Skip to main content
Open Access Publications from the University of California

Dermatology Online Journal

Dermatology Online Journal bannerUC Davis

Utility of myeloperoxidase stain in the diferential diagnosis of leukemia cutis vs. hystiocitoid Sweet syndrome

Main Content

Utility of myeloperoxidase stain in the diferential diagnosis of leukemia cutis vs. hystiocitoid Sweet syndrome
Pedro Valerón-Almazán, Jesús Bastida, Jaime Vilar, Néstor Santana, Carolina Medina, Gregorio Carretero
Dermatology Online Journal 17 (4): 11

Service of Dermatology, Hospital Universitario de Gran Canaria Dr. Negrín. Las Palmas de Gran Canaria, Spain


Leukemia cutis is defined as a skin infiltration by leukemic cells. The diagnosis of myeloid leukemia cutis (MLC) can represent a challenge, especially in those cases without symptoms of systemic disease. The clinical appearance, histopathological analysis and inmunohistochemical profile can be indistinguishable from those observed in cases of hystiocitoid Sweet syndrome (HSS). We present a case of MLC in which the cutaneous affectation was the first sign of the systemic leukemia. In this setting, the myeloperoxidase stain was the clue to rule out the possibility of HSS. We discuss the role and the utility of the myeloperoxidase stain in the differentiation of these two entities.


Leukemia Cutis is defined as a skin infiltration by leukemic cells. Most cases occur after a diagnosis of systemic leukemia (55%), but concurrent (38%) or previous involvement (7%) can also be seen [1]. It is most often associated with acute myeloblastic leukemia (AML), varying from 2 percent to 20 percent [2, 3, 4], especially in monocytic and myelomonocytic subtypes [3, 5].

Case report

Figure 1
Figure 1. Multiple, erythematous, round and arciform plaques, confluent on the back.

A 66-year-old man presented with a history of intermittent fever and skin rash, which was present for 12 weeks. Personal antecedents included diabetes, hypertension, and thalassemia minor. He was a chronic smoker. Upon examination, the rash appeared as erythematous, infiltrated plaques located principally on the back (Figure 1). Similar circumscribed papules were also seen on the upper thighs. The first diagnostic impression was of Sweet syndrome.

Figure 2Figure 3
Figure 2. Superficial and deep dermal infiltration. Periadnexal arrangement is observed, with no epidermotropism (HE, x40)

Figure 3. Multiple cells with granular and eosinophilic cytoplasm. Irregular round or oval nuclei are evident, sometimes with prominent nucleoli (HE, x400).

Figure 4
Figure 4. Negativity myeloperoxidase staining in the infltration

His blood workup showed low levels of hemoglobin (9.2 mg/dL), but no other alterations were detected. The biopsy revealed a dense infiltration of cells involving the dermis with perivascular and periadnexal arrangements. Epidermotropism was absent and the upper papillary dermis was spared. The cells were monomorphic, medium-sized, with a relatively eosinophilic granular cytoplasm and large nuclei with occasional prominent nucleoli (Figures 2 and 3). Immunohistochemistry was performed and these cells expressed lysozyme, CD68, CD45 and focally CD15. They were negative for myeloperoxidase (MPO) (Figure 4), CD3, CD4, CD20, CD34, CD56 and TDT (terminal deoxynucleotidyl transferase).

The patient was diagnosed with leukemia cutis, and futher tests were performed. The cytologic study of the peripheral blood showed 12 percent blasts cells. The bone marrow biopsy revealed a strong infiltration by blast cells (60% of total), compatible with AML. The kariotype was 46XY. Considering these findings, the clinical picture was finally diagnosed as leukemia cutis secondary to AML M1, using the French-American-British (FAB) classification. The patient was treated with the 3+7 idarubicin and cytarabine schedule, but this induction was not effective. In this context, intravenous high doses of cytarabine were administered. After 8 months of follow up, no blast cells were found in the peripheral blood and the bone marrow biopsy showed a normocellular composition, with 10 percent blast cells. The cutaneous lesions resolved to residual hyperpigmented macules.


The diagnosis of myeloid leukemia cutis (MLC) can represent a challenge, especially in those cases without symptoms of systemic disease, as occurred in our patient. Histiocytoid Sweet syndrome (HSS), originally described by Requena et al [6], can resemble all the features found in MLC, leading to diagnostic pitfalls [7].

First of all, the clinical appearance based on erythematous, infiltrated plaques, typically seen in Sweet syndrome, is found in 26 percent of patients with leukemia cutis of all types [1]. Moreover, the histopathological analysis in MLC and HSS can present overlapping features. As is suggested in the first description of HSS, neither the cytologic characteristics of the cells nor the inmunohistochemical profile can exclude the possibility of MLC [6]. The inmunohistochemical similarities between MLC and HSS are shown in Table 1.

