J Cutan Pathol 2015: 42: 730–738 doi: 10.1111/cup.12513 John Wiley & Sons. Printed in Singapore

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Journal of Cutaneous Pathology

Mycosis fungoides with epidermal mucinosis: a variant of mycosis fungoides with a spongiosis-like pattern Background: The histopathologic diagnosis of mycosis fungoides (MF) has classically relied on the presence of atypical epidermotropic T-lymphocytes predominating over spongiosis. However, in some cases of MF, prominent epidermal mucinosis in a spongiosis-like pattern mimics a spongiotic dermatitis. To our knowledge, only one series in the literature has thus far recognized the presence of epidermal mucinosis in MF. Methods: We evaluated 30 skin biopsies from 18 patients with the clinical diagnosis of MF, which fulfilled all histopathologic criteria for patch- or plaque-stage MF, but also showed epidermal mucinosis in a spongiosis-like pattern. A total of 15 specimens were studied by immunohistochemistry, and seven were tested for T-cell receptor (TCR) gene rearrangements. Twenty biopsies of spongiotic dermatitides were included as controls. Results: We confirmed the presence of epidermal mucinosis in all 30 cases of MF with a spongiosis-like pattern based on histopathologic criteria and the colloidal iron stain for mucin. Immunohistochemistry in 15 specimens showed significant loss of pan-T-cell antigens CD5 (10/15) and CD7 (14/15); and TCR clonality was detected in 7 specimens from 6 patients, supporting the diagnosis of MF. Conclusions: We report helpful histopathologic criteria for distinguishing MF with epidermal mucinosis in a spongiosis-like pattern from spongiotic dermatitis.

Stephanie W. Hu1 , Howard Ratech2 , Rizwan Naeem2 , Jo-Ann Latkowski3 and Hideko Kamino1,4

Keywords: cutaneous lymphomas, cutaneous T cell lymphoma, epidermal mucinosis, mycosis fungoides

Hideko Kamino, MD, Dermatopathology Section, Ronald O. Perelman Department of Dermatology, New York University School of Medicine, 330 East 38th Street, 42D New York, NY 10016, USA Tel: +212 867 5501 Fax: (212)684-2991 e-mail: [email protected]

Hu SW, Ratech H, Naeem R, Latkowski J-A, Kamino H. Mycosis fungoides with epidermal mucinosis: a variant of mycosis fungoides with a spongiosis-like pattern. J Cutan Pathol 2015; 42: 730–738. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Cutaneous T cell lymphomas (CTCLs) are a clinically and histopathologically diverse group of lymphoproliferative disorders that

730

1

Dermatopathology Section, Ronald O. Perelman Department of Dermatology, New York University School of Medicine, New York, NY, USA, 2 Department of Pathology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY, USA, 3 Ronald O. Perelman Department of Dermatology, New York University School of Medicine, New York, NY, USA, and 4 Department of Pathology, New York University School of Medicine, New York, NY, USA

Accepted for publication March 22, 2015

are characterized by the clonal accumulation of neoplastic T-lymphocytes in the skin.1 The World Health Organization and European

Mycosis fungoides with epidermal mucinosis Organization for Research and Treatment of Cancer (WHO–EORTC) classification of cutaneous lymphomas with primary cutaneous manifestations lists 13 entities.2 Mycosis fungoides (MF) represents the most common type of CTCL, accounting for 50% of patients and demonstrating a male predominance of approximately 2 : 1.2 Despite the early clinical clues to the diagnosis of MF, it is common for years to elapse before a histopathologic diagnosis can be established. Improvements in our ability to diagnose early MF continue to be made. In 1979, Sanchez and Ackerman studied 46 biopsy specimens of patch-stage lesions from patients in whom MF was unequivocally established and found that the critical feature for histopathologic diagnosis was ‘the presence of an increased number of mononuclear cells distributed singly or in small collections within an epidermis devoid of spongiotic microvesiculation.’3 In 1994, Shapiro and Pinto studied 222 skin biopsies of MF and Sezary syndrome to document the vast histopathologic spectrum and to evaluate the earliest histopathologic changes.4 Their results indicated that CTCL reproduced practically all of the histopathologic patterns used for diagnosing inflammatory skin disease,5,6 and that a common clue to the diagnosis of CTCL was epidermotropism with little spongiosis. The authors noted that even in the ‘spongiotic only’ subgroup of their histopathologic classification, ‘the spongiosis was always just slight.’4 The predominance of lymphocytes over spongiosis has long been an essential clue for the histopathologic diagnosis of MF, often serving as a key factor for distinguishing between early MF and the eczematous dermatitides, with which it is commonly confused. Interestingly, in 1986, Nickoloff reported a series of cases of patch- and plaque-stage MF in which significant spongiosis was observed in the epidermis, and documented the presence of colloidal iron-positive material in these zones of ‘spongiosis.’7 Nickoloff proposed that MF should be added to the list of diseases in which there is epidermal mucinosis. Since then, several unusual variants of MF have been identified, including a particularly uncommon variant associated with vesicles or blisters, termed vesicular, bullous or vesiculobullous MF.8 However, no further studies have reported epidermal mucinosis in MF. Classical teaching maintains that minimal spongiosis is a diagnostic clue to MF. As recently as 2005, the International Society for Cutaneous Lymphoma included ‘epidermotropism without spongiosis’ as one of the

