- Email: [email protected]
CD30+ cutaneous lymphoproliferative disorders with pseudocarcinomatous hyperplasia are associated with a T-helper-17 cytokine profile and infiltrating granulocytes
CD301 cutaneous lymphoproliferative disorders with pseudocarcinomatous hyperplasia are associated with a T-helper-17 cytokine profile and infiltrating granulocytes Joan Guitart, MD,a Maria Estela Martinez-Escala, MD,a Janyana M. D. Deonizio, MD,a Pedram Gerami, MD,a and Marshall E. Kadin, MDb,c Chicago, Illinois; Boston, Massachusetts; and Providence, Rhode Island Background: The pathogenetic mechanism of CD301 cutaneous lymphoproliferative disorders (CLPD) associated with pseudocarcinomatous hyperplasia (PCH) and granulocytic inflammation surrounding atypical CD301 lymphocytes remains unclear. Objective: We sought to characterize clinical and pathological findings of a cohort of patients with PCH associated with CD301 CLPD and to analyze the cytokine profile of the atypical lymphocytes. Methods: We retrospectively reviewed medical records and pathological material of CD301 CLPD with PCH. Immunohistochemistry for T-helper (Th)17 cytokine profile was performed. Results: In all, 25 patients with a median age of 52 years were included. The median follow-up was 3.7 years. Histologically, an infiltrating pattern of PCH was observed in 14 cases with a neutrophilic-rich infiltrate (P = .21), and epidermal pattern in 11 cases with eosinophil-rich infiltrate (P = .03). Th17 or Th22 cytokines were detected in tumor cells in 81% cases tested. Tumor cells expressed Th17 transcription factor retinoic acid receptor (ROR)-related orphan receptor gamma-2 in 2 of 7 samples tested and 1 was positive for aryl hydrocarbon receptor. Limitations: This is a retrospective study of a small sample. Conclusions: PCH in CD301 CLPD is associated with Th17/Th22 cytokine expression in the atypical lymphocytes. Although these lesions commonly regress spontaneously and are associated with an indolent course, some cases develop a generalized process and tumor progression. ( J Am Acad Dermatol 2015;72:508-15.) Key words: CD301 cutaneous lymphoproliferative disorders; cytokine profile; granulocytes; pseudocarcinomatous hyperplasia; squamous cell carcinoma; T-helper-17/22.
P
and trauma.1 PCH has also been reported in CD301 cutaneous lymphoproliferative disorders (CLPD) and less commonly with other cutaneous T-cell lymphomas.2 The direct link between these events and epithelial hyperplasia is most likely inflammation. T-helper (Th)17 and Th22 cells link the immune
From the Departments of Dermatology and Pathology, Northwestern University Feinberg School of Medicine, Chicagoa; Department of Dermatology, Boston University School of Medicineb; and Roger Williams Medical Center, Providence.c Dr Kadin is supported by the Norko-Jon Family Foundation. Conflicts of interest: None declared. Accepted for publication November 19, 2014. Reprint requests: Joan Guitart, MD, Department of Dermatology, Northwestern University Feinberg School of Medicine, 676 N St
Clair St, Suite 1600, Chicago, IL 60091. E-mail: j-guitart@ northwestern.edu. Published online January 10, 2015. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.11.021
seudocarcinomatous hyperplasia (PCH) is a reactive epithelial proliferation that histologically and clinically may resemble squamous cell carcinoma (SCC). PCH has been associated with several triggering events including infections, neoplasias, toxic chemicals, autoimmune reactions,
508
J AM ACAD DERMATOL
Guitart et al 509
VOLUME 72, NUMBER 3
we analyzed 10 patients with a CD301 CLPD but response to tissue inflammation and interleukin (IL)-17A and IL-17F and IL-22 are their respective without PCH for Th17 cytokine expression. prototypic cytokines as key regulators of homeostasis and epithelial barrier function.3 The role of IL-22 Statistical analysis and IL-17 in the pathogenesis of psoriasis is well The inflammatory infiltrate was compared with known. This hyperplastic process is characterized by the groups obtained from the 2 types of PCH, cytokeratin 17 (CK17) expression.4 Therefore, epidermal and infiltrating. We used Mann Whitney we hypothesized that overU test for independent expression of IL-17 and/or groups to determine the staCAPSULE SUMMARY IL-22 by neoplastic lymtistical significance. The phocytes may provide a expression of Th17 Pseudocarcinomatous hyperplasia with rationale to link epithelial cytokines by lesions with a neutrophil- and eosinophil-rich hyperplasia and the granuloPCH were compared with infiltrate is a histologic feature of some cytic infiltrate. lesions without PCH. We CD301 cutaneous lymphoproliferative used x 2 exact test for indedisorders. pendent groups to determine METHODS Epidermal hyperplasia and the the statistical significance. Upon obtaining approval associated granulocytic infiltrate are from the institutional review linked to a T-helper-17 cytokine profile. board at Northwestern RESULTS Pseudocarcinomatous hyperplasia in University, we search our derDemographic data of the CD301 cutaneous lymphoproliferative matopathology service and sample are detailed in Table I. disorders can be misdiagnosed as cutaneous lymphoma dataFive cases were clinically squamous cell carcinoma and is bases for cases diagnosed as consistent with LyP with mulassociated with an indolent course with CD301 CLPD including lymtiple papulonodules, mean often spontaneous involution. less than 1 cm; 14 cases ALCL phomatoid papulosis (LyP) with tumor lesions of more and anaplastic large-cell lymthan 2 cm in greatest dimenphoma (ALCL) with assocision up to 8 cm; and 6 cases were classified as ated PCH. Clinical charts were reviewed when borderline tumors, as often seen in the spectrum of available and standard clinical and laboratory data CD301 CLPD conditions. Clusters of papulonodular along with clinical photographs were reviewed. Histopathological material was reviewed and several lesions or a large tumor with adjacent smaller satellite parameters, including the pattern of PCH (epidermal lesions were commonly reported. Most lesions had or infiltrating), density of the infiltrate, presence or central hemorrhagic and corrugated keratotic crust absence of epidermal or dermal necrosis, ulceration, often with craterlike ulceration (21/25). The lesions hemorrhage, adnexotropism, and presence of other were polypoid and exophytic in 7 and smooth without inflammatory cell components were assessed. ulceration in 8 cases (Fig 1). The clinical differential Density of eosinophils and neutrophils was semidiagnoses included infectious pyoderma, pyogenic quantitatively rated from 0 to 3 (0: not detected, 3: granuloma, pyoderma vegetans, SCC, and blastomyabundant) and the predominant size and morphology cosis. Single lesions (T1) were reported in 7 cases, 5 of the tumor cells was recorded. patients had lesions localized to 1 anatomic