Histiocytoid Sweet syndrome may indicate leukemia cutis: A novel application of fluorescence in situ hybridization Rahul N. Chavan, MD, PhD,a Mark A. Cappel, MD,e Rhett P. Ketterling, MD,b,c David A. Wada, MD,f Nicole M. Rochet, MD,d Ryan Knudson, BA, MBA,b and Lawrence E. Gibson, MDa,b Rochester, Minnesota; Jacksonville, Florida; and Salt Lake City, Utah Background: In patients with malignancy-associated Sweet syndrome, a thorough evaluation for leukemia cutis should be considered. Objective: We sought to describe the clinicopathologic characteristics of histiocytoid Sweet syndrome. Methods: We retrospectively identified patients with histiocytoid Sweet syndrome at our institution from January 1992 through December 2010. We evaluated the underlying cutaneous infiltrate using immunohistochemistry and fluorescence in situ hybridization. Results: We re-evaluated all 22 patients with hematologic malignancyeassociated Sweet syndrome. Six patients had a monocytoid infiltrate that was consistent with histiocytoid Sweet syndrome; subsequent evaluation of these patients demonstrated cytogenetic abnormalities on prior bone-marrow biopsy specimens. Fluorescence in situ hybridization analysis was feasible in cutaneous specimens from 5 of the 6 patients and demonstrated the same cytogenetic abnormalities that were identified on prior bonemarrow biopsy specimens in 4 patients. Therefore, these 4 patients may have had a form of leukemia cutis. Limitations: This was a retrospective study. Conclusion: For patients with histiocytoid Sweet syndrome, an underlying hematologic malignancy, and a monocytoid infiltrate on biopsy specimen, fluorescence in situ hybridization of the cutaneous infiltrate may be beneficial to identify cytogenetic abnormalities that may indicate leukemia cutis. ( J Am Acad Dermatol 2014;70:1021-7.) Key words: cytogenetic abnormalities; fluorescence in situ hybridization analysis; histiocytoid Sweet syndrome; leukemia cutis; malignancy-associated Sweet syndrome; Sweet syndrome.

I

n 1964, Sweet syndrome, also known as acute febrile neutrophilic dermatosis, was initially described by Robert Douglas Sweet.1 Since then, Sweet syndrome has become part of a larger group of dermatologic diseases with an underlying neutrophilic infiltrate.2 The diagnostic criteria of Sweet syndrome were delineated by Su and Liu,3 and one of the major criteria required for diagnosis is histopathologic evidence of a neutrophilic infiltrate, typically in the absence of leukocytoclastic vasculitis.2,4,5 Sweet syndrome can be associated with an From the Departments of Dermatology,a Laboratory Medicine and Pathology,b and Medical Genetics,c and College of Medicine,d Mayo Clinic, Rochester; Department of Dermatology, Mayo Clinic, Jacksonvillee; and Department of Dermatology, University of Utah, and Huntsman Cancer Institute.f Funding sources: None. Conflicts of interest: None declared. Accepted for publication January 25, 2014.

Abbreviations used: FISH: MLL:

fluorescence in situ hybridization mixed lineage leukemia gene

underlying hematologic myeloid disorder, solid tumor malignancies, inflammatory bowel disease, and gastrointestinal tract and upper respiratory tract infections.6-10 A number of medications are also known to cause Sweet syndrome.11,12 Reprint requests: Lawrence E. Gibson, MD, Department of Dermatology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. E-mail: [email protected]. Published online March 17, 2014. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.01.874