standard histopathologic criteria in their diagnostic algorithm for early MF.9 We have observed cases of MF in which epidermal mucinosis in a spongiosis-like pattern predominated over epidermotropic T-lymphocytes in some foci, thus causing potential histopathologic confusion with spongiotic (eczematous) dermatitis. This retrospective study was undertaken in an attempt to histopathologically characterize this variant of MF with epidermal mucinosis. Materials and methods Case selection A total of 30 biopsies from 18 patients were collected from the files of the Dermatopathology Section at New York University Medical Center from 1994 to 2012 in which the clinical diagnoses included MF and/or CTCL, and the histopathologic diagnoses were patch-or plaque-stage MF with epidermal mucinosis in a spongiosis-like pattern. Some patients had multiple biopsies, either because multiple biopsies were submitted simultaneously or at different points in time. This study was approved by the New York University Medical Center Institutional Review Board. With the exception of epidermal mucinosis, the histopathologic diagnosis of patch- and plaque-stages of MF was based on previously published criteria by Sanchez and Ackerman,3 the WHO-EORTC2 and the International Society of Cutaneous Lymphomas.9 Diagnosis of MF required the presence of an epidermotropic, cytologically atypical lymphoid infiltrate. Clinicopathologic correlation was obtained for each biopsy. More detailed clinical history and follow-up were obtained when available. A total of 20 biopsies of spongiotic dermatitides from 18 patients were included as controls. The clinical diagnoses included allergic contact dermatitis, nummular dermatitis, atopic dermatitis, tinea and dyshidrotic dermatitis. In 19 cases, the corresponding histopathologic diagnoses were spongiotic dermatitis, consistent with the clinical diagnosis. In one case of spongiotic tinea, the presence of fungal hyphae was confirmed with the periodic acid-Schiff with diastase (PAS-D) stain. Histopathologic and immunohistochemical methods The tissues were fixed in 10% formalin, routinely processed and embedded in paraffin. In addition to routine hematoxylin and eosin (H&E) stain, epidermal and dermal mucin

731

Hu et al. (acidic mucopolysaccharides) were identified with the colloidal iron stain. A PAS-D stain was used to evaluate serum in the areas of spongiosis, vesiculation and crusting. For each biopsy, all sections were examined by two authors (SH and HK) and assessed for the following characteristics: the presence and degree of epidermal mucinosis, vesiculation, follicular mucinosis, dermal mucinosis, degree of epidermotropism of lymphocytes, lymphoid cytologic atypia and the presence of Pautrier microabscesses, intercellular edema (spongiosis), serum in the epidermis and cornified layer, Langerhans cell granulomas, eosinophils, dermal edema, dermal fibrosis and solar elastosis. The degree of mucinosis was graded as 1, mild; 2, moderate; and 3, significant. In 15 specimens from 15 patients with mild to significant epidermal mucinosis, immunohistochemical studies were performed on 4-μm-thick paraffin sections using the avidin-biotin horseradish peroxidase-antiperoxidase complex (ABC) method, preceded by microwave antigen retrieval of tissue sections. The antibodies used were CD3, CD4, CD5, CD7, CD8 and CD20 (Dako, Carpinteria, CA, USA). The immunohistochemical studies were either performed at the time of diagnosis or retrospectively obtained in those specimens with sufficient remaining tissue blocks. Detection of clonal T-cell receptor gene rearrangements Seven specimens from six patients, with mild to significant mucinosis, were tested for TCR-β and/or γ gene rearrangements. Two specimens also had vesiculation. Testing for T-cell receptor (TCR) clonality was performed at the time of diagnosis or retrospectively on available remaining tissue blocks. DNA was purified from formalin-fixed paraffin-embedded skin biopsies using the QIAamp DNA minikit (Qiagen Valencia, CA). Master mixes contained fluorescently labeled primers based on published BIOMED-2 sequences.10 All polymerase chain reactions (PCR) were performed on a GeneAmp PCR System 9700 (Applied Biosystems, Foster City, CA). The PCR products were detected by automated capillary gel electrophoresis using a 3130XL Genetic Analyzer (Applied Biosystems, Grand Island, NY, USA). Product sizes were calibrated with Gene Scan-500 LIZ size standards (Applied Biosystems). DNA from reactive tonsil was used as a polyclonal control; DNA from Jurkat T-cell line (American Type Culture Collection,

