible mechanisms for HRS cells to escape a cytotoxic response are by deletion of reactive cytotoxic T cells and NK cells, or by deletion of Thl cells, resulting in the absence of help to the effector cells. Thl cells are more sensitive than Th2 cells to CD95 mediated activation induced cell death [42]. This difference in susceptibility is already present in THO cells [43]. In Hodgkin's disease, the surrounding lymphocytes predominantly express aTh2 phenotype. A polyclonal antibody reactive with membrane bound as well as secreted CD95L and a monoclonal antibody reactive with a cytoplasmic determinant of CD95L give strong staining of HRS cells. CD95L expression has also been detected on malignant cells in several other malignancies, giving these malignant cells yet another possible mechanism to escape the immune response [44]. It is therefore possible that HRS cells escape a cytotoxic and Thl immune response through the induction of apoptosis in surrounding cytotoxic Tand NK lymphocytes and Thl cells.

24

7.

8.

9. 10.

11.

12.

14. 15.

16.

17.

18.

19.

20.

21.

22 23.

24. 25.

26.

27.

28.

29.

30.

31.

32.

33.

34. 35.

36

37.

38.

39.

40.

41. 42.

43.

44.

factor beta in the pathogenesis of nodular sclerosing Hodgkin's disease. Am J Pathol 1990; 136: 1209-14. Hsu S-M. Lin J. Xie S-S et al. Abundant expression of transforming growth factor-(51 and -(?2 by Hodgkin's Reed-Sternberg cells and by reactive T lymphocytes in Hodgkin's disease Hum Pathol 1993; 24: 249-55. Newcom SR, Kadin ME, Ansari AA. Production of transforming growth factor-beta activity by Ki-1 positive lymphoma cells and analysis of its role in the regulation of Ki-1 positive lymphoma growth Am J Pathol 1988;1 31. 569-77. Reinhold D, Bank U. Buhling F et al. Transforming growth factor-pl (TGF-pi) inhibits DNA synthesis of PMW-stimulated PBMC via suppression of IL-2 and IL-6 production. Cytokine 1994; 6: 382-8. Potters M, Diepstra A, Meulenaar R et al. The absence of effective T-cell activation in Hodgkin's disease may be the result of TGFP production by Reed-Slernberg cells. Blood 1997, 10: 265b (Abstr). Newcom SR, Kadin ME, Ansari AA. Diehl V. L-428 nodular sclerosing Hodgkin's cell secrete a unique transforming growth factor-beta active at physiologic pH J Clin Invest 1988b, 82: 1915-21. Lawrence DA. Transforming growth factor-p: A general review. Eur Cytokine Netw 1996: 7: 363-74. Newcom SR, Tagra KK. High molecular weight transforming growth factor-p is excreted in the urine in active nodular sclerosing Hodgkin's disease. Cancer Res 1992; 52: 6768-73. Schwab U, Stein H, Gerdes J et al. Production of a monoclonal antibody specific for Hodgkin and Reed-Sternberg cells of Hodgkin's disease and a subset of normal lymphoid cells. Nature 1982; 299. 65-7. Smith CA, Farrah T, Goodwin RG The TNF receptor superfamily of cellular and viral proteins: Activation, costimulation, and death. Cell 1994; 76. 959-62. Gruss H-J, Pinto A, Gloghini A et al. CD30 ligand expression in nonmalignant and Hodgkin's disease-involved lymphoid tissues. Am J Pathol 1996, 149. 469-81. Carbone A, Gloghini A, Gruss H, Pinto A. CD40 ligand is constitutively expressed in a subset of T-cell lymphomas and on the microenvironmental reactive T cells of follicular lymphomas and Hodgkin's disease. Am J Pathol 1995; 147: 912-22. Grewal I, Flavell RA. A central role of CD40 ligand in the regulation of CD4+ T-cell responses. Immunol Today 1996; 17: 410-4. Abbas AK. Die and let live: Eliminating dangerous lymphocytes. Cell 1996; 84: 655-7. Zhang X, BrunnerT, Carter L et al. Unequal death in T-helper cell (Th)i and Th 2 effectors: Th|, but not Th 2 , effectors undergo rapid Fas/FasL-mediated apoptosis. J Exp Med 1997; 185: 1837-49. Varadhachary AS. Perdow SN, Hu C et al. Differential ability of T-cell subsets to undergo activation-induced cell death. Proc Natl Acad Sci USA 1997: 94: 5778-83. Walker PR, Saas P, Dietrich PY. Role of Fas ligand (CD95L) in immune escape: The tumor cell strikes back. J Immunol 1997; 158:4521-4.