area (T2), An extensive immunohistochemistry panel and 13 had widespread lesions (T3), in 1 case the including CD3, CD4, CD5, CD7, CD8, CD30, T-cell extent of disease was unknown; the most common site restricted intracellular antigen-1, CD15, anaplastic were legs (19) followed by arms (12), torso (9), and lymphoma receptor tyrosine kinase-1, CK17, head/neck (4); no mucosal lesions were observed. In 3 transacting T-cell-specific transcriptor factor of the T3 patients we noted a tendency for the lesions guanine-adenosine-thyamine-adenosine-3 (GATA3), to be more prominent in 1 limb. Table II compares the ROR-related orphan receptor gamma-2, aryl hydroclinical data between both types of PCH pattern carbon receptor, and cytokines IL-17F and IL-22 (epidermal and infiltrating). Pruritus was reported in were performed. Epstein-Barr virus RNA expression 5 of 20 and 1 patient with localized disease (T1) was evaluated using in situ hybridization (Epsteinreported the tumor onset after arthropod bite. No Barr encoding region) and T-cell receptor gene other patient reported trauma, bites, or other precedclonality analysis with polymerase chain reaction ing lesion at the tumor site. No subject reported any methods using BIOMED-2 protocol was reported significant lymphadenopathy at presentation. Three when available from skin and blood samples. To children were included in the cohort, 2 boys and 1 girl. compare the results obtained with a control group, All 3 were 11 years old, 1 with a single lesion (T1) and d
d
d
J AM ACAD DERMATOL
510 Guitart et al
Abbreviations used: ALCL: CK17: CLPD: IL: KA: LyP: MF: PCH: SCC: Th:
anaplastic large-cell lymphoma cytokeratin 17 cutaneous lymphoproliferative disorders interleukin keratoacanthoma lymphomatoid papulosis mycosis fungoides pseudocarcinomatous hyperplasia squamous cell carcinoma T-helper
the other 2 with extensive disease (T3) and all 3 resolved spontaneously. Blood parameters were for the most part within normal limits with normal serum lactate dehydrogenase in 17 known cases. One patient had increased serum lactate dehydrogenase to 1.2 times the upper normal limit, but this patient also had breast cancer and the cutaneous nodules resolved in a few months without further recurrence. Positron emission tomography and computed tomography scans were negative except for increased glucose avidity at the tumor site in 16 patients. Medical history included 2 patients with remote history of systemic Hodgkin and non-Hodgkin lymphoma in remission, 2 patients with history of breast cancer, 1 with polycythemia vera, 1 with psoriasis, and 4 with atopic eczema. No patient had mycosis fungoides (MF). With a median follow-up of 3.7 years, 12 patients are in remission (11 alive without disease, 1 dead without disease), 12 have persistent disease (1 dead with disease, 11 alive with disease), and 1 was lost to follow-up. Progression of cutaneous lesions over time with larger ulcerated tumors and shotty and transient lymphadenopathy were noted in 4 cases, whereas the remaining 8 patients had stable disease. The T stage of the patients who achieved a complete resolution was lower (7 T1, 2 T2, 3 T3) than patients with persistent or progressive disease (0 T1, 3 T2, 8 T3). No patient developed nodal or systemic involvement. Nine patients reported 1 or more episodes of skin lesions followed by spontaneous remission. Nine patients achieved complete remission after surgery, radiation, phototherapy, oral methotrexate or in some cases spontaneously. The total treatments received with variable results included intralesional or oral steroids (15), oral methotrexate (8), surgery (7), localized electron beam radiation (7), systemic chemotherapy (3), phototherapy (2), and observation (1). Two surgical excisions were performed under the erroneous diagnosis of SCC. However local recurrence
MARCH 2015
Table I. Summary of demographic data Patients
Total, n M:F gender Age, y (range) Race White Asian Hispanic Black CD301 CLPD LyP ALCL Borderline Duration before diagnosis, (range) Follow-up duration, (range)
25 13:12 52 (11-89) 20 3 1 1 5 14 6 3.2 y (1 mo to 24 y) 3.7 y (5 mo to 13 y)
ALCL, Anaplastic large-cell lymphoma; CLPD, cutaneous lymphoproliferative disorder; F, female; LyP, lymphomatoid papulosis; M, male.
was observed at the surgical site in 1 of the cases (Fig 1, D). Histopathology PCH was extensive in most of the representative case biopsy specimens (21) and 4 cases were more localized. We noted 2 patterns of PCH. The infiltrating pattern observed in 14 cases was characterized by expansive infiltrating strands of large squamous cells resembling keratoacanthoma (KA) and radially arranged often arising from follicular infundibula with bulbous appearance and lesser involvement of the epidermis (Fig 2, A). The keratinocytes had abundant pale and glassy cytoplasm with prominent nucleoli but without hyperchromasia or pleomorphism (Fig 2, B). The epidermal pattern, observed in 11 cases, was a broader and more horizontal process frequently with extensive irregular acanthosis with broad strands of epithelium pushing into the dermis often accompanied by spongiosis (Fig 2, C and D). Admixed granulocytes within the lymphoid infiltrate were noted in most cases (23/25). The neutrophilic infiltrate noted in 22 of 25 tended to be brisk (111 in 12, 11 in 5, 1 in 5) and more commonly associated with the infiltrating pattern of PCH, but not statistically significant (P = .21), whereas eosinophils were observed in 15 cases (111 in 8, 11 in 3, 1 in 4) and more commonly associated with the epidermal pattern of PCH (P = .03) (Table III). Neutrophilic microabscesses within the epithelial structures were often noted. Rare plasma cells were observed in 6 of 25 cases (Fig 3). Areas of necrosis within the tumor were also observed in 13 of 25 cases (2 extensive, 11 focal) without specific association
J AM ACAD DERMATOL
Guitart et al 511
VOLUME 72, NUMBER 3
Fig 1. CD301 cutaneous lymphoproliferative disorders. A, Erythematous plaque with central erosion and 2 satellite lesions. B, Erythematous plaque with central necrotic crust on the right side of the jaw. C, Corrugated and eroded tumoral plaque on the right elbow. D, Erythematous and eroded plaque growing over skin graft of a previous lesion diagnosed as squamous cell carcinoma. E, Crusted nodule surrounded on erythematous plaque on the right arm. F, Smoothed nodule on the left elbow.
Table II. Comparison of demographic and clinical data based on histologic pattern Features
M:F gender Age, y (range) Duration before diagnosis, y (range) Spontaneous regression Diagnosis LyP Borderline ALCL Stage* T1 T2 T3
Epidermal (n = 11)
Infiltrating (n = 14)*
6:5 52.9 (11-89) 2.9 (0-7) 4
7:7 50.6 (11-87) 3.7 (0-20) 5
3 2 6
4 2 8
3 3 5
4 1 8
ALCL, Anaplastic large-cell lymphoma; F, female; LyP, lymphomatoid papulosis; M, male. *Clinical information of 1 patient was missing.
