Clinical, histopathologic, and immunophenotypic features of lymphomatoid papulosis with CD8 predominance in 14 pediatric patients

Clinical, histopathologic, and immunophenotypic features of lymphomatoid papulosis with CD8 predominance in 14 pediatric patients Aieska de Souza, MD, Michael J. Camilleri, MD, David A. Wada, MD, David L. Appert, MD, Lawrence E. Gibson, MD, and Rokea A. el-Azhary, MD, PhD Rochester, Minnesota Background: Lymphomatoid papulosis (LyP) is a cyclic papulonodular eruption that is clinically benign and histologically malignant. Association with hematologic neoplasias has been reported in 5% to 20% of all cases. Objective: We sought to review the clinical and histopathologic features of LyP in pediatric patients. Methods: We searched for the records of all patients with a clinical and histopathologic diagnosis of LyP seen at our clinic from January 1991 through April 2008. The cases of pediatric patients (aged \ 20 years) were reviewed in detail. Results: Of 123 patients with LyP identified, 14 (11%) were in the pediatric age group. Most were male (64%); mean age of onset was 12 years. Type A LyP was identified in 12 patients, one patient had type B, and none had type C (type not determined in one case). Ten cases showed CD8 predominance by immunohistochemistry. T-cell intracytoplasmic antigen staining was positive in 3 cases of CD81 LyP type A and the one case of LyP type B. Lesional T-cell receptor gene rearrangement studies were negative in 9 of 10 patients with LyP type A. The average follow-up time was 5.5 years. Lesions improved with treatment in most cases, and none of the cases were associated with hematologic malignancies. Limitations: This was a retrospective review. Conclusions: Among our pediatric patients, we noted a predominance of CD81 LyP, which does not seem to have an aggressive course. Further longitudinal studies are necessary to evaluate prognostic differences between CD41 and CD81 LyP and their biological significance. ( J Am Acad Dermatol 2009;61:993-1000.) Key words: CD8 disorders; immunophenotypic features; lymphomatoid papulosis; pediatric patients; T-cell receptor gene rearrangement.

L

ymphomatoid papulosis (LyP) is a rhythmic papular eruption of unknown cause that has malignant histologic features and a benign clinical course.1 Classically, it is characterized by

From the Department of Dermatology, Mayo Clinic. Funding sources: None. Conflicts of interest: None declared. Accepted for publication May 14, 2009. Reprint requests: Rokea A. el-Azhary, MD, PhD, Department of Dermatology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. E-mail: [email protected]. Published online July 6, 2009. 0190-9622/$36.00 ª 2009 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2009.05.014

Abbreviations used: EORTC: European Organization for Research and Treatment of Cancer LyP: lymphomatoid papulosis TCRGR: T-cell receptor gene rearrangement TIA-1: T-cell intracytoplasmic antigen WHO: World Health Organization

cyclic eruptions of papulonodular lesions with occasional central ulceration, followed by spontaneous healing and scar formation.2 LyP is classified according to the World Health Organization (WHO)eEuropean Organization for Research and Treatment of Cancer (EORTC) as a CD301 lymphoproliferative disorder3 and, although it is not an 993

