- Email: [email protected]
Melanoma skews dendritic cells to facilitate a T helper 2 profile
Melanoma skews dendritic cells to facilitate a T helper 2 profile Martin McCarter, MD,a Jason Clarke, MD,a Donald Richter, MS,a and Cara Wilson, MD,b Denver, Colo
Background. Patients with progressing melanoma have a circulating cytokine profile reflecting a T helper cell type 2 (Th2) imbalance, while patients responding to therapy favor a Th1 profile. The aim of this study was to determine the role of circulating dendritic cells (DCs) in mediating this imbalance. Methods. Isolated human peripheral blood mononuclear cells (PBMCs) were exposed to cell-free melanoma--conditioned medium (MCM) or control fibroblast-conditioned medium before stimulation. In separate experiments, isolated circulating DCs were exposed to MCM before addition of T cells. DC maturation and function were determined. Mixed leukocyte response T-cell proliferation was quantified and supernatants were assayed for Th1 (interleukin [IL]-2 and interferon c) and Th2 (IL-4, IL-5, and IL-10) cytokines. Results. PBMCs exposed to MCM produced significantly more Th2-type cytokines (IL-4, IL-5, and IL10) over time than those exposed to control medium. DCs exposed to MCM before addition of T cells, produced a similar pattern of a sustained longer term Th2 response after an initial burst of IL-2. Exposure to MCM did not significantly affect DC maturation or IL-12 production. T-cell proliferation did not change significantly in the mixed leukocyte response, however, the percentage of viable CD4+ T cells in the MCM-treated group was significantly less than control (37 vs 50%, P < .05). Conclusions. Exposure of PBMCs to melanoma produces a Th2-type cytokine profile, which may be, in part, facilitated by DCs. (Surgery 2005;138:321-8.) From the Departments of Surgerya and Medicine, Division of Immunology,b University of Colorado at Denver and Health Sciences Center, Denver
UPON ANTIGENIC STIMULATION, naı¨ve T helper cells enter a T helper 1 or T helper 2 (Th1 or Th2) differentiation pathway. The type of generated effector T-helper response is reflected by the predominant type of cytokines produced such that interferon c (IFN-c) and interleukin-2 (IL-2) are associated with a Th1 response, and IL-4, IL-5, and IL-10 are associated with a Th2 response.1 These responses are believed to serve the hosts immediate needs, with Th1 responses tending to be more activating and Th2 responses tending to be more counterregulatory.2 Although initially described for its role in combating certain infections, the importance of a differentiated T-helper response in fighting tumors has been recognized only recently. For example, a vigorous Th1 response is Presented at the 66th Annual Meeting of the Society of University Surgeons, Nashville, Tennessee, February 9-12, 2005. Supported, in part, by a 5 K12 CA86913 clinical oncology research career development program award. Reprint requests: Martin McCarter, MD, Assistant Professor of Surgery, University of Colorado Health Sciences Center, 4200 Ninth Ave, C-311, Denver, CO 80262. E-mail: martin.mccarter@ uchsc.edu. 0039-6060/$ - see front matter Ó 2005 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2005.06.011
required for the destruction of tumor cells, while a Th2 response has been linked to tolerance of fetal and transplanted tissues.3-5 In addition, patients with advanced melanoma have circulating cytokine levels, reflecting a shift towards a Th2-type profile, compared with early stage melanoma or normal patients.6 Conversely, other investigators found that melanoma patients who have no measurable disease or who are responding to therapy had circulating cytokine levels reflecting a Th1 profile.7,8 DCs are potent antigen-presenting cells that help coordinate the immune response and have been employed as vectors for immunologic therapy.9,10 Depending on the local environmental conditions and stimuli, DCs skew the T-helper response toward Th1 or Th2.11-13 Melanoma cells are known to produce a variety of immunosuppressive products that, combined with conditions of the local environment, serve to dampen the immunologic response.14,15 We therefore hypothesized that DCs exposed to melanoma will facilitate the generation of a Th2 cytokine profile from naı¨ve T cells. MATERIAL AND METHODS Cells and conditioned medium. Human peripheral blood mononuclear cells (PBMCs) were SURGERY 321
322 McCarter et al
Surgery August 2005
Fig 1. A, Th1 cytokines from PBMCs 1 to 7 days after exposure to MCM. B, Th2 cytokines from PBMCs 1-7 days after exposure to MCM. Control medium (diamonds) and MCM (squares). *P < .05, compared with control medium (n = 6). IL, Interleukin; IFN-g, interferon c.
isolated with the use of Ficoll-Paque (Amersham Biosciences, Piscataway, NJ) following an institutional review board--approved protocol. Cells were washed twice in Dulbecco phosphate-buffered saline (DPBS) and resuspended in RPMI with 10% human serum. The human melanoma cell line (WM793) and human foreskin fibroblast cells (ATCC no. CRL-7014) were cultured to 80% to 90% confluence in RPMI with human serum. The cells were washed in DPBS, and fresh RPMI medium was applied. The conditioned medium was collected 24 hours later. This melanoma (MCM)or fibroblast-conditioned medium (control) was collected, centrifuged at 10,000g, passed through a 20-micron filter, and stored at ÿ80°C until needed.
For use in experiments, the control and MCM was diluted to a final concentration of 20% conditioned medium and 80% fresh RPMI with human serum. In this and prior experiments, this concentration has not affected the viability of the cells, yet it has preserved the biological activity of the conditioned medium.14 Isolated PBMCs (2 3 106) were exposed to control medium or MCM with and without anti-CD3 (OKT3; Upstate Biotech, Waltham, Mass) as a stimulus for periods of 1 to 7 days. Culture supernatants were harvested and assayed for secretion of Th1 (IL-2, IFN-c) and Th2 (IL-4, IL-5, IL-10) cytokines by a human Th1/Th2 cytometric bead array kit (BD Biosciences, Franklin Lakes, NJ).