J AM ACAD DERMATOL
512 Guitart et al
MARCH 2015
Fig 2. CD301 cutaneous lymphoproliferative disorders. A, Pseudocarcinomatous hyperplasia (PCH) infiltrating type, prominent horn cysts are observed. B, High-power view of atypical large lymphocytes intermixed with neutrophils and strands of pale keratinocytes. C, PCH epidermal type with strands of epidermis extending to the mid dermis. D, High-power view of the strands of keratinocytes, with atypical lymphocytes intermixed with eosinophils.
Table III. Comparison between pseudocarcinomatous hyperplasia types and histologic details Histologic features
Average extent of PCH, n 1 (Focal) 11 (Moderate) 111 (Extensive) Neutrophils, n 0 1 2 3 Semiquantitative average Eosinophils, n 0 1 2 3 Semiquantitative average Type of cells, n Large Reed-Sternberg-like Small-medium Necrosis, n
Epidermal Infiltrating P PCH (n = 11) PCH (n = 14) value
3 3 5
1 3 10
2 3 2 4 1.7
1 2 3 8 2.4
.21
3 1 0 7 2.0
6 4 3 1 1.0
.03
6 3 2 5
7 7 0 9
PCH, Pseudocarcinomatous hyperplasia.
with certain cell types, pattern of PCH, or nature of the granulocytic infiltrate. Large lymphoid cells were the predominant cell type in all but 2 cases, which were characterized by medium/small cells. The large cells had a predominant immunoblastic appearance with characteristic multinucleated cells in 10 cases and mostly anaplastic or pleomorphic morphology in 15 cases. The distribution of positive immunomarkers was as follows: CD3 in 15 of 23, CD4 in 20 of 25, CD8 in 3 of 25, CD5 in 5 of 11, CD2 in 8 of 10, CD7 in 2 of 8, T-cell restricted intracellular antigen-1 in 8 of 13, cytoplasmic anaplastic lymphoma receptor tyrosine kinase-1 in 1 of 16, nuclear anaplastic lymphoma receptor tyrosine kinase-1 in 0 of 16, Epstein-Barr encoding region 0 of 8, and CD15 in 2 of 14. None of the cases with CD151 had a history of Hodgkin or non-Hodgkin lymphoma. Double-negative staining for CD4 and CD8 was observed in 3 of 25 cases. IL-17F was expressed in 12 of 16 cases and IL-22 in 11 of 13 (Fig 4, A and B). Combined expression of both cytokines were as follows: IL-17F1/IL-221 10 cases; IL-17F1/IL-22 1 case; IL-17F /IL-221 1 case; and IL-17F /IL-22 1 case. Thus, Th17 or Th22
J AM ACAD DERMATOL
Guitart et al 513
VOLUME 72, NUMBER 3
Fig 3. CD301 cutaneous lymphoproliferative disorders. A and B show microabscesses of neutrophils within the epidermis.
ROR-related orphan receptor gamma-2, and 1 was positive for aryl hydrocarbon receptor, which enhances IL-22 expression (Fig 4, C ). All 14 cases tested had strong expression of CK17 known to be inducible by Th17 cytokines (Fig 5).4 Ten control samples of CD301 CLPD but without PCH were assessed for the expression of Th17 cytokines. Eight of 10 controls were negative for Th17 cytokines. Differences were highly significant (P \ .001) by x2 exact test. T-cell receptor gene rearrangement analysis revealed positive clonality in 11 of 17 skin biopsy specimens and 2 of 4 peripheral blood.
DISCUSSION
Fig 4. CD301 cutaneous lymphoproliferative disorders. Tumor cells express interleukin (IL)-17 (A), IL-22 (B), and T-helper-17 transcriptor factor, ROR-related orphan receptor gamma-2 (C).
cytokines were detected in tumor cells in 13 of 16 (81%) cases tested. Th2 transcription factor GATA3 was negative in all 7 cases tested, 2 of 7 expressed
PCH has been previously reported in cutaneous T-cell lymphoma, mostly in CD301 CLPD cases, but also in MF and other cutaneous T-cell lymphoma variants.2 We have not observed PCH associated with MF and none of our CD301 CLPD with PCH had any evidence of plaque or plaque MF lesions. One exception is pagetoid reticulosis, considered a variant of MF that can present with marked verrucous epithelial hyperplasia resembling a verrucous carcinoma. Such cases also tend to express CD30 raising the possibility of a common pathogenetic mechanism. The prevalence of PCH in CD301 CLPD is not precisely known but has been reported to range between 14% and 55%.2,5,6 Our cohort of 25 cases is, to our knowledge, the largest series published to date, comprising approximately 10% to 15% of all our CD301 CLPD cases. The pathogenesis of PCH associated with CD301 CLPD remains unknown. Increased expression of epidermal growth factor and transforming growth factor-a by the tumor cells and epidermal growth factor receptor by epidermis has been implicated.2 However, this observation was not confirmed by others.7 Moreover, increased epidermal growth factor receptor expression may be a byproduct to be expected in a variety of PCH, regardless of the driving mechanism for epithelial growth. Instead,
514 Guitart et al
J AM ACAD DERMATOL
MARCH 2015
Fig 5. CD301 cutaneous lymphoproliferative disorders. A, Cytokeratin 17 is strongly positive on the hyperplastic epithelia. B, Tumoral cells express CD30.
