Lymphoproliferative Disorders

Lymphoproliferative Disorders

Lymphomatoid Papulosis in Children

Introduction.—Lymphomatoid papulosis (LyP) is a continuing, self-healing eruption characterized by a benign clinical course and malignant histology. The initial lesions are small papules that progress to papulonecrotic lesions, then regress. About 10% to 20% of patients progress to systemic lymphomas. Most LyP lesions are observed in adults, but it has also been reported in children. Described is a child with disease onset at age 2. A literature review is included.

Case Report.—Boy, 12 years, had a 10-year history of asymptomatic papules on his face, trunk, and 4 extremities. The lesions typically became necrotic, then healed spontaneously in 2—4 weeks with slight scarring and hypopigmentation. He experienced good health otherwise and had an unremarkable family history. He had multiple, erythematous papules that were 5-10 mm in diameter mainly on his trunk and extremities. He had no lymphadenopathy or hepatosplenomegaly. A biopsy specimen from an active lesion on his back showed a dense perivascular infiltrate of large atypical cells, small lymphoid cells, and eosinophils. The atypical cells had a large nucleus. Immunohistochemical analysis revealed large atypical cells positive for CD3, CD4, CD7, CD25, CD30, CD45RO,CD71, and HLA-DR.

Review of Literature.—A review of literature between 1956 and 1995 revealed 44 reported patients with LyP with disease onset before age 20. Onset in childhood occurred most frequently before the age of 5 years. Two patients with lesions appearing in childhood went on to develop systemic malignant melanomas in adulthood. One patient developed Hodgkin’s disease at age 57, and the other patient developed undifferentiated lymphoma at about age 50. Most patients in this series were treated with topical applications of corticosteroids alone and in combination. Most treatments were considered ineffective.

Lymphoproliferative DisordersConclusion.—Lymphomatoid papulosis usually affects adults, but 44 patients with onset before age 20 have been reported. Of patients with childhood onset, 42.9% had disease manifestations before age 5 years. The T-cell receptor rearrangement in lesional skin is not indicative of malignant change, because clonal expansion of T-cells was also found in patients with a self-healing clinical course.

► The authors’ review of the literature is most helpful. Of 44 reported cases of lymphomatoid papulosis with age of onset younger than age 20, only 2 developed malignancy, one at the age of 50 and the other at the age of 57. The overall incidence of malignancy in lymphomatoid papulosis is estimated at 10% to 20%. Demonstration of T-cell receptor gene rearrangement in lesional skin is not irrefutable evidence of malignant change. Clonal expansion of T-cells may be found in patients with benign, self-healing lymphomatoid papulosis as well as in the variant which is associated with malignancy.

Lymphomatoid Papulosis: Successful Weekly Pulse Superpotent Topical Corticosteroid Therapy in Three Pediatric Patients

Introduction.—Lymphomatoid papulosis is a T-cell proliferation with histopathologic features suggestive of a malignant process, despite its benign behavior. This cutaneous disorder is characterized by the appearance of crops of erythematous papules or nodules that can develop ulcers or crusts. About 10% of reported patients develop a malignant lymphoma subsequent to the diagnosis of lymphomatoid papulosis. Most patients are adults, but lymphomatoid papulosis has been reported in children. De-scribed are 3 children with lymphomatoid papulosis without systemic involvement who had excellent control of lesions with superpotent topical corticosteroids.

Methods.—Three children free of systemic disease were evaluated at baseline and annually by a pediatric oncologist. Cutaneous lesions were treated twice daily for 2-3 weeks with a superpotent topical corticosteroid, such as halobetasol propionate 0.05% ointment or cream or clobetasol propionate 0.05% cream, followed by weekly pulsed therapy (application every 12 hours for a total of 3 doses each week). Median follow-up was 36 months.

Results.—All 3 patients experienced resolution of lesions at a range of 2-6 months. During follow-up 1 patient had no new lesions, 1 patient developed 1 new lesion every 6 months, and 1 patient had occasional new lesions. No patients developed evidence of systemic disease.

Discussion/Conclusion.—The lesions of lymphomatoid papulosis may appear for weeks to decades but tend to regress spontaneously. Treatment is usually initiated because scarring is of concern. There are several treat-ments available for lymphomatoid papulosis, but their invasive nature and serious side effects limit their use in children. The use of superpotent topical corticosteroids was helpful in this patient series. The cytotoxic actions of this therapy seemed to outweigh any immunosuppressive effect.