994 de Souza et al

J AM ACAD DERMATOL DECEMBER 2009

aggressive malignant process, the association with inclusion criteria required a clinical and histopathohematologic neoplasias can range from 5% to 20% of logic diagnosis of LyP. all cases.4-6 All available biopsy specimens from the pediatric LyP rarely presents in childhood. Only a few patients were reviewed. Original laboratory reports previous reports have detailed the clinical and laband slides were reviewed; additional stains were oratory features of LyP in pediatric patients.4-10 The requested if they were not previously performed. most common histopathologic subtype described for Cases were evaluated using the WHO-EORTC3 criteria for LyP and classified as adult-onset and pediatric-ontype A, B, or C. Tissue was set LyP is type A, which is CAPSULE SUMMARY prepared for light microscharacterized by large, atypcopy by fixation in 10% neuical, CD301 lymphocytes, reLymphomatoid papulosis (LyP) is rare in sembling Reed-Sternberg tral buffered formaldehyde pediatric patients; we present 14 cells of Hodgkin lymphoma, and by staining with hemapediatric cases, one of the largest series present in a wedge-shaped toxylin and eosin. Immunoin the literature. distribution throughout the peroxidase staining was also Pediatric LyP is clinically similar to adult dermis and admixed with performed with antileukoLyP, but T-cell clonality is less common in various numbers of inflamcyte monoclonal antibodies pediatric LyP, and no associated matory cells. Fewer cases are against CD3 and CD4 (Novohematologic malignancies developed in classified as type B, which is castra, Leica Microsystems, our pediatric patients. characterized by small, Wetzlar, Germany), T-cell inCD30e lymphocytic cells tracytoplasmic antigen (TIAImmunophenotyping studies showed a with cerebriform nuclei in a 1) (Immunotec, Marseilles, previously unreported predominance of 1 bandlike pattern, along with France), and CD8 and CD30 CD8 LyP that is not aggressive and epidermotropism. Type C is (Dako, Carpinteria, CA). cannot be clinically distinguished from a more rare form of LyP Slides of immunophenothe CD4-predominant subtype. consisting of a monotonous typing studies were evaluated population of large, atypical, by two different observers (A. CD301 cells diffusely infiltrating the dermis, with d. S. and M. J. C.). Anti-CD30 and anti-CD3 antibody fewer associated inflammatory cells than observed in studies were considered positive if 5% or more of the the other types.3,6 Immunophenotyping studies of cells in the infiltrate were positive. CD4 and CD8 LyP lesions have shown a predominance of CD4 subpopulations were analyzed according to the total expression1-8; rare cases of CD8-predominant LyP CD31 population present on each slide. CD4 or CD8 10,11 and have been previously observed in children predominance was defined as the subpopulation of adults.12-14 cells that constituted the majority of the CD31 and T-cell receptor gene rearrangement (TCRGR) CD301 large cells in cases of type A LyP and the 5 studies may show positive results for LyP lesions. majority of the atypical cells in type B LyP. However, they must be interpreted with caution The results of TCRGR studies were extracted from because clonality can be present in other diseases the patient records. TCRGR studies were performed such as cutaneous T-cell lymphomas, parapsoriasis, by Southern blotting, polymerase chain reaction, or lymphomatoid lupus profundus, atopic dermatitis, both, on samples of skin lesion (snap-frozen tissue or pityriasis lichenoides chronica, pigmentary purpura, paraffin-fixed material), peripheral blood, or bone and drug-associated lymphomatoid hypersensitivmarrow, singly or in combination, as previously ity.15-17 described.5 Here, we describe the clinical, histopathologic, A follow-up questionnaire was sent to all patients and molecular features of LyP in pediatric patients with questions regarding continued eruptions, cur(aged \ 20 years). rent disease status, and treatments used. All patients included had signed the authorization form for research. d

d

d

METHODS After institutional review board approval, we searched our electronic patient database for the records of all patients with a diagnosis of LyP from January 1991 through April 2008. Of the cases retrieved, we reviewed only those in the pediatric age group, defined as age younger than 20 years. The

RESULTS A total of 123 patients with LyP were seen at our institution during the study period, of which 14 (11.4%) were in the pediatric age group. The clinical characteristics of these 14 patients are shown in Table I. Most patients were male (9 male, 5 female)

de Souza et al 995

J AM ACAD DERMATOL VOLUME 61, NUMBER 6

Table I. Clinical and pathologic features of lymphomatoid papulosis in pediatric patients with therapy and follow-up LyP lesion characteristics Patient No./age, y/sex

Type/T-cell type

Lesion/size, mm

Location

Duration, wk/scar

Peripheral blood findings*

Followup, y

2/N

np

2

1/5/M

A/CD8

Papules/1-3

Trunk, extremities

2/11/F

A/null

Papules/2-6

Trunk, extremities

3/12/M

A/CD4

Papules/ 0.5-10

Trunk, neck, extremities

4/19/F

A/CD8

Papules/2-5

Trunk, extremities

5/1/M

A/CD8

Papules/2-4

Trunk, extremities

4-6/Y

np

6/8/M

A/CD8

Papules/3

Trunk, extremities

8/Y

np

7/11/F

A/CD8

3-4/N

np

8/12/M

A/CD8

9/14/M

A/CD8

Papules with Extremities scales/4-15

10/15/M

A/CD8

Papules/3

11/18/F

A/CD8

Papules/5-7

12/19/F

A/CD8

Papules/2

Nodules/ND Hemorrhagic Extremities papules/5 Ulcerated Trunk, papules/3-12 extremities

