Atkins, Kristin A. M.D.; Dahlem, Monica M. M.D.; Kohler, Sabine M.D.
From the Departments of Pathology (K.A.A., S.K.) and Dermatology (S.K.), Stanford University Medical Center, Stanford; and Dermatology Medical Group (M.M.D.), San Francisco, California
Address correspondence and reprint requests to Dr. Sabine Kohler, Stanford University Medical Center, Department of Pathology, L‐235, 300 Pasteur Drive, Stanford, CA 94305‐5302. E‐mail: firstname.lastname@example.org
Lymphomatoid papulosis (LyP) represents the indolent end of a spectrum of CD30‐positive lymphoproliferative disorders of the skin (1,2). This eruption consists of papules and nodules that occur in crops lasting several weeks to months before regressing spontaneously. The disease is chronic/recurrent and may last for decades. In most patients its course is benign, but in a small subset of patients LyP is associated with a hematolymphoid malignancy, most commonly mycosis fungoides, non‐Hodgkin's lymphoma, or Hodgkin's lymphoma. Histopathologically, LyP is commonly divided into three subtypes (3). Type A is most frequent and consists of a wedge‐shaped infiltrate with CD30+ large atypical RS‐like cells set in a background of mixed inflammation with lymphocytes, eosinophils, and neutrophils. Type B lesions closely resemble mycosis fungoides. The infiltrate often is bandlike, exhibits epidermotropism, and consists mainly of cerebriform lymphocytes, with fewer CD30+ large cells and fewer inflammatory cells. Type C LyP resembles large‐cell anaplastic lymphoma. Large atypical cells fill the entire dermis, and only scant inflammatory cells are admixed.
As a rule, definitive diagnosis of lymphomatoid papulosis requires close clinicopathologic correlation. Rare cases of follicular, pleomorphic, and giant cell variants of type A and C LyP have been reported (4). Although there does not seem to be any prognostic significance to these rare histologic subtypes, the variants can pose diagnostic problems and can lead to misdiagnosis. We report an unusual case of myxoid LyP that was initially mistaken for a mesenchymal neoplasm.
A 14‐year‐old girl presented because of a 3‐month history of slightly tender, violaceous, firm papules and nodules on the extremities, with a waxing and waning course. The nodules ranged from 0.5 to 2 cm in diameter, and many had central crusting and ulceration. General physical examination, hematologic workup, and chest radiography yielded unremarkable findings; no lymphadenopathy or splenomegaly was detected. The clinical differential diagnosis at the time of presentation was pityriasis lichenoides et varioliformis acuta versus lymphomatoid papulosis. A single biopsy specimen was taken from a papule on her right arm. She declined systemic antibiotic treatment, and the papules resolved spontaneously within 6 weeks. She has been free of disease during a follow‐up period of more than 1 year.
Immunohistochemistry was performed as previously described (5) with the following antibodies on formalinfixed paraffin‐embedded tissue: CD3 (Dako, Ilstrup, Denmark; 1:400), CD20 (Dako; 1:500), CD68 (Dako; 1:400), ALK 1 (Dako; 1:80), CD30 (Dako; 1:20), keratin AE1 (Boehringer‐Mannheim, Indianapolis, IN; 1:200), and S100 (Dako; 1:400). Antibody reactivity was detected with streptavidin‐biotin complex and visualized with diaminobenzidine.
Histology and Immunohistochemistry
Hematoxylin and eosin‐stained sections demonstrated a myxoid neoplasm centered on the deep dermis and extending into the subcutaneous fat (Fig. 1A). Numerous discohesive medium and large cells with irregular, folded hyperchromatic nuclei were embedded in a myxoid stroma and accompanied by a prominent neutrophilic and eosinophilic infiltrate and occasional reactiveappearing lymphocytes and histiocytes (Fig. 1B). Morphologically normal mitoses ranged from 2 to 3 mitoses per 10 high‐power fields. Focal epidermotropism by small lymphocytes was present, but most of the biopsy demonstrated a distinct Grenz zone. No apoptotic keratinocytes, basal vacuolar alteration, or other epidermal changes that are common in pityriasis lichenoides were present. The prominent mucinous stroma was highlighted by a colloidal iron stain. The large cells reacted with antibodies directed against CD30 (Fig. 2) and CD3 but not with antibodies against CD15, CD20, ALK‐1, S100, or keratin AE1. Over 50% of the cellular infiltrate and virtually all of the large atypical cells expressed CD30.
