Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

FoxP3 and indoleamine 2,3-dioxygenase immunoreactivity in sentinel nodes from melanoma patients

American Journal of Otolaryngology



1) Assess FoxP3/indoleamine 2,3-dioxygenase immunoreactivity in head and neck melanoma sentinel lymph nodes and 2) correlate FoxP3/indoleamine 2,3-dioxygenase with sentinel lymph node metastasis and clinical recurrence.

Study Design

Retrospective cohort study.


Patients with sentinel lymph node biopsy for head and neck melanoma between 2004 and 2011 were identified. FoxP3/indoleamine 2,3-dioxygenase prevalence and intensity were determined from the nodes. Poor outcome was defined as local, regional or distant recurrence. The overall immunoreactivity score was correlated with clinical recurrence and sentinel lymph node metastasis using the chi-square test for trend.


Fifty-six sentinel lymph nodes were reviewed, with 47 negative and 9 positive for melanoma. Patients with poor outcomes had a statistically significant trend for higher immunoreactivity scores (p = 0.03). Positive nodes compared to negative nodes also had a statistically significant trend for higher immunoreactivity scores (p = 0.03). Among the negative nodes, there was a statistically significant trend for a poor outcome with higher immunoreactivity scores (p = 0.02).


FoxP3/indoleamine 2,3-dioxygenase immunoreactivity correlates with sentinel lymph node positivity and poor outcome. Even in negative nodes, higher immunoreactivity correlated with poor outcome. Therefore higher immunoreactivity may portend a worse prognosis even without metastasis in the sentinel lymph node. This could identify a subset of patients that may benefit from future trials and treatment for melanoma through Treg and IDO suppression.

1. Introduction

Melanoma is an aggressive cancer with 76,250 new cases of invasive melanoma and 55,560 cases of melanoma in situ in 2012 [1] . The incidence continues to steadily rise at the fastest rate of any malignancy in the U.S and abroad [1] . Head and neck melanoma, which comprises 18% of primary lesions, has a worse prognosis compared to other locations [2], [3], and [4]. Ten-year survivability is dependent on various factors including primary location [5] and [6], vertical growth phase, thickness, level of invasion, ulceration, and density of mitotic figures [7] and [8]. According to the American Joint Committee on Cancer (AJCC) the most important prognostic factor is the number of nodal metastases [9] . Overall 35%–5% of patients with regional metastases develop distant disease [6] . Furthermore, the status of the sentinel lymph nodes (SLNs) in patients with clinically negative lymph nodes (LNs) is the most important prognostic factor for recurrence [10] . In patients with nodal disease, but no distant metastases (AJCC stage III), the outcomes are remarkably varied in terms of ultimate distant metastases and mortality [9] . This demonstrates the challenge in prognosticating and managing melanoma.

The SLN is in the regional LN basin that receives direct lymphatic flow from the primary lesion and therefore the first metastases. It can be identified by dye, gamma probe, lymphoscintigraphy or a combination at a rate of 94% [11] . SLNs may contain macrometastases, which are evident on exam or imaging, or sub-clinical micrometasteses. Micrometastases occur in 15%–20% of head and neck melanomas [7] . These can be identified on standard hematoxylin and eosin staining or may require specific immunohistochemical (IHC) staining or reverse transcriptase polymerase chain reactions (PCR).

SLN biopsy has become the standard of care to determine subclinical LN metastases and has replaced elective lymph node dissection (ELND) in melanoma. This is supported by a lack of survival benefit from up-front ELND versus SLN biopsy [12] . This has also been confirmed in the head and neck population [13] . A positive SLN leads to a completion neck dissection (CND) that still results in a 13.1% nodal recurrence and a 39.7% total recurrence probability [11] . This suggests that a positive SLN biopsy is an indicator that the melanoma has already spread systemically. Despite the use of SLN biopsy, the death rate from melanoma continued to rise from 1990 to 2006 [11] . Therefore, the presence of positive SLN suggests not only the need for CND, but also the need to consider adjuvant therapies and closer surveillance to detect recurrence.