Face, trunk, extremities Trunk, extremities Upper extremities, axillas

No Se´zary cells; CD19: 3%, CD5: 55%, CD2: 75%, CD3: 61%, CD4: 49%, CD8: 20%, CD161/CD3e (NK cells): 23%, H/S ratio: 2 3-6/Y No Se´zary cells; normal immunophenotyping; no clones detected by PCR 8-12/Y np 3-4/Y

4-6/Y

ND/Y

No Se´zary cells; CD3: 85%, CD19: 8%; CD161/CD561 (NK cells): 8%, CD4: 52%, CD8: 28%, H/S ratio: 1.9 No Se´zary cells

9

Treatment and response

Total improvement with UV treatment; recurrence in winter Partial improvement with UV treatment; less frequent eruptions

2

Partial improvement with UV treatment; less frequent eruptions 7 No improvement with oral antibiotics and UV treatment 13 Total improvement with UV treatment; recurrence with treatment interruption 0.2 No improvement with oral antibiotics and topical corticosteroids; total improvement with sun exposure 1

No improvement with UV treatment None UV treatment

7

4/Y

np

No improvement with topical corticosteroids and UV treatment None UV treatment

ND/N

np

None Oral antibiotics

2/N

np

9

Partial improvement with oral antibiotics and topical corticosteroids; less frequent eruptions

Plaques/ND Continued

996 de Souza et al

J AM ACAD DERMATOL DECEMBER 2009

Table I. Cont’d LyP lesion characteristics Patient No./age, y/sex

Type/T-cell type

Lesion/size, mm

Location

13/17/M

B/CD4

Papules/3

Face, trunk, axillas

14/11/M

B/CD8 np/npy

Papules/3

Trunk, extremities

Duration, wk/scar

Peripheral blood findings*

ND/N

np

2/N

Lymphocytes with convoluted nuclei present; CD3: 60%, CD19: 19%, CD161/CD561 (NK cells): 20%, CD4: 32%, CD8: 26%, H/S ratio: 1.23

Followup, y

Treatment and response

None Topical corticosteroids 2

No improvement with topical corticosteroids and UV treatment

F, Female; H/S, helper/suppressor CD4/CD8 ratio; LyP, lymphomatoid papulosis; M, male; N, no; ND, not described on chart; NK, natural killer; np, not performed; PCR, polymerase chain reaction; UV, ultraviolet; Y, yes. *T-cell receptor gene rearrangement studies were positive for only one patient (patient 6), not performed for 3 patients (patients 10, 12, and 13), and negative for all other patients. y Biopsy specimen of lesion showed mixed infiltrate with large, atypical, CD301 cells. No other tissue was available for further subtyping.

with a mean and median age of onset of 12 years (range, 1-19 years). The average lesion size was 3.5 mm; lesions were described as erythematous papules (some with scales and ulceration) or hemorrhagic papules (Fig 1). Nodules or plaques were present in combination with papules in two cases. Lesions were present in multiple locations for all patients. The areas predominantly affected were the upper and lower extremities and the trunk. No mucosal lesions were observed. The clinical course in all cases was characterized by recurrent eruption of lesions, with an average duration of 3 weeks, lasting up to 12 weeks in some cases (Table I). The number of lesions varied from 1 to 75 at a time (mean, 23). In two cases, the main symptom was pruritus, and in one case the lesions were tender. Scars were present in 8 patients (57%). The scar was hyperpigmented in 4 cases, a combination of hyperpigmented and hypopigmented in one case, and not specified in 3 cases. In all 14 cases, biopsy results were compatible with a diagnosis of LyP. In general, a mixed, predominantly lymphocytic inflammatory infiltrate with admixed neutrophils, eosinophils, and intravascular neutrophils was noted. The cases were divided according to histopathologic findings characterizing LyP as type A, B, or C (Table II). Biopsy results for patient 14 confirmed LyP with CD301 cells, as documented in his history. Unfortunately, no tissue samples were available to do any further immunohistochemical staining. None of the biopsy specimens reviewed had features compatible with LyP

Fig 1. Typical lymphomatoid papulosis lesions in pediatric patient showing discrete erythematous papules and hemorrhagic necrotic center.