► This article summarizes the current thinking about lymphomatoid papulosis and underscores the importance of using the least toxic therapies available in any disease treatment, especially in children. Lymphomatoid papulosis is thought to represent a delicate balance between host defenses and T-cell malignancy. The authors suggest that while the use of corticosteroids might appear contraindicated in this situation, in these 3 pediatric patients the therapeutic benefit of the superpotent topical corticosteroid outweighed any potential immunosuppressive effect.

Aggressive Angiodestructive Variant of Primary Cutaneous CD30+ Lym-phoma: How to Classify?

Introduction.—The European Organization for Research and Treatment of Cancer presented a new classification of cutaneous lymphomas in October 1994. The primary cutaneous CD30+ lymphomas were classified as belonging to the low-grade lymphomas, and recent studies confirm a generally favorable prognosis for these lesions. However, the 2 cases presented here (primary cutaneous large-cell anaplastic CD30+ lymphoma, 1 of the B cell type and 1 of the T cell type), exhibited fulminant clinical courses.

Case Report 1.—Woman, 65, was seen with blue-reddish nodules in the left axilla that extended to the buttocks and upper legs. Other findings were progressive lassitude, edema of the right lower leg, and intermittent fever. Lesions showed typical histological features of a leucocytoclastic, nodular vasculitis. Chest x-ray examination revealed a small lung tumor, which was removed and identified as B cell lymphoma, immunoblastic subtype. Four months after thoracotomy followed by chemotherapy, new lesions developed that were similar to those at initial examination. Microscopic findings included angiocentricity and angiodestruction, with anaplastic large cells expressing CD30(Kil) antigen on their cell membrane. One year after chemotherapy was administered for the lymphoma, the patient had a cutaneous and pulmonary relapse. Remission was achieved with etoposide.

Case Report 2.—The second patient, a 57-year-old man, was seen with edema and vesicles and erosions on the right forearm. A new lesion examined microscopically demonstrated a perivascular infiltrate consisting of large polymorphous, anaplastic cells with pathologic mitotic figures. The anaplastic cells in this case were of T cell origin, and were CD30+ with angiodestructive and angiocentric behavior. The patient died of fluconazole-resistant Candida sepsis before specific antitumoral chemotherapy could be administered.

Discussion.—A precise distinction must be made between large-cell anaplastic CD30+ lymphoma, with manifestations primarily in the lymph nodes, which has a markedly poor prognosis, and the rather rare form of the primarily cutaneous CD30+ lymphoma, which has a more favorable prognosis.

► Recently lymphoma classification has been redefined in various consensus conferences. As a result primary cutaneous CD30+ lymphomas have at times been grouped as low-grade lymphomas (European Organization of Research and Treatment of Cancer) or as high-grade lymphomas (revised Kiel classification). Only the Revised European-American Lymphoma Study Group classification system recognizes angiodestructive lymphomas as a separate entity. Jappe et al. present 2 cases of primary cutaneous anaplastic CD30+ lymphomas, both of which were characterized by an angiodestructive growth pattern with a fulminant clinical course. The authors maintain that angiocentric and/or angiodestructive manifestations of primary cutaneous lymphomas deserve special consideration within the classification scheme of lymphomas.

Persistent Pigmented Purpuric Dermatitis and Mycosis Fungoides: Simulant, Precursor, or Both? A Study by Light Microscopy and Molecular Methods

Background.—Mycosis fungoides is one of the most common non- Hodgkin’s lymphomas and has a variety of clinical and histologic features. This variation may result from the admixture of non-neoplastic inflam-matory cells in many lesions of mycosis fungoides and the wide range of reaction patterns of the skin. In 1 variant of mycosis fungoides, purpuric areas develop within lesional skin.

The relationship between mycosis fungoides and persistent pigmented purpuric dermatitis was examined for several reasons: Purpuric lesions can occur in mycosis fungoides, the first patient in the United States reported as having lichen aureus turned out to have mycosis fungoides; the diagnosis of several patients referred to the authors’ cutaneous lymphoma clinic was changed from mycosis fungoides to persistent pigmented purpuric dermatitis after a review of biopsy specimens; and the authors had experience with 2 patients who had both conditions. In 1 of these patients, persistent pigmented purpuric dermatitis preceded mycosis fungoides.