McCarter et al 323
Surgery Volume 138, Number 2
Mixed lymphocyte response. Dendritic cells (DCs) were isolated from circulating PBMCs with the use of a magnetic cell–sorting isolation protocol (Miltenyi Biotech, Auburn, Calif). Briefly, PBMCs were incubated with a non-DC depletion cocktail specific for CD14 and CD19 cells. The labeled cells were then removed by magnetic separation. The resulting DC cell population was then purified with a DC-enrichment cocktail that selects cells expressing a BDCA-1, BDCA-3, and BDCA-4 phenotype. This protocol results in greater than 80% DCs, as confirmed by flow cytometery. Isolated DCs were cultured with or without control or MCM for 24 hours before the addition of allogenic T cells at a ratio of 1:20 (DC:T cells). T cells were isolated by a single-step magnetic enrichment of cells expressing CD3 with the use of magnetic cell–sorting separation (Miltenyi Biotech). CD3+ cells were labeled with carboxyfluorescein diacetate succinimidyl ester (Molecular Probes, Eugene, Ore) for 15 minutes before being washed with RPMI + human serum. The MLR was allowed to incubate for 7 days. Samples of supernatant were removed for cytokine analysis from some experiments at days 1, 3, and 7. The T-cell proliferative response was assessed by flow cytometry as described (Molecular Probes). Briefly, the presence of dividing cells was determined by flow cytometry as a decrease in fluorescence intensity of the FL1 channel, which typically was seen as a stepwise reduction in fluorescence representing individual cellular divisions. T-cell populations were identified by staining for CD3-APC, CD4-TC, and CD8-FITC (Caltag Laboratories, Burlingame, Calif) cells. T-cell populations were identified by first gating on CD3+ cells then by differential identification of the CD4+ and CD8+ within that population of cells. Derivation of DC from PBMC and DC maturation. For some experiments, DCs were derived from circulating PBMCs by plating the PBMC at 2 3 106 cells per well of a 24-well plate in RPMI + 10% human serum and allowing the cells to adhere for 2 hours at 37°C. Nonadherent cells were removed, and the remaining cells were incubated in a cytokine cocktail consisting of RPMI + 10% human serum + 800 U/mL GMCSF + 104 U/mL IL-4 (R&D Systems, Minneapolis, Minn) for 7 days. This maturation cocktail was replaced every other day. On the seventh day, the maturation cocktail was replaced with a maturation cocktail consisting of 20% control or MCM, and 24 hours later the cells were stimulated with 2 lg/mL CD40 ligand (R&D Systems) or 100 ng/mL of lipopolysaccharide (Sigma, St. Louis, Mo) or 10 ng/mL of tumor
Table I. Additional cytokines (pg/mL) present in control and MCM Cytokine
Control
Melanoma
IL-6 IL-8 IL-12 TNF-a PGE2 TGF-ß
2461 1630 ND ND 161 619
28,860 4565 ND ND 95 853
IL, Interleukin; ND, none detected; TNF-a, tumor necrosis factor a; PGE2, prostaglandin E2; TGF- ß, transforming growth factor ß.
necrosis factor a (TNF-a; Sigma) for an additional 24 hours. For DC maturation analysis, the cells were bathed in ice-cold DPBS, then stained with a lineage-negative cocktail (fluorescein isothiocyanate staining for CD3, CD14, CD16, CD19, CD20, and CD 56; BD Biosciences, San Jose, Calif), HLA-DRPE, CD11c (BD Biosciences), CD123 (Miltenyi Biotec) then with anti–CD40 PE or anti–CD86 PE (both from Ancell, Bayport, Minn). DCs were identified as being lineage-negative and HLA-DRpositive. The DCs were further analyzed as being positive for either CD11c or CD123, and maturation was identified by enhanced fluorescence of either CD40 or CD86. Secretion of IL-12p70 was assayed in supernatants by enzyme-linked immunosorbent assay (ELISA) (ElisaTech, Denver, Colo, and R&D Systems). RESULTS MCM effect on PBMCs. Within 24 hours of stimulation with OKT3, isolated PBMCs exposed to MCM produced significantly more IL-2 (1467 ± 211 vs 6960 ± 545 pg/mL), IL-4 (80 ± 14 vs 17 ± 12 pg/mL), and IL-10 (829 ± 57 vs 1479 ± 53 pg/mL) than PBMCs exposed to control medium (Fig 1, A and B). At 3 days, the amount of IL-2 present is trending back down, yet there still is more IL-2 present in the MCM-treated PBMCs, compared with the control group. Meanwhile there is a doubling of the Th2-type cytokines IL-4 (38 ± 21 vs 143 ± 76 pg/mL), IL-5 (156 ± 39 vs 369 ± 68 pg/mL), and IL-10 (3364 ± 686 vs 6310 ± 771 pg/mL) in the melanoma-treated group. By 7 days, there was significantly more IL-5 (338 ± 174 vs 1165 ± 433 pg/mL) and IL-10 (2251 ± 1181 vs 8065 ± 3610 pg/mL) produced in the MCM-treated groups with no difference in IL-2, IL-4 or IFN-c. In the absence of a stimulating agent, no significant Th1 or Th2 cytokine production was identified at any of the time points. Cytokines in melanoma and control media. Control medium and MCM contained very low
324 McCarter et al
Surgery August 2005
Fig 2. A, Th1 cytokines from MLR supernatants 1 to 7 days after DC exposure to MCM. B, Th2 cytokines from MLR supernatants 1 to 7 days after DC exposure to MCM. Control medium represented (diamonds) and MCM (squares). *P < .05, compared with control medium (n = 6). C, IL-12p70 production from DCs. DCs exposed to MCM for 24 hours followed by MLR for 7 days. There was no significant difference in IL-12p70 production between the groups. IL, Interleukin; IFN-g, interferon c.