732

Manassas, VA, USA) was used as a clonal control for TCR-β and TCR-γ gene rearrangements. Results A total of 30 shave and punch biopsies from 18 patients with MF were studied. The clinical information, histopathologic findings, immunohistochemical results and TCR gene rearrangement studies are detailed in Table 1. The patients included 11 men and 7 women, ranging in age from 34 to 86 years. There were 10 patients with one biopsy, and 8 patients with multiple biopsies: 4 patients had two biopsies; and 4 patients had three biopsies. In six out of eight patients with multiple biopsies, the biopsies were taken during the same visit. In the two remaining patients, multiple biopsies were taken at different times, with intervals ranging from 1 to 15 years. Anatomic sites of biopsies included nine on the upper extremities, eight on the trunk, seven on the lower extremities, four on the buttocks and two on the axilla. In all 30 biopsies, the clinical diagnoses listed included MF and/or CTCL. In eight cases in which detailed clinical histories and follow-up were provided, seven had typical findings of patch-stage MF, with characteristic wrinkled patches. One patient presented with erythroderma. In all eight patients there was no clinical evidence of vesiculation or follicular lesions. Six of these eight patients were subsequently diagnosed with stage I or II disease, and all were well-controlled with topical corticosteroids, topical bexarotene or narrow-band ultraviolet B therapy. Of the remaining two, one patient developed extensive patch/plaque disease and small tumors that were ultimately responsive to topical nitrogen mustard, while the other patient developed stage III disease and, at most recent follow-up, was treated with romidepsin with little improvement. In all 30 skin biopsies diagnosed as patch or plaque stage MF, there was a superficial perivascular to moderately dense band-like infiltrate composed of atypical lymphocytes with enlarged, hyperchromatic nuclei and irregular nuclear contours. The atypical lymphocytes infiltrated the epidermis predominantly as single cells. In each biopsy, lymphocytes predominated over the spongiosis-like pattern in some foci. Pautrier microabscesses were present in three specimens. All 30 biopsies revealed widening of the intercellular spaces between keratinocytes, giving a spongiotic appearance to the epidermis, and vesicles were identified in six cases (Figs. 1A,B and 2A,B). The degree of

53

72

74

58 57

82

73 79

73

86

83

84

74

71

68

65 53

2

3

4

5 6

7

8 9

10

11

12

13

14

15

16

17 18

M M

M

F

F

M

F

F

M

F M

M

F M

F

M

M

M

Gender

R upper arm L upper arm L axilla L flank

R breast R buttock L arm

R thigh L thigh L upper back

R back R chest L hip L buttock R leg R arm L buttock

R thigh R forearm R medial thigh L flank Buttocks R arm R upper arm L elbow R knee R shin

Back

L forearm

Axilla

Biopsy site

CTCL MF vs. tinea versicolor vs. seborrheic dermatitis

MF

MF

MF

MF vs. Sezary syndrome

MF vs. allergic contact dermatitis MF

MF vs. contact dermatitis

MF MF vs. psoriasis

MF vs. lyphomatoid papulosis vs. drug reaction

MF MF

Recurrent MF

T-cell lymphoma vs. eczematous dermatitis MF

MF

Clinical diagnosis

MF, plaque stage MF, patch stage MF, plaque stage MF, patch stage

MF, plaque stage MF, plaque stage MF, patch stage

MF, patch stage MF, plaque stage MF, plaque stage

MF, plaque stage MF, patch stage MF, patch stage MF, patch stage MF, patch stage MF, plaque stage MF, patch stage

MF, patch stage MF, patch stage MF, plaque stage MF, plaque stage MF, patch stage MF, patch stage MF, patch stage MF, patch stage MF, patch stage MF, plaque stage

MF, plaque stage

MF, patch stage

MF, patch stage

H&E, hematoxylin and eosin stain; MF, mycosis fungoides; TCR, T-cell receptor. *1, mild mucinosis, 2, moderate mucinosis, 3, significant mucinosis.