Methods.—Specimens from 56 patients with persistent pigmented pur-puric dermatitis were examined to determine the incidence of mycosis fungoides-like histologic patterns, such as psoriasiform lichenoid, psori-asiform spongiotic lichenoid, and atrophic lichenoid patterns. The degree of spongiosis, epidermotropism, papillary dermal fibrosis, lymphocytic atypia, epidermal hyperplasia, number of extravasated erythrocytes and siderophages, and the distribution of lymphocytic infiltrate within the epidermis were also analyzed.

Results.—Among the 56 patients, patterns typically seen in mycosis fungoides were seen in 29 patients. In persistent pigmented purpuric dermatitis, lymphocytes aligned along the epidermal side of the dermoepidermal junction were seen, with few necrotic keratinocytes, as seen in mycosis fungoides. Papillary dermal edema was common in persistent pigmented purpuric dermatitis but not in mycosis fungoides. In mycosis fungoides but not in persistent pigmented purpuric dermatitis, lympho-cytes had significantly atypical nuclei and had moved into the upper spinous layer.

Because of these similarities, clonality of the T-cell population was tested with a polymerase chain reaction assay for 7-chain rearrangements. Clonal populations were seen in 3 of 3 specimens and 1 of 2 specimens from 2 patients with both persistent pigmented purpuric dermatitis and mycosis fungoides. Clonal populations were also seen in 8 of 12 specimens typical of lichenoid patterns of persistent pigmented purpuric dermatitis.

Discussion.—These findings indicate that the lichenoid variants of persistent pigmented purpuric dermatitis may be biologically related to mycosis fungoides. The high rate of clonality in persistent pigmented purpuric dermatitis invalidates the usefulness of gene rearrangement studies—especially with current methods of 7-chain probes and polymerase chain reaction assays—to distinguish mycosis fungoides from persistent pigmented purpuric dermatitis because either condition can demonstrate a clonal T-cell population. Studies of fresh tissue specimens for (3-chain gene rear-rangements using Southern blotting are recommended.

Progressive pigmentary dermatosis (Schamberg’s disease), purpura annularis telangiectoides (Majocchi’s disease), pigmented purpuric lichenoid dermatitis (Gougerot-Blum syndrome), and lichen- aureus are commonly grouped together under the rubric “pigmentary purpuric eruptions.” However, dermatologists are well aware that these 4 conditions have distinctive clinical morphology that enables them to be distinguished from one other.

Furthermore, because the etiology and pathogenesis of these conditions are unknown, it seems premature to conclude that they are different manifestations of the same basic process.

The article by Toro et al. found that the lesions in 29 of their 56 patients with persistent pigmented purpuric eruptions had histologic patterns that can also be seen in mycosis fungoides. The polymerase chain reaction assay for -y-chain rearrangements was performed on 12 specimens with lichenoid patterns of persistent pigmented purpuric eruptions and showed that 8 of 12 specimens had clonal populations of lymphocytes.

Only long-term follow-up of these patients will answer the question as to whether this finding is biologically meaningful. Meanwhile, stigmatizing patients with Schamberg’s disease or purpura annularis telangiectoides is probably not indicated as these conditions are not only clinically, but also histologically, different from the lichenoid variants of persistent pigmented purpuric dermatitis.

Folliculotropic Mycosis Fungoides With Large-Cell Transformation Pre-senting as Dissecting Cellulitis of the Scalp

Background.—Follicular mycosis fungoides is a rare variant of cutaneous T-cell lymphoma. It is characterized by malignant lymphocytes that preferentially infiltrate hair follicles with minimal epidermal involvement. Patients can have follicular papules, erythematous scaling boggy plaques with plugged follicles, and comedo-like lesions. A rare case of folliculo- tropic mycosis fungoides that presented in a manner similar to dissecting cellulitis of the scalp is described.

Case Report.—Man, 41, had experienced “eczema” and chronic draining nodular lesions on his neck, and scalp and in the beard area for several years. The patient also had erythematous and hyperpigmented patches and plaques on the chest and back, with diffuse follicular prominence. There was also significant cervical, axillary, inguinal, and epitrochlear adenopathy. There was no hi- dradenitis suppurativa. This African-American man had received topical corticosteroids and oral antibiotics for treatment of presumed, dissecting cellulitis of the scalp. Although the lesions partially responded, there was gradual progression, with recurrent flares, and new cutaneous areas of involvement. Large-cell transformation of cutaneous T-cell lymphoma was seen in scalp and axillary lymph node specimens. Although the patient received electron beam radiation, topical mechlorethamine, interferon-a, and systemic chemotherapy, he died of progressive cutaneous T-cell lymphoma 9 months after diagnosis.