1021

J AM ACAD DERMATOL

1022 Chavan et al

JUNE 2014

More recently, Requena et al13 described a variant and duration of the lesion; underlying hematologic of Sweet syndrome, histiocytoid Sweet syndrome, in malignancy; duration of disease; bone-marrow biwhich a predominantly histiocytoid-appearing infilopsy specimen abnormalities; and treatment and trate was identified on histopathologic evaluation. response to treatment. The mononuclear cells were immunoreactive for In 6 patients, a chromosomal abnormality was CD15, CD43, CD45, CD68, MAC-386, HAM56, myenoted in the bone-marrow biopsy specimen. In 5 loperoxidase, and lysozyme, consistent with patients, FISH analysis was performed to evaluate for histiocytoid-appearing the presence of a chromoimmature myeloid cells.13 In somal abnormality in the CAPSULE SUMMARY a subset of the cases, fluorescutaneous infiltrate of their cence in situ hybridization initial biopsy specimen. In 1 Histiocytoid Sweet syndrome is (FISH) studies to determine patient, it was not possible to characterized by a predominance of the presence of BCR/ABL design a FISH strategy to mononuclear cells on histopathologic gene fusion was performed specifically evaluate the cutaevaluation. and was negative in all the neous infiltrate for the Patients given the diagnosis of cases tested in their study.13 chromosomal abnormality histiocytoid Sweet syndrome may have identified in the bone The BCR/ABL FISH analysis leukemia cutis. marrow. was used in an effort to Briefly, formalin-fixed, exclude leukemia cutis, but When a diagnosis of histiocytoid Sweet paraffin-embedded sections only from chronic myelogesyndrome is being considered, additional were prepared for FISH analnous leukemia, which was histopathologic investigations may be ysis as described previnot a confirmed diagnosis in needed to evaluate for leukemia cutis, ously.18 Probes used in this the majority of patients in because these mononuclear cells may that study. The observations study were centromere 8 and harbor corresponding cytogenetic by Requena et al13 have also cMYC (8q24.1), DEK (6p23) abnormalities. and CAN (9q34), ABL (9q34) been reported by other and BCR (22q11.2), centroinvestigators.14-17 mere 9 and CTD-244M18 (9q21), 7q (D7S486), and These findings have prompted both clinicians and MLL (11q). pathologists to consider a broader definition of Sweet syndrome in cases where the traditional diagnostic criteria3 are not fully met. Using FISH RESULTS analysis in 5 cases, we demonstrated that the Clinical data cytogenetic abnormalities identified on prior boneClinical characteristics and appropriate laboratory marrow biopsy specimens were also present in skin data for each patient are listed in Table I. Eight biopsy specimens. In this study, we demonstrate the female patients ranged in age from 40 to 76 years importance of evaluating patients with histiocytoid (median, 64 years), and 14 male patients ranged in Sweet syndrome for leukemia cutis. age from 51 to 80 years (median, 72 years). The most common cutaneous presentation was erythematous plaques (Fig 1), with underlying tenderness, fever, METHODS arthralgia, and fatigue. Eighteen patients were After Mayo Clinic Institutional Review Board treated with systemic corticosteroids, usually folapproval, we identified patients with a diagnosis of lowed by resolution of their skin lesion. In 1 patient, Sweet syndrome at Mayo Clinic, Rochester, MN, topical corticosteroids were used; in 3 patients, the between January 1, 1992, and December 31, 2010. treatments were not described. Three patients did Patients who denied research authorization for renot respond to treatment with systemic corticosteview of their medical records were not included in roids and were subsequently treated with colchicine. this study. We identified 21 patients with an underThe most common hematologic malignancies lying hematologic malignancy and concomitant were myelodysplastic syndrome in 10 patients and diagnosis of Sweet syndrome. One patient who acute myelogenous leukemia in 3 patients. was not identified in our initial search was added Myeloproliferative disorder, multiple myeloma, and to the 21 identified patients. All cases selected were chronic lymphocytic leukemia occurred in 2 patients evaluated by 2 reviewers to confirm the diagnostic each. Non-Hodgkin lymphoma, refractory anemia criteria for Sweet syndrome3 were met. For the 22 with excess blasts, and chronic myelogenous patients, the following clinical data were abstracted leukemia each occurred in 1 patient. In 6 patients, from the medical records using the institutional bone-marrow biopsy specimens demonstrated a patient database: age; sex; location, description, d

d

d

1/F/60 2/F/76

SS SS

SS SS

AML CML

6;9 Translocation 9;22 Translocation

3/F/64

SS

SS

None

4/F/69 5/F/55 6/F/66 7/F/40 8/F/48

SS SS SS SS SS

SS SS SS SS SS

Diffuse large-cell, non-Hodgkin lymphoma CLL Multiple myeloma Multiple myeloma AML AML