34

1

Age

Histopathologic diagnosis

Yes/Yes 2 Yes/Yes 2 Yes/Yes 3 Yes/Yes 1 (focal 3)

Yes/Yes 1 Yes/Yes 2 Yes/Yes 2

Yes/Yes 3 Yes/Yes 3 (vesiculation) Yes/Yes 2

Yes/Yes 2 Yes/Yes 3 Yes/Yes 3 (vesiculation) Yes/Yes 2 Yes/Yes 3 Yes/Yes 2 Yes/Yes 3 Yes/Yes 3 Yes/Yes 3 Yes/Yes 3 (vesiculation and scale-crust) Yes/Yes 3 Yes/Yes 3 Yes/Yes 2 Yes/Yes 3 Yes/Yes 3 Yes/Yes 3 (vesiculation) Yes/Yes 1

Yes/Yes 3 (vesiculation)

Yes/Yes 3

Yes/Yes 2 (vesiculation)

Epidermal mucin (H&E/colloidal iron)*

Yes/Yes 2 Yes/Yes 1 Yes/Yes 3 Yes/Yes 1

Yes/Yes 1 Yes/Yes 1 Yes/Yes 2

Yes/Yes 3 Yes/Yes 3 Yes/Yes 1

Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 Yes/Yes 2

Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 Yes/Yes 3 Yes/Yes 2 Yes/Yes 2 Yes/Yes 3 Yes/Yes 2 Yes/Yes 2

Yes/Yes 1

Yes/Yes 3

Yes/Yes 2

Dermal mucin (H&E/colloidal iron)*

No No Yes/Yes 3 No

No No No

No No No

No No No No No No No

No No Yes/Yes 3 No Yes/Yes 2 Yes/Yes 2 Yes/Yes 2 No No No

Yes/Yes 2

No

No

CD3+, CD4:CD8 4:1, CD5+, CD7 (70% loss) CD3+, CD4:CD8 8:1, CD5 (40% loss), CD7 (80% loss)

CD3+, CD4:CD8 10:1, CD5+, CD7 (90% loss) CD3+, CD4:CD8 10:1, CD5 and CD7 (70% loss) CD3+, CD4+, CD8+ rare, CD5 and CD7 (80% loss) CD3+, CD4+, CD8−, CD5 and CD7 (80% loss) CD3+, CD4+, CD8−, CD5+, CD7 (80% loss)

CD3+, CD4:CD8 4:1, CD5 (80% loss), CD7 (60% loss) CD3+, CD4:CD8 1:1, CD5+, CD7 (70% loss)

CD3+, CD4:CD8 2:1, CD5+, CD7+ CD3+, CD4:CD8 1:1, CD5+, CD7 (70% loss)

CD3+, CD4:CD8 8:1, CD5 (90% loss), CD7 (90% loss)

CD3+, CD4:CD8 8:1, CD5 (95% loss), CD7 (90% loss) CD3+, CD4:CD8 8:1, CD5 and CD7 (50% loss)

CD3+, CD4:CD8 10:1, CD5(90% loss), CD7 (80% loss)

Follicular mucinosis (H&E/colloidal iron)* Immunohistochemistry

Table 1. Clinical, histopathological, immunohistochemical and molecular diagnostic data for cases of mycosis fungoides with epidermal mucinosis

𝛃, 𝛄

𝛃, 𝛄

𝛄

𝛃, 𝛄 𝛃, 𝛄

β, γ

Clonal TCR gene rearrangement

Mycosis fungoides with epidermal mucinosis

733

Hu et al.