There are likely other critical factors in SLNs that correlate with clinical outcomes. One factor may be regulatory T cells (Tregs). Tregs exhibit a CD4 + CD25 + FoxP3 + phenotype and are immunosuppressive. Forkhead box P3 (FoxP3) is a transcription repressor expressed on activated Tregs [14] . Tregs can become part of the tumor microenvironment and suppress proliferation and activity of effector T cells and thereby allow tumor immune escape. Increased infiltration of Tregs has been reported in various types of tumors and SLNs [15] and [16]. There are no studies comparing Treg expression in SLN with outcome specifically in the head and neck melanoma population. Increased Treg in SLN may identify a subset of patients that have worse clinical outcomes after CND.

Measuring Treg prevalence in SLN may identify patients that would benefit from treatments targeting Treg. One such agent is Ipilimumab, a human monoclonal antibody that blocks cytotoxic T lymphocyte antigen 4 (CTLA-4). CTLA-4 down-regulates T cell activation and is found on cytotoxic T cells and Tregs. Ipilumumab has recently shown an improvement in overall survival in previously treated metastatic melanoma [17] . Several other agents active against Tregs are being studied in the basic science, translational and clinical settings [17], [18], and [19].

Ipilumumab has a substantial benefit in some metastatic melanoma patients, but many derive no therapeutic benefit from it [17] . Other immunologic factors can affect Treg function. Both tumors and native immune cells can express indoleamine 2,3-dioxygenase (IDO), an enzyme that activates Tregs to promote tumor immune escape. IDO acts as an intracellular enzyme that catabolizes tryptophan degradation. IDO knockout mammalian models and those given IDO inhibitors have slower melanoma growth and longer survival when treated with Ipilumumab [16], [20], [21], and [22]. New agents against IDO are actively being studied, some in conjunction with Treg inhibitors [14], [20], [22], and [23].

Our objectives were to 1) assess FoxP3/IDO immunoreactivity in head and neck melanoma SLNs and 2) correlate FoxP3/IDO with SLN metastasis positivity and clinical outcome. This could identify a subset of patients that may benefit from future trials and treatment for melanoma through Treg and IDO suppression.

2. Materials and methods

Duke University School of Medicine institutional review board approved this study. Patients with a new diagnosis of primary malignant melanoma of the head or neck between 1990 and 2005 were identified. Those who underwent complete excision with negative surgical margins and SLN biopsy were included. Those who were lost to follow up or died from a cause unrelated to melanoma during the study period were excluded. A review of their medical records was done to determine demographics, histologic subtype, location, stage and recurrence up to five years. Recurrence was classified as local, regional or distant. Poor outcome was defined as local, regional or distant recurrence.

SLN samples from the Duke University Medical Center Pathology archives were obtained and new sections were made. Sectioning was done by a pathology department researcher (Z.S.) who was blinded to LN status and clinical information. Tissue sections were deparaffinized, rehydrated, subjected to antigen retrieval, incubated and then counterstained. IHC staining for expression of FoxP3 and IDO was performed. The FoxP3 antibody was a rat monoclonal antibody from eBioscience (Catalog# 14-5773). The IDO antibody was a rat monoclonal antibody from Santa Cruz (Catalog# sc53978). For both stains the antigen retrieval was by citrate buffer in pH of 6.1 at 100 C for 20 minutes. The primary antibody incubation was for 45 minutes at room temperature. The secondary biotinylated rabbit anti-rat antibody was from Vector Laboratories (Catalog# BA-4001) and was incubated for 30 minutes. Tertiary detection was then performed with ABC Elite from Vector Laboratories (Catalog# PK-7100) for 30 minutes with a substrate chromogen of liquid DAB + from Dako (Catalog# k3468) for 5 minutes. Staining for CD3 and CD4 was also initially performed to confirm that the FoxP3 + cells represented T cells, rather than another FoxP3 + cell type. FoxP3 positivity alone has previously been used as the most specific marker of Tregs [15], [24], [25], and [26].