9/M/51 10/M/79 11/M/67

SS SS SS

SS SS SS

12/M/74 13/M/70

SS SS

14/M/75 15/M/73 16/M/80 17/M/79 18/M/71 19/M/67 20/M/68 21/M/66 22/M/76

Malignancy

Chromosome aberration

Treatment

Response

Prognosis

Symptomatic Systemic corticosteroids and colchicine Systemic corticosteroids

Resolved Resolved

Deceased 12 mo Deceased 71 mo

Resolved

Deceased 7 mo

None Topical corticosteroids Systemic corticosteroids Systemic corticosteroids Systemic corticosteroids

Lost to follow-up Resolved Resolved Resolved Resolved

Deceased 9 mo Alive [10 y after diagnosis Alive [5 y after diagnosis Deceased 9 mo Deceased 1 mo

MDS MDS Probable MDS

None None IgGk None 7q Deletion, MLL amplifications None None None

Systemic corticosteroids Systemic corticosteroids Systemic corticosteroids

Resolved Resolved Lost to follow-up

SS SS

RAEB MPD

Trisomy 8 None

Resolved Resolved

SS SS SS Probable SS

SS SS SS SS

CLL MDS MPD MDS

None None None 9q Deletion

Resolved Resolved Resolved Relapsed

Deceased 5 wk Lost to follow-up Deceased 7 mo Deceased 19 mo

SS SS SS SS SS

SS SS SS SS SS

MDS MDS MDS MDS MDS

None None None None 5q Deletion

Systemic corticosteroids Systemic corticosteroids and colchicine Systemic corticosteroids Systemic corticosteroids Systemic corticosteroids Systemic corticosteroids and colchicine None Systemic corticosteroids Systemic corticosteroids Systemic corticosteroids None

Lost to follow-up Deceased 1.5 mo Lost to follow-up, deceased 48 mo Deceased 32 mo Lost to follow-up

Lost to follow-up Resolved Resolved Relapsed Resolved

Deceased 1.5 mo Lost to follow-up Lost to follow-up Lost to follow-up Lost to follow-up

AML, Acute myelogenous leukemia; CLL, chronic lymphocytic leukemia; CML, chronic myelogenous leukemia; F, female; M, male; MDS, myelodysplastic syndrome; MPD, myeloproliferative disorder; RAEB, refractory anemia with excess blasts; SS, Sweet syndrome.

Chavan et al 1023

Histologic evaluation

J AM ACAD DERMATOL

Clinical description

Patient/sex/age, y

VOLUME 70, NUMBER 6

Table I. Patient characteristics and laboratory data

1024 Chavan et al

J AM ACAD DERMATOL

JUNE 2014

Fig 1. Leukemia cutis in a patient initially given the diagnosis of histiocytoid Sweet syndrome. Erythematous to violaceous tender plaques on the back of right (A) and left (B) hands of patient 12 at the time of presentation to the clinic. Pathologically this was diagnosed as histiocytoid Sweet syndrome, but leukemia cutis was confirmed with further molecular evaluation.