Fig. 1. Mycosis fungoides with epidermal mucinosis, case #3. The epidermis shows significant mucinosis in a spongiosis-like pattern and vesiculation. There is epidermotropism of lymphocytes with hyperchromatic nuclei and irregular contours (A,B). The mucin is amphophilic with the H&E stain (A,B) and positive with the colloidal iron stain (C). PAS-D stain does not react with mucin and is negative for serum (D). Immunohistochemistry shows epidermotropism of CD3 + lymphocytes in the areas of mucinosis (E), with predominance of CD4+ lymphocytes (F). There is significant loss of CD5 (G) and CD7 expression (H).

spongiosis-like pattern did not necessarily correspond to the degree of epidermotropism by atypical lymphocytes. In addition, in 22 biopsies there was no crusting of the cornified layer overlying most of the areas with prominent spongiosis-like pattern and/or vesiculation.

734

In eight biopsies, minimal eosinophilic serum deposits were found in the cornified layer. In all H&E-stained sections, the widened intercellular spaces and vesiculation showed amphophilic, finely granular deposits consistent with mucin. In all 30 biopsies, colloidal iron stain confirmed

Mycosis fungoides with epidermal mucinosis

Fig. 2. Mycosis fungoides with epidermal mucinosis, case # 10. The epidermis shows significant mucinosis in a spongiosis-like pattern and early vesiculation. There is epidermotropism of lymphocytes with hyperchromatic nuclei and irregular contours (A,B). The mucin is positive with the colloidal iron stain (C). PAS-D stain is negative for serum (D).

the presence of epidermal mucinosis in these areas (Figs. 1C and 2C). The appearance of the mucin varied from fine and evenly distributed deposits between elongated intercellular bridges to small droplets to large pools of mucin in cases with a vesicular pattern. Follicular mucinosis accompanying epidermal mucinosis was seen in 6 of 30 biopsies; almost all occurred in cases with significant epidermal mucin deposits and/or vesiculation. PAS-D stain was negative for serum in 20 cases and only minimally positive for serum in foci within the epidermis and scale-crusts in eight cases (Figs. 1D and 2D). Langerhans cell granulomas were absent within vesicles and in areas of mucinosis in 29 out of 30 cases, and only one case showed a single Langerhans cell granuloma. In addition, papillary dermal fibroplasia was present in 26 out of 30 biopsies. Eosinophils were absent in 20 out of 30 biopsies. There was less than one eosinophil per high power field (HPF) in eight biopsies, one to two eosinophils/HPF in one biopsy, and three eosinophils/HPF focally in one biopsy associated with scale-crust. Of note, little or no edema of the papillary dermis was observed, but colloidal iron stain showed that the amount of dermal mucin was largely proportional to the degree of epidermal mucinosis. In 22 out of 30

biopsies, however, epidermal mucin was slightly more abundant than dermal mucin. Solar elastosis was present in biopsy specimens taken from sun-exposed areas of the body, but largely did not correlate with the degree of epidermal or dermal mucinosis. In 15 biopsies from 15 patients, immunohistochemistry demonstrated that the infiltrate was comprised predominantly of CD3-positive T-lymphocytes, and highlighted the epidermotropism. There was a predominance of CD4-positive over CD8-positive T-lymphocytes. In 10 out of 15 specimens there was significant loss of CD5 expression, ranging from 40 to 95%, and in 14 out of 15 specimens there was loss of CD7 expression, ranging from 50 to 90% (Figs. 1E–H). In all 15 biopsies, only rare dermal lymphocytes were positive for B-cell marker CD20. A total of seven specimens from six patients were tested for TCR gene rearrangements. Table 1 shows that in five specimens from five patients, both TCR-β and γ chain clonal gene rearrangements were detected. In the other two specimens, which were biopsied from two separate anatomic sites from the same patient, identical clonal TCR γ chain gene rearrangements were detected; TCR-β was not performed because of technical reasons. The demonstration

735

Hu et al.

