ORIGINAL STUDY
Phenotypic Variability in Primary Cutaneous Anaplastic Large T-cell Lymphoma: A Study on 35 Patients Cesare Massone, MD and Lorenzo Cerroni, MD
Abstract: Primary cutaneous anaplastic large T-cell lymphoma (pcALCL) is a well-defined entity characterized by neoplastic cells expressing CD30, CD2, CD3, CD4, and CD5. Cases with different phenotype have been reported, including variable loss of CD2, CD3, and CD5, and expression of cytotoxic phenotype (CD8+) and/or of cytotoxic proteins. Aberrant phenotypes represent a diagnostic pitfall and may be the cause of misdiagnoses. We reviewed 35 cases of pcALCL (M:F = 19:16; mean age, 50.8 years; range, 14–92 years), to better characterize the immunophenotypic spectrum of the disease. Twelve cases (34%) had a T-helper phenotype (CD4+/CD82), and TIA-1 was positive in 5 of 8 stained cases. Six cases (18%) had a T-cytotoxic phenotype (CD42/CD8+) and were also positive for TIA-1. Positivity for both CD4 and CD8 was observed in 7 cases (20%), 4 of which were stained for TIA-1 and found to be positive, whereas both CD4 and CD8 were negative in 9 cases (26%, only 1/8 tested cases being TIA-1 positive). CD2 was positive in 21 of 27 tested cases (78%), CD3 in 21 of 34 cases (62%), and CD5 in 15 of 31 cases (48%). Interestingly, 11 cases (31%) showed a profoundly aberrant phenotype lacking simultaneously several T-cell markers. Our data allow a better characterization of pcALCL with aberrant phenotypes, showing the remarkable variability in expression of different markers. Key Words: primary cutaneous anaplastic large T-cell lymphoma, CD30+ cutaneous lymphoproliferative disorders, cutaneous T-cell lymphoma, immunophenotype (Am J Dermatopathol 2014;36:153–157)
INTRODUCTION Primary cutaneous anaplastic large T-cell lymphoma (pcALCL) is a well-defined entity with prognostic and biological differences from the nodal counterpart. It is usually characterized by large cells with anaplastic, pleomorphic, or immunoblastic morphology, but several histopathologic variants have been described.1–3 Neutrophils (“inflammatory-type” pattern), lymphocytes (“lymphohistiocytic” pattern), or even eosinophils can predominate in the infiltrate, and cases with small- to medium-sized cells (“small-cell” pattern) or signetring cell morphology have also been reported.4,5 From the Research Unit Dermatopathology, Department of Dermatology, Medical University of Graz, Austria. The authors declare no conflicts of interest. Reprints: Lorenzo Cerroni, MD, Department of Dermatology, Medical University of Graz, Auenbruggerplatz 8, Graz A-8036, Austria (e-mail: [email protected]). © 2014 Lippincott Williams & Wilkins
Am J Dermatopathol Volume 36, Number 2, February 2014
Neoplastic cells in conventional cases of pcALCL express CD30 and CD2, CD3, CD4, and CD5 (T-helper phenotype), and are negative for CD8, CD56, and cytotoxic proteins.1–3 Cases with different phenotype have been reported, including variable loss of the T-cell markers CD2, CD3, and CD5, and expression of cytotoxic phenotype (CD8+) and/or of cytotoxic proteins (granzyme B, TIA-1, perforin).1–3,6–10 Aberrant phenotypes represent a diagnostic pitfall and may be the cause of misdiagnoses. Although pcALCL is traditionally considered as a lymphoma of T-helper lymphocytes, the exact proportion of cases with aberrant phenotype is unknown. In this study, we analyzed the clinicopathologic and immunophenotypic features of 35 patients with pcALCL, to better characterize the immunophenotypic spectrum of this cutaneous lymphoma.