Fig 2. Leukemia cutis resembling histiocytoid Sweet syndrome from patient 1. A, At scanning magnification, dermal edema and dense dermal infiltrate composed of mononuclear cells is noted. B, Cytologic atypia of the cells (arrows). (A and B, Hematoxylin-eosin stain; original magnifications: A, 32; B, 320.)

chromosomal aberration and were therefore selected for histopathologic evaluation and FISH analysis. Histopathology In all 6 patients, no epidermal involvement was noted, and the underlying infiltrate was characterized by mononuclear cells with pale and vesicular nuclei with a mildly eosinophilic cytoplasm (Fig 2). In all specimens, some neutrophils were interspersed with the mononuclear cells, and in most, focal leukocytoclasia was noted. Lymphocytes were usually also present in a perivascular distribution. On the basis of the histopathologic evaluation, diagnosis

in the 6 patients was considered to be consistent with histiocytoid Sweet syndrome. Immunohistochemical studies Pertinent immunohistochemical findings for the 5 patients with a known chromosomal aberration who also underwent FISH testing are reviewed in Table II. All specimens evaluated for CD20 were expectantly negative. Three of the 5 specimens had a sparse CD31 lymphocytic infiltrate in a perivascular distribution. CD68 (KP1) is useful because it is known to be positive in both histiocytes and cells of the myeloid lineage. In all cases the mononuclear cells demonstrated positive staining with CD68 (KP1) in a

J AM ACAD DERMATOL

Chavan et al 1025

VOLUME 70, NUMBER 6

Table II. Histopathologic evaluation of specimens from patients with bone-marrow chromosomal abnormalities Patient

1 2 8 12 17

CD3

CD20

Focal positive Negative Negative Focal positive Focal positive

Negative Negative Negative Negative Negative

CD68 (KP1)

Negative Negative Positive Positive Positive

MPO

Positive Positive Negative Positive Positive

trisomy 8 (Fig 3, D), and the patient subsequently died in 32 months. Patient 17 had myelodysplastic syndrome with a 9q deletion, and the results of cutaneous FISH analysis were equivocal. Therefore, in this patient it was not possible to determine if the infiltrate contained leukemic cells.

DISCUSSION MPO, Myeloperoxidase.

cytoplasmic distribution, and there was also cytoplasmic staining of the polymorphonuclear cells. To further demonstrate a myeloid lineage of the infiltrate, myeloperoxidase was used. Four specimens had diffuse staining with myeloperoxidase, and in a single specimen, the infiltrate was CD681 and myeloperoxidase negative. FISH analysis Five patients with chromosomal abnormalities identified in the bone marrow underwent FISH analysis of their initial cutaneous biopsy specimens (Fig 3). Patient 1 had acute myeloid leukemia with a 6;9 chromosomal translocation that results in DEK/ CAN fusion, and FISH analysis demonstrated DEK/ CAN fusions in a considerable number of cells, indicative of leukemia cutis (Fig 3, A). Presence of DEK/CAN fusion is associated with a poor prognosis, the patient subsequently died from her underlying hematologic malignancy in 12 months, and there was no report of any additional cutaneous lesions after treatment with systemic corticosteroids. In patient 2, chronic myelogenous leukemia was diagnosed with a 9;22 translocation resulting in BCR/ ABL fusion, and BCR/ABL fusion was noted in numerous cells with cutaneous FISH, indicative of leukemia cutis (Fig 3, B). The patient had a relapse of her cutaneous disease and required further treatment with colchicine, after which the lesions resolved. She was also treated with imatinib, blocking phosphorylation of BCR/ABL, and died 71 months later. In patient 8, acute myeloid leukemia was diagnosed with 7q/AML1 deletion and amplification of 11q/MLL. Cutaneous FISH analysis demonstrated 7q deletion (Fig 3, C ) and amplification of 11q in a considerable number of cells. Shortly after the appearance of her cutaneous lesions, the patient elected hospice care and died 1 month later. Patient 12 had refractory anemia with excess blasts with complex clone including trisomy 8 when his cutaneous lesions developed. Cutaneous FISH analysis demonstrated at least 10% of cells with