The European Organization for Research and Treatment of Cancer (EORTC) classifies LyP as an indolent lymphoma and divides it into three categories: A, B, and C (6). The infiltrate in type A LyP is characteristically wedge‐shaped and composed of a variable percentage of CD30+ large atypical cells and histiocytic cells, with an associated prominent, mixed inflammatory infiltrate. Type B has perivascular to bandlike infiltrates of medium‐sized irregular and hyperchromatic lymphocytes with focal epidermotropism with fewer CD30+ large atypical cells and is morphologically similar to mycosis fungoides. Type C is characterized by a more monotonous population of cytologically atypical CD30+ cells with only scant associated inflammation (6). Differentiation of cutaneous anaplastic large cell lymphoma from type C LyP is problematic because they can be identical histologically and immunophenotypically, and borderline cases exist. In addition, approximately 50% to 60% of lymphomatoid papulosis cases tested demonstrate clonal rearrangements of T‐cell receptor (TCR) genes (7).
The t(2;5) translocation is present in only a subset of systemic CD30+ malignant lymphomas but not in LyP, primary cutaneous anaplastic large cell lymphoma, or the majority of cases of systemic CD30+ lymphoma and usually does not help in the differential diagnosis (8,9). Histologic features that support LyP are a prominent inflammatory component and an infiltrate limited to the dermis, with only focal involvement of the subcutaneous fat. In addition the atypical cells are discohesive, forming small clusters rather than sheets. However, close clinical correlation is paramount for definitive diagnosis. LyP typically presents with numerous papules and nodules measuring less than 2 cm each, which spontaneously regress. In contrast, CD30+ lymphoma is characterized by one or a few tumors measuring more than 2 cm in diameter. However, strict clinical requirements for size, distribution, and duration of lesions are not defined, and borderline cases exist in which the histology and clinical presentation do not correlate (e.g., regressing anaplastic large cell lymphoma) or show features intermediate between LyP and CD30+ lymphoma (2).
The clinical presentation of our patient, who had numerous crusted papules that spontaneously regressed, is typical for LyP. The histopathologic features, however, are unusual: diffuse and deep subcutaneous involvement, a high number of CD30+ atypical cells, and a prominent myxoid stroma. The high number of CD30+ cells is reminiscent of type C LyP, yet the prominent inflammatory component is more in keeping with type A LyP. The most confounding features in this case, however, are the myxoid stroma and deep location, features that first caused misdiagnosis as a malignant mesenchymal neoplasm by the referring pathologist.
Several inflammatory dermatoses are commonly associated with dermal mucin, such as cutaneous lupus erythematosus, dermatomyositis, pretibial myxedema, and reticular erythematous mucinosis. Conversely, some inflammatory conditions such as pityriasis lichenoides et varioliformis acuta are almost never associated with mucin. The literature also describes rare cases of neoplastic cutaneous lymphoid infiltrates with myxoid stroma, including primary cutaneous anaplastic large cell lymphoma and three cases of extranodal B cell lymphoma (10–13). All cases mimicked myxoid mesenchymal neoplasms such as myxoid chondrosarcoma and myxoid malignant fibrous histiocytoma. We have also observed cases of subcutaneous panniculitic T‐cell lymphoma with prominent dermal mucin (unpublished observation). Mucin associated with lymphomatoid papulosis is an uncommon finding. Mucin in the dermis was noted in three (7.5%) of 40 cases of lymphomatoid papulosis reviewed by Weinman and Ackerman (14). The quantity or distribution of the mucin was not described in this study.
The pathomechanism responsible for dermal mucin deposition is unclear. Tse et al. speculated that the myxoid change in cutaneous lymphomas was secondary to stimulation of stromal cells and lymphatic obstruction (11). A report of mucin in a cutaneous lesion of malignant lymphoma but not in an involved lymph node in the same patient suggests local stimulation of mesenchymal cells rather than an inherent feature of the lymphoma (10).
This case was referred to us for consultation with a diagnosis of “mesenchymal neoplasm.” Pleomorphic variants of lymphoma and LyP can be confused with mesenchymal tumors and carcinomas. In our patient, malignant melanoma, carcinoma, and malignant fibrous histiocytoma were excluded by the absence of reactivity with S100, cytokeratins, and anti‐CD68. The strong expression of anti‐CD30 and anti‐CD3 and the absence of anti‐CD20 and anti‐CD15 in conjunction with the clinical course supported a diagnosis of lymphomatoid papulosis and argued against a diagnosis of Hodgkin's lymphoma and pityriasis lichenoides et varioliformis acuta.