MPO staining has been described as very sensitive in cases of HSS [6, 7], but it is not a constant feature in MLC (43-58%) [8, 9]. In our case, the absence of MPO staining in the infiltrate was the clue to suspect MLC instead of HSS, so additional blood and bone marrow examinations were performed to confirm this diagnosis. However, MPO is often positive in MLC, so this marker is not always useful to differentiate between these two entities.

Even before the description of HSS, other previous reports have investigated the similar findings in MLC and Sweet syndrome in the setting of malignant hemopathies. These parallelisms include the similar clinical presentation, the predomination in AML 4 and 5, the common triggers (granulocye colony-stimulating factor, all-trans-retinoic acid) [10] as well as the presence of atypical cells in the infiltrate [11, 12], among others. Interestingly, the concomitant occurrence of Sweet syndrome and MLC has also been reported [13, 14], but we cannot rule out that these cases constitute examples of HSS before the original description.

Because of the similarities among MLC and HSS, additional blood cytologic studies have been employed for the differential diagnosis, even in patients with unknown antecedents of myeloproliferative disease. This observation is supported by Requena et al in their study, in which almost all the patients diagnosed with HSS were first evaluated for peripheral blood analysis (no myeloid cells detected in 27 patients) [6].

In conclusion, we present a case of leukemia cutis secondary to AML M1, presenting as the first sign of the disease. In this setting, the differential diagnosis with HSS could be problematic. The skin infiltration was clinical and histopathologically compatible with HSS, but the negativity of the MPO stain was definitive to confirm the diagnosis. In cases of MLC in which MPO is positive, peripheral blood cytologic studies have been suggested as a useful tool to rule out HSS. Although blast cells are not identified and the diagnosis of HSS is confirmed, close follow up of these cases should always be considered.


1. Su WP, Buechner SA, Li CY. Clinicopathologic correlations in leucemia cutis. J Am Acad Dermatol. 1984;11(1):121-8. [PubMed]

2. Sepp N, Radaskiewicz T, Meiger CJ, Smolle J, Seewan H, Fritsch P, Kerl H. Specific skin manifestations in acute leukemia with monocytic differentiation. A morphologic and inmunohistochemical study of 11 cases. Cancer. 1993;71(1):124-32. [PubMed]

3. Baer MR, Barcos M, Farrell H, Raza A, Preisler HD. Acute myelogenous leucemia with leucemia cutis. Eighteen cases between 1969 and 1986. Cancer. 1989;63(11):2192-200. [PubMed]

4. Desch JK, Smoller BR. The spectrum of cutaneous disease in leukemias. J Cutan Pathol. 1993;20(5):407-10. [PubMed]

5. Ratnam KV, Khor CJ, Su WP. Leukemia Cutis. Dermatol Clin. 1994;12(2):419-431. [PubMed]

6. Requena L, Kutzner H, Palmedo G, Pascual M, Fernández-Herrera J, Fraga J, García-Díez A, Sanchez-Yus E. Histiocytoid Sweet Syndrome: a dermal infiltration of inmature neutrophilic granulocytes. Arch Dermatol. 2005;141(7):834-42. [PubMed]

7. Chow S, Pasternak S, Green P, Tremaine R, Reardon M, Murray S, Northgrave S, Walsh N. Histiocytoid neutrophilic dermatoses and panniculitides: variations on a theme. Am J Dermatopathol. 2007;29(4):334-41. [PubMed]

8. Cronin DM, George TI, Sundram UN. An updated approach to the diagnosis of myeloid Leukemia cutis. Am J Clin Pathol. 2009;132(1):101-10. [PubMed]

9. Cibull TL, Thomas AB, O'Malley DP, Billings SD. Myeloid leukemia cutis: a histologic and immunohistochemical review. J Cutan Pathol. 2008;35(2):180-5. [PubMed]

10. Vignon-Pennamen MD, Aractingi S. Sweet's syndrome and leukemia cutis: a common skin homing mechanism?. Dermatology. 2003;206(2):81-4. [PubMed]

11. Deguchi M, Tsunoda T, Yuda F, Tagami H. Sweet's syndrome in acute myelogenous leukemia showing dermal infiltration on leukemic cells. Dermatology. 1997;194(2):182-4. [PubMed]

12. Urano Y, Miyaoka Y, Kosaka M, Kabe K, Uchida N, Arase S. Sweet's syndrome associated with chronic myelogenous leukemia: demonstration of leukemic cells within a skin lesion. J Am Acad Dermatol. 1999;40(2 Pt 1):275-9. [PubMed]

13. Dominguez ER, Greene JN, Sandin RL, DeGregorio R, Glass LF. Sweet syndrome and leukemia cutis in the same patient: a case report and review. Cancer Control. 1995;2(4):343-6. [PubMed]

14. del Pozo J, Martínez W, Pazos JM, Yebra-Pimentel MT, García-Silva J, Fonseca E. Concurrent Sweet's syndrome and leukemia cutis in patients with myeloid disorders. Int J Dermatol. 2005;44(8):677-80. [PubMed]

© 2011 Dermatology Online Journal