type C. Most cases (n = 12) were LyP type A, which characteristically demonstrated a wedgelike dermal infiltrate composed of a mixture of cells, including large atypical lymphocytes, histiocytes, neutrophils, and occasional eosinophils (Fig 2, A). Ten of these 12 patients had CD8-predominant LyP. Only one patient had LyP type B; the histopathologic analysis showed a bandlike pattern of the infiltrate in the papillary dermis, with smaller, cerebriform lymphocytes and nuclear atypia. In addition, this patient had both CD41 and CD81 cells. Other than the known features that distinguish LyP type A from type B, no notable differences distinguished CD8-predominant from CD4-predominant LyP in the A or B subtypes (Table II). One patient with LyP type A was negative for both CD4 and CD8 (Fig 3).

de Souza et al 997

J AM ACAD DERMATOL VOLUME 61, NUMBER 6

Table II. Histopathologic features according to lymphomatoid papulosis type and predominant T-lymphocyte population (n = 13) Type A Histopathologic features

Epidermotropism Vacuolar changes Pautrier abscess Eosinophilic spongiosis Papillary dermal edema Eosinophils Neutrophils Intravascular neutrophils Thrombi Fibrinoid necrosis Vasculitic changes Perieccrine atypical lymphocytic infiltration Perineural atypical lymphocytic infiltration Pilosebaceous atypical lymphocytic infiltration Adipose tissue atypical lymphocytic infiltration Red blood cell extravasation Ulceration/erosion

1

CD4 (n = 1)

1 1

1 1 1 1 1 1 1

1

CD8 (n = 10)*

Type B CD4 /CD8 (n = 1) CD4 (n = 1)y CD81 (n = 1)y

6 (60) 6 (60) 0 0 5 (50) 5 (50) 5 (50) 4 (40) 3 (30) 3 (30) 0 6 (75) (n = 8) 3 (37.5) (n = 8) 8 (80) 1 (12.5) (n = 8) 5 (50) 3 (30)

e

e

1

1

1 1

1 1 1 1 1

1 1 1

1

1 1

1 1

1

*Values are No. of patients (%) with histopathologic feature. y Represents same patient, who had lymphomatoid papulosis type B and was CD41/CD81.

For the 12 patients with LyP type A lesions, TCRGR studies were performed in 10 patients (two of whom had testing of a blood sample in addition to a skin lesion). All showed negative results except for one case (patient 6), in which a minor T-cell clonal population (gamma chain J-region) was detected on the lesional skin biopsy specimen by Southern blotting. The two blood samples tested for TCRGR were negative. Table I summarizes the immunophenotyping studies. The infiltrate consisted of lymphocytes that were reactive for CD3 in all specimens and for CD30 in all LyP type A cases (12/12). Lymphocytes did not stain for CD30 in LyP type B specimens (0/2 specimens, both from the same patient). The CD31 T-lymphocyte subpopulation predominantly coexpressed CD4 ( $ 50% of cells) in one of the LyP type A biopsy specimens and in one of the two biopsy specimens of the type B case. The large atypical cells stained positive for CD8, as did most ( $ 50%) of the CD31 T lymphocytes in 10 of the LyP type A cases (Fig 2) and the one case of LyP type B, for a total of 11 of the 14 cases. The TIA-1 stain was positive in the infiltrate in 3 cases of CD81 LyP type A (patients 7, 8, and 12) and the one case of CD81 LyP type B (patient 13) (data not shown). TIA-1 staining was negative in two cases of CD81 LyP type A. The interpretation of TIA-1 staining in the other cases could not be performed because of limitations of tissue sampling and availability.

Peripheral blood smear and flow cytometry analysis were performed in 4 patients with LyP type A lesions (patients 2, 3, 8, and 9). Peripheral blood immunophenotyping showed unremarkable CD4 and CD8 counts with normal CD4:CD8 ratios in all 4 cases. Lymphocytes with convoluted nuclei were present in only one case, the patient (patient 14) with unknown LyP type. Treatment, which was prescribed immediately after the diagnosis in all cases (Table I), consisted of ultraviolet light therapy alone in 7 cases or a combination of ultraviolet treatment and oral antibiotics (patient 4), or topical corticosteroids (patients 9 and 14). Oral antibiotics were prescribed alone in one case (patient 11) and in combination with topical corticosteroids in two cases (patients 6 and 12). Topical hydrocortisone alone was prescribed for one case (patient 13). Follow-up information, gathered by chart review, response to letter questionnaire, or both, was available for 10 cases. The average time between the first appointment and last updated information was 5.5 years (range, 0.2-13 years). Lesions improved with treatment in most cases. Total clearing of the lesions without recurrence followed treatment in 3 cases. The others had no change (4 cases) or partial improvement (3 cases), with worsening of lesions with treatment interruption or during the winter (Table I). None of the cases were associated with hematologic malignancies. However, one patient (patient 8)