we propose the hypothesis that PCH is caused by Th17/Th22 cytokines produced by tumor cells. IL-17 is secreted predominantly by Th17 cells, a special subset of Th CD4 cells. IL-17 synergizes with interferon-g to enhance proinflammatory cytokine production by human keratinocytes.8 IL-22 is produced by Th17 cells and in a more restricted fashion by Th22 cells. Th17 or Th22 cytokines were detected in tumor cells in 81% of the cases tested. Conversely none of the cases tested had Th2 transcription factor GATA3. In addition strong CK17 expression, which may be induced by T cells with a Th17 and/or Th1 cytokine profile, was noted in all cases tested. Therefore, our results suggest a key role for Th17 cytokines, particularly IL-22 in the pathogenesis of PCH. The brisk neutrophilic and to a lesser extent eosinophilic infiltrates observed in 92% of our cases supports this hypothesis. It cannot be assumed that granulocytes within the infiltrate directly cause the epithelial hyperplasia because most neutrophil-rich and eosinophil-rich CD301 CLPD in our experience and that of others show no evidence of PCH.9,10 The so-called pyogenic or neutrophil-rich ALCL has been reported to occur more frequently in immunocompromised patients, mostly patients with HIV and or those posttransplantation.10 Besides the intrinsic immune-senescence in elderly patients, only 1 patient was status postallohematopoietic stem cell transplantation and 4 had other lymphoproliferative or myeloproliferative syndromes. Spontaneous regression is common in LyP, but even larger tumors that are best classified as borderline lesions, type C LyP, or ALCL often show signs of regression with decrease size, lesser erythema, and extensive areas of necrosis. The CD30 antigen is a protein member of the tumor necrosis factor receptor superfamily that is involved in control of cell survival.11 The binding of CD30 and its ligand CD30L may provoke either apoptosis or
proliferation, depending of the microenvironment and nature of the cell expressing this antigen.12 Consequently, CD30 may play a role in the growth and involution phases of CD301 CLPD lesions. In particular, PCH associated with CD301 CLPD has been reported to predict spontaneous regression.6,13 Recently Zayour and Lazova6 proposed the term ‘‘oligolesional nodules with PCH followed by spontaneous resolution’’ as a variant of CD301 CLPD based on their experience with 4 cases that met such description. Our data confirm that spontaneous resolution is common, particularly in cases with limited disease. All patients with a single lesion (T1) in our cohort resolved spontaneously or after therapy without recurrence. However, PCH in CD301 CLPD tended to have a more protracted course in patients with more extensive disease and complete resolution was exceptional and only observed in 3 of 11 T3 patients. Interestingly, patients with extensive disease often demonstrated PCH in 1 of the skin biopsy specimens, but not on others, suggesting local interactions of cytokines with microenvironment. Our cohort includes 3 children, proving that this presentation is not limited to adult or elderly patients as previously suggested.6 Two of our cases had been misinterpreted as SCC. The diagnosis was corrected when other lesions recurred allowing for reinterpretation of the original material. Likewise, there are several reports in the literature of misdiagnosis of PCH as SCC or KA.6,7 This is not surprising because the infiltrative pattern may resemble SCC and in some cases the atypical lymphocytes are sparse or obscured by the neutrophilic and eosinophilic infiltrate.14,15 However, in our experience, the nuclear detail of keratinocytes lacked significant atypia and was instead composed of mature squamous cells with abundant pale cytoplasm and open chromatin, features more typical of KA. Conversely, in other cases the squamous component
J AM ACAD DERMATOL
Guitart et al 515
VOLUME 72, NUMBER 3
may be difficult to appreciate with narrow inconspicuous squamous strands obscured by the dense mixed infiltrate and areas of necrosis. We believe that some PCH with CD301 CLPD cases have been reported in the past as the follicular variant of LyP, characterized by a perifollicular or folliculotropic atypical T-cell infiltrate occasionally associated with hypertrophic follicular epithelium or pustules.16 Several years ago, we reported a case of KA associated with LyP, where the large atypical T cells were in close proximity to the squamous strands.17 At the time, we proposed that the neoplastic T cells may play a role in the limited epithelial growth of KA and that such hypothesis could also apply to cases previously reported as follicular LyP. This observation was recently supported in a study by Fernandez-Flores18 showing that scattered CD301 T cells throughout the lesion are a common component of KA. But overall these cells comprise a small proportion of the lesion and fail to show tumoral growth pattern or significant nuclear atypia as seen in our cases. In conclusion, we propose that the cytokine profile observed in CD301 CLPD with PCH is consistent with a Th17 neoplastic T-cell population, which may provide a rationale to link epithelial hyperplasia with the associated neutrophilic infiltrates and CD30 expression of neoplastic T cells. We also confirm that PCH in CD301 CLPD is associated with an indolent course and may forecast involution, but not necessarily predict a limited and self-contained lymphoproliferative process. REFERENCES 1. El-Khoury J, Kibbi AG, Abbas O. Mucocutaneous pseudoepitheliomatous hyperplasia: a review. Am J Dermatopathol. 2012;34:165-175. 2. Courville P, Wechsler J, Thomine E, et al. Pseudoepitheliomatous hyperplasia in cutaneous T-cell lymphoma. A clinical, histopathological and immunohistochemical study with particular interest in epithelial growth factor expression. The French Study Group on Cutaneous Lymphoma. Br J Dermatol. 1999;140:421-426. 3. Eyerich S, Eyerich K, Cavani A, Schmidt-Weber C. IL-17 and IL-22: siblings, not twins. Trends Immunol. 2010;31:354-361. 4. Depianto D, Kerns ML, Dlugosz AA, Coulombe PA. Keratin 17 promotes epithelial proliferation and tumor growth by
5.
6. 7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
polarizing the immune response in skin. Nat Genet. 2010;42: 910-914. Krishnan J, Tomaszewski MM, Kao GF. Primary cutaneous CD30-positive anaplastic large cell lymphoma. Report of 27 cases. J Cutan Pathol. 1993;20:193-202. Zayour M, Lazova R. Pseudoepitheliomatous hyperplasia: a review. Am J Dermatopathol. 2011;33:112-122. quiz 23-26. Scarisbrick JJ, Calonje E, Orchard G, Child FJ, Russell-Jones R. Pseudocarcinomatous change in lymphomatoid papulosis and primary cutaneous CD301 lymphoma: a clinicopathologic and immunohistochemical study of 6 patients. J Am Acad Dermatol. 2001;44:239-247. Teunissen MB, Koomen CW, de Waal Malefyt R, Wierenga EA, Bos JD. Interleukin-17 and interferon-gamma synergize in the enhancement of proinflammatory cytokine production by human keratinocytes. J Invest Dermatol. 1998;111:645-649. Kong YY, Dai B, Kong JC, Lu HF, Shi DR. Neutrophil/ eosinophil-rich type of primary cutaneous anaplastic large cell lymphoma: a clinicopathological, immunophenotypic and molecular study of nine cases. Histopathology. 2009;55: 189-196. Papalas JA, Van Mater D, Wang E. Pyogenic variant of primary cutaneous anaplastic large-cell lymphoma: a lymphoproliferative disorder with a predilection for the immunocompromised and the young. Am J Dermatopathol. 2010;32:821-827. Cespedes YP, Rockley PF, Flores F, Ruiz P, Kaiser MR, Elgart GW. Is there a special relationship between CD30-positive lymphoproliferative disorders and epidermal proliferation? J Cutan Pathol 2000;27:271-275. Gruss HJ, Boiani N, Williams DE, Armitage RJ, Smith CA, Goodwin RG. Pleiotropic effects of the CD30 ligand on CD30-expressing cells and lymphoma cell lines. Blood. 1994; 83:2045-2056. Bernier M, Bagot M, Broyer M, Farcet JP, Gaulard P, Wechsler J. Distinctive clinicopathologic features associated with regressive primary CD30 positive cutaneous lymphomas: analysis of 6 cases. J Cutan Pathol. 1997;24:157-163. Biswas A, Gey van Pittius D, Stephens M, Smith AG. Recurrent primary cutaneous lymphoma with florid pseudoepitheliomatous hyperplasia masquerading as squamous cell carcinoma. Histopathology. 2008;52:755-758. Lin JH, Lee JY. Primary cutaneous CD30 anaplastic large cell lymphoma with keratoacanthoma-like pseudocarcinomatous hyperplasia and marked eosinophilia and neutrophilia. J Cutan Pathol. 2004;31:458-461. Kempf W, Kazakov DV, Baumgartner HP, Kutzner H. Follicular lymphomatoid papulosis revisited: a study of 11 cases, with new histopathological findings. J Am Acad Dermatol. 2013;68: 809-816. Guitart J, Gordon K. Keratoacanthomas and lymphomatoid papulosis. Am J Dermatopathol. 1998;20:430-433. Fernandez-Flores A. CD301 cell population in common keratoacanthomas: a study of 21 cases. Rom J Morphol Embryol. 2008;49:159-162.