Surgery Volume 138, Number 2
levels (0-5 g/mL) of Th1 and Th2 cytokines at baseline. Other potentially bioactive cytokines are present in both control medium and MCM. In particular, the MCM contained increased amounts of IL-6 and IL-8 and slightly higher levels of transforming growth factor ß (TGF-ß) relative to control medium (Table I). DC-based mixed lymphocyte response. Supernatants from the MLR experiments, in which DCs were preincubated with MCM, followed a similar pattern to those seen with whole PBMCs. There was an early burst in IL-2 (68 ± 23 vs 259 ± 87 pg/mL) production accompanied by a trend toward increased early IL-4, IL-5, IL-10 production in the MCM-treated group. By day 7, IL-2 was no longer detectable, and there was significant enhancement of persistent Th2 production of IL-5 (67 ± 22 vs 139 ± 37) and IL-10 (46 ± 19 vs 106 ± 34 pg/mL) in the MCM-treated group (Fig 2, A and B). There was no significant difference in IL-12 production between groups at day 7 (Fig 2, C). T-cell proliferation in the MLR. Overall, there was no significant difference in CD4+, CD8+, or total T-cell proliferation induced by MCM-treated DCs. However, after 7 days in the MLR, there were significantly fewer CD4+ staining T cells present in the MCM-treated group (37 vs 50%, P < .05). This was accompanied by a trend towards an increased percentage of CD8+ T cells in the MCM-treated group at 7 days. (Fig 3) DC maturation. Exposure of DCs (in PBMCs) to MCM did not appear to alter their ability to mature, as judged by the similar percentages and expression of maturation markers CD40 and CD86 (Fig 4, A). This observation remained the same even with larger numbers of derived DCs and variable lengths of exposure to MCM up to 7 days. Use of known DC-maturing agents such as CD40L and TNF-a in the setting of MCM did not change the results (Fig 4, B) Finally, there was no significant difference in the production of IL-12 with any of the treatment groups or maturation agents used (data not shown). DISCUSSION Patients with advanced or progressing melanoma frequently have circulating cytokine levels reflecting a Th2 pattern.6,16 It had been assumed that T cells primarily were responsible for this observation, but how and why this pattern evolved is not well understood. From a teleologic point of view, a sustained Th2 response would create a tolerogenic environment in which melanoma could continue to grow and spread unencumbered
McCarter et al 325
by the immune system. Melanoma does have some direct effects on T cells; however, the mechanisms involved in the genesis of a sustained Th2 response in cancer patients is not known.17,18 Results from the present experiments indicate that melanoma cells themselves do not produce Th2-type cytokines, yet they do create an environment in which stimulated PBMCs induce a Th2 response. When whole PBMCs were exposed to MCM and then stimulated, we observed an initial burst of both Th1 and Th2 cytokines. However, longer-term cultures suggested a preponderance of Th2 cytokines. This observation corresponds to the clinical findings from the sera of patients with advanced melanoma.6 Although it is presumed that these observed differences in cytokines measured from the supernatants are due to an increased production of cytokines, similar results could be observed if there was a decrease in select cytokine consumption by other cells in culture. The coordination of a durable immunologic response requires the activation of antigenpresenting cells, such as DCs, to stimulate specific T helper cells.9 An effective antitumor response involves antigen processing, DC cell maturation, and specific costimulatory signals to T cells that subsequently produce a Th1 response.10 To better understand the potential role of DCs in generating the observed Th2 response, we isolated DCs, exposed them to MCM, then added back allogenic T cells in an MLR. The global trend toward a sustained longer-term Th2 response in the MLR suggests that the DCs may be, in part, responsible for maintaining this effect. Exactly how the melanoma is conditioning the DCs to facilitate a Th2 response remains to be determined. Cultured human melanoma cell lines vary considerably in their activity on cells from the immune system although we have identified at least one other cell line (A375) that produces similar effects on the T-helper response (data not shown). Previous experience with colon and breast cancer cell lines found variable immunosuppressive effects on macrophages, depending on the characteristics of the individual cell line.14 As DCs go through the maturation process, they transition from antigen uptake and processing to T-cell stimulating machines.9 Some investigators have suggested that patients with advanced cancers tend to have fewer or less mature DCs.19,20 However, in these experiments, exposure to MCM did not appear to significantly influence the level of maturation, as determined by typical phenotypic maturation markers. Furthermore, there was no significant difference in IL-12 production in MCM
326 McCarter et al
Surgery August 2005
Fig 3. T-cell populations and proliferation after DC exposure to MCM. Control medium (solid bars) and MCM-treated cells (stippled bars). *P < .05, compared with control medium (n=6). Allogenic T cells added to DCs for 7 days after 24 hours of exposure to MCM.
Fig 4. A, Expression of DC maturation markers after exposure to MCM. PBMCs exposed to control (solid bars) or MCM (stippled bars). Graph represents percent of DCs (CD11c+, CD123+) expressing markers of maturation (CD86 and CD40). B, Derived DCs exposed to control or MCM and assessed for mean florescent intensity (MFI) surface expression of CD40 after stimulation with CD40L (solid bars) and TNF-a (hashed bars). Representative data from several experiments with different MCM incubation times and stimulating agents.
DCs. IL-12 is a key cytokine secreted by DCs to promote a Th1 response21; therefore, the absence of significant IL-12 may have allowed a Th2 response to dominate.
The T-cell proliferative response is used as a measure of DC activation of T cells.13 Cancer patients are reported to have impaired T-cell proliferative capacity.17 In these experiments, we did
McCarter et al 327
Surgery Volume 138, Number 2
not observe any significant differences in total T-cell proliferation in the MLR. One explanation for this observation may be found in the pronounced early burst of IL-2 in MCM-treated groups since IL-2 is a critical cytokine for T-cell proliferation and growth.22 It may be that the relatively early burst in IL-2 at day 3 obscured any observable effects on proliferation at day 7. When analyzing the CD4+ and CD8+ cells, we noted a significant decrease in the number of CD4+ cells at 7 days in the MCM-treated group. This suggests that MCMtreated DCs could not keep CD4+ cells alive and well for the full duration of the MLR. Other investigators have found that cancer patients with progressive disease tend to have lower CD4 counts, and fewer CD4 cells were present within metastatic deposits of those patients who failed therapy, compared with those who were responding to therapy.23,24 Furthermore, CD4 cells are shorter lived, and IL-2 tends to sustain CD8 proliferative capacity more than CD4 cells do.25 Additional experiments are needed to determine the effect of MCM on T-cell apoptosis and DC apoptosis. Several melanoma-derived cytokines (IL-6, IL-8, prostaglandin E2 [PGE2], TGF-b) are known to influence the immunologic response.15,26-28 Although commonly implicated as an immunosuppressive agent, PGE2 is not likely to play a major role in these experiments since there was relatively low levels of PGE2 in any of the tested media.29,30 TGF-b is an immunosuppressive agent present in most serum-containing media.30,31 While it could have played a role in these experiments, this explanation seems unlikely since only slightly higher concentrations were present in the MCM than in the control medium. Of the tested cytokines, it would seem that IL-6 may be the most biologically active since it was markedly elevated in the MCM, compared with control. Furthermore, IL-6 is known to impede the Th1 response and is a marker of progressing metastatic melanoma.32,33 Other as yet partially characterized factors are secreted by melanoma and produce a variety of immunosuppressive effects.14 A Th2 cytokine profile is associated with downregulation of the immunologic response and progression of melanoma. The interactions between melanoma, components of the immune system (such as DCs and Tcells), and the local environment are complex. These results suggest that melanoma may skew DCs to facilitate a more-tolerogenic Th2 response rather than an antitumor Th1 response. Understanding how this pattern evolves may help unravel the mechanisms of tumor-induced immune
suppression and lead to improved targeted immunologic therapies.