The stained slides were then reviewed and scored by two independent study team members (M.R., J.C.) according to a modified German Immunoreactive Scoring System ( Table 1 ). Both were blinded to the patient data, SLN status and clinical outcome at the time of the scoring. A third physician (W.L.) was available to review the slides of any scores that differed by 2 points or more. The intensity and prevalence were determined under microscopic review as an overall average over each entire section.

Table 1 Modified German Immunoreactive Scoring System based on the intensity of staining and the prevalence of positive cells.

Intensity of staining FoxP3 + prevalence IDO + prevalence
0 No staining No positive cells No staining
1 Weak staining 1%–5% positive cells 1%–50%
2 Moderate staining 6%–10% positive cells > 50%
3 Intense staining 11%–20% positive cells  
4   21%–40% positive cells  

FoxP3 = forkhead box P3; IDO = indoleamine 2,3-dioxygenase.

The raw number scores for the intensity and prevalence were determined for both FoxP3 and IDO. The intensity was calculated on the same scale for both markers, but there was a different range of prevalence and therefore a different modification of the German Immunoreactive Scoring System were used for each marker. The two reviewer's raw scores were averaged and then the product of this averaged FoxP3 and IDO score was calculated. This gave a final FoxP3/IDO immunoreactivity score that was either low, intermediate or high. A chi-square test for trend on Graphpad (Version 6) was performed with immunoreactivity, SLN metastasis positivity and clinical outcome.

3. Results

We analyzed 56 SLNs from a total of 30 patients. Forty-seven SLNs were associated with a good clinical outcome (i.e. no evidence of disease) and 9 with a poor clinical outcome. There were a total of 47 negative SLNs and 9 positive SLNs. There was not exact correlation between the 9 positive SLNs and the 9 associated with bad outcome, as 5 of the SLNs associated with a poor clinical outcome were negative for metastasis ( Table 2 ). In total, 19 SLNs had low, 26 intermediate and 11 high immunoreactivity scores.

Table 2 Patient and tumor characteristics.

  + SLN − SLN
Patients (n = 30) n = 7 n = 23
Age 47.4 (17–67) 57.4 (29–76)
Male gender 86% 83%
SLN (n = 56) n = 9 n = 47
Tumor stage    
 1a   1
 1b 1  
 2a   7
 2b   3
 3a 1 1
 3b 2 3
 4a 1 3
 4b 1 5
 Scalp 2 5
 Face 1 4
 Ear   4
 Nose 1 2
 Lip   1
 Neck 3 7
 Superficial spreading 2 8
 Nodular 1 4
 Lentigo maligna   3
 Desmoplastic   2
 Spindle cell   2
 Unspecified 4 4
Good outcome 4 20
Poor outcome 3 3
 Local, then distant   1
 Regional 1  
 Regional, then distant 1 1
 Distant 1 1

SLN = sentinel lymph node.

SLNs with a good clinical outcome (N = 47) had 38% (n = 18) low, 47% (n = 22) intermediate and 15% (n = 7) high immunoreactivity. Those with poor clinical outcome (N = 9) had a statistically significant trend of higher scores than those with good outcomes, with 11% (n = 1) low, 44% (n = 4) intermediate and 44% (n = 4) high (p = 0.03) ( Fig. 1 ).


Fig. 1 Percentage of each immunoreactivity score by clinical outcome. SLN = sentinel lymph node; FoxP3 = forkhead box P3; IDO = indoleamine 2,3-dioxygenase.

