Secondary Logo

Journal Logo

Original articles

Expression of Langerhans cells, dermal dendritic cells, and plasmacytoid dendritic cells in early stage mycosis fungoides: a case–control study

Fawzy, Marwa M.a; Abd El-latif, Mostafa I.a; Hegazy, Rehaba; Tawdy, Amiraa; Katayama, Ichirob

Author Information
Journal of the Egyptian Women's Dermatologic Society: September 2014 - Volume 11 - Issue 3 - p 208-214
doi: 10.1097/01.EWX.0000445907.70209.91
  • Free



Mycosis fungoides (MF) is the most common type of primary cutaneous T-cell lymphoma 1. Possible contributing stimuli to the pathogenesis of MF involve chronic inflammation and clonal abnormalities in the neoplastic T cells 2. To date, it is unknown why this lymphoma remains skin bound for a considerable time, and the exact mode responsible for the slow tumor progression is as yet obscure. One possible mechanism explaining the prolonged disease development could be that, in early stage MF, the patient’s immune system controls tumor progression by an antitumor immune response 3 in which dendritic cells (DCs) are thought to actively participate.

Miyagaki and Sugaya 4 have mentioned that certain cytokines and chemokines that are overexpressed in MF patients’ sera, such as thymic stromal protein and CC chemokine ligand 18, potently activate immature myeloid DC and induce DC chemotaxis, respectively. Moreover, the interleukin 22-producing T cells, which are overexpressed in MF lesional skin, induce chemotaxis of dermal dendritic cells (DDCs) and Langerhans cells (LCs) through CC chemokine ligand 20 expression by keratinocytes.

DCs are potent antigen-presenting cells that help orchestrate the innate and adaptive immune systems to induce tolerance and immunity. They are diversified in their phenotypes, stages of maturation, degrees of activation, and functions 5. DCs are believed to play a pivotal role in the immunobiology of cutaneous lymphoma 2. However, their role in MF progression is still not well understood 6.

Several subtypes of DCs exist among human lymphoid tissues, nonlymphoid tissues, and in peripheral blood. In the skin, three types of DCs are described: LCs, DDCs, and plasmacytoid dendritic cells (pDCs) 5.

Among the most frequently studied was the LCs; yet, controversial reports have emerged regarding their expression, and thereby their possible role in MF 2,5–13. Both DDCs 5–7,10 and pDCs 5–7 had their share in relatively fewer studies in the literature, yielding obscure and conflicting results.

The aim of the current study was to focus on evaluating the three different lineages of DCs, LCs, DDCs, and pDCs, through determining the expression of their significant markers CD1a, factor XIIIa, and CD123, respectively, particularly in the early stage MF (IA, IB, IIA) to clarify the possible role played by DCs in the complex pathogenesis of MF. We also aimed to determine if any correlation exists between them and the clinical as well as the histopathological severity of such cases.

Patients and methods

This prospective case–control study was approved by the Dermatology Research Ethics Committee (REC) Office, Faculty of Medicine, Cairo University. The Declaration of Helsinki principles were followed in our study. The MF patients were recruited from the Cairo University Hospital, Dermatology Outpatient Clinic over 6 months from January to June during 2009, and written informed consent was signed by each participant before initiating the study. Skin samples of healthy controls were obtained from the archives of Dermatology Department, Faculty of Medicine, Osaka University (Japan).

The study included 16 patients with early stage MF (IA, IB, and IIA). Six age and sex matched healthy controls were also included in the study. All included patients were off any systemic or topical treatment apart from emollients.

Each patient was subjected to complete history taking and full clinical examination, and the diagnosis was confirmed by histopathological examination of a 5 mm punch skin biopsy taken from the suspected lesion and stained by hematoxylin and eosin stain. Staging of MF was further completed by lymph node examination and biopsy (when indicated) in the surgery outpatient clinic, Faculty of Medicine, Cairo University. Laboratory investigations (complete blood picture, liver and kidney function tests, fasting and 2 h postprandial blood sugar, and lactate dehydrogenase) and radiological investigations (chest radiograph and abdominal ultrasound) were carried out.