REFERENCES 1. Pulendran B. Modulating TH1/TH2 responses with microbes, dendritic cells, and pathogen recognition receptors. Immunol Res 2004;29:187-96. 2. Del Prete G. The concept of type-1 and type-2 helper T cells and their cytokines in humans. Int Rev Immunol 1998;16: 427-55. 3. Bennett WA, Lagoo-Deenadayalan S, Whitworth NS, Brackin MN, Hale E, Cowan BD. Expression and production of interleukin-10 by human trophoblast: relationship to pregnancy immunotolerance. Early Pregnancy 1997;3: 190-8. 4. Rankin EB, Yu D, Jiang J, et al. An essential role of Th1 responses and interferon gamma in infection-mediated suppression of neoplastic growth. Cancer Biol Ther 2003; 2:687-93. 5. Zelenika D, Adams E, Humm S, Lin CY, Waldmann H, Cobbold SP. The role of CD4+ T-cell subsets in determining transplantation rejection or tolerance. Immunol Rev 2001; 182:164-79. 6. Lauerova L, Dusek L, Simickova M, et al. Malignant melanoma associates with Th1/Th2 imbalance that coincides with disease progression and immunotherapy response. Neoplasma 2002;49:159-66. 7. Winter H, Hu HM, Poehlein CH, et al. Tumour-induced polarization of tumour vaccine-draining lymph node T cells to a type 1 cytokine profile predicts inherent strong immunogenicity of the tumour and correlates with therapeutic efficacy in adoptive transfer studies. Immunology 2003;108: 409-19. 8. Schultz ES, Schuler-Thurner B, Stroobant V, et al. Functional analysis of tumor-specific Th cell responses detected in melanoma patients after dendritic cell-based immunotherapy. J Immunol 2004;172:1304-10. 9. Mellman I, Steinman RM. Dendritic cells: specialized and regulated antigen processing machines. Cell 2001;106:2558. 10. Banchereau J, Schuler-Thurner B, Palucka AK, Schuler G. Dendritic cells as vectors for therapy. Cell 2001;106:271-4. 11. Anderson CF, Lucas M, Gutierrez-Kobeh L, Field AE, Mosser DM. T cell biasing by activated dendritic cells. J Immunol 2004;173:955-61. 12. Kaufman HL, Disis ML. Immune system versus tumor: shifting the balance in favor of DCs and effective immunity. J Clin Invest 2004;113:664-7. 13. Mazzoni A, Segal DM. Controlling the Toll road to dendritic cell polarization. J Leukoc Biol 2004;75:721-30. 14. Naama HA, McCarter MD, Mack VE, et al. Suppression of macrophage nitric oxide production by melanoma: mediation by a melanoma-derived product. Melanoma Res 2001; 11:229-38. 15. Biggs MW, Eiselein JE. Suppression of immune surveillance in melanoma. Med Hypotheses 2001;56:648-52. 16. Tatsumi T, Kierstead LS, Ranieri E, et al. Disease-associated bias in T helper type 1 (Th1)/Th2 CD4(+) T cell responses against MAGE-6 in HLA-DRB10401(+) patients with renal cell carcinoma or melanoma. J Exp Med 2002;196: 619-28. 17. Giacomoni D, Ben-Efraim S, Najmabadi F, Dray S. Inhibitors of lymphocyte activation secreted by human melanoma cell lines. Med Oncol Tumor Pharmacother 1990;7:273-80.
328 McCarter et al
18. Zhang XM, Xu Q. Metastatic melanoma cells escape from immunosurveillance through the novel mechanism of releasing nitric oxide to induce dysfunction of immunocytes. Melanoma Res 2001;11:559-67. 19. Onishi H, Morisaki T, Baba E, et al. Dysfunctional and shortlived subsets in monocyte-derived dendritic cells from patients with advanced cancer. Clin Immunol 2002;105: 286-95. 20. Almand B, Clark JI, Nikitina E, et al. Increased production of immature myeloid cells in cancer patients: a mechanism of immunosuppression in cancer. J Immunol 2001;166:678-89. 21. Knutson KL, Disis ML. IL-12 enhances the generation of tumour antigen-specific Th1 CD4 T cells during ex vivo expansion. Clin Exp Immunol 2004;135:322-9. 22. Seder RA, Ahmed R. Similarities and differences in CD4+ and CD8+ effector and memory T cell generation. Nat Immunol 2003;4:835-42. 23. Hakansson A, Gustafsson B, Krysander L, Hakansson L. Tumour-infiltrating lymphocytes in metastatic malignant melanoma and response to interferon alpha treatment. Br J Cancer 1996;74:670-6. 24. Kuss I, Hathaway B, Ferris RL, Gooding W, Whiteside TL. Imbalance in absolute counts of T lymphocyte subsets in patients with head and neck cancer and its relation to disease. Adv Otorhinolaryngol 2005;62:161-72. 25. Iezzi G, Karjalainen K, Lanzavecchia A. The duration of antigenic stimulation determines the fate of naive and effector T cells. Immunity 1998;8:89-95.
Surgery August 2005
26. Ijland SA, Jager MJ, Heijdra BM, Westphal JR, Peek R. Expression of angiogenic and immunosuppressive factors by uveal melanoma cell lines. Melanoma Res 1999;9:445-50. 27. Ugurel S, Rappl G, Tilgen W, Reinhold U. Increased serum concentration of angiogenic factors in malignant melanoma patients correlates with tumor progression and survival. J Clin Oncol 2001;19:577-83. 28. Redondo P, Sanchez-Carpintero I, Bauza A, Idoate M, Solano T, Mihm MC Jr. Immunologic escape and angiogenesis in human malignant melanoma. J Am Acad Dermatol 2003;49:255-63. 29. Denkert C, Kobel M, Berger S, et al. Expression of cyclooxygenase 2 in human malignant melanoma. Cancer Res 2001;61:303-8. 30. Botti C, Seregni E, Ferrari L, Martinetti A, Bombardieri E. Immunosuppressive factors: role in cancer development and progression. Int J Biol Markers 1998;13:51-69. 31. Gold LI. The role for transforming growth factor-beta (TGF-beta) in human cancer. Crit Rev Oncog 1999;10:30360. 32. Dodge IL, Carr MW, Cernadas M, Brenner MB. IL-6 production by pulmonary dendritic cells impedes Th1 immune responses. J Immunol 2003;170:4457-64. 33. Mouawad R, Rixe O, Meric JB, Khayat D, Soubrane C. Serum interleukin-6 concentrations as predictive factor of time to progression in metastatic malignant melanoma patients treated by biochemotherapy: a retrospective study. Cytokines Cell Mol Ther 2002;7:151-6.