LyP in the pediatric population is extremely rare: fewer than 30 cases are reported in the English‐language literature (15). However, on the basis of these small series, LyP in children and adolescents seems to follow a course similar to that seen in adults—chronic, with lessening symptoms and eventual remission—although in 5% to 10% of patients a malignant hematolymphoid neoplasm develops (15).
We have described a case of LyP with a classic clinical presentation but unusual histology. The unusual myxoid histology and deep dermal involvement initially led to diagnosis as a malignant mesenchymal neoplasm. A diagnosis of LyP was supported immunophenotypically with the finding of CD30 T cells and clinically by the recurrent and regressing nature of the lesions. Although the presence of myxoid stroma did not seem to affect the clinical presentation in this case, it is important to recognize myxoid variants of lymphoid processes to ensure proper diagnosis and management.
1. LeBoit PE. Lymphomatoid papulosis and cutaneous CD30+ lymphoma. Am J Dermatopathol 1996;18:221-35.
2. Demierre MF, Goldberg LJ, Kadin ME, et al. Is it lymphoma or lymphomatoid papulosis? J Am Acad Dermatol 1997;36:765-72.
3. Willemze R, Beljaards RC. Spectrum of primary cutaneous CD30 (Ki-1)-positive lymphoproliferative disorders. J Am Acad Dermatol 1993;28:973-80.
4. Kato N, Matsue K. Follicular lymphomatoid papulosis. Am J Dermatopathol 1997;19:189-96.
5. Natkunam Y, Smoller BR, Zehnder JL, et al. Aggressive cutaneous NK and NK-like T-cell lymphomas: clinicopathologic, immunohistochemical, and molecular analyses of 12 cases. Am J Surg Pathol 1999;23:571-81.
6. Willemze R, Kerl H, Sterry W, et al. EORTC classification for primary cutaneous lymphomas: a proposal from the Cutaneous Lymphoma Study Group of the European Organization for Research and Treatment of Cancer. Blood 1997;90:354-71.
7. Weiss LM, Wood GS, Trela M, et al. Clonal T-cell populations in lymphomatoid papulosis: evidence of a lymphoproliferative origin for a clinically benign disease. N Engl J Med 1986;315:475-9.
8. Herbst H, Sander C, Tronnier M, et al. Absence of anaplastic lymphoma kinase (ALK) and Epstein-Barr virus gene products in primary cutaneous anaplastic large cell lymphoma and lymphomatoid papulosis. Br J Dermatol 1997;137:680-6.
9. DeCoteau J, Butmarc J, Kinney M, et al. The t(2:5) chromosomal translocation is not a common feature of primary cutaneous CD30+ lymphoproliferative disorders: comparison with anaplastic largecell lymphoma of nodal origin. Blood 1996;15:3437-41.
10. Chan J, Buchanan R, Fletcher C. Sarcomatoid variant of anaplastic large-cell Ki-1 lymphoma. Am J Surg Pathol 1990;14:983-8.
11. Tse CC, Chan JK, Yuen RW, et al. Malignant lymphoma with myxoid stroma: a new pattern in need of recognition. Histopathology 1991;18:31-5.
12. Fung DT, Chan JK, Tse CC, et al. Myxoid change in malignant lymphoma. Pathogenetic considerations. Arch Pathol Lab Med 1992;116:103-5.
13. Nonomura A, Mizukami Y, Hirone T, et al. Malignant lymphoma with myxoid change and sarcomatous features. Intern Med 1993; 32:139-44.
14. Weinman V, Ackerman A. Lymphomatoid papulosis: a critical review and new findings. Am J Dermatopathol 1981;2:129-163.
15. Zirbel GM, Gellis SE, Kadin ME, et al. Lymphomatoid papulosis in children. J Am Acad Dermatol 1995;33:741-8.
Key Words:: CD30‐positive lymphoproliferative disorder; Cutaneous anaplastic large cell lymphoma; Cutaneous lymphoma; Lymphomatoid papulosis.
© 2003 Lippincott Williams & Wilkins, Inc.