P
and trauma.1 PCH has also been reported in CD301 cutaneous lymphoproliferative disorders (CLPD) and less commonly with other cutaneous T-cell lymphomas.2 The direct link between these events and epithelial hyperplasia is most likely inflammation. T-helper (Th)17 and Th22 cells link the immune
From the Departments of Dermatology and Pathology, Northwestern University Feinberg School of Medicine, Chicagoa; Department of Dermatology, Boston University School of Medicineb; and Roger Williams Medical Center, Providence.c Dr Kadin is supported by the Norko-Jon Family Foundation. Conflicts of interest: None declared. Accepted for publication November 19, 2014. Reprint requests: Joan Guitart, MD, Department of Dermatology, Northwestern University Feinberg School of Medicine, 676 N St
Clair St, Suite 1600, Chicago, IL 60091. E-mail: j-guitart@ northwestern.edu. Published online January 10, 2015. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.11.021
seudocarcinomatous hyperplasia (PCH) is a reactive epithelial proliferation that histologically and clinically may resemble squamous cell carcinoma (SCC). PCH has been associated with several triggering events including infections, neoplasias, toxic chemicals, autoimmune reactions,
508
J AM ACAD DERMATOL
Guitart et al 509
VOLUME 72, NUMBER 3
we analyzed 10 patients with a CD301 CLPD but response to tissue inflammation and interleukin (IL)-17A and IL-17F and IL-22 are their respective without PCH for Th17 cytokine expression. prototypic cytokines as key regulators of homeostasis and epithelial barrier function.3 The role of IL-22 Statistical analysis and IL-17 in the pathogenesis of psoriasis is well The inflammatory infiltrate was compared with known. This hyperplastic process is characterized by the groups obtained from the 2 types of PCH, cytokeratin 17 (CK17) expression.4 Therefore, epidermal and infiltrating. We used Mann Whitney we hypothesized that overU test for independent expression of IL-17 and/or groups to determine the staCAPSULE SUMMARY IL-22 by neoplastic lymtistical significance. The phocytes may provide a expression of Th17 Pseudocarcinomatous hyperplasia with rationale to link epithelial cytokines by lesions with a neutrophil- and eosinophil-rich hyperplasia and the granuloPCH were compared with infiltrate is a histologic feature of some cytic infiltrate. lesions without PCH. We CD301 cutaneous lymphoproliferative used x 2 exact test for indedisorders. pendent groups to determine METHODS Epidermal hyperplasia and the the statistical significance. Upon obtaining approval associated granulocytic infiltrate are from the institutional review linked to a T-helper-17 cytokine profile. board at Northwestern RESULTS Pseudocarcinomatous hyperplasia in University, we search our derDemographic data of the CD301 cutaneous lymphoproliferative matopathology service and sample are detailed in Table I. disorders can be misdiagnosed as cutaneous lymphoma dataFive cases were clinically squamous cell carcinoma and is bases for cases diagnosed as consistent with LyP with mulassociated with an indolent course with CD301 CLPD including lymtiple papulonodules, mean often spontaneous involution. less than 1 cm; 14 cases ALCL phomatoid papulosis (LyP) with tumor lesions of more and anaplastic large-cell lymthan 2 cm in greatest dimenphoma (ALCL) with assocision up to 8 cm; and 6 cases were classified as ated PCH. Clinical charts were reviewed when borderline tumors, as often seen in the spectrum of available and standard clinical and laboratory data CD301 CLPD conditions. Clusters of papulonodular along with clinical photographs were reviewed. Histopathological material was reviewed and several lesions or a large tumor with adjacent smaller satellite parameters, including the pattern of PCH (epidermal lesions were commonly reported. Most lesions had or infiltrating), density of the infiltrate, presence or central hemorrhagic and corrugated keratotic crust absence of epidermal or dermal necrosis, ulceration, often with craterlike ulceration (21/25). The lesions hemorrhage, adnexotropism, and presence of other were polypoid and exophytic in 7 and smooth without inflammatory cell components were assessed. ulceration in 8 cases (Fig 1). The clinical differential Density of eosinophils and neutrophils was semidiagnoses included infectious pyoderma, pyogenic quantitatively rated from 0 to 3 (0: not detected, 3: granuloma, pyoderma vegetans, SCC, and blastomyabundant) and the predominant size and morphology cosis. Single lesions (T1) were reported in 7 cases, 5 of the tumor cells was recorded. patients had lesions localized to 1 anatomic area (T2), An extensive immunohistochemistry panel and 13 had widespread lesions (T3), in 1 case the including CD3, CD4, CD5, CD7, CD8, CD30, T-cell extent of disease was unknown; the most common site restricted intracellular antigen-1, CD15, anaplastic were legs (19) followed by arms (12), torso (9), and lymphoma receptor tyrosine kinase-1, CK17, head/neck (4); no mucosal lesions were observed. In 3 transacting T-cell-specific transcriptor factor of the T3 patients we noted a tendency for the lesions guanine-adenosine-thyamine-adenosine-3 (GATA3), to be more prominent in 1 limb. Table II compares the ROR-related orphan receptor gamma-2, aryl hydroclinical data between both types of PCH pattern carbon receptor, and cytokines IL-17F and IL-22 (epidermal and infiltrating). Pruritus was reported in were performed. Epstein-Barr virus RNA expression 5 of 20 and 1 patient with localized disease (T1) was evaluated using in situ hybridization (Epsteinreported the tumor onset after arthropod bite. No Barr encoding region) and T-cell receptor gene other patient reported trauma, bites, or other precedclonality analysis with polymerase chain reaction ing lesion at the tumor site. No subject reported any methods using BIOMED-2 protocol was reported significant lymphadenopathy at presentation. Three when available from skin and blood samples. To children were included in the cohort, 2 boys and 1 girl. compare the results obtained with a control group, All 3 were 11 years old, 1 with a single lesion (T1) and d