Discussion.—Large-cell transformation of folliculotropic mycosis fungoides is an aggressive form of cutaneous T-cell lymphoma. Folliculotropic mycosis fungoides can result in lesions that resemble dissecting cellulitis of the scalp. It is important to be aware of folliculotropic mycosis fungoides so that cutaneous T-cell lymphoma can be considered in the differential diagnosis of dissecting cellulitis of the scalp and perifolliculitis capitis abscedens et suffodiens.

► Folliculotropic mycosis fungoides is an interesting new variant of cutaneous T-cell lymphoma (CTCL) that has diverse manifestations. Previous cases have presented as follicular papules, erythematous scaling boggy plaques with plugged follicles, and comedo-like lesions. Gilliam et al. describe a patient with clinical features of dissecting cellulitis of the scalp. We have observed a case of folliculotropic CTCL that presented with clinical features simulating alopecia areata. Therefore, based on the small number of reported cases and the diverse clinical presentations, there are no reliable clues to the prospective diagnosis. The diagnosis of folliculotropic CTCL must be made on a histopathologic basis, which means that the dermatologist must have a heightened awareness of this condition, as well as other unusual variants of CTCL such as alopecia mucinosa, bullous CTCL, CTCL resembling chronic hand dermatitis, and CTCL simulating persistent pigmented purpuric dermatitis.

Histologic Criteria for the Diagnosis of Erythrodermic Mycosis Fungoides and Sezary Syndrome: A Critical Reappraisal

Introduction.—The diagnosis of early stage mycosis fungoides (MF) and Sezary syndrome (SS) is clinically and histologically challenging. The histologic parameters for determining a diagnosis of MF with well-developed patch and plaque stage lesions are clearly defined, but the same criteria seem to be less relevant for determining MF in patients with erythroderma secondary to the disease. The histologic features of erythrodermic MF and SS were evaluated and compared in 28 biopsy specimens from 17 patients.

Methods.—Two dermatopathologists independently reviewed each slide and scored them semiquantitatively using 24 parameters. These data were compared with data previously reported for 64 patients with limited patch and plaque stage lesions of MF. The comparison revealed that biopsies of limited patch and plaque lesions had significantly more parakeratosis and acanthosis, less disproportionate epidermotropism, fewer lymphocytes aligned within the basal layer, fewer hyperconvoluted cells in the epidermis, more dermal hyperconvoluted cells, more papillary dermal fibrosis, more prominent telangiectasis, and more mitotic figures than biopsies from patients with erythrodermic MF or SS.

Conclusion.—The histologic features of erythrodermic MF and SS were more subtle than those of patch and plaque stage MF and were difficult to diagnose histologically. The histologic features of erythrodermic MF and SS were similar and could not be distinguished reliably. A dermal infiltrate with over 5 hyperconvoluted dermal lymphoid cells per high power field in the presence of epidermotropism without accompanying spongiosis may be suggestive of erythrodermic MF or SS. This finding should prompt the initiation of immunophenotyping and genotyping to help define the diagnosis.

► Erythrodermic mycosis fungoides and Sezary syndrome cannot be distinguished from each other histologically. The authors have defined criteria that may help to distinguish erythrodermic mycosis fungoides and Sezary syndrome from patch and plaque stage disease. A more pressing problem is to distinguish them from erythroderma resulting from other causes. Kohler et al. found that epidermotropism of atypical lymphocytes and formation of Pautrier microabscesses are not consistent features of the generalized forms of cutaneous T-cell lymphoma, but they indicate that more than 5 hyperconvoluted lymphocytes per high power field (x300 total magnification) should raise the suspicion of erythrodermic mycosis fungoides or Sezary syndrome in erythrodermic patients. It awaits others to evaluate the sensitivity, specificity, and predictive value of this criterion.