MATERIALS AND METHODS From a total of 114 cases of pcALCL documented in the lymphoma database of the Research Unit of Dermatopathology, Department of Dermatology, Medical University of Graz (Graz, Austria), 35 cases were selected for the study. The other 79 cases were excluded because of lack of complete data or of available material for further stainings. Patients with history of mycosis fungoides (MF) or lymphomatoid papulosis (LyP) and “borderline cases” of CD30+ cutaneous lymphoproliferative disorder (cases in which a definitive distinction between pcALCL and LyP could not be made) and cases of cutaneous Hodgkin disease have been excluded from the study. Diagnosis of pcALCL has been made according to the WHO classification of lymphomas.1,2 Primary skin involvement was defined as the presence of cutaneous lymphoma without nodal and/or visceral involvement after complete staging procedures.1,2 Details of some of the cases have been published previously.4,11
Histology
Biopsy specimens were fixed in 10% buffered formalin and subsequently embedded in paraffin. Sections were stained with hematoxylin–eosin for routine histopathologic evaluation. All cases and stainings have been reviewed independently by both the authors.
Immunohistology Detailed immunophenotypic analyses were performed on routinely fixed paraffin-embedded tissue sections according to a previously described 3-step immunoperoxidase method. Microwave enhancement was used for all the antibodies. Second and third antibodies were obtained from Dako (Dakopatts; Glostrup, Denmark). The first antibody were from www.amjdermatopathology.com |
153
Massone and Cerroni
Am J Dermatopathol Volume 36, Number 2, February 2014
Novocastra-Leica Microsystems (Wetzlar, Germany; CD2, CD3, CD4, CD5), Dako (Dakopatts, CD8, CD30, ALK-1), and Immunotech (Prague, Czech Republic; TIA-1).12
phenotype (CD42/CD8+) and were also positive for TIA-1. Positivity for both CD4 and CD8 was observed in 7 cases (20%), 4 of which were stained for TIA-1 and found to be positive. In 9 cases, both CD4 and CD8 were negative (26%), with only 1 of 8 cases stained for TIA-1 being positive in this group. Analysis of the phenotypic pattern in 24 cases with complete stainings revealed many cases with profound aberrations. In addition, it is striking that no case showed a “conventional” T-helper phenotype without loss of pan–T-cell markers and without expression of cytotoxic proteins. Five cases were positive for CD2 only and negative for all other T-cell markers and for cytotoxic proteins (Fig. 1). Four cases lacked expression of all tested pan–T-cell markers and were positive only for CD8 and TIA-1 (2 cases) (Fig. 2), CD4 (1 case), or CD4, CD8, and TIA-1 (1 case).
RESULTS Patients were 19 males and 16 females. The median age was 59 years (mean age: 50.8 years; range 14–92 years). All cases revealed strong expression of CD30 and were negative for ALK-1. No case was associated with HIV infection or iatrogenic immunosuppression. Immunohistochemical data are reported in Table 1. CD2 was positive in 21 of 27 (78%), CD3 in 21 of 34 (62%), and CD5 in 15 of 31 (48%) tested cases. Twelve cases (34%) had a T-helper phenotype (CD4+/CD82), and TIA-1 was positive in 5 of 8 stained cases. Six cases (18%) had a T-cytotoxic
TABLE 1. Clinical Data and Immunophenotype of Our Series of Patients Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Sex
Age
CD2
CD3
CD5
CD4
CD8
TIA-1
M F M M M F F M F F M F F M M M F F M M F M M M M M M F F F F M M F F
82 58 70 80 92 54 14 60 59 40 78 62 16 67 58 48 69 46 43 18 57 68 83 55 60 72 82 61 64 82 77 66 14 71 35
+ + + + 2 2 2 + + ND 2 + + + 2 + ND + 2 ND ND + + ND ND + + + ND + + + ND + +
2 2 2 2 2 2 2 2 2 + + + 2 2 2 + + + + + + + + + + + + + ND + + 2 + + +
2 2 2 2 2 2 2 2 + + + + 2 2 2 + + 2 + ND ND 2 + ND + + 2 + + + + + ND 2 2
2 2 2 2 + 2 2 2 2 2 2 2 2 2 + 2 2 + + + + + + + + + + + + + + + + 2 +
2 2 2 2 2 + + 2 2 2 2 2 + + + + ND 2 + + + 2 2 2 2 2 + + 2 2 + 2 2 + 2
2 2 2 2 2 + + 2 ND 2 2 + + + + + ND 2 + ND ND 2 + ND + + + + ND + ND ND ND + +
F, female; M, male; ND, not done.