The negative SLNs (N = 47) had 36% (n = 17) low, 51% (n = 24) intermediate, 13% (n = 6) high immunoreactivity. The positive SLNs (N = 9) showed a statistically significant trend of higher scores than the negative ones with 22% (n = 2) low, 22% (n = 2) intermediate and 56% (n = 5) high (p = 0.03) ( Fig. 2 ).


Fig. 2 Percentage of each immunoreactivity score by SLN status (either negative or positive for melanoma metastasis). SLN = sentinel lymph node; FoxP3 = forkhead box P3; IDO = indoleamine 2,3-dioxygenase.

Among negative SLNs, 42 correlated with a good outcome and 5 with a bad outcome. Those with a good outcome had 40% (n = 17) low, 50% (n = 21) intermediate and 10% (n = 4) high immunoreactivity. Those with a poor outcome had a statistically significant trend of higher scores than those with a good outcome, with 0% (n = 0) low, 60% (n = 3) intermediate and 40% (n = 2) high (p = 0.02) ( Fig. 3 ).


Fig. 3 Percentage of each immunoreactivity score for only SLNs that are negative for melanoma metastasis. SLN = sentinel lymph node; FoxP3 = forkhead box P3; IDO = indoleamine 2,3-dioxygenase.

4. Discussion

These results are the first to show a statistically significant trend toward higher FoxP3/IDO immunoreactivity in positive SLNs compared to negative SLNs in patients with head and neck melanoma. This supports previous studies of Tregs in SLN in melanoma as well as different tumors. For example, in a study of SLNs in breast cancer, the majority of the FoxP3 +/IDO + SLNs were also positive for metastasis [27] . Lee et al. [28] showed a correlation of SLN tumor burden with prevalence of FoxP3 and IDO expressing cells in melanoma SLNs by PCR. Despite attempts to correlate degree of FoxP3 + Treg infiltration in and around the primary cutaneous melanoma with clinical outcome, but no correlation was found [25] . In contrast, on our IHC review we found increased FoxP3 expression in cells surrounding the areas of metastatic melanoma compared to other areas of the SLN. Additionally, in the positive SLNs in all-site melanoma. Brody et al. used IHC to show an association between FoxP3 and IDO expression and shorter survival [29] . Our positive SLNs had a trend toward higher FoxP3/IDO immunoreactivity, regardless of the clinical outcome.

Furthermore, we showed a trend toward higher FoxP3/IDO immunoreactivity with a poor clinical outcome. This is congruent with previous results in uveal melanoma, all-site melanoma and non-melanoma tumors. Intratumoral Treg prevalence has been shown to be an independent negative prognostic marker for uveal melanoma [30] . Knol et al. [24] showed that Tregs decrease progression free survival in all-site metastatic melanoma. The prognostic significance of IDO immunoreactivity in addition to metastasis has also been shown in SLNs of 160 patients in all-site melanoma by Speeckaert et al. [31] . These patients had a worse overall survival and progression free survival with high IDO expression  [31] . IDO expression correlated with CTLA-4 expression in Tregs. Laimer et al. [32] showed that IDO expression predicted shorter survival in oral squamous cell carcinoma, which suggests a potential to evaluate SLNs in these patients and treat according to IDO expression. IDO expression also correlated with FoxP3 expression, which is not surprising given IDOs known function in Treg activation [31] . Interestingly, this correlation was significant in SLNs that were negative, but not positive for metastases [31] . This suggests a significant role of Tregs and IDO, even in negative SLNs.