Background. Patients with progressing melanoma have a circulating cytokine profile reflecting a T helper cell type 2 (Th2) imbalance, while patients responding to therapy favor a Th1 profile. The aim of this study was to determine the role of circulating dendritic cells (DCs) in mediating this imbalance. Methods. Isolated human peripheral blood mononuclear cells (PBMCs) were exposed to cell-free melanoma--conditioned medium (MCM) or control fibroblast-conditioned medium before stimulation. In separate experiments, isolated circulating DCs were exposed to MCM before addition of T cells. DC maturation and function were determined. Mixed leukocyte response T-cell proliferation was quantified and supernatants were assayed for Th1 (interleukin [IL]-2 and interferon c) and Th2 (IL-4, IL-5, and IL-10) cytokines. Results. PBMCs exposed to MCM produced significantly more Th2-type cytokines (IL-4, IL-5, and IL10) over time than those exposed to control medium. DCs exposed to MCM before addition of T cells, produced a similar pattern of a sustained longer term Th2 response after an initial burst of IL-2. Exposure to MCM did not significantly affect DC maturation or IL-12 production. T-cell proliferation did not change significantly in the mixed leukocyte response, however, the percentage of viable CD4+ T cells in the MCM-treated group was significantly less than control (37 vs 50%, P < .05). Conclusions. Exposure of PBMCs to melanoma produces a Th2-type cytokine profile, which may be, in part, facilitated by DCs. (Surgery 2005;138:321-8.) From the Departments of Surgerya and Medicine, Division of Immunology,b University of Colorado at Denver and Health Sciences Center, Denver
UPON ANTIGENIC STIMULATION, naı¨ve T helper cells enter a T helper 1 or T helper 2 (Th1 or Th2) differentiation pathway. The type of generated effector T-helper response is reflected by the predominant type of cytokines produced such that interferon c (IFN-c) and interleukin-2 (IL-2) are associated with a Th1 response, and IL-4, IL-5, and IL-10 are associated with a Th2 response.1 These responses are believed to serve the hosts immediate needs, with Th1 responses tending to be more activating and Th2 responses tending to be more counterregulatory.2 Although initially described for its role in combating certain infections, the importance of a differentiated T-helper response in fighting tumors has been recognized only recently. For example, a vigorous Th1 response is Presented at the 66th Annual Meeting of the Society of University Surgeons, Nashville, Tennessee, February 9-12, 2005. Supported, in part, by a 5 K12 CA86913 clinical oncology research career development program award. Reprint requests: Martin McCarter, MD, Assistant Professor of Surgery, University of Colorado Health Sciences Center, 4200 Ninth Ave, C-311, Denver, CO 80262. E-mail: martin.mccarter@ uchsc.edu. 0039-6060/$ - see front matter Ó 2005 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2005.06.011
required for the destruction of tumor cells, while a Th2 response has been linked to tolerance of fetal and transplanted tissues.3-5 In addition, patients with advanced melanoma have circulating cytokine levels, reflecting a shift towards a Th2-type profile, compared with early stage melanoma or normal patients.6 Conversely, other investigators found that melanoma patients who have no measurable disease or who are responding to therapy had circulating cytokine levels reflecting a Th1 profile.7,8 DCs are potent antigen-presenting cells that help coordinate the immune response and have been employed as vectors for immunologic therapy.9,10 Depending on the local environmental conditions and stimuli, DCs skew the T-helper response toward Th1 or Th2.11-13 Melanoma cells are known to produce a variety of immunosuppressive products that, combined with conditions of the local environment, serve to dampen the immunologic response.14,15 We therefore hypothesized that DCs exposed to melanoma will facilitate the generation of a Th2 cytokine profile from naı¨ve T cells. MATERIAL AND METHODS Cells and conditioned medium. Human peripheral blood mononuclear cells (PBMCs) were SURGERY 321
322 McCarter et al
Surgery August 2005
Fig 1. A, Th1 cytokines from PBMCs 1 to 7 days after exposure to MCM. B, Th2 cytokines from PBMCs 1-7 days after exposure to MCM. Control medium (diamonds) and MCM (squares). *P < .05, compared with control medium (n = 6). IL, Interleukin; IFN-g, interferon c.
isolated with the use of Ficoll-Paque (Amersham Biosciences, Piscataway, NJ) following an institutional review board--approved protocol. Cells were washed twice in Dulbecco phosphate-buffered saline (DPBS) and resuspended in RPMI with 10% human serum. The human melanoma cell line (WM793) and human foreskin fibroblast cells (ATCC no. CRL-7014) were cultured to 80% to 90% confluence in RPMI with human serum. The cells were washed in DPBS, and fresh RPMI medium was applied. The conditioned medium was collected 24 hours later. This melanoma (MCM)or fibroblast-conditioned medium (control) was collected, centrifuged at 10,000g, passed through a 20-micron filter, and stored at ÿ80°C until needed.
For use in experiments, the control and MCM was diluted to a final concentration of 20% conditioned medium and 80% fresh RPMI with human serum. In this and prior experiments, this concentration has not affected the viability of the cells, yet it has preserved the biological activity of the conditioned medium.14 Isolated PBMCs (2 3 106) were exposed to control medium or MCM with and without anti-CD3 (OKT3; Upstate Biotech, Waltham, Mass) as a stimulus for periods of 1 to 7 days. Culture supernatants were harvested and assayed for secretion of Th1 (IL-2, IFN-c) and Th2 (IL-4, IL-5, IL-10) cytokines by a human Th1/Th2 cytometric bead array kit (BD Biosciences, Franklin Lakes, NJ).