998 de Souza et al

J AM ACAD DERMATOL DECEMBER 2009

Fig 3. Histopathologic findings in patient 2 with lymphomatoid papulosis type A and null (CD4e/CD8e) phenotype. (Hematoxylin-eosin stain.) Inset A, Anti-CD30 immunostain highlights large atypical cells. Inset B, Anti-CD4 immunostain is negative in large cells. Inset C, Anti-CD8 immunostain is also negative in large cells. (Main image and insets A, B, and C, Original magnifications: 3400.)

had a medulloepithelioma at age 2 years, followed by enucleation of the left eye and glaucoma, and LyP at age 12 years.

DISCUSSION

Fig 2. Histopathologic findings in patient with lymphomatoid papulosis type A. A, Sample of lesion showing typical histologic features, including large anaplastic lymphoid cells in dermis with mixed inflammatory background. (Hematoxylin-eosin stain.) B, Anti-CD30 immunostain highlights many large atypical cells. C, Anti-CD4 immunostain is negative in large cells. D, Many large atypical lymphoid cells are positive for CD8 (antiCD8 antibody). (A to D, Original magnifications: 3400.)

LyP, a relatively rare disorder, is even less commonly encountered in the pediatric population. Childhood LyP constituted 11.4% of the total population of patients with LyP in our study. The average age of onset in our group of 14 patients was 12 years; presenting features included papules and nodules, which most often persisted in a cyclic eruptive pattern, which is similar to the pattern in adults. Previous studies of childhood LyP reported a younger mean age of onset (9 years), a similar presentation with cyclic papulonodular eruptions, and variable response to treatment.2,4,5,7-11 The histopathologic features showed a predominance of type A LyP with a polymorphous infiltrate, with or without eosinophils and large atypical CD301 lymphocytes, similar to features in adults.5,6 Previous reports of TCRGR studies showed positive results in two cases without association with lymphoproliferative disorders.10,11 In our series, one of 11 patients had a positive TCRGR resultea minor gamma-chain clone in an 8-year-old boy. Although follow-up was limited, LyP in this patient did show a typical cyclic pattern and improvement with sun exposure. Our series showed a high incidence of CD8predominant cases, the significance of which

J AM ACAD DERMATOL

de Souza et al 999

VOLUME 61, NUMBER 6

remains to be elucidated. No clinical or pathologic features distinguished CD4-predominant cases from CD8-predominant cases. A CD81 phenotype in lymphoproliferative disorders commonly associated with cytotoxic features may be associated with aggressive lymphoma,14 but this was not observed in our CD81 LyP group. The expression of the cytotoxic marker TIA-1 in CD41 LyP has been well documented previously.18,19 TIA-1 expression was documented in several of our cases of CD81 LyP, a pattern that is less commonly recognized, perhaps owing to the rarity of the CD81 subset. Many CD301 lymphoproliferative disorders are associated with a cytotoxic profile, and this finding is consistent with the phenotypic similarities shared by this group of diseases, including LyP. The clinical and histopathologic features, and the behavior of the eruption, in our patients were typical of LyP on the basis of the clinical follow-up (up to 13 years). The paucity of CD81 pediatric LyP cases reported previously in the literature might be explained by the fact that this marker was not reported in many of the large series.10,11 CD8 expression has been reported in related diseases such as hypopigmented mycosis fungoides and pityriasis lichenoides varioliformis acuta.20-22 Histopathologic analysis of 15 cases of hypopigmented mycosis fungoides, a cutaneous T-cell lymphoma with characteristic onset in childhood or adolescence, demonstrated a predominance of CD8 expression in most of the cases (9/15) in addition to the presence of clonal TCRGR.20 A CD8-predominant T-cell infiltrate has also been noted in pityriasis lichenoides varioliformis acuta, an inflammatory disease characterized by cyclic, benign, self-healing eruptions clinically indistinguishable from those of LyP.21,22 A previously published report on CD81 LyP in a series of 4 patients (aged 14-43 years) demonstrated histologic features that were distinct from our cases.12 This group observed a predominance of small uniform or monotypic lymphocytes, the presence of a granulomatous inflammation around eccrine sweat coils, and an absence of eosinophils and neutrophils. We did not observe these features in our specimens; no granulomas were seen, and the infiltrate was polymorphous rather than monomorphous. The authors hypothesized that the immunologic mechanism leading to the prevalence of the suppressor phenotype was responsible for vasculitic changes in LyP, a feature not observed in our patients.12 We observed a high frequency of epidermotropism, vacuolar changes, papillary dermal edema, thrombi, and fibrinoid necrosis, but no true vasculitis. Perieccrine and pilosebaceous