Patients’ assessment as being mild, moderate, or severe was carried out by two fixed blinded observers. The clinical score was calculated through evaluating the degrees of erythema, induration, scaling, extent, and poikiloderma. Skin sections were evaluated for epidermotropism, density of dermal infiltrate, and presence of individual haloed atypical lymphocytes, according to which the histopathological score was determined. All biopsies were examined by two qualified dermatopathologists. Both clinical and pathological scores were calculated by the same protocol performed by El Mofty et al. 14.


Skin biopsies were taken from lesional skin of MF patients and from normal skin of normal control individuals and fixed with 10% formalin.

Paraffin sections of 3 μm were cut and dried, followed by deparaffinization and rehydration. Sections were stained according to the protocol mentioned by Abd El-Latif et al.15. Briefly, sections were heated for antigen retrieval, blocked with 3% hydrogen peroxide, washed twice with PBS containing 0.05% Tween 20 (PBS/Tween 20), and incubated in blocking solution (Protein Block, Serum-Free; Dako, Carpinteria, California, USA) for 15 min. Sections were then incubated for 1 h at room temperature with mouse monoclonal anti-CD1a antibody (M3571, Clone 010; Dako), mouse monoclonal anti-CD123 antibody (14-1239, Clone 6H6; eBioscience, San Diego, California, USA), or mouse monoclonal anti-factor XIIIa antibody (ab1834, Clone AC-1A1; Abcam, Cambridge, UK), with dilution ratio of 1/50, 1/200, and 1/300, respectively. All sections were then washed twice and processed using LSAB+System-AP kit (Dako) according to the manufacturer’s protocol. Reaction products of the new fuchsin substrate system were visualized and sections were counterstained with hematoxylin. The localization of the antibody was determined as membranous and cytoplasmic for CD1a, membranous for CD123, and cytoplasmic for factor XIIIa. Our main concern was the exact number of different types of DCs in the stained sections. Accordingly, we assessed the markers expression regardless of the intensity of the staining. The number of DC markers positive cells was counted independently by two researchers in three high-power fields in the epidermis and dermis together 16–18.

Statistical analysis

Data were statistically described in terms of mean±standard deviation (SD) and range for quantitative data, or frequencies (number of cases) and percentages for qualitative data. Comparison of numerical variables between the study groups was performed using the Mann–Whitney U-test for independent samples. Within-group comparison of numerical variables was performed using the Wilcoxon signed-rank test for paired (matched) samples. Correlation between various variables was determined using the Spearman rank correlation equation. Accuracy was represented using the terms sensitivity and specificity. Receiver operator characteristic (ROC) analysis was used to determine the optimum cutoff value for the studied diagnostic markers. P value less than 0.05 was considered statistically significant. All statistical calculations were performed using computer program statistical package for the social science (SPSS Inc., Chicago, Illinois, USA) version 15 for Microsoft Windows.


The current prospective study included 16 patients (10 men, 63%, and six women, 37%) with early stage MF (IA, IB, IIA). Their age ranged between 21 and 62 years (46.19±10.99 years). Six healthy controls (two men, 33%, four women, 67%) with age range between 37 and 74 years (58.67±13.5 years) were also included. The demographic and clinical data of the included patients are elaborated in Table 1.

Table 1
Table 1:
Demographic and clinical data of the included patients

Dendritic cells expression

As the CD123 and factor XIIIa were negative in the epidermis in both groups, and as the CD1a was almost negative in the dermis of healthy controls, we assessed the DC markers in both epidermis and dermis together.