Fig. 3. Spongiotic dermatitis, control. The epidermis shows significant spongiosis and spongiotic vesicles, with scattered small lymphocytes and eosinophils. A perivascular mixed-cell infiltrate is present within the papillary dermis (A). PAS-D stain highlights the positive serum within the spongiotic vesicles (B). Colloidal iron stain shows mild intercellular deposits of mucin (C).

of TCR clonality in seven out of seven specimens from six patients further supported the diagnosis of MF with mild to significant epidermal mucinosis. In all 20 biopsy control specimens of spongiotic/eczematous dermatitis, the areas of spongiosis showed widening of the epidermal intercellular spaces with deposits of eosinophilic serum, crusting and dermal edema. Clusters of Langerhans cells were present near the foci of spongiosis, and more than three eosinophils per tissue section were identified in all controls (Fig. 3A). Moreover, positive PAS-D staining was consistent with serum in areas of spongiosis, vesiculation and within crusts of the cornified layer (Fig. 3B). In 18 biopsies, these areas were negative to minimally positive with the colloidal iron stain (Fig. 3C); and moderately positive in two cases, one of which was a case of spongiotic tinea. Table 2 lists the histopathologic changes of MF with epidermal mucinosis that distinguish it from spongiotic dermatitis. Discussion The histopathologic diagnosis of MF has classically relied on the presence of atypical epidermotropic T-lymphocytes predominating over spongiosis, seen as widening of the intercellular spaces between keratinocytes. However, there

736

are cases of MF in which prominent epidermal mucinosis in a spongiosis-like pattern causes diagnostic uncertainty. We report the finding of epidermal mucinosis in 30 biopsy specimens from 18 patients with MF, in which the colloidal iron stain identified intercellular mucin deposition within the spongiosis-like pattern. We emphasize the possible diagnostic dilemma that could arise in distinguishing this variant of MF from spongiotic dermatitides with exocytosis of lymphocytes. To our knowledge, only one prior study has reported the presence of epidermal mucinosis in MF.7 The histopathologic points useful for distinguishing MF with epidermal mucinosis from spongiotic dermatitis are listed in Table 2. In particular, the amphophilic granular characteristics of the mucin deposits with the H&E stain in spongiosis-like areas, the positive reaction with the colloidal iron stain, the predominance of epidermal lymphocytes in some foci, the lack of Langerhans cell granulomas and the absence or minimal serum deposits in the cornified layer with the PAS-D stain support the diagnosis of MF with epidermal mucinosis, and not a spongiotic dermatitis. The paucity of eosinophils in most of the cases further decreases the likelihood of an eczematous dermatitis because less than three eosinophils per tissue section has been shown to be statistically significant in differentiating cases

Mycosis fungoides with epidermal mucinosis Table 2. Histopathologic clues to distinguish MF with epidermal mucinosis from spongiotic dermatitis. MF with epidermal mucinosis

Spongiotic dermatitis

Amphophilic, granular mucin (HE stain) Positive reaction with the colloidal iron stain

Spongiotic eosinophilic serum (HE stain) Positive reaction with the PAS-D stain Rarely mucin deposits with the colloidal iron stain Epidermal scattered lymphocytes

Epidermal lymphocytes predominate in some foci and a few Pautrier microabscesses Lymphocytes with hyperchromatic nuclei and irregular contours Rare Langerhans cell granulomas and eosinophils Lack of or minimal serum and scale crust in most cases Papillary dermal fibrosis

No atypical lymphocytes Frequent Langerhans cell granulomas and eosinophils Presence of serum and scale crust above areas of spongiosis Papillary dermal edema

MF, mycosis fungoides; PAS-D, periodic acid-Schiff with diastase stain.

diagnostic or suspicious for patch or plaque stage MF from an eczematous dermatitis.11 Papillary dermal fibroplasia was present in 26 out of 30 skin biopsies, a finding that is characteristic of MF and of the chronicity of the lesions. This contrasts with the papillary dermal edema that usually is present in a spongiotic dermatitis. There are cases of eczematized MF due to a superimposed allergic or irritant contact dermatitis. In these cases, there is spongiosis with eosinophilic serum in the epidermis, crusting on the surface and eosinophils and/or neutrophils in the infiltrate. Since the original documentation of vesicular variants of MF, no additional studies have been performed to evaluate the contents of the vesicles. This variant has been reported to show severe spongiosis and intraepidermal vesiculation, causing diagnostic confusion with inflammatory disorders.12 – 14 In one reported case, mild epidermal infiltration of unremarkable lymphocytes and absence of Pautrier microabscesses suggested a subacute eczematous dermatitis in the initial biopsy.14 However, in view of the clinical features, a repeat biopsy was performed, which showed an intense lymphohistiocytic infiltrate in the papillary dermis with extensive exocytosis of large cells with convoluted nuclei, and early Pautrier microabscesses, thus allowing for diagnosis of a spongiotic, vesicular variant of MF.14 Of note, seven out of eight of our patients, for whom detailed clinical histories were available, showed no clinical evidence of vesiculation, but rather harbored the characteristic wrinkled patches of early MF and one of these eight patients presented with erythroderma. This is in stark contrast to reported cases of the vesicular variant of MF, which have largely presented clinically with exudative lesions and vesicular morphologies.8,14 Additionally, 6 of 30 biopsies in the current study showed follicular mucinosis accompanying