In this study, we have used FISH to further evaluate cutaneous specimens of 5 patients with an underlying hematologic malignancy and an initial diagnosis of Sweet syndrome. There have been numerous case reports since Requena et al’s13 initial description of histiocytoid Sweet syndrome in 2005, and it is important for dermatologists and dermatopathologists to consider other entities that may mimic histiocytoid Sweet syndrome. At first glance, the infiltrate in most of these biopsy specimens is not typical of what is classically described in Sweet syndrome, which includes a neutrophil-dense dermal infiltrate. However, the clinicians considered Sweet syndrome as the likely diagnosis in most patients. Therefore, it is necessary to question the reason for atypical infiltrate in patients who present with clinical features of Sweet syndrome. Histiocytoid Sweet syndrome is a fairly recently described diagnosis and was not considered in these patients during the initial interpretation of their biopsy findings. However, our re-evaluation of the biopsy specimens led us to consider these cells as histiocytoid or cells of myelomonocytic origin. To further identify the lineage of the cells, immunohistochemical studies were used to differentiate between myeloid cells and histiocytes. In 4 of the 5 patients in our study, we were able to detect the presence of the leukemic cells in the cutaneous biopsy specimens by using the appropriate malignancy-specific FISH probes. The FISH analysis was equivocal in 1 patient, and this result highlights the difficulty in excluding leukemia cutis with cutaneous FISH, unless appropriate probes can be designed. A limitation of our study is that FISH is only useful when there is a known hematologic malignancy with an identifiable chromosomal abnormality. In our study, clinical practice, and review of the literature, patients with classic Sweet syndrome typically have biopsy specimens that demonstrate neutrophils. However, when the pathologic studies demonstrate a mixed infiltrate, then other diagnoses should be also considered. A subset of our patients had leukemia cutis, but they still presented with clinical features suggestive of Sweet syndrome. Therefore, one should consider various causes of

1026 Chavan et al

J AM ACAD DERMATOL

JUNE 2014

Fig 3. Evaluation of initial cutaneous biopsy specimens with fluorescence in situ hybridization to identify chromosomal abnormalities, which confirmed leukemia cutis in 4 patients. A, Arrows indicate fusion of DEK/CAN cells expressing the 6;9 translocation (DEK 6p23, red; CAN 9q34, green) in patient 1, consistent with leukemia cutis. B, Cells express the 9;22 translocation resulting in BCR/ABL fusion (ABL 9q34, red; BCR 22q11.2, green) in patient 2, indicative of leukemia cutis. C, Cells express the 7q deletion (7q32, red; centromere 7, green) in patient 8. Arrows indicate nuclei with deletion of D7S486, indicative of leukemia cutis in this patient. D, Approximately 10% of the cells in patient 12 demonstrated trisomy 8, consistent with leukemia cutis (centromere 8, green; cMYC 8q24.1, red ).

Sweet syndrome, including infections, medications, or malignancies. Three of 5 patients in this study had complete response to systemic corticosteroids, which at the time of initial evaluation was consistent with the rendered diagnosis of Sweet syndrome. In retrospect, 2 of these patients required additional secondline treatment with colchicine, which is typically not required in patients with classic Sweet syndrome. The resolution of the cutaneous lesions with treatment would generally be consistent with Sweet syndrome, despite the presence of an underlying hematologic malignancy. However, as our study demonstrates, it is important to consider leukemia cutis, as this diagnostic pitfall can be resolved with cutaneous FISH analysis. The ability to clinically follow up the patient course was important, because the patients died from progression of their underlying hematologic malignancy within a time frame of 1 month to 71 months. Diagnostic criteria for histiocytoid Sweet syndrome represent a challenge, as histopathologic evaluation demonstrates an ‘‘atypical’’ infiltrate that