The mean number of positive cells of both CD1a-positive LCs and factor XIIIa DDcs was significantly higher in the MF patients (3.5–8.7, 6.08±1.5 and 6–15.5, 8.6±2.4, respectively) in comparison with the controls (0.3–4.8, 2.7±1.6 and 1.3–9.5, 4.8±2.9, P=0.001 and 0.013, respectively). However, no significant difference was documented regarding the mean number of CD123-positive pDCs between the MF patients (0–3.7, 0.8±0.95) and the controls (0–1.2, 0.2±0.47) (P=0.13) (Figs 1–3) (Table 2).

Figure 1
Figure 1:
Photomicrograph of a section in the skin stained for CD1a. (a, b) Patients with mycosis fungoides (MF) showing both dermal and epidermal CD1a immunostaining. (c, d) Control skin showing mainly epidermal CD1a immunostaining. The number of positive signals is much more in the MF patients than in healthy controls. CD1a expression was membranous or cytoplasmic in both groups (CD1a immunostaining with Mayer’s hematoxylin counterstain, magnification ×100).
Figure 2
Figure 2:
Photomicrograph of a section in the skin stained for factor XIIIa. (a, b) Mycosis fungoides (MF) patient showing only dermal factor XIIIa immunostaining. (c, d) Control skin showing only dermal factor XIIIa immunostaining. The number of positive signals is much more in the MF patients than in healthy controls. Factor XIIIa expression was cytoplasmic in both groups (factor XIII immunostaining with Mayer’s hematoxylin counterstain, magnification ×100).
Figure 3
Figure 3:
Photomicrograph of a section in the skin for CD123. (a, b) Mycosis fungoides patient showing almost negative CD123 immunostaining. The few detected signals were only dermal. (c, d) Control skin showing almost negative CD123 immunostaining. The few detected signals were only dermal. The number of positive signals is almost equal in both groups. CD123 expression was membranous in both groups (CD123 immunostaining with Mayer’s hematoxylin counterstain, ×100).
Table 2
Table 2:
Comparison between the studied groups with respect to the assessed markers

In the MF lesions as well as in control skin, the expression of the CD123 pDCs was significantly less than both CD1a-positive LCs and factor XIIIa DDcs (P=0.001, 0.027, and 0.001, 0.028, respectively); in addition, factor XIIIa DDcs showed significant higher expression than CD1a-positive LCs in MF lesions only (P=0.002).

We used ROC curve to verify the possibility of these DC markers to be used as diagnostic markers for MF. ROC analysis determined the best cutoff values of both CD1a-positive LCs and factor XIIIa DDcs in diagnosing MF. The area under the curve of CD1a-positive LCs and factor XIIIa DDcs was 0.948 and 0.844 (95% confidence interval 0.858–1.037, 0.582–1.106), with the best cutoff values of 3.167 and 5.833, which achieved sensitivity of 100%, specificity of 67%, and sensitivity of 100%, specificity 67%, respectively (Figs 4 and 5). None of these markers showed high specificity at the best cutoff values.

Figure 4
Figure 4:
Graph showing receiver operator characteristic (ROC) analysis determining the best cutoff values of CD1a-positive Langerhans cells (LCs) in diagnosing mycosis fungoides. The area under the curve of CD1a-positive LCs was 0.948 (95% confidence interval 0.858–1.037), with the best cutoff values of 3.167 that achieved sensitivity of 100% and specificity of 67%.
Figure 5
Figure 5:
Graph showing receiver operator characteristic (ROC) analysis determining the best cutoff values of factor XIIIa dermal dendritic cells (DDCs) in diagnosing mycosis fungoides. The area under the curve of factor XIIIa DDCs was 0.844 (95% confidence interval 0.582–1.106), with the best cutoff values of 5.833 that achieved sensitivity of 100% and specificity of 67%.

A significant correlation was only found between CD123 pDCs and XIIIa DDcs expression in MF patients (r=0.538, P=0.039). There were no significant correlations between the three DC lineages and the different variables (age, sex, stage, duration, and clinical and histopathological scoring) (Table 3).