infiltration of atypical lymphocytes, as well as dermal eosinophils and neutrophils, were observed in almost all CD81 LyP type A cases but not in the LyP type B case. Our findings support the previously published impression that granulomas and a monomorphous lymphoid infiltrate are uncommon in LyP.3-7,10,11,23 None of the patients in our study had development of an associated lymphoma. Although 3 previous series of case reports failed to detect associated hematologic malignancies (mean follow-up: 12.2 years,5 4 years,7 and 4.6 years8), Nijsten et al4 described 3 cases in a cohort of 35 (9%) with associated non-Hodgkin lymphoma (median follow-up: 9 years). Dupont24,25 described a pediatric patient with LyP, in whom undifferentiated lymphoma developed after 40 years of follow-up; Weinman and Ackerman26 reported a pediatric case of LyP associated with a diagnosis of Hodgkin disease at age 57 years. The longest follow-up in our series was 13 years. Continued close monitoring of these patients is warranted. Longitudinal follow-up is particularly important for patient 6, the 8-year-old boy with CD8-predominant LyP and a positive TCRGR result. In summary, LyP in the pediatric population is uncommon, and constitutes approximately 10% of all LyP cases. Our pediatric cohort of cases demonstrated similar clinical features, including a male predominance, to those previously reported in adult LyP; however, clonality studies were mostly negative in our series. Similar to previous reports on pityriasis lichenoides varioliformis acuta and hypopigmented mycosis fungoides, we noted a predominance of CD8 positivity in LyP type A, the biological significance of which is not currently known. However, the lesions behaved more like benign LyP clinically rather than an aggressive CD81 variant. Type C LyP was not observed in our group. None of our patients had development of lymphoma, but many continued to have cyclic recurrences of LyP eruptions. Further longitudinal studies with larger numbers of cases and prolonged follow-up are necessary to validate these findings and to evaluate whether prognostic differences arise between CD41 and CD81 LyP cases. REFERENCES 1. Macaulay WL. Lymphomatoid papulosis: a continuing selfhealing eruption, clinically benignehistologically malignant. Arch Dermatol 1968;97:23-30. 2. Bories N, Thomas L, Phan A, Balme B, Salameire D, ThurotGuillou C, et al. Lymphomatoid papulosis in childhood: six case reports and a literature review [in French]. Ann Dermatol Venereol 2008;135:657-62. 3. Slater DN. The new World Health Organization-European Organization for Research and Treatment of Cancer

1000 de Souza et al

J AM ACAD DERMATOL DECEMBER 2009

4.

5.

6.

7. 8.

9.

10. 11.

12.

13.

14.

15.