Clinicopathological Aspects of HTLV-1 Positive and Negative Cutaneous T-cell Lymphoma

Introduction.—Human T-cell lymphotropic virus type 1 (HTLV-1) has been identified in patients with cutaneous T-cell lymphoma (CTCL) and adult T-cell leukemia/lymphoma (ATL). About 40% of patients with ATL have cutaneous lesions that may resemble other types of CTCL. Clinical and histopathologic differences have been observed between CTCL and ATL. The clinical, histopathologic, and immunohistochemical characteristics of 17 patients with HTLV-1+ and 17 patients with HTLV-1 CTCL from Bahia, Brazil, were compared to determine whether there were any significant between-group differences.

Methods.—Patients underwent serologic testing, histopathologic evaluation, and immunohistochemical assessment. Clinical data were obtained from medical records. Survival data were collected.

Results.—Cutaneous lesions were the initial clinical manifestation of disease in all 34 patients. In both the HTLV-1and HTLV-1 groups, 12 and 5 patients respectively, had pleomorphic lymphoma and mycosis fungoides (MF). The clinical subtypes in the HTLV-1+ group included 5, 2, and 9, and 1 cases respectively, of acute, chronic, smoldering, and lymphoma type ATL. Patients who were HTLV-1+ were more likely to have a more aggressive course; to have a greater frequency of erythroderma and disseminated skin lesions; and to be black, mulatto, and male. Nevertheless, nearly 40% of patients in the HTLV-1+ group had an indolent course of disease with an evolution of up to 20 years. There were no between- group differences in intensity of cellular pleomorphism, cellular size, epidermotropism, frequency and size of Pautrier’s abscesses, fibrosis of the papillary derma, and eosinophilic infiltrate.

Conclusion.—Clinical and histological features of HTLV-1+ CTCL are variable and have no specific characteristics that distinguish them from other types of CTCL. It may be appropriate to designate CTCL as HTLV-1 or HTLV-1 with a morphologic classification as pleomorphic, MF, large cell anaplastic lymphoma, or other histologic types.

► These investigators found no histopathologic or immunohistochemical features that distinguish HTLV-1+ CTCL from HTLV-1 patients. Eleven of the 17 HTLV-1+ patients had erythrodermic disease vs. only 3 of 17 in the HTLV-1negative group. This tendency for erythroderma has not been previously noted in HTLV-1+ CTCL. The only reliable method to distinguish HTLV-1+ cases is by serologic testing. It may be important to identify HTLV-1+ cases because they might have a more aggressive course. Special therapeutic measures, such as interferon-α, or interferon-β, and zidovudine, may improve the prognosis for these patients.

No Evidence of HTLV-1 Proviral Integration in Lymphoproliferative Disorders Associated With Cutaneous T-cell Lymphoma

Objective.—Cutaneous T-cell lymphoma (CTCL) is a CD4+ T-cell neoplasm of unknown cause; it includes mycosis fungoides and Sezary syndrome. Human T-lymphotropic virus type 1 (HTLV-1) genetic sequences have been found in patients with CTCL. Various molecular biology techniques were used to seek the presence of HTLV-1 proviral DNA sequences in lesions from patients with CTCL-associated diseases.

Methods.—A total of 35 CTCL-associated lymphoproliferative disorders drawn from the United States, Switzerland, and Japan were studied. Diagnoses included lymphomatoid papulosis, Hodgkin’s disease, and primary cutaneous CD30+ large-cell lymphoma of T-cell lineage. Concurrent CTCL was present in 6 patients. All assays performed—including genomic polymerase chain reaction (PCR)/Southern blotting, dot blotting, and Southern blotting—were specific for HTLV-1 provirus, with sensitivities ranging from 10-6 to 10-2.

Results.—A genomic PCR for the HTLV-1 proviral pX region, followed by confirmatory Southern blot analysis with a nested oligonucleotide probe, was uniformly negative. All cases were also negative on dot blot analysis of genomic DNA performed with a full-length HTLV-1 proviral DNA probe that included the entire HTLV-1 genome. When the same probe was used in Southern blot analysis of EcoRI-digested genomic DNA from cases of Hodgkin’s disease, the results were also negative.

Conclusions.—No HTLV-1 genetic sequences were identified in patients with CTCL-associated lymphoproliferative disorders. This is despite the use of sensitive assays specific for HTLV-1 in patients with a variety of diagnoses. Thus the HTLV-1 retrovirus does not appear to play a pathogenic role in these CTCL-associated disorders.