Table 3
Table 3:
Spearman’sρ correlations between CD1, CD123, and factor XIIIa and clinical and pathological data of the mycosis fungoides patients


The current study reestablishes the possible role played by different types of DCs in the complex pathogenesis of MF. This was evident through the significantly higher mean number of positive cells of both CD1a-positive LCs and factor XIIIa DDcs in the MF patients in comparison with the controls. In contrast, in our study, no clear role could be deduced for the pDcs at least in the early stages of the disease, as no significant difference was documented regarding the mean number of CD123-positive pDCs between the MF patients and the controls.

In agreement with our study, several previous reports documented CD1a-positive LCs and factor XIIIa DDcs to be increased in MF 10,19–23. Pautrier microabscesses, the pathognomonic feature of epidermotropic early MF, have been shown to be composed of intraepidermal collections of stimulated and proliferating malignant cells, adherent to the dendrites of intraepidermal dendritic antigen-presenting LCs 9, thereby indicating a dynamic communication between the two cell types. On the basis of evidence from few studies showing the in-vivo association of the MF cells with LCs, a hypothesis for the LCs in MF proposed that cutaneous T cell lymphoma are diseases of chronic T-cell stimulation by LC-mediated antigen presentation 2. All these factors point to the important role suggested to be played by LCs in the progression of MF through this ‘cross-talk’ between both the LCs and the tumor cells.

In contrast, others suggested that CD1a-positive LCs appear to be reactive against tumor cells in MF because their numbers increase as the number of neoplastic cells increases 8. Bani and Moretti 12 demonstrated no close apposition between LCs and tumor cells. Furthermore, Meissner et al.24 examined retrospectively the influence of LCs on the survival of 35 MF patients and found a better prognosis in patients with high CD1a-positive epidermal cells. There is no clear explanation to the huge discrepancy between the results of the different studies, and whether the LCs appear in the MF lesions to help in the initiation and progression of the disease or to halt it. This is a crucial point that needs further studying as it could be a pivotal point for the future therapeutic plan.

In a trial to answer this dilemma, several authors proposed the ‘dual-role’ for the DCs. It is becoming increasingly evident that DCs play a key role in the cancer immunoediting taking place in MF. This could be attributed to the fact that, on one hand, DCs are capable of mounting effective immune responses to tumors. On the other hand, they may promote tumor immune escape by inducing tumor antigen-specific anergy. It appears that the distribution of DC subsets within the tumor and their maturation state have a profound impact on the clinical outcome of the disease 25. This refers to their participation in the cancer immunoediting hypothesis, which defines tumor outgrowth as the result of an imbalance between immunosurveillance and tumor immune escape 26.

Surprisingly, our study documented no upregulation for the pDCs in contrast to other studies 6,7. This discrepancy might be attributed to the fact that we focused only on early stage MF, whereas these studies dealt with more advanced stages and sezary syndrome. The dynamic microenvironment with the variable cellular expression changing with the different disease stages is a rule in MF 27, which could be hypothesized to apply to the pDCs as well – that is, having a more pronounced role in the later stages. However, this was not what was expected, specially that pDCs represent a key component in innate antiviral immunity because of their capacity to produce large amounts of IFN-α 28, and one of the hypotheses placed for the pathogenesis of MF is the chronic response to a viral agent. Furthermore, the IFN-α reflects the Th1 cytokine profile that is the signature of early stage MF 29.

The significant correlation between CD123 pDCs and XIIIa DDcs expression in MF patients, despite the nonsignificant CD123 pDCs confirms the possible role of pDCs that is more evident in the advanced stages and sezary syndrome. Our study showed no significant correlations between any of the three DC lineages and different variables (age, sex, stage, duration, and clinical and histopathological scoring). Schlapbach 6 reported similar findings as well, showing no correlation between the density of the DCs and the clinical staging of the disease. We find no surprise in that finding, as MF is a multifactorial disease and no one factor could significantly influence its severity. However, Schwingshackl et al.7 showed a positive correlation between the histological intensity of the tumor infiltrate and DC numbers.