McCarter et al 323
Surgery Volume 138, Number 2
Mixed lymphocyte response. Dendritic cells (DCs) were isolated from circulating PBMCs with the use of a magnetic cell–sorting isolation protocol (Miltenyi Biotech, Auburn, Calif). Briefly, PBMCs were incubated with a non-DC depletion cocktail specific for CD14 and CD19 cells. The labeled cells were then removed by magnetic separation. The resulting DC cell population was then purified with a DC-enrichment cocktail that selects cells expressing a BDCA-1, BDCA-3, and BDCA-4 phenotype. This protocol results in greater than 80% DCs, as confirmed by flow cytometery. Isolated DCs were cultured with or without control or MCM for 24 hours before the addition of allogenic T cells at a ratio of 1:20 (DC:T cells). T cells were isolated by a single-step magnetic enrichment of cells expressing CD3 with the use of magnetic cell–sorting separation (Miltenyi Biotech). CD3+ cells were labeled with carboxyfluorescein diacetate succinimidyl ester (Molecular Probes, Eugene, Ore) for 15 minutes before being washed with RPMI + human serum. The MLR was allowed to incubate for 7 days. Samples of supernatant were removed for cytokine analysis from some experiments at days 1, 3, and 7. The T-cell proliferative response was assessed by flow cytometry as described (Molecular Probes). Briefly, the presence of dividing cells was determined by flow cytometry as a decrease in fluorescence intensity of the FL1 channel, which typically was seen as a stepwise reduction in fluorescence representing individual cellular divisions. T-cell populations were identified by staining for CD3-APC, CD4-TC, and CD8-FITC (Caltag Laboratories, Burlingame, Calif) cells. T-cell populations were identified by first gating on CD3+ cells then by differential identification of the CD4+ and CD8+ within that population of cells. Derivation of DC from PBMC and DC maturation. For some experiments, DCs were derived from circulating PBMCs by plating the PBMC at 2 3 106 cells per well of a 24-well plate in RPMI + 10% human serum and allowing the cells to adhere for 2 hours at 37°C. Nonadherent cells were removed, and the remaining cells were incubated in a cytokine cocktail consisting of RPMI + 10% human serum + 800 U/mL GMCSF + 104 U/mL IL-4 (R&D Systems, Minneapolis, Minn) for 7 days. This maturation cocktail was replaced every other day. On the seventh day, the maturation cocktail was replaced with a maturation cocktail consisting of 20% control or MCM, and 24 hours later the cells were stimulated with 2 lg/mL CD40 ligand (R&D Systems) or 100 ng/mL of lipopolysaccharide (Sigma, St. Louis, Mo) or 10 ng/mL of tumor
Table I. Additional cytokines (pg/mL) present in control and MCM Cytokine
Control
Melanoma
IL-6 IL-8 IL-12 TNF-a PGE2 TGF-ß
2461 1630 ND ND 161 619
28,860 4565 ND ND 95 853
IL, Interleukin; ND, none detected; TNF-a, tumor necrosis factor a; PGE2, prostaglandin E2; TGF- ß, transforming growth factor ß.
necrosis factor a (TNF-a; Sigma) for an additional 24 hours. For DC maturation analysis, the cells were bathed in ice-cold DPBS, then stained with a lineage-negative cocktail (fluorescein isothiocyanate staining for CD3, CD14, CD16, CD19, CD20, and CD 56; BD Biosciences, San Jose, Calif), HLA-DRPE, CD11c (BD Biosciences), CD123 (Miltenyi Biotec) then with anti–CD40 PE or anti–CD86 PE (both from Ancell, Bayport, Minn). DCs were identified as being lineage-negative and HLA-DRpositive. The DCs were further analyzed as being positive for either CD11c or CD123, and maturation was identified by enhanced fluorescence of either CD40 or CD86. Secretion of IL-12p70 was assayed in supernatants by enzyme-linked immunosorbent assay (ELISA) (ElisaTech, Denver, Colo, and R&D Systems). RESULTS MCM effect on PBMCs. Within 24 hours of stimulation with OKT3, isolated PBMCs exposed to MCM produced significantly more IL-2 (1467 ± 211 vs 6960 ± 545 pg/mL), IL-4 (80 ± 14 vs 17 ± 12 pg/mL), and IL-10 (829 ± 57 vs 1479 ± 53 pg/mL) than PBMCs exposed to control medium (Fig 1, A and B). At 3 days, the amount of IL-2 present is trending back down, yet there still is more IL-2 present in the MCM-treated PBMCs, compared with the control group. Meanwhile there is a doubling of the Th2-type cytokines IL-4 (38 ± 21 vs 143 ± 76 pg/mL), IL-5 (156 ± 39 vs 369 ± 68 pg/mL), and IL-10 (3364 ± 686 vs 6310 ± 771 pg/mL) in the melanoma-treated group. By 7 days, there was significantly more IL-5 (338 ± 174 vs 1165 ± 433 pg/mL) and IL-10 (2251 ± 1181 vs 8065 ± 3610 pg/mL) produced in the MCM-treated groups with no difference in IL-2, IL-4 or IFN-c. In the absence of a stimulating agent, no significant Th1 or Th2 cytokine production was identified at any of the time points. Cytokines in melanoma and control media. Control medium and MCM contained very low
324 McCarter et al
Surgery August 2005
Fig 2. A, Th1 cytokines from MLR supernatants 1 to 7 days after DC exposure to MCM. B, Th2 cytokines from MLR supernatants 1 to 7 days after DC exposure to MCM. Control medium represented (diamonds) and MCM (squares). *P < .05, compared with control medium (n = 6). C, IL-12p70 production from DCs. DCs exposed to MCM for 24 hours followed by MLR for 7 days. There was no significant difference in IL-12p70 production between the groups. IL, Interleukin; IFN-g, interferon c.