Screening Results for HTLV-1 and Epstein-Barr Virus DNA Sequences in Mycosis Fungoides and Sezary Syndrome in Japan

Background.—Recent reports have suggested that human T-cell lymphotropic virus type 1 (HTLV-1) and Epstein-Barr virus (EBV) play a pathogenic role in mycosis fungoides (MF) and Sezary syndrome (SS). Some reports have suggested that MF and SS may have etiologic features in common with adult T-cell leukemia/lymphoma (ATLL), although not all studies have confirmed this association. The relationship of HTLV-1 and EBV with MF and SS was studied in Japan, an endemic area for infections with these 2 viruses.

Methods.—A group of Japanese patients with cutaneous lymphoproliferative disorders were studied, including 14 with classic MF, 4 with SS, 6 with ATLL, and 33 with other conditions. Tissue or blood specimens were tested for the presence of HTLV-l-related genes—gag, pol, or pX-using the polymerase chain reaction techniques. Samples from 9 healthy controls were studied as well. Studies for the EBV gene were performed in 55 patients with lymphoproliferative disorders.

Results.—All 6 patients with ATLL tested positive for the gag, pol, and pX genes, compared with none of the 14 patients with MF. One of 4 patients with SS and 2 of 13 patients with non-MF T-cell lymphomas were positive for the pol gene. The sequence of the polymerase chain reaction product was the same as that registered for the HTLV-1 pol gene. Thirteen of 55 patients with cutancous lymphoproliferative disorders—but none of 17 patients with MF or SS—were found to have EBV DNA sequences.

Conclusions.—The findings do not support the suggestion that infection with HTLV-1 or EBV plays a pathogenic role in classic MF or SS in Japanese patients. Patients with other cutaneous lymphoproliferative disorders may show evidence of EBV involvement. More research is needed to confirm the presence of latent EBV infection, and to define the clinicopathologic findings in EBV-related disorders.

Anti-HTLV-1 Antibody Positive Cutaneous T-cell Lymphoma

Introduction.—Human T-lymphotropic virus type 1 (HTLV-1) is the cause of adult T-cell leukemia/lymphoma (ATLL), a neoplasm of the mature helper T lymphocyte. Recent reports have described the discovery of the HTLV-1 genome in some patients with cutaneous T-cell lymphoma (CTCL); thus HTLV-1 may have a causal relationship with CTCL. However, there is some diagnostic confusion between these 2 neoplasms. Two patients with CTCL and consistently positive serum anti-HTLV-1 antibodies were examined in an attempt to clarify the differences between CTCL and ATLL.

Methods.—The patients were a Japanese man and woman aged 54 and 59 years, respectively. Both had a diagnosis of CTCL; and both tested positive for serum anti-HTLV-1 antibodies. One patient had multiple subcutaneous nodules with internal organ involvement, whereas the other had typical mycosis fungoides. The 2 cases were studied in detail, with Southern blotting, inverse polymerase chain reaction (IPCR), and PCR with 4 sets of primers for the gag, pol, env, and pX regions of HTLV-1.

Findings.—In both patients, IPCR and Southern blotting showed no integration of HTLV-1 DNA. All but 1 of the PCRs with 4 sets of primers were positive for HTLV-1.

Conclusions.—The findings in CTCL patients positive for anti-HTLV-1 antibody may help to differentiate ATLL from CTCL. Human T-lymphotropic virus type 1 may be responsible for promoting or maintaining CTCL, although the HTLV-1 genome also may be indirectly involved in the initiation of CTCL.

► Evidence is accumulating that HTLV-1 does not play an important role in the pathogenesis of the vast majority of cases classified under the CTCL umbrella (Abstracts 21-8 and 21-9). Nevertheless, differentiation of adult T-cell leukemia/lymphoma from CTCL can be difficult, and it is still possible, as suggested by Kikuchi et al., (Abstract 21-10) that the HTLV-1 genome may play an indirect role in initiating CTCL.

Granulomatous Slack Skin in Childhood

Introduction.—Granulomatous slack skin (GSS) is thought to be a cutaneous lymphoma of T-helper cells. It has a progressive course that is sometimes characterized by transformation to disseminated lymphoma. Of 20 published cases in the literature, none of the patients have been children. The clinical, histopathologic, immunohistochemical, and genotypic findings are reported in a child with GSS that initially simulated “pure cutaneous histiocytosis X.”