Most interestingly, our FoxP3/IDO immunoreactivity correlated with poor outcome even when the SLN was negative for melanoma metastasis. This supports other studies showing higher prevalence of FoxP3 + with high IDO expression in negative SLNs in all-site melanoma [31] . If negative SLNs are biopsied then CND is not typically indicated. Aggressive follow up and adjuvant treatment are unlikely to be pursued, unless otherwise indicated. This seems appropriate for patients with negative SLNs and low immunoreactivity, since they all had a good clinical outcome. The outcomes are less clear with intermediate or high immunoreactivity. Even negative SLNs with an intermediate or high compared to a low immunoreactivity score, had a statistically significant trend (p = 0.02) of a worse clinical outcome. This suggests that even without regional metastasis at the time of SLN, IHC may be able to more precisely prognosticate outcome. Patients with high or even intermediate FoxP3/IDO immunoreactivity could be considered for adjuvant treatments and close follow up, similar to those recommended for patients with positive SLNs. Furthermore, traditional detection of melanoma micrometastasis in SLN is limited by the number of sections taken and the histology techniques applied. This varies by institution and false negatives occur. de Rosa et al. [11] calculated a false negative rate of 20.4% in their review. The addition of FoxP3/IDO IHC to the melanoma SLN protocols may help identify patients that need more aggressive management. Using FoxP3/IDO as an immune biomarker may relieve some of the burden of the technical aspect of staining SLN to detect tumor metastases.

If a high FoxP3/IDO microenvironment is detected, prognosis may be worse, but this negative prognostic factor may not be insurmountable. Activated Tregs contribute to a pro-tumor immunologic condition in SLNs, but their immune suppression is not constitutive. Tregs maintain plasticity even after being activated. IDO keeps Tregs in an immune suppressor mode, but when IDO is blocked in mammalian models by 1-MT, Tregs can respond to other factors in the microenvironment and convert to a non-suppressor phenotype [33] . This supports the future use of FoxP3/IDO IHC analysis on head and neck melanoma, regardless of the presence of SLN metastasis, to determine who would benefit most from therapeutics that target Treg, IDO and related immunologic factors.

The prior referenced studies have included few head and neck SLNs and no data analysis on this sub-population. Given the complex SLN drainage pathways in the head and neck, as well as generally worse outcomes, analysis of this difficult to adequately power sub-population has been needed. Future studies need to continue addressing this sub-population and additionally could include mucosal melanomas, given their poor prognosis.

IHC is easily and commonly done on pathologic specimens, but IHC for FoxP3 + and IDO + cells is a surrogate marker for activated Tregs. The process of Treg activation by IDO and subsequent inhibition of antitumor activity is a more elaborate, functional process than IHC can fully characterize. Unlike prior studies in melanoma that have assessed only FoxP3 or IDO immunoreactivity, the combination of the two in our study captures the dynamic functionality of Tregs better. There are other components that may be uncovered to better understand the immunology that inhibits the host's antitumor microenvironment.

This IHC analysis was not precisely quantitative. Quantitative scanners are available and could be used in future studies, although they are not necessarily used in routine clinical IHC. This study was also limited by its retrospective nature. Future directions include a prospective study validating the correlation defined between FoxP3/IDO immunoreactivity and outcome from this study. This prospective study should be done in a blinded fashion and used to predict clinical outcomes. Furthermore, Treg and IDO analysis on SLN could be an inclusion criterion for future clinical trials investigating new therapies targeting Tregs, IDO and immune system tumor escape.

5. Conclusion

In this retrospective cohort study of sentinel lymph node biopsy in head and neck melanoma subjects, FoxP3/indoleamine 2,3-dioxygenase immunoreactivity was correlated with presence of melanoma and clinical recurrence. This suggests a potential role of these immune markers in sentinel nodes for prognostication.


Support received from AAO-HNSF CORE Grant and American Medical Association Seed Grant (M.R.). This project was supported by Career Development Award (IK2BX001398) from the Biomedical Laboratory Research and Development Service of the Department of Veterans Affairs Office of Research and Development (W.T.L.). The views expressed in this article are those of the author (W.T.L.) and do not necessarily represent the views of the Department of Veterans Affairs or United States Government.