epidermal mucinosis. Almost all occurred in cases with significant epidermal mucin deposition and/or vesiculation. In all eight patients with detailed clinical histories, there was no clinical evidence of follicular lesions. Follicular mucinosis has been implicated as an independent risk factor for rapid disease progression and higher mortality rates in MF.15 To our knowledge, the presence of epidermal mucinosis in cases of MF with follicular mucinosis has not been previously reported in peer reviewed series. However, there is a brief description and illustration of pilotropic MF with epidermal mucinosis in the textbook of ‘Skin Lymphoma’ by Cerroni.16 The mechanism of epidermal mucinosis, its immunologic properties and its role in MF require further study. From a pathophysiologic standpoint, perhaps most compelling is the hypothesis that the mucinous epithelial interstitium could harbor critical metabolic exchange properties and function in the concealment of surface antigens.17 The degree of acidic mucopolysaccharide deposition might alter the pathways through which inflammatory and malignant cells migrate within the epithelium, and might facilitate metabolic exchange between the internal and external environments. In fact, hyaluronic acid (HA), the major component of acid mucopolysaccharides, has a bifunctional role, in that it not only controls solute transport and micro-circulatory exchanges, due to its influence on interstitial volume, hydraulic conductibility and macromolecule diffusion but it is also involved in inflammatory aspects. HA fragments induce the expression of cytokines and chemokines as well as inducible nitric oxide synthase by macrophages, upregulates chemokines,18 affects extracellular matrix turnover in alveolar murine macrophages19 and stimulates transforming growth factor-β1 synthesis,20 which is considered to be a major pro-inflammatory pro-fibrotic

737

Hu et al. factor. It is thus conceivable that an expanded, mucinous interstitial space in the epidermis provides easy access for the immigration of transformed T-lymphocytes, and that this qualitatively altered chemical environment may also contribute to the survival and proliferation of the intraepidermal neoplastic T-cells. Given that chronic antigenic stimulation is hypothesized to be an instigator of MF,21 the potential immunomodulatory role of epidermal mucinosis in MF is a possible area for further investigation. Follicular mucinosis appears to result from the deposition of a stromal, dermal-type mucin composed of hyaluronic acid (staining positively with colloidal iron and the Alcian blue stain at pH 2.5, and non-reactive with PAS-D).22 The mucin production in follicular mucinosis has been also attributed to the secretory function of follicular keratinocytes as a possible response to stimulation by cytokines released from perifollicular T-lymphocytes, and is characterized by the mucinous degeneration

of follicular epithelium.23 – 25 While it appears that the epidermal mucinosis in our cases may consist of a similar acid mucin with similar staining properties, it is not clear why a dermal mucin should be deposited selectively within such epithelial structures in MF. Whereas the mechanisms of epidermal mucinosis and the immunologic and biochemical properties of the epithelial mucinous interstitium require further clarification, its role in MF may provide new insight into the cause and treatment of these conditions. The data presented in this study highlight the presence of epidermal mucin in cases of MF with a significant spongiosis-like pattern on H&E stain. This is emphasized because of the possible diagnostic dilemma that could arise in distinguishing this variant of MF from spongiotic/eczematous dermatitides that have lymphocytic infiltrates. We report helpful histopathologic criteria for distinguishing MF with epidermal mucinosis in a spongiosis-like pattern from spongiotic dermatitis.