can be challenging to differentiate from leukemia cutis. If feasible, additional investigation of the cutaneous infiltrate, including FISH studies that may link it to an underlying hematologic malignancy, should be considered. REFERENCES 1. Sweet RD. An acute febrile neutrophilic dermatosis. Br J Dermatol 1964;76:349-56. 2. Ratzinger G, Burgdorf W, Zelger BG, Zelger B. Acute febrile neutrophilic dermatosis: a histopathologic study of 31 cases with review of literature. Am J Dermatopathol 2007;29: 125-33. 3. Su WP, Liu HN. Diagnostic criteria for Sweet’s syndrome. Cutis 1986;37:167-74. 4. Corazza M, Lauriola MM, Borghi A, Marzola A, Virgili A. Sweet’s syndrome: a retrospective clinical, histopathological and immunohistochemical analysis of 11 cases. Acta Derm Venereol 2008;88:601-6. 5. Rochael MC, Pantaleao L, Vilar EA, Zacaron LH, Spada EQ, Xavier MH, et al. Sweet’s syndrome: study of 73 cases, emphasizing histopathological findings. An Bras Dermatol 2011;86:702-7. 6. Fett DL, Gibson LE, Su WP. Sweet’s syndrome: systemic signs and symptoms and associated disorders. Mayo Clin Proc 1995; 70:234-40.

J AM ACAD DERMATOL

Chavan et al 1027

VOLUME 70, NUMBER 6

7. Anavekar NS, Williams R, Chong AH. Sweet’s syndrome in an Australian hospital: a retrospective analysis. Australas J Dermatol 2007;48:161-4. 8. Cohen PR. Sweet’s syndrome: a comprehensive review of an acute febrile neutrophilic dermatosis. Orphanet J Rare Dis 2007;2:34. 9. Cohen PR, Holder WR, Tucker SB, Kono S, Kurzrock R. Sweet syndrome in patients with solid tumors. Cancer 1993;72:2723-31. 10. Borges da Costa J, Silva R, Soares de Almeida L, Filipe P, Marques Gomes M. Sweet’s syndrome: a retrospective study of 42 admitted patients in a Portuguese hospital. Int J Dermatol 2009;48:953-5. 11. Thompson DF, Montarella KE. Drug-induced Sweet’s syndrome. Ann Pharmacother 2007;41:802-11. 12. Walker DC, Cohen PR. Trimethoprim-sulfamethoxazole-associated acute febrile neutrophilic dermatosis: case report and review of drug-induced Sweet’s syndrome. J Am Acad Dermatol 1996;34:918-23. 13. Requena L, Kutzner H, Palmedo G, Pascual M, FernandezHerrera J, Fraga J, et al. Histiocytoid Sweet syndrome: a dermal

14.

15.

16.

17.

18.

infiltration of immature neutrophilic granulocytes. Arch Dermatol 2005;141:834-42. Apalla Z, Kanatli L, Sotiriou E, Manousari A, Papagarifallou I, Calonje E. Histiocytoid Sweet syndrome. Clin Exp Dermatol 2011;36:562-3. Chow S, Pasternak S, Green P, Tremaine R, Reardon M, Murray S, et al. Histiocytoid neutrophilic dermatoses and panniculitides: variations on a theme. Am J Dermatopathol 2007;29: 334-41. Hunermund A, Wendel AM, Geissinger E, Brocker EB, Stoevesandt J. Typically atypical: histiocytoid Sweet syndrome, associated with malignancy. J Dtsch Dermatol Ges 2011;9: 666-9. Liu CI, Hsiao CH, Wu JT, Tsai TF. Sweet syndrome with histiocytoid infiltrate and neutropenia: a rare combination. J Am Acad Dermatol 2009;61:882-4. Shao H, Xi L, Raffeld M, Feldman AL, Ketterling RP, Knudson R, et al. Clonally related histiocytic/dendritic cell sarcoma and chronic lymphocytic leukemia/small lymphocytic lymphoma: a study of seven cases. Mod Pathol 2011;24:1421-32.

Histiocytoid Sweet syndrome may indicate leukemia cutis: a novel application of fluorescence in situ hybridization.

In patients with malignancy-associated Sweet syndrome, a thorough evaluation for leukemia cutis should be considered...
1MB Sizes 0 Downloads 3 Views