d
d
J AM ACAD DERMATOL
510 Guitart et al
Abbreviations used: ALCL: CK17: CLPD: IL: KA: LyP: MF: PCH: SCC: Th:
anaplastic large-cell lymphoma cytokeratin 17 cutaneous lymphoproliferative disorders interleukin keratoacanthoma lymphomatoid papulosis mycosis fungoides pseudocarcinomatous hyperplasia squamous cell carcinoma T-helper
the other 2 with extensive disease (T3) and all 3 resolved spontaneously. Blood parameters were for the most part within normal limits with normal serum lactate dehydrogenase in 17 known cases. One patient had increased serum lactate dehydrogenase to 1.2 times the upper normal limit, but this patient also had breast cancer and the cutaneous nodules resolved in a few months without further recurrence. Positron emission tomography and computed tomography scans were negative except for increased glucose avidity at the tumor site in 16 patients. Medical history included 2 patients with remote history of systemic Hodgkin and non-Hodgkin lymphoma in remission, 2 patients with history of breast cancer, 1 with polycythemia vera, 1 with psoriasis, and 4 with atopic eczema. No patient had mycosis fungoides (MF). With a median follow-up of 3.7 years, 12 patients are in remission (11 alive without disease, 1 dead without disease), 12 have persistent disease (1 dead with disease, 11 alive with disease), and 1 was lost to follow-up. Progression of cutaneous lesions over time with larger ulcerated tumors and shotty and transient lymphadenopathy were noted in 4 cases, whereas the remaining 8 patients had stable disease. The T stage of the patients who achieved a complete resolution was lower (7 T1, 2 T2, 3 T3) than patients with persistent or progressive disease (0 T1, 3 T2, 8 T3). No patient developed nodal or systemic involvement. Nine patients reported 1 or more episodes of skin lesions followed by spontaneous remission. Nine patients achieved complete remission after surgery, radiation, phototherapy, oral methotrexate or in some cases spontaneously. The total treatments received with variable results included intralesional or oral steroids (15), oral methotrexate (8), surgery (7), localized electron beam radiation (7), systemic chemotherapy (3), phototherapy (2), and observation (1). Two surgical excisions were performed under the erroneous diagnosis of SCC. However local recurrence
MARCH 2015
Table I. Summary of demographic data Patients
Total, n M:F gender Age, y (range) Race White Asian Hispanic Black CD301 CLPD LyP ALCL Borderline Duration before diagnosis, (range) Follow-up duration, (range)
25 13:12 52 (11-89) 20 3 1 1 5 14 6 3.2 y (1 mo to 24 y) 3.7 y (5 mo to 13 y)
ALCL, Anaplastic large-cell lymphoma; CLPD, cutaneous lymphoproliferative disorder; F, female; LyP, lymphomatoid papulosis; M, male.
was observed at the surgical site in 1 of the cases (Fig 1, D). Histopathology PCH was extensive in most of the representative case biopsy specimens (21) and 4 cases were more localized. We noted 2 patterns of PCH. The infiltrating pattern observed in 14 cases was characterized by expansive infiltrating strands of large squamous cells resembling keratoacanthoma (KA) and radially arranged often arising from follicular infundibula with bulbous appearance and lesser involvement of the epidermis (Fig 2, A). The keratinocytes had abundant pale and glassy cytoplasm with prominent nucleoli but without hyperchromasia or pleomorphism (Fig 2, B). The epidermal pattern, observed in 11 cases, was a broader and more horizontal process frequently with extensive irregular acanthosis with broad strands of epithelium pushing into the dermis often accompanied by spongiosis (Fig 2, C and D). Admixed granulocytes within the lymphoid infiltrate were noted in most cases (23/25). The neutrophilic infiltrate noted in 22 of 25 tended to be brisk (111 in 12, 11 in 5, 1 in 5) and more commonly associated with the infiltrating pattern of PCH, but not statistically significant (P = .21), whereas eosinophils were observed in 15 cases (111 in 8, 11 in 3, 1 in 4) and more commonly associated with the epidermal pattern of PCH (P = .03) (Table III). Neutrophilic microabscesses within the epithelial structures were often noted. Rare plasma cells were observed in 6 of 25 cases (Fig 3). Areas of necrosis within the tumor were also observed in 13 of 25 cases (2 extensive, 11 focal) without specific association
J AM ACAD DERMATOL
Guitart et al 511
VOLUME 72, NUMBER 3
Fig 1. CD301 cutaneous lymphoproliferative disorders. A, Erythematous plaque with central erosion and 2 satellite lesions. B, Erythematous plaque with central necrotic crust on the right side of the jaw. C, Corrugated and eroded tumoral plaque on the right elbow. D, Erythematous and eroded plaque growing over skin graft of a previous lesion diagnosed as squamous cell carcinoma. E, Crusted nodule surrounded on erythematous plaque on the right arm. F, Smoothed nodule on the left elbow.
Table II. Comparison of demographic and clinical data based on histologic pattern Features
M:F gender Age, y (range) Duration before diagnosis, y (range) Spontaneous regression Diagnosis LyP Borderline ALCL Stage* T1 T2 T3
Epidermal (n = 11)
Infiltrating (n = 14)*
6:5 52.9 (11-89) 2.9 (0-7) 4
7:7 50.6 (11-87) 3.7 (0-20) 5
3 2 6
4 2 8
3 3 5
4 1 8
ALCL, Anaplastic large-cell lymphoma; F, female; LyP, lymphomatoid papulosis; M, male. *Clinical information of 1 patient was missing.
J AM ACAD DERMATOL
512 Guitart et al
MARCH 2015
Fig 2. CD301 cutaneous lymphoproliferative disorders. A, Pseudocarcinomatous hyperplasia (PCH) infiltrating type, prominent horn cysts are observed. B, High-power view of atypical large lymphocytes intermixed with neutrophils and strands of pale keratinocytes. C, PCH epidermal type with strands of epidermis extending to the mid dermis. D, High-power view of the strands of keratinocytes, with atypical lymphocytes intermixed with eosinophils.