Correspondence to S. Poppema. MD, PhD Department of Pathology & Laboratory Medicine University Hospital Groningen Hanzeplein I 9700RB Groningen The Netherlands E-mail: s.poppema y path.azg.nl

Downloaded from http://annonc.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 13, 2015

13.

blood to lymphoid organs in untreated patients with Hodgkin's disease. Scan J Haematol 1983: 31: 305-14. Lavender JP, Goldman JM, Arnot RN, Thakur ML. Kinetics of indium-Ill labelled lymphocytes in normal subjects and patients with Hodgkin's disease. BMJ 1977; 2: 797-9. Oudejans JJ. Jiwa NM, Kummer JA et al. Activated cytotoxic T cells as prognostic marker in Hodgkin"s disease. Blood 1997; 89: 1376-82. Poppema S. Immunology of Hodgkin's disease. Baillere's Clin Haematol 1996; 9: 447-57. Poppema S. The nature of the lymphocytes surrounding ReedSternberg cells in nodular lymphocyte predominance and in other types of Hodgkin's disease. Am J Pathol 1989; 135: 351-7. Poppema S. Hepperle B. Restricted V gene usage in T-cell lymphomas as detected by anti-T-cell receptor variable region reagents. Am J Pathol 1991; 138: 1479-84. Rubin B, Martin EPG. Arnaud J et al. Expression and signal transduction of T-cell antigen receptor (TCR)/CD3 complexes on fresh or in vitro expanded T lymphocytes from patients with Hodgkin's and non-Hodgkin's lymphomas. Scand J Immunol 1997; 45: 715-25. Morimoto C, Schlossman SF. CD26: A key costimulatory molecule on CD4 memory T cells. Immunologist 1994; 2: 4-10. Fleischer B. CD26: A surface protease involved in T-cell activation. Immunol Today 1994; 15: 180-4. Poppema S, Lai R,Visser L, Yan XJ. CD45 (Leucocyte common antigen) expression inT- and B-lymphocyte subsets. Leuk Lymph 1995; 20: 217-22. Gruss H-J, Pinto A. Duyster J. Poppema S. Hodgkin's disease: A tumor with disturbed immunological pathways Immunol Today 1997; 18: 156-63. Brinkmann V, Kristofic C. Massive production of Th 2 cytokines by human CD4+ effector T cells transiently expressing the natural killer cell marker CD57/HNKI. Immunology 1997; 91: 541-7. Poppema S,Visser L, de Leij LFMH. Reactivity of presumed antinatural killer cell antibody Leu 7 with intrafollicular T lymphocytes. Clin Exp Immunol 1983; 54- 834-7. Munro JM, Freedman AS, Aster JC et al. //) vivo expression of the B7 costimulatory molecule by subsets of antigen-presenting cells and the malignant cells of Hodgkin's disease. Blood 1994; 83: 793-8. Fox DA, Hussey RE, Fitzgerald KA et al. Ta,, a novel 105 kD human T-cell activation antigen defined by a monoclonal antibody. J Immunol 1984; 133: 1250-6. MatternT, ScholzW. Feller et al. Expression of CD26 (dipeptidyl peptidase IV) on resting and activated human T-lymphocytes. Scand J Immunol 1991; 33: 737-48. Fields PE, Gajewski TF, Fitch FW. Blocked Ras activation in anergic CD4+ T cells. Science 1996: 271. 1276-8. Li W, Whaley D. Mondino A. Mueller DL. Blocked signal transduction to the ERK and JNK protein kinases in anergic CD4+ T cells. Science 1996: 271: 1272-6. Kang S, Beverly B. Tran A et al. Transactivation by AP-I is a molecular target of T-cell clonal anergy. Science 1992; 257: 1134-8. Groux H. Bigler M, de Vries JE. Roncarolo M. Interleukin-10 induces a long-term antigen-specific anergic state in human CD4+ T cells. J Exp Med 1996: 184: 19-29. Kehrl JH.Wakefield LM, Roberts AB. Production of transforming growth factor beta by human T lymphocytes and its potential role in the regulation of T-cell growth. J Exp Med 1986: 163: 1037-50. Hsu S. Waldron JW. Hsu P, Hough J. Cytokines in malignant lymphomas: Review and prospective evaluation. Hum Pathol 1993: 24: 1040 57. Kadin ME. Agnarsson BA. Ellingsworth LR. Newcom SR Immunohistochemical evidence of a role for transforming growth