Surgery Volume 138, Number 2
levels (0-5 g/mL) of Th1 and Th2 cytokines at baseline. Other potentially bioactive cytokines are present in both control medium and MCM. In particular, the MCM contained increased amounts of IL-6 and IL-8 and slightly higher levels of transforming growth factor ß (TGF-ß) relative to control medium (Table I). DC-based mixed lymphocyte response. Supernatants from the MLR experiments, in which DCs were preincubated with MCM, followed a similar pattern to those seen with whole PBMCs. There was an early burst in IL-2 (68 ± 23 vs 259 ± 87 pg/mL) production accompanied by a trend toward increased early IL-4, IL-5, IL-10 production in the MCM-treated group. By day 7, IL-2 was no longer detectable, and there was significant enhancement of persistent Th2 production of IL-5 (67 ± 22 vs 139 ± 37) and IL-10 (46 ± 19 vs 106 ± 34 pg/mL) in the MCM-treated group (Fig 2, A and B). There was no significant difference in IL-12 production between groups at day 7 (Fig 2, C). T-cell proliferation in the MLR. Overall, there was no significant difference in CD4+, CD8+, or total T-cell proliferation induced by MCM-treated DCs. However, after 7 days in the MLR, there were significantly fewer CD4+ staining T cells present in the MCM-treated group (37 vs 50%, P < .05). This was accompanied by a trend towards an increased percentage of CD8+ T cells in the MCM-treated group at 7 days. (Fig 3) DC maturation. Exposure of DCs (in PBMCs) to MCM did not appear to alter their ability to mature, as judged by the similar percentages and expression of maturation markers CD40 and CD86 (Fig 4, A). This observation remained the same even with larger numbers of derived DCs and variable lengths of exposure to MCM up to 7 days. Use of known DC-maturing agents such as CD40L and TNF-a in the setting of MCM did not change the results (Fig 4, B) Finally, there was no significant difference in the production of IL-12 with any of the treatment groups or maturation agents used (data not shown). DISCUSSION Patients with advanced or progressing melanoma frequently have circulating cytokine levels reflecting a Th2 pattern.6,16 It had been assumed that T cells primarily were responsible for this observation, but how and why this pattern evolved is not well understood. From a teleologic point of view, a sustained Th2 response would create a tolerogenic environment in which melanoma could continue to grow and spread unencumbered
McCarter et al 325
by the immune system. Melanoma does have some direct effects on T cells; however, the mechanisms involved in the genesis of a sustained Th2 response in cancer patients is not known.17,18 Results from the present experiments indicate that melanoma cells themselves do not produce Th2-type cytokines, yet they do create an environment in which stimulated PBMCs induce a Th2 response. When whole PBMCs were exposed to MCM and then stimulated, we observed an initial burst of both Th1 and Th2 cytokines. However, longer-term cultures suggested a preponderance of Th2 cytokines. This observation corresponds to the clinical findings from the sera of patients with advanced melanoma.6 Although it is presumed that these observed differences in cytokines measured from the supernatants are due to an increased production of cytokines, similar results could be observed if there was a decrease in select cytokine consumption by other cells in culture. The coordination of a durable immunologic response requires the activation of antigenpresenting cells, such as DCs, to stimulate specific T helper cells.9 An effective antitumor response involves antigen processing, DC cell maturation, and specific costimulatory signals to T cells that subsequently produce a Th1 response.10 To better understand the potential role of DCs in generating the observed Th2 response, we isolated DCs, exposed them to MCM, then added back allogenic T cells in an MLR. The global trend toward a sustained longer-term Th2 response in the MLR suggests that the DCs may be, in part, responsible for maintaining this effect. Exactly how the melanoma is conditioning the DCs to facilitate a Th2 response remains to be determined. Cultured human melanoma cell lines vary considerably in their activity on cells from the immune system although we have identified at least one other cell line (A375) that produces similar effects on the T-helper response (data not shown). Previous experience with colon and breast cancer cell lines found variable immunosuppressive effects on macrophages, depending on the characteristics of the individual cell line.14 As DCs go through the maturation process, they transition from antigen uptake and processing to T-cell stimulating machines.9 Some investigators have suggested that patients with advanced cancers tend to have fewer or less mature DCs.19,20 However, in these experiments, exposure to MCM did not appear to significantly influence the level of maturation, as determined by typical phenotypic maturation markers. Furthermore, there was no significant difference in IL-12 production in MCM
326 McCarter et al
Surgery August 2005
Fig 3. T-cell populations and proliferation after DC exposure to MCM. Control medium (solid bars) and MCM-treated cells (stippled bars). *P < .05, compared with control medium (n=6). Allogenic T cells added to DCs for 7 days after 24 hours of exposure to MCM.
Fig 4. A, Expression of DC maturation markers after exposure to MCM. PBMCs exposed to control (solid bars) or MCM (stippled bars). Graph represents percent of DCs (CD11c+, CD123+) expressing markers of maturation (CD86 and CD40). B, Derived DCs exposed to control or MCM and assessed for mean florescent intensity (MFI) surface expression of CD40 after stimulation with CD40L (solid bars) and TNF-a (hashed bars). Representative data from several experiments with different MCM incubation times and stimulating agents.
DCs. IL-12 is a key cytokine secreted by DCs to promote a Th1 response21; therefore, the absence of significant IL-12 may have allowed a Th2 response to dominate.
The T-cell proliferative response is used as a measure of DC activation of T cells.13 Cancer patients are reported to have impaired T-cell proliferative capacity.17 In these experiments, we did
McCarter et al 327
Surgery Volume 138, Number 2
not observe any significant differences in total T-cell proliferation in the MLR. One explanation for this observation may be found in the pronounced early burst of IL-2 in MCM-treated groups since IL-2 is a critical cytokine for T-cell proliferation and growth.22 It may be that the relatively early burst in IL-2 at day 3 obscured any observable effects on proliferation at day 7. When analyzing the CD4+ and CD8+ cells, we noted a significant decrease in the number of CD4+ cells at 7 days in the MCM-treated group. This suggests that MCMtreated DCs could not keep CD4+ cells alive and well for the full duration of the MLR. Other investigators have found that cancer patients with progressive disease tend to have lower CD4 counts, and fewer CD4 cells were present within metastatic deposits of those patients who failed therapy, compared with those who were responding to therapy.23,24 Furthermore, CD4 cells are shorter lived, and IL-2 tends to sustain CD8 proliferative capacity more than CD4 cells do.25 Additional experiments are needed to determine the effect of MCM on T-cell apoptosis and DC apoptosis. Several melanoma-derived cytokines (IL-6, IL-8, prostaglandin E2 [PGE2], TGF-b) are known to influence the immunologic response.15,26-28 Although commonly implicated as an immunosuppressive agent, PGE2 is not likely to play a major role in these experiments since there was relatively low levels of PGE2 in any of the tested media.29,30 TGF-b is an immunosuppressive agent present in most serum-containing media.30,31 While it could have played a role in these experiments, this explanation seems unlikely since only slightly higher concentrations were present in the MCM than in the control medium. Of the tested cytokines, it would seem that IL-6 may be the most biologically active since it was markedly elevated in the MCM, compared with control. Furthermore, IL-6 is known to impede the Th1 response and is a marker of progressing metastatic melanoma.32,33 Other as yet partially characterized factors are secreted by melanoma and produce a variety of immunosuppressive effects.14 A Th2 cytokine profile is associated with downregulation of the immunologic response and progression of melanoma. The interactions between melanoma, components of the immune system (such as DCs and Tcells), and the local environment are complex. These results suggest that melanoma may skew DCs to facilitate a more-tolerogenic Th2 response rather than an antitumor Th1 response. Understanding how this pattern evolves may help unravel the mechanisms of tumor-induced immune
suppression and lead to improved targeted immunologic therapies.