Table III. Comparison between pseudocarcinomatous hyperplasia types and histologic details Histologic features
Average extent of PCH, n 1 (Focal) 11 (Moderate) 111 (Extensive) Neutrophils, n 0 1 2 3 Semiquantitative average Eosinophils, n 0 1 2 3 Semiquantitative average Type of cells, n Large Reed-Sternberg-like Small-medium Necrosis, n
Epidermal Infiltrating P PCH (n = 11) PCH (n = 14) value
3 3 5
1 3 10
2 3 2 4 1.7
1 2 3 8 2.4
.21
3 1 0 7 2.0
6 4 3 1 1.0
.03
6 3 2 5
7 7 0 9
PCH, Pseudocarcinomatous hyperplasia.
with certain cell types, pattern of PCH, or nature of the granulocytic infiltrate. Large lymphoid cells were the predominant cell type in all but 2 cases, which were characterized by medium/small cells. The large cells had a predominant immunoblastic appearance with characteristic multinucleated cells in 10 cases and mostly anaplastic or pleomorphic morphology in 15 cases. The distribution of positive immunomarkers was as follows: CD3 in 15 of 23, CD4 in 20 of 25, CD8 in 3 of 25, CD5 in 5 of 11, CD2 in 8 of 10, CD7 in 2 of 8, T-cell restricted intracellular antigen-1 in 8 of 13, cytoplasmic anaplastic lymphoma receptor tyrosine kinase-1 in 1 of 16, nuclear anaplastic lymphoma receptor tyrosine kinase-1 in 0 of 16, Epstein-Barr encoding region 0 of 8, and CD15 in 2 of 14. None of the cases with CD151 had a history of Hodgkin or non-Hodgkin lymphoma. Double-negative staining for CD4 and CD8 was observed in 3 of 25 cases. IL-17F was expressed in 12 of 16 cases and IL-22 in 11 of 13 (Fig 4, A and B). Combined expression of both cytokines were as follows: IL-17F1/IL-221 10 cases; IL-17F1/IL-22 1 case; IL-17F /IL-221 1 case; and IL-17F /IL-22 1 case. Thus, Th17 or Th22
J AM ACAD DERMATOL
Guitart et al 513
VOLUME 72, NUMBER 3
Fig 3. CD301 cutaneous lymphoproliferative disorders. A and B show microabscesses of neutrophils within the epidermis.
ROR-related orphan receptor gamma-2, and 1 was positive for aryl hydrocarbon receptor, which enhances IL-22 expression (Fig 4, C ). All 14 cases tested had strong expression of CK17 known to be inducible by Th17 cytokines (Fig 5).4 Ten control samples of CD301 CLPD but without PCH were assessed for the expression of Th17 cytokines. Eight of 10 controls were negative for Th17 cytokines. Differences were highly significant (P \ .001) by x2 exact test. T-cell receptor gene rearrangement analysis revealed positive clonality in 11 of 17 skin biopsy specimens and 2 of 4 peripheral blood.
DISCUSSION
Fig 4. CD301 cutaneous lymphoproliferative disorders. Tumor cells express interleukin (IL)-17 (A), IL-22 (B), and T-helper-17 transcriptor factor, ROR-related orphan receptor gamma-2 (C).
cytokines were detected in tumor cells in 13 of 16 (81%) cases tested. Th2 transcription factor GATA3 was negative in all 7 cases tested, 2 of 7 expressed
PCH has been previously reported in cutaneous T-cell lymphoma, mostly in CD301 CLPD cases, but also in MF and other cutaneous T-cell lymphoma variants.2 We have not observed PCH associated with MF and none of our CD301 CLPD with PCH had any evidence of plaque or plaque MF lesions. One exception is pagetoid reticulosis, considered a variant of MF that can present with marked verrucous epithelial hyperplasia resembling a verrucous carcinoma. Such cases also tend to express CD30 raising the possibility of a common pathogenetic mechanism. The prevalence of PCH in CD301 CLPD is not precisely known but has been reported to range between 14% and 55%.2,5,6 Our cohort of 25 cases is, to our knowledge, the largest series published to date, comprising approximately 10% to 15% of all our CD301 CLPD cases. The pathogenesis of PCH associated with CD301 CLPD remains unknown. Increased expression of epidermal growth factor and transforming growth factor-a by the tumor cells and epidermal growth factor receptor by epidermis has been implicated.2 However, this observation was not confirmed by others.7 Moreover, increased epidermal growth factor receptor expression may be a byproduct to be expected in a variety of PCH, regardless of the driving mechanism for epithelial growth. Instead,
514 Guitart et al
J AM ACAD DERMATOL
MARCH 2015
Fig 5. CD301 cutaneous lymphoproliferative disorders. A, Cytokeratin 17 is strongly positive on the hyperplastic epithelia. B, Tumoral cells express CD30.
we propose the hypothesis that PCH is caused by Th17/Th22 cytokines produced by tumor cells. IL-17 is secreted predominantly by Th17 cells, a special subset of Th CD4 cells. IL-17 synergizes with interferon-g to enhance proinflammatory cytokine production by human keratinocytes.8 IL-22 is produced by Th17 cells and in a more restricted fashion by Th22 cells. Th17 or Th22 cytokines were detected in tumor cells in 81% of the cases tested. Conversely none of the cases tested had Th2 transcription factor GATA3. In addition strong CK17 expression, which may be induced by T cells with a Th17 and/or Th1 cytokine profile, was noted in all cases tested. Therefore, our results suggest a key role for Th17 cytokines, particularly IL-22 in the pathogenesis of PCH. The brisk neutrophilic and to a lesser extent eosinophilic infiltrates observed in 92% of our cases supports this hypothesis. It cannot be assumed that granulocytes within the infiltrate directly cause the epithelial hyperplasia because most neutrophil-rich and eosinophil-rich CD301 CLPD in our experience and that of others show no evidence of PCH.9,10 The so-called pyogenic or neutrophil-rich ALCL has been reported to occur more frequently in immunocompromised patients, mostly patients with HIV and or those posttransplantation.10 Besides the intrinsic immune-senescence in elderly patients, only 1 patient was status postallohematopoietic stem cell transplantation and 4 had other lymphoproliferative or myeloproliferative syndromes. Spontaneous regression is common in LyP, but even larger tumors that are best classified as borderline lesions, type C LyP, or ALCL often show signs of regression with decrease size, lesser erythema, and extensive areas of necrosis. The CD30 antigen is a protein member of the tumor necrosis factor receptor superfamily that is involved in control of cell survival.11 The binding of CD30 and its ligand CD30L may provoke either apoptosis or