References 1. Girardi M, Heald PW, Wilson LD. The pathogenesis of mycosis fungoides. N Engl J Med 2004; 350: 1978. 2. Willemze R, Jaffe ES, Burg G, et al. WHO–EORTC classification for cutaneous lymphomas. Blood 2005; 105: 3768. 3. Sanchez JL, Ackerman AB. The patch stage of mycosis fungoides: criteria for histologic diagnosis. Am J Dermatopathol 1979; 1: 5. 4. Shapiro PE, Pinto FJ. The histologic spectrum of mycosis fungoides/Sezary syndrome. Am J Surg Pathol 1994; 18: 645. 5. Ackerman AB. Histologic diagnosis of inflammatory skin diseases. Philadelphia: Lea and Febiger, 1978. 6. Ackerman AB. Supplement to the fourth printing of histologic diagnosis of inflammatory skin diseases. Philadelphia: Lea and Febiger, 1988. 7. Nickoloff BJ. Epidermal mucinosis in mycosis fungoides. J Am Acad Dermatol 1986; 15: 83. 8. Kazakov DV, Burg G, Kempf W. Clinicopathological spectrum of mycosis fungoides. J Eur Acad Dermatol Venereol 2004; 18: 397. 9. Pimpinelli N, Olsen EA, Santucci M, et al. Defining early mycosis fungoides. J Am Acad Dermatol 2005; 53: 1053. 10. Van Dongen JJ, Langerak AW, Bruggemann M, et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report

738

11.

12.

13.

14.

15.

16.

17.

18.

of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia 2003; 17: 2257. Dalton SR, Chandler WM, Abuzeid M, et al. Eosinophils in mycosis fungoides: an uncommon finding in the patch and plaque stages. Am J Dermatopathol 2012; 34: 586. McBride SR, Dahl MG, Slater DN, Sviland L. Vesicular mycosis fungoides. Br J Dermatol 1998; 138: 141. Gantcheva M, Lalova A, Broshtilova V, Negenzova Z, Tsankov N. Vesicular mycosis fungoides. J Dtsch Dermatol Ges 2005; 3: 898. Bowman PH, Hogan DJ, Sanusi ID. Mycosis fungoides bullosa: report of a case and review of the literature. J Am Acad Dermatol 2001; 45: 934. Van Doorn R, Van Haselen CW, Van Voorst Vader PC, et al. Mycosis fungoides: disease evolution and prognosis of 309 Dutch patients. Arch Dermatol 2000; 136: 504. Cerroni L. Mycosis fungoides. In Cerroni L ed. Skin lymphoma. The illustrated guide, 4th ed. Hoboken: Wiley-Blackwell, 2014; 36. Reed RJ. The T-lymphocyte, the mucinous epithelial interstitium and immunoestimulation. Am J Dermatopathol 1981; 3: 207. Horton MR, McKee CM, Bao C, et al. Hyaluronan fragments synergize with interferon-gamma to induce the CX-C chemokines mig and interferon inducible protein-10 in mouse macrophages. J Biol Chem 1998; 273: 35088.

19. McKee CM, Penno MB, Cowman M, et al. Hyaluronan (HA) fragments induce chemokine gene expression in alveolar macrophages: the role of HA size and CD44. J Clin Invest 1996; 98: 2403. 20. Horton MR, Olman MA, Noble PW. Hyaluronan fragments induce plasminogen activator inhibitor-1 and inhibit urokinase activity in mouse alveolar macrophages: a potential mechanism for impaired fibrinolytic activity in acute lung injury. Chest 1999; 116: 17S. 21. Cohen SR, Stenn KS, Braverman IM, Beck GJ. Mycosis fungoides: clinicopathologic relationships, survival, and therapy in 59 patients with observations on occupation as a new prognostic factor. Cancer 1985; 46: 2654. 22. Abel E. Inflammatory diseases of the epidermal appendages and of cartilage. In Elder D, Elenitsas R, Jaworsky C, Johnson B, eds. Lever’s histopathology of the skin, 8th ed. Philadelphia: Lippincott-Raven, 1997; 415. 23. Kaya G, Augsburger E, Chavaz P, Saurat JH. CD44 and hyaluronate expression in follicular mucinosis. J Cutan Pathol 2006; 33: 227. 24. Lancer HA, Bronstein BR, Nakagawa H, Bhan AK, Mihm MC Jr. Follicular mucinosis: a detailed morphologic and immunopathologic study. J Am Acad Dermatol 1984; 10: 760. 25. Ishibashi A. Histogenesis of mucin in follicular mucinosis: an electron microscopic study. Acta Derm Venereol 1976; 56: 163.

Mycosis fungoides with epidermal mucinosis: a variant of mycosis fungoides with a spongiosis-like pattern.

The histopathologic diagnosis of mycosis fungoides (MF) has classically relied on the presence of atypical epidermotropic T-lymphocytes predominating ...
1MB Sizes 0 Downloads 12 Views