  • [1] A. Jemal, R. Siegel, J. Xu, et al. Cancer statistics, 2010. CA Cancer J Clin. 2010;60:277-300 Crossref
  • [2] A.M. Lachiewicz, M. Berwick, C.L. Wiggins, et al. Survival differences between patients with scalp or neck melanoma and those with melanoma of other sites in the Surveillance, Epidemiology, and End Results (SEER) program. Arch Dermatol. 2008;144:515-521
  • [3] M.J. White, H.C. Polk Jr. Therapy of primary cutaneous melanoma. Med Clin North Am. 1986;70:71-87
  • [4] P.A. Gimotty, D. Guerry. Prognostication in thin cutaneous melanomas. Arch Pathol Lab Med. 2010;134:1758-1763
  • [5] D.L. Larson, J.D. Larson. Head and neck melanoma. Clin Plast Surg. 2010;37:73-77 Crossref
  • [6] S.R. Fisher. Elective, therapeutic, and delayed lymph node dissection for malignant melanoma of the head and neck: analysis of 1444 patients from 1970 to 1998. Laryngoscope. 2002;112:99-110 Crossref
  • [7] S.P. Davison, M.S. Clifton, L. Kauffman, et al. Sentinel node biopsy for the detection of head and neck melanoma: a review. Ann Plast Surg. 2001;47:206-211 Crossref
  • [8] D.E. Elder, P.A. Gimotty, D. Guerry. Cutaneous melanoma: estimating survival and recurrence risk based on histopathologic features. Dermatol Ther. 2005;18:369-385 Crossref
  • [9] C.M. Balch, S.J. Soong, J.E. Gershenwald, et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol. 2001;19:3622-3634
  • [10] J.E. Gershenwald, W. Thompson, P.F. Mansfield, et al. Multi-institutional melanoma lymphatic mapping experience: the prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol. 1999;17:976-983
  • [11] N. de Rosa, G.H. Lyman, D. Silbermins, et al. Sentinel node biopsy for head and neck melanoma: a systematic review. Otolaryngol Head Neck Surg. 2011;145:375-382 Crossref
  • [12] C.M. Balch, S. Soong, M.I. Ross, et al. Long-term results of a multi-institutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm). Intergroup Melanoma Surgical Trial. Ann Surg Oncol. 2000;7:87-97 Crossref
  • [13] W.J. Kane, P. Yugueros, R.P. Clay, et al. Treatment outcome for 424 primary cases of clinical I cutaneous malignant melanoma of the head and neck. Head Neck. 1997;19:457-465 Crossref
  • [14] W. Chen, X. Liang, A.J. Peterson, et al. The indoleamine 2,3-dioxygenase pathway is essential for human plasmacytoid dendritic cell-induced adaptive T regulatory cell generation. J Immunol. 2008;181:5396-5404 Crossref
  • [15] M. Beyer, J.L. Schultze. Regulatory T cells in cancer. Blood. 2006;108:804-811 Crossref
  • [16] D.A. Wainwright, M. Dey, A. Chang, et al. Targeting Tregs in malignant brain cancer: overcoming IDO. Front Immunol. 2013;4:116
  • [17] F.S. Hodi, S.J. O'Day, D.F. McDermott, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363:711-723 Crossref
  • [18] J. Guo, J. Zhu, X. Sheng, et al. Intratumoral injection of dendritic cells in combination with local hyperthermia induces systemic antitumor effect in patients with advanced melanoma. Int J Cancer. 2007;120:2418-2425 Crossref
  • [19] B. Weide, S. Pascolo, B. Scheel, et al. Direct injection of protamine-protected mRNA: results of a phase 1/2 vaccination trial in metastatic melanoma patients. J Immunother. 2009;32:498-507 Crossref
  • [20] A. Poljak, R. Grant, C.J. Austin, et al. Inhibition of indoleamine 2,3 dioxygenase activity by H2O2. Arch Biochem Biophys. 2006;450:9-19 Crossref