Case Report.—Boy, 11 years, was evaluated for soft, painless skin masses in his neck, right axilla and arm, anterior wall of the abdomen, both inguinal regions, and malleolar and dorsal regions of both feet. The disease began 3 years earlier with erythematous lesions of the neck and wrists that were diagnosed as chronic granulomatous dermatitis. A diagnosis of cutaneous histi-ocytosis X was given after histopathologic examination of a skin biopsy specimen from the abd.ominal mass. The tumor masses diminished with corticoid therapy. Twenty-eight months later, abundant new masses appeared. Only part of the abdominal masses were removed because the surgical specimen of left-sided masses weighed 2,760 g. Histologic examination showed an extensive lymphoid infiltrate with scattered multinucleated giant cells extending from the papillary dermis to the subcutis. The immunophenotype of the lymphoid cells was CD43+ (MT1), DC45+, CD45RO+, and CD20—. For the giant cells, the phenotype was lysozyme positive, CD68+, and Mac387—. Genotyping revealed the tumoral lymphoid cells had clonal rearrangement for the gene of the (3 chain of the T-cell receptor (CβTCR). The disease was controlled with systemic glucocorticoids.

Conclusion.—This patient was initially diagnosed with a cutaneous form of histiocytosis because of the presence of many histocytes arranged in aggregates in the papillary and mid-dermis. Deep and extensive biopsies are advised in patients with slack skin disease.

► This is the first report of granulomatous siack skin occurring in a child. The clinical differential diagnosis includes other diseases associated with skin laxity including anetoderma and acquired cutis laxa. Microscopic observation of a lymphocytic infiltrate with prominent giant cells and phagocytosis of elastic fibers suggests a histologic differential diagnosis that includes granulomatous mycosis fungoides, granuloma annulare, xanthogranuloma, and sarcoidosis. The lymphocytic infiltrate is composed of T-helper cells and clonal rearrangement of the gene for the T-cell receptor β chain is characteristic. The authors emphasize that the unique clinical and histopathologic picture of granulomatous slack skin makes misdiagnosis unlikely even in children.

Lymphoproliferative DisordersEvaluation of Classic Architectural Criteria in Non-mycosis Fungoides Cutaneous Lymphomas

Objective.—It can be very difficult to make the pathologic diagnosis of cutaneous lymphoid infiltrates—not only to distinguish lymphomas from pseudolymphomas but also to subclassify the lymphoid infiltrates. The B-cell and T-cell patterns for evaluating cutaneous lymphoid malignancies are well recognized. However, exceptions can occur, and there have been no studies analyzing the architectural criteria of non-mycosis fungoides (non-MF) cutaneous lymphomas. The architectural features of a series of cutaneous B-cell lymphomas were analyzed to determine the frequency with which B-cell lymphomas mimic a T-cell pattern.

Methods.—Using published criteria for B- and T-cell cutaneous malignancies, 97 cases of non-MF cutaneous lymphomas were reviewed. A total of 77 primary and secondary cutaneous B-cell lymphomas were included, all with immunohistochemical studies identifying their lineage. The interpretations were made, along with 20 cases of non-MF and T-cell lymphoma, in randomized fashion from hematoxylin and eosin-stained slides. At least 2 dermatopathologists reviewed each slide, without knowledge of the previous diagnosis. Among the architectural criteria associated with B-cell patterns were the presence of dense perivascular, periappendageal, and/or nodular collections of lymphocytes, centered in the deep dermis, and separated from the epidermis by a grenz zone. For T-cell patterns, the criteria were location mainly in the upper dermis, an interstitial pattern, the presence of epidermotropism, and the lack of a grenz zone.

Results.—No clear pattern could be assigned to 3 of the 77 B-cell lymphomas. Epidermotropism in B-cell lymphomas was a rare but striking finding. Of the remaining 74, architectural criteria led to an incorrect diagnosis of T-cell lymphoma in 4 cases. Of these 4 misclassified specimens, 3 were large-cell lymphomas. The non-MF T-cell lymphomas studied often showed a B-cell pattern.

Conclusions.—Some rare B-cell lymphomas will have architectural features more consistent with those of T-cell lymphoma. This finding is particularly likely among B-large-cell lymphomas. As noted previously, some non-MF T-cell lymphomas have architectural features more characteristic of B-cell lymphomas. Although these patterns must be interpreted cautiously, most non-MF lymphomas will have a B-cell pattern, regardless of their true lineage.