Our study confirms the active participation of the two lineages of the studied DCs in the early stage MF; however, several limitations prevented us from reaching a clear answer to the preset question, which is what is the exact role played by the different subsets of DCs in the complex story of MF. Whether it plays ‘good cup’ or ‘bad cup’ remains unclear. Whether the DCs appeared in reaction to the tumor cells to antagonize their progression, or the DCs represented a source for chronic antigenic stimulation resulting in the immune escape of the tumor cells and thereby the perpetuation of the disease needs to be established. Larger scale studies are still needed to clarify these controversies, as further understanding of the DCs and their role in MF can help us to uncover the pathogenesis of this disease and to further explore the therapeutic uses of DCs.


Conflicts of interest

There are no conflicts of interest.


1. Willemze R1, Jaffe ES, Burg G, Cerroni L, Berti E, Swerdlow SH, et al.. WHO-EORTC classification for cutaneous lymphomas. Blood 2005; 105:3768–3785.
2. Der-Petrossian M1, Valencak J, Jonak C, Klosner G, Dani T, Müllauer L, et al.. Dermal infiltrates of cutaneous T-cell lymphomas with epidermotropism but not other cutaneous lymphomas are abundant with langerin dendritic cells. J Eur Acad Dermatol Venereol 2011; 25:922–927.
3. Lüftl M1, Feng A, Licha E, Schuler G. Dendritic cells and apoptosis in mycosis fungoides. Br J Dermatol 2002; 147:1171–1179.
4. Miyagaki T, Sugaya M. Immunological milieu in mycosis fungoides and Sézary syndrome. J Dermatol 2014; 41:11–18.
5. Ni X1, Duvic M. Dendritic cells and cutaneous T-cell lymphomas. G Ital Dermatol Venereol 2011; 146:103–113.
6. Schlapbach C1, Ochsenbein A, Kaelin U, Hassan AS, Hunger RE, Yawalkar N. High numbers of DC-SIGN+ dendritic cells in lesional skin of cutaneous T-cell lymphoma. J Am Acad Dermatol 2010; 62:995–1004.
7. Schwingshackl P1, Obermoser G, Nguyen VA, Fritsch P, Sepp N, Romani N. Distribution and maturation of skin dendritic cell subsets in two forms of cutaneous T-cell lymphoma: mycosis fungoides and Sézary syndrome. Acta Derm Venereol 2012; 92:269–275.
8. Goteri G1, Filosa A, Mannello B, Stramazzotti D, Rupoli S, Leoni P, Fabris G. Density of neoplastic lymphoid infiltrate, CD8+ T cells, and CD1a+ dendritic cells in mycosis fungoides. J Clin Pathol 2003; 56:453–458.
9. Edelson RL. Cutaneous T cell lymphoma: the helping hand of dendritic cells. Ann N Y Acad Sci 2001; 941:1–11.
10. Fivenson DP1, Nickoloff BJ. Distinctive dendritic cell subsets expressing factor XIIIa, CD1a, CD1b and CD1c in mycosis fungoides and psoriasis. J Cutan Pathol 1995; 22:223–228.
11. Fujita M1, Horiguchi Y, Miyachi Y, Furukawa F, Kashihara-Sawami M, Imamura S. A subpopulation of Langerhans cells (CD1a+Lag-) increased in the dermis of plaque lesions of mycosis fungoides. J Am Acad Dermatol 1991; 25:491–499.
12. Bani D, Moretti S. Are Langerhans cells usual components of the dermal infiltrate of mycosis fungoides? Arch Dermatol Res 1987; 279:561–563.
13. Cox NH, Turbitt ML, Ashworth J, Mackie RM. Distribution of T cell subsets and Langerhans cells in mycosis fungoides, and the effect of PUVA therapy. Clin Exp Dermatol 1986; 11:564–568.