  • [21] H. Fujigaki, K. Saito, F. Lin, et al. Nitration and inactivation of IDO by peroxynitrite. J Immunol. 2006;176:372-379 Crossref
  • [22] G.D. Basu, T.L. Tinder, J.M. Bradley, et al. Cyclooxygenase-2 inhibitor enhances the efficacy of a breast cancer vaccine: role of IDO. J Immunol. 2006;177:2391-2402 Crossref
  • [23] R.B. Holmgaard, D. Zamarin, D.H. Munn, et al. Indoleamine 2,3-dioxygenase is a critical resistance mechanism in antitumor T cell immunotherapy targeting CTLA-4. J Exp Med. 2013;210:1389-1402 Crossref
  • [24] A.C. Knol, J.M. Nguyen, G. Quereux, et al. Prognostic value of tumor-infiltrating Foxp3 + T-cell subpopulations in metastatic melanoma. Exp Dermatol. 2011;20:430-434 Crossref
  • [25] A. Ladanyi, A. Mohos, B. Somlai, et al. FOXP3 + cell density in primary tumor has no prognostic impact in patients with cutaneous malignant melanoma. Pathol Oncol Res. 2010;16:303-309 Crossref
  • [26] A. Mohos, T. Sebestyen, G. Liszkay, et al. Immune cell profile of sentinel lymph nodes in patients with malignant melanoma—FOXP3 + cell density in cases with positive sentinel node status is associated with unfavorable clinical outcome. J Transl Med. 2013;11:43 Crossref
  • [27] A.S. Mansfield, P.S. Heikkila, A.T. Vaara, et al. Simultaneous Foxp3 and IDO expression is associated with sentinel lymph node metastases in breast cancer. BMC Cancer. 2009;9:231 Crossref
  • [28] J.H. Lee, Y. Chen, J.L. Chan, et al. Molecular analysis of melanoma-induced sentinel lymph node immune dysfunction. Cancer Immunol Immunother. 2011;60:685-692 Crossref
  • [29] J.R. Brody, C.L. Costantino, A.C. Berger, et al. Expression of indoleamine 2,3-dioxygenase in metastatic malignant melanoma recruits regulatory T cells to avoid immune detection and affects survival. Cell Cycle. 2009;8:1930-1934 Crossref
  • [30] D. Mougiakakos, C.C. Johansson, E. Trocme, et al. Intratumoral forkhead box P3-positive regulatory T cells predict poor survival in cyclooxygenase-2-positive uveal melanoma. Cancer. 2010;116:2224-2233
  • [31] R. Speeckaert, K. Vermaelen, N. van Geel, et al. Indoleamine 2,3-dioxygenase, a new prognostic marker in sentinel lymph nodes of melanoma patients. Eur J Cancer. 2012;48:2004-2011 Crossref
  • [32] K. Laimer, B. Troester, F. Kloss, et al. Expression and prognostic impact of indoleamine 2,3-dioxygenase in oral squamous cell carcinomas. Oral Oncol. 2011;47:352-357 Crossref
  • [33] M.D. Sharma, D.Y. Hou, Y. Liu, et al. Indoleamine 2,3-dioxygenase controls conversion of Foxp3 + Tregs to TH17-like cells in tumor-draining lymph nodes. Blood. 2009;113:6102-6111 Crossref


a Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, Duke University Medical Center, Durham, NC, USA

b Department of Pathology, Duke University Medical Center, Durham, NC, USA

c Surgical Service, Section of Otolaryngology–Head and Neck Surgery, Durham VA Medical Center, Durham, NC, USA

lowast Corresponding author at: DUMC 3805, Durham NC 27710, USA. Tel.: + 1 919 681 8449.

Each of the authors has contributed to, read and approved this manuscript. None of the authors have any conflict of interest, financial or otherwise. This manuscript is original and it, or any part of it, has not been previously published; nor is it under consideration for publication elsewhere.

☆☆ Presented at the American Academy of Otolaryngology–Head and Neck Surgery Foundation Annual Meeting, September 29–October 2 2013, Vancouver, Canada.