154
| www.amjdermatopathology.com
DISCUSSION This is one of the largest studies performed on immunophenotype of pcALCL, showing that profoundly aberrant phenotypes are more common than previously reported and that a conventional T-helper phenotype is a rare exception rather than the rule. In the WHO-EORTC and WHO classifications of lymphomas pcALCL is defined as a tumor composed of large cells with anaplastic, pleomorphic, or immunoblastic morphology expressing CD30, T-cell markers, CD4, and frequently cytotoxic proteins (granzyme B, TIA-1, perforin).1–3 Some cases (,5%) have a CD8+ T-cell phenotype.1–3,10 Variable loss of T-cell markers is common, but the exact frequency is unknown. In our study, we were able to identify 4 of 35 cases (11%) lacking simultaneously all pan–T-cell markers (CD2, CD3, and CD5), and 7 more (20%) positive only for CD2 and lacking CD3 and CD5. Of these 11 cases, 5 also lacked CD4 and CD8 expression. These data show that almost onethird of the cases of pcALCL present with profound phenotypic aberrations concerning pan-T-cell markers expression, thus representing pitfalls in the histopathologic diagnosis of this lymphoma. Although several studies reported on loss of T-cell markers expression, detailed phenotypic aberrations and loss of all T-cell markers were never described. In a small series of 7 cases, Plaza et al10 found positivity for CD3 in all cases, and CD2 and CD5 were not expressed in 2 cases only (28%). De Bruin et al previously reported negativity of CD2 in 2 of 14 (14%), CD3 in 5 of 13 (38%), and CD5 in 7 of 10 (70%) cases, respectively, whereas Beljaards et al described loss of CD2 expression in 5 of 34 (15%), CD3 in 9 of 34 (26%), and CD5 in 15 of 28 (54%) cases, respectively.13–15 In our experience, a CD4 helper phenotype was observed only in 12 of 35 (34%) cases. Five out of 8 cases that were stained for TIA-1 were positive for this marker, showing that cytotoxic proteins are expressed in the majority of cases with a T-helper phenotype. These data demonstrate that a TIA-1– negative T-helper phenotype is rare in pcALCL. Six out of 34 stained cases (18%) expressed CD8 and TIA-1 only, whereas 7 cases (20%) revealed a coexpression of CD4 and CD8 (with expression of TIA-1 in 4 of 4 tested cases). To our knowledge, Ó 2014 Lippincott Williams & Wilkins
Am J Dermatopathol Volume 36, Number 2, February 2014
Phenotypic Variability in pcALCL
FIGURE 1. Case 4. A, Anaplastic lymphoid tumor with positivity for (B) CD2 and (C) CD30, and negativity for (D) CD3, (E) CD5, (F) CD4, (G) CD8, and (H) TIA-1.
only 17 cases of CD8+ pcALCLs have been previously reported in the literature.8–10,14,16–20 Our study expands on previous experience, showing that a cytotoxic phenotype is probably more common than previously thought. Interestingly, 9 cases (26%) were negative for both CD4 and CD8. The existence of a T/null-cell phenotype is well known for the nodal ALCL and represents a very rare variant,21 but a double CD4/CD8 negativity has been reported only rarely in pcALCL and we were able to find only 10 similar cases in the literature.7,9,21–26 Only 1 of our CD42/ CD82 cases expressed TIA-1, demonstrating that most of these double-negative cases do not express a cytotoxic phenotype. The synthesis of our data shows that pcALCL can present with 4 different and similarly common profiles, namely, positivity for CD4 only, for CD8 only, for both CD4 and CD8, or negativity for both markers. Interestingly, irrespective of the CD4/CD8 phenotypic profile, cytotoxic granules are expressed in the majority of tested cases (16/ 26, 62%) and were found to be absent mostly in the T/nullcell phenotypic variant. In fact, 7 of 10 TIA-1-negative cases Ó 2014 Lippincott Williams & Wilkins
were characterized by double negativity for both CD4 and CD8, suggesting that lack of cytotoxic proteins is mostly restricted to this phenotypic variant. Although it may be that these cases start at the onset as either CD4+ or CD8+ and subsequently undergo loss of marker expression, the frequent positivity of TIA-1 in both helper and cytotoxic variants may suggest that the T/null-cell phenotype starts as such from the beginning. The presence of different immunophenotypic subtypes and frequent aberrations in pcALCLs, and the possible expression of CD30 by neoplastic lymphocytes of other lymphoma types, implies that several differential diagnoses should be considered when evaluating biopsy specimens of putative pcALCL. Expression of alpha/beta T-cell receptor (bF1) excludes primary cutaneous gamma/delta T-cell lymphoma, and negativity for Epstein–Barr virus allows to exclude extranodal NK/T-cell lymphoma, nasal type. Cutaneous small– medium pleomorphic CD4+ T-cell lymphoma does not express CD30 and is not composed of large anaplastic cells. Primary cutaneous peripheral T-cell lymphoma not otherwise specified www.amjdermatopathology.com |
155
Massone and Cerroni
Am J Dermatopathol Volume 36, Number 2, February 2014
FIGURE 2. Case 7. A, Large dermal tumor with (B) anaplastic cells expressing (C) CD30, (D) CD8, and (E) TIA-1, but being negative for (F) CD2, (G) CD3, (H) CD4, and (I) CD5.