14. El Mofty M, Ramadan S, Fawzy MM, Hegazy RA, Sayed S. Broad band UVA: a possible reliable alternative to PUVA in the treatment of early-stage mycosis fungoides. Photodermatol Photoimmunol Photomed 2012; 28:274–277.
15. Abd El-Latif MI1, Murota H, Terao M, Katayama I. Effects of a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor and low-density lipoprotein on proliferation and migration of keratinocytes. Br J Dermatol 2010; 163:128–137.
16. Wada N, Zaki MA, Kohara M, Ogawa H, Sugiyama H, Nomura S, et al.. Diffuse large B cell lymphoma with an interfollicular pattern of proliferation shows a favourable prognosis: a study of the Osaka Lymphoma Study Group. Histopathology 2012; 60:924–932.
17. Murota H, El-latif MA, Tamura T, Amano T, Katayama I. Olopatadine hydrochloride improves dermatitis score and inhibits scratch behavior in NC/Nga mice. Int Arch Allergy Immunol 2010; 153:121–132.
18. Zaki MA, Wada N, Ikeda J, Shibayama H, Hashimoto K, Yamagami T, et al.. Prognostic implication of types of tumor-associated macrophages in Hodgkin lymphoma. Virchows Arch 2011; 459:361–366.
19. Pigozzi B, Bordignon M, Belloni Fortina A, Michelotto G, Alaibac M. Expression of the CD1a molecule in B- and T-lymphoproliferative skin conditions. Oncol Rep 2006; 15:347–351.
20. Fivenson DP, Hanson CA, Nickoloff BJ. Localization of clonal T cells to the epidermis in cutaneous T-cell lymphoma. J Am Acad Dermatol 1994; 315 Pt 1717–723.
21. Chu A, Berger CL, Kung P, Edelson RL. In situ identification of Langerhans cells in the dermal infiltrate of cutaneous T cell lymphoma. J Am Acad Dermatol 1982; 6:350–354.
22. Cerio R, Spaull J, Oliver GF, Jones WE. A study of factor XIIIa and MAC 387 immunolabeling in normal and pathological skin. Am J Dermatopathol 1990; 12:221–233.
23. Tirumalae R, Panjwani PK. Origin use of CD4, CD8, and CD1a immunostains in distinguishing mycosis fungoides from its inflammatory mimics: a pilot study. Indian J Dermatol 2012; 57:424–427.
24. Meissner K, Löning T, Rehpenning W. Epidermal Langerhans cells and prognosis of patients with mycosis fungoides and Sezary syndrome. In Vivo 1993; 7:277–280.
25. Chaput N, Conforti R, Viaud S, Spatz A, Zitvogel L. The Janus face of dendritic cells in cancer. Oncogene 2008; 27:5920–5931.
26. Dunn GP, Old LJ, Schreiber RD. The three Es of cancer immunoediting. Annu Rev Immunol 2004; 22:329–360.
27. Wong HK, Mishra A, Hake T, Porcu P. Evolving insights in the pathogenesis and therapy of cutaneous T-cell lymphoma (mycosis fungoides and Sezary syndrome). Br J Haematol 2011; 155:150–166.
28. Cella M, Jarrossay D, Facchetti F, Alebardi O, Nakajima H, Lanzavecchia A, Colonna M. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nat Med 1999; 5:919–923.
29. Whittemore AS, Holly EA, Lee IM, Abel EA, Adams RM, Nickoloff BJ, et al.. Mycosis fungoides in relation to environmental exposures and immune response: a case–control study. J Natl Cancer Inst 1989; 81:1560–1567.

dermal dendritic cells; Langerhans cells; mycosis fungoides; plasmacytoid dendritic cells

© 2014 Egyptian Women's Dermatologic Society