proliferation, depending of the microenvironment and nature of the cell expressing this antigen.12 Consequently, CD30 may play a role in the growth and involution phases of CD301 CLPD lesions. In particular, PCH associated with CD301 CLPD has been reported to predict spontaneous regression.6,13 Recently Zayour and Lazova6 proposed the term ‘‘oligolesional nodules with PCH followed by spontaneous resolution’’ as a variant of CD301 CLPD based on their experience with 4 cases that met such description. Our data confirm that spontaneous resolution is common, particularly in cases with limited disease. All patients with a single lesion (T1) in our cohort resolved spontaneously or after therapy without recurrence. However, PCH in CD301 CLPD tended to have a more protracted course in patients with more extensive disease and complete resolution was exceptional and only observed in 3 of 11 T3 patients. Interestingly, patients with extensive disease often demonstrated PCH in 1 of the skin biopsy specimens, but not on others, suggesting local interactions of cytokines with microenvironment. Our cohort includes 3 children, proving that this presentation is not limited to adult or elderly patients as previously suggested.6 Two of our cases had been misinterpreted as SCC. The diagnosis was corrected when other lesions recurred allowing for reinterpretation of the original material. Likewise, there are several reports in the literature of misdiagnosis of PCH as SCC or KA.6,7 This is not surprising because the infiltrative pattern may resemble SCC and in some cases the atypical lymphocytes are sparse or obscured by the neutrophilic and eosinophilic infiltrate.14,15 However, in our experience, the nuclear detail of keratinocytes lacked significant atypia and was instead composed of mature squamous cells with abundant pale cytoplasm and open chromatin, features more typical of KA. Conversely, in other cases the squamous component
J AM ACAD DERMATOL
Guitart et al 515
VOLUME 72, NUMBER 3
may be difficult to appreciate with narrow inconspicuous squamous strands obscured by the dense mixed infiltrate and areas of necrosis. We believe that some PCH with CD301 CLPD cases have been reported in the past as the follicular variant of LyP, characterized by a perifollicular or folliculotropic atypical T-cell infiltrate occasionally associated with hypertrophic follicular epithelium or pustules.16 Several years ago, we reported a case of KA associated with LyP, where the large atypical T cells were in close proximity to the squamous strands.17 At the time, we proposed that the neoplastic T cells may play a role in the limited epithelial growth of KA and that such hypothesis could also apply to cases previously reported as follicular LyP. This observation was recently supported in a study by Fernandez-Flores18 showing that scattered CD301 T cells throughout the lesion are a common component of KA. But overall these cells comprise a small proportion of the lesion and fail to show tumoral growth pattern or significant nuclear atypia as seen in our cases. In conclusion, we propose that the cytokine profile observed in CD301 CLPD with PCH is consistent with a Th17 neoplastic T-cell population, which may provide a rationale to link epithelial hyperplasia with the associated neutrophilic infiltrates and CD30 expression of neoplastic T cells. We also confirm that PCH in CD301 CLPD is associated with an indolent course and may forecast involution, but not necessarily predict a limited and self-contained lymphoproliferative process. REFERENCES 1. El-Khoury J, Kibbi AG, Abbas O. Mucocutaneous pseudoepitheliomatous hyperplasia: a review. Am J Dermatopathol. 2012;34:165-175. 2. Courville P, Wechsler J, Thomine E, et al. Pseudoepitheliomatous hyperplasia in cutaneous T-cell lymphoma. A clinical, histopathological and immunohistochemical study with particular interest in epithelial growth factor expression. The French Study Group on Cutaneous Lymphoma. Br J Dermatol. 1999;140:421-426. 3. Eyerich S, Eyerich K, Cavani A, Schmidt-Weber C. IL-17 and IL-22: siblings, not twins. Trends Immunol. 2010;31:354-361. 4. Depianto D, Kerns ML, Dlugosz AA, Coulombe PA. Keratin 17 promotes epithelial proliferation and tumor growth by
5.
6. 7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
polarizing the immune response in skin. Nat Genet. 2010;42: 910-914. Krishnan J, Tomaszewski MM, Kao GF. Primary cutaneous CD30-positive anaplastic large cell lymphoma. Report of 27 cases. J Cutan Pathol. 1993;20:193-202. Zayour M, Lazova R. Pseudoepitheliomatous hyperplasia: a review. Am J Dermatopathol. 2011;33:112-122. quiz 23-26. Scarisbrick JJ, Calonje E, Orchard G, Child FJ, Russell-Jones R. Pseudocarcinomatous change in lymphomatoid papulosis and primary cutaneous CD301 lymphoma: a clinicopathologic and immunohistochemical study of 6 patients. J Am Acad Dermatol. 2001;44:239-247. Teunissen MB, Koomen CW, de Waal Malefyt R, Wierenga EA, Bos JD. Interleukin-17 and interferon-gamma synergize in the enhancement of proinflammatory cytokine production by human keratinocytes. J Invest Dermatol. 1998;111:645-649. Kong YY, Dai B, Kong JC, Lu HF, Shi DR. Neutrophil/ eosinophil-rich type of primary cutaneous anaplastic large cell lymphoma: a clinicopathological, immunophenotypic and molecular study of nine cases. Histopathology. 2009;55: 189-196. Papalas JA, Van Mater D, Wang E. Pyogenic variant of primary cutaneous anaplastic large-cell lymphoma: a lymphoproliferative disorder with a predilection for the immunocompromised and the young. Am J Dermatopathol. 2010;32:821-827. Cespedes YP, Rockley PF, Flores F, Ruiz P, Kaiser MR, Elgart GW. Is there a special relationship between CD30-positive lymphoproliferative disorders and epidermal proliferation? J Cutan Pathol 2000;27:271-275. Gruss HJ, Boiani N, Williams DE, Armitage RJ, Smith CA, Goodwin RG. Pleiotropic effects of the CD30 ligand on CD30-expressing cells and lymphoma cell lines. Blood. 1994; 83:2045-2056. Bernier M, Bagot M, Broyer M, Farcet JP, Gaulard P, Wechsler J. Distinctive clinicopathologic features associated with regressive primary CD30 positive cutaneous lymphomas: analysis of 6 cases. J Cutan Pathol. 1997;24:157-163. Biswas A, Gey van Pittius D, Stephens M, Smith AG. Recurrent primary cutaneous lymphoma with florid pseudoepitheliomatous hyperplasia masquerading as squamous cell carcinoma. Histopathology. 2008;52:755-758. Lin JH, Lee JY. Primary cutaneous CD30 anaplastic large cell lymphoma with keratoacanthoma-like pseudocarcinomatous hyperplasia and marked eosinophilia and neutrophilia. J Cutan Pathol. 2004;31:458-461. Kempf W, Kazakov DV, Baumgartner HP, Kutzner H. Follicular lymphomatoid papulosis revisited: a study of 11 cases, with new histopathological findings. J Am Acad Dermatol. 2013;68: 809-816. Guitart J, Gordon K. Keratoacanthomas and lymphomatoid papulosis. Am J Dermatopathol. 1998;20:430-433. Fernandez-Flores A. CD301 cell population in common keratoacanthomas: a study of 21 cases. Rom J Morphol Embryol. 2008;49:159-162.