► The authors’ message is quite clear: non-MF T-cell lymphomas may demonstrate a “B-cell pattern” and some B-cell lymphomas demonstrate a “T-cell pattern.” Hence, caution should be exercised in the interpretation of architectural criteria.

Epstein-Barr Virus-associated Lymphoproliferative Disease During Methotrexate Therapy for Psoriasis

Introduction.—A standard treatment for severe psoriasis and rheumatoid arthritis is low-dose methotrexate. However, high-dose methotrexate causes marked immunosuppression and toxic hematological reactions. Treatment of rheumatoid arthritis and dermatomyositis can be complicated by immunosuppression-associated lymphoproliferative disorders, which also have been observed in immunosuppressed organ transplant recipients. A patient who had long-term treatment with low-dose methotrexate for psoriasis developed a lymphoproliferative disease.

Case Study.—Man, 51, with a 15-year history of severe psoriasis, was treated with methotrexate at a weekly dose of 25 mg for 9 years. He developed a B-cell lymphoproliferative disorder and was treated with 4 cycles of combination chemotherapy. A year later on reexamination, he had no evidence of relapse. However, the following year he had erythrodermic psoriasis, malnutrition, and arthropathic involvement, and he died of septicemia.

Results.—Epstein-Barr virus (EBV) latent membrane protein 1 was expressed by the lymphoid cells and the EBV viral genome was present. The B-cell lymphoproliferative disorder was mainly monoclonal, according to the findings of polymerase chain reaction studies. These findings suggest that the disorder was the result of a single EBV-infected B-cell clone. In this patient, the EBV appears to have been critical in the pathogenesis of the lymphoproliferative disorder.

Conclusion.—In patients with psoriasis treated with methotrexate, lym-phoproliferative disorders associated with EBV infection may be seen. The clinicopathologic presentation is similar to that occurring in transplant patients who receive immunosuppression therapy.

► The authors describe a patient who developed a lymphoproliferative disease during long-term treatment with low-dose methotrexate. They argue that immunosuppression secondary to the drug led to infection with EBV and subsequent neoplasia. Unfortunately, they offer no laboratory evidence that the patient was immunosuppressed! Most frequently, EBV-associated lymphomas occur in patients receiving methotrexate therapy for rheumatoid arthritis or dermatomyositis. As is the case with Kaposi’s sarcoma occurring in immunosuppressed individuals, spontaneous remission may occur following discontinuation or reduction of the immunosuppressive drug(s).

Widespread Skin-limited Langerhans Cell Histiocytosis: Complete Remission With Interferon Alfa

Purpose.—The diagnosis of Langerhans cell histiocytosis (LCH) can be applied to a spectrum of diseases that range from a disseminated proliferative disorder in infants to a benign, solitary, lytic lesion of bone. In the middle are indolent types of LCH affecting the skin, mucous membranes, bones, and organs. Histologic examination reveals tissue infiltration with histiocytes that look like Langerhans cells. Few patients are seen with isolated LCH of the skin; usually they are children, in whom the condition is self-resolving. A case of disseminated LCH limited to the skin in an adult was reported.

Case.—Man, 29, had generalized skin nodules develop during a 2- month period. He had diffuse papular to nodular lesions located over his entire body, except for the face. The diagnosis of LCH was confirmed by skin biopsy and immunophenotyping studies. There were no signs of lymphadenopathy or organomegaly, no radio- graphic or CT abnormalities, and no bone marrow abnormalities.

Without treatment, the size and number of skin nodules continued to increase. The patient was started on subcutaneous interferon (IFN)-α, 6 megaunits/day. By 9 months, the skin nodules had largely disappeared, with skin biopsy confirming complete regression. The patient received continued IFN-α at a dosage of 3 megaunits/day for another 9 months, achieving complete remission. He was recurrence free on an IFN-a regimen of 3 megaunits twice weekly.

Discussion.—This patient’s skin lesions resolved with systemic IFN-a therapy. The mechanism of IFN-induced remission is unclear.

► Langerhans cell histiocytosis may be a clonal neoplastic disorder. If so, the mechanisms by which IFN induces remission may be similar to those observed in other clonal neoplastic disorders, such as chronic myelogenous leukemia.