REFERENCES 1. Pulendran B. Modulating TH1/TH2 responses with microbes, dendritic cells, and pathogen recognition receptors. Immunol Res 2004;29:187-96. 2. Del Prete G. The concept of type-1 and type-2 helper T cells and their cytokines in humans. Int Rev Immunol 1998;16: 427-55. 3. Bennett WA, Lagoo-Deenadayalan S, Whitworth NS, Brackin MN, Hale E, Cowan BD. Expression and production of interleukin-10 by human trophoblast: relationship to pregnancy immunotolerance. Early Pregnancy 1997;3: 190-8. 4. Rankin EB, Yu D, Jiang J, et al. An essential role of Th1 responses and interferon gamma in infection-mediated suppression of neoplastic growth. Cancer Biol Ther 2003; 2:687-93. 5. Zelenika D, Adams E, Humm S, Lin CY, Waldmann H, Cobbold SP. The role of CD4+ T-cell subsets in determining transplantation rejection or tolerance. Immunol Rev 2001; 182:164-79. 6. Lauerova L, Dusek L, Simickova M, et al. Malignant melanoma associates with Th1/Th2 imbalance that coincides with disease progression and immunotherapy response. Neoplasma 2002;49:159-66. 7. Winter H, Hu HM, Poehlein CH, et al. Tumour-induced polarization of tumour vaccine-draining lymph node T cells to a type 1 cytokine profile predicts inherent strong immunogenicity of the tumour and correlates with therapeutic efficacy in adoptive transfer studies. Immunology 2003;108: 409-19. 8. Schultz ES, Schuler-Thurner B, Stroobant V, et al. Functional analysis of tumor-specific Th cell responses detected in melanoma patients after dendritic cell-based immunotherapy. J Immunol 2004;172:1304-10. 9. Mellman I, Steinman RM. Dendritic cells: specialized and regulated antigen processing machines. Cell 2001;106:2558. 10. Banchereau J, Schuler-Thurner B, Palucka AK, Schuler G. Dendritic cells as vectors for therapy. Cell 2001;106:271-4. 11. Anderson CF, Lucas M, Gutierrez-Kobeh L, Field AE, Mosser DM. T cell biasing by activated dendritic cells. J Immunol 2004;173:955-61. 12. Kaufman HL, Disis ML. Immune system versus tumor: shifting the balance in favor of DCs and effective immunity. J Clin Invest 2004;113:664-7. 13. Mazzoni A, Segal DM. Controlling the Toll road to dendritic cell polarization. J Leukoc Biol 2004;75:721-30. 14. Naama HA, McCarter MD, Mack VE, et al. Suppression of macrophage nitric oxide production by melanoma: mediation by a melanoma-derived product. Melanoma Res 2001; 11:229-38. 15. Biggs MW, Eiselein JE. Suppression of immune surveillance in melanoma. Med Hypotheses 2001;56:648-52. 16. Tatsumi T, Kierstead LS, Ranieri E, et al. Disease-associated bias in T helper type 1 (Th1)/Th2 CD4(+) T cell responses against MAGE-6 in HLA-DRB10401(+) patients with renal cell carcinoma or melanoma. J Exp Med 2002;196: 619-28. 17. Giacomoni D, Ben-Efraim S, Najmabadi F, Dray S. Inhibitors of lymphocyte activation secreted by human melanoma cell lines. Med Oncol Tumor Pharmacother 1990;7:273-80.
328 McCarter et al
18. Zhang XM, Xu Q. Metastatic melanoma cells escape from immunosurveillance through the novel mechanism of releasing nitric oxide to induce dysfunction of immunocytes. Melanoma Res 2001;11:559-67. 19. Onishi H, Morisaki T, Baba E, et al. Dysfunctional and shortlived subsets in monocyte-derived dendritic cells from patients with advanced cancer. Clin Immunol 2002;105: 286-95. 20. Almand B, Clark JI, Nikitina E, et al. Increased production of immature myeloid cells in cancer patients: a mechanism of immunosuppression in cancer. J Immunol 2001;166:678-89. 21. Knutson KL, Disis ML. IL-12 enhances the generation of tumour antigen-specific Th1 CD4 T cells during ex vivo expansion. Clin Exp Immunol 2004;135:322-9. 22. Seder RA, Ahmed R. Similarities and differences in CD4+ and CD8+ effector and memory T cell generation. Nat Immunol 2003;4:835-42. 23. Hakansson A, Gustafsson B, Krysander L, Hakansson L. Tumour-infiltrating lymphocytes in metastatic malignant melanoma and response to interferon alpha treatment. Br J Cancer 1996;74:670-6. 24. Kuss I, Hathaway B, Ferris RL, Gooding W, Whiteside TL. Imbalance in absolute counts of T lymphocyte subsets in patients with head and neck cancer and its relation to disease. Adv Otorhinolaryngol 2005;62:161-72. 25. Iezzi G, Karjalainen K, Lanzavecchia A. The duration of antigenic stimulation determines the fate of naive and effector T cells. Immunity 1998;8:89-95.
Surgery August 2005
26. Ijland SA, Jager MJ, Heijdra BM, Westphal JR, Peek R. Expression of angiogenic and immunosuppressive factors by uveal melanoma cell lines. Melanoma Res 1999;9:445-50. 27. Ugurel S, Rappl G, Tilgen W, Reinhold U. Increased serum concentration of angiogenic factors in malignant melanoma patients correlates with tumor progression and survival. J Clin Oncol 2001;19:577-83. 28. Redondo P, Sanchez-Carpintero I, Bauza A, Idoate M, Solano T, Mihm MC Jr. Immunologic escape and angiogenesis in human malignant melanoma. J Am Acad Dermatol 2003;49:255-63. 29. Denkert C, Kobel M, Berger S, et al. Expression of cyclooxygenase 2 in human malignant melanoma. Cancer Res 2001;61:303-8. 30. Botti C, Seregni E, Ferrari L, Martinetti A, Bombardieri E. Immunosuppressive factors: role in cancer development and progression. Int J Biol Markers 1998;13:51-69. 31. Gold LI. The role for transforming growth factor-beta (TGF-beta) in human cancer. Crit Rev Oncog 1999;10:30360. 32. Dodge IL, Carr MW, Cernadas M, Brenner MB. IL-6 production by pulmonary dendritic cells impedes Th1 immune responses. J Immunol 2003;170:4457-64. 33. Mouawad R, Rixe O, Meric JB, Khayat D, Soubrane C. Serum interleukin-6 concentrations as predictive factor of time to progression in metastatic malignant melanoma patients treated by biochemotherapy: a retrospective study. Cytokines Cell Mol Ther 2002;7:151-6.