classification for cutaneous lymphomas: a practical marriage of two giants. Br J Dermatol 2005;153:874-80. Nijsten T, Curiel-Lewandrowski C, Kadin ME. Lymphomatoid papulosis in children: a retrospective cohort study of 35 cases. Arch Dermatol 2004;140:306-12. el-Azhary RA, Gibson LE, Kurtin PJ, Pittelkow MR, Muller SA. Lymphomatoid papulosis: a clinical and histopathologic review of 53 cases with leukocyte immunophenotyping, DNA flow cytometry, and T-cell receptor gene rearrangement studies. J Am Acad Dermatol 1994;30:210-8. Sanchez NP, Pittelkow MR, Muller SA, Banks PM, Winkelmann RK. The clinicopathologic spectrum of lymphomatoid papulosis: study of 31 cases. J Am Acad Dermatol 1983;8:81-94. Zirbel GM, Gellis SE, Kadin ME, Esterly NB. Lymphomatoid papulosis in children. J Am Acad Dermatol 1995;33:741-8. Van Neer FJ, Toonstra J, Van Voorst Vader PC, Willemze R, Van Vloten WA. Lymphomatoid papulosis in children: a study of 10 children registered by the Dutch cutaneous lymphoma working group. Br J Dermatol 2001;144:351-4. Ghiggeri GM, Bleid D, Garaventa A, Coccia C, Gambini C, Caridi G, et al. Recurrent lymphomatoid papulosis associated with nephrotic syndrome: an occurrence of uncertain origin. Pediatr Nephrol 2009;24:189-92. Aoki E, Aoki M, Kono M, Kawana S. Two cases of lymphomatoid papulosis in children. Pediatr Dermatol 2003;20:146-9. Hellman J, Phelps RG, Baral J, Fasy TM, Ahern CM, Strauchen JA. Lymphomatoid papulosis with antigen deletion and clonal rearrangement in a 4-year-old boy. Pediatr Dermatol 1990;7: 42-7. Magro CM, Crowson AN, Morrison C, Merati K, Porcu P, Wright ED. CD81 lymphomatoid papulosis and its differential diagnosis. Am J Clin Pathol 2006;125:490-501. Weinstein A, Mirzabeigi M, Withee M, Vincek V. Lymphomatoid papulosis in an HIV-positive man. AIDS Patient Care STDS 2004;18:563-7. Berti E, Tomasini D, Vermeer MH, Meijer CJ, Alessi E, Willemze R. Primary cutaneous CD8-positive epidermotropic cytotoxic T cell lymphomas: a distinct clinicopathological entity with an aggressive clinical behavior. Am J Pathol 1999;155:483-92. Gellrich S, Wilks A, Lukowsky A, Wernicke M, Muller A, Marcus Muche J, et al. T cell receptor-gamma gene analysis of CD301 large atypical individual cells in CD301 large

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

primary cutaneous T cell lymphomas. J Invest Dermatol 2003;120:670-5. Magro CM, Nuovo GJ, Crowson AN. The utility of the in situ detection of T-cell receptor beta rearrangements in cutaneous T-cell-dominant infiltrates. Diagn Mol Pathol 2003;12:133-41. Kazakov DV, Kutzner H, Palmedo G, Boudova L, Michaelis S, Michal M, et al. Primary cutaneous lymphoproliferative disorders with dual lineage rearrangement. Am J Dermatopathol 2006;28:399-409. Boulland ML, Wechsler J, Bagot M, Pulford K, Kanavaros P, Gaulard P. Primary CD30-positive cutaneous T-cell lymphomas and lymphomatoid papulosis frequently express cytotoxic proteins. Histopathology 2000;36:136-44. Kummer JA, Vermeer MH, Dukers D, Meijer CJ, Willemze R. Most primary cutaneous CD30-positive lymphoproliferative disorders have a CD4-positive cytotoxic T-cell phenotype. J Invest Dermatol 1997;109:636-40. El-Shabrawi-Caelen L, Cerroni L, Medeiros LJ, McCalont TH. Hypopigmented mycosis fungoides: frequent expression of a CD81 T1cell phenotype. Am J Surg Pathol 2002;26:450-7. Jang KA, Choi JC, Choi JH. Expression of cutaneous lymphocyte-associated antigen and TIA-1 by lymphocytes in pityriasis lichenoides et varioliformis acuta and lymphomatoid papulosis: immunohistochemical study. J Cutan Pathol 2001;28:453-9. Varga FJ, Vonderheid EC, Olbricht SM, Kadin ME. Immunohistochemical distinction of lymphomatoid papulosis and pityriasis lichenoides et varioliformis acuta. Am J Pathol 1990;136: 979-87. Kadin ME. Characteristic immunologic profile of large atypical cells in lymphomatoid papulosis: possible implications for histogenesis and relationship to other diseases. Arch Dermatol 1986;122:1388-90. Dupont A. La longue evolution d’une reticulose papuleuse apparement benigne et sa transformation malinge tardive. Bull Soc Fr Dermatol Syph 1973;80:153-4. Dupont A. Very late malignant transformation of papular reticulosis with a prolonged development (lymphomatoid papulosis) [in French]. Ann Dermatol Syph (Paris) 1973;100: 141-6. Weinman VF, Ackerman AB. Lymphomatoid papulosis: a critical review and new findings. Am J Dermatopathol 1981; 3:129-63.