characteristically shows loss of T-cell markers but usually does not express CD30 in the majority of neoplastic cells. Tumor-stage MF may show CD30 expression together with loss of CD2, CD3, and/or CD5. In contrast to MF, a pertinent history is missing in patients with pcALCLs, and lesions are solitary or localized rather than generalized. LyP type C presents with multiple chronic, recurrent, self-healing papulonodular lesions that show on histopathology a nodular infiltrate with anaplastic cell morphology. Although the phenotype may show similarities, the clinical history is diagnostic. Primary cutaneous aggressive epidermotropic CD8+ cytotoxic T-cell lymphoma is characterized by multiple ulcerated lesions and usually by negativity for CD30, and morphological features are different from those of pcALCL.1,3,27,28 In conclusion, our data show that many cases of pcALCL present with a phenotype that deviates from the conventional profile described in the literature. Simultaneous loss of several T-cell markers is a relatively frequent event. An accurate diagnosis is possible only with precise
156
| www.amjdermatopathology.com
morphologic and phenotypic studies and with clinicopathologic correlation. REFERENCES 1. Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas. Blood. 2005;105:3768–3785. 2. Ralfkiaer E, Willemze R, Paulli M, et al. Primary cutaneous anaplastic large cell lymphoma. In: Swerdlow SH, Campo E, Harris NE, et al. eds. WHO Classification of Tumours of Haemotopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008:300–301. 3. Cerroni L, Gatter K, Kerl H. An Illustrated Guide to Skin Lymphoma. 2nd ed. Malden, Oxford, Carlton: Blackwell Publishing; 2004. 4. Massone C, El-Shabrawi-Caelen L, Kerl H, et al. The morphologic spectrum of primary cutaneous anaplastic large T-cell lymphoma: a histopathologic study on 66 biopsy specimens from 47 patients with report of rare variants. J Cutan Pathol. 2008;35:46–53. 5. Burg G, Kempf W, Kazakov DV, et al. Pyogenic lymphoma of the skin: a peculiar variant of primary cutaneous neutrophil-rich CD30+ anaplastic large-cell lymphoma. Clinicopathological study of four cases and review of the literature. Br J Dermatol. 2003;148:580–586. 6. Kaudewitz P, Stein H, Dallenbach F, et al. Primary and secondary cutaneous Ki-1+ (CD30+) anaplastic large cell lymphomas. Morphologic,
Ó 2014 Lippincott Williams & Wilkins
Am J Dermatopathol Volume 36, Number 2, February 2014
Phenotypic Variability in pcALCL
immunohistologic, and clinical-characteristics. Am J Pathol. 1989;135: 359–367. Kummer JA, Vermeer MH, Dukers DF, et al. Most primary cutaneous CD30-positive lymphoproliferative disorders have a CD4-positive cytotoxic T-cell phenotype. J Invest Dermatol. 1997;109:636–640. Krenacs L, Wellmann A, Sorbara L, et al. Cytotoxic cell antigen expression in anaplastic large cell lymphomas of T- and null-cell type and Hodgkin’s disease: evidence for distinct cellular origin. Blood. 1997; 89:980–989. Boulland ML, Wechsler J, Bagot M, et al. Primary CD30-positive cutaneous T-cell lymphomas and lymphomatoid papulosis frequently express cytotoxic proteins. Histopathology. 2000;36:136–144. Plaza JA, Feldman AL, Magro C. Cutaneous CD30-positive lymphoproliferative disorders with CD8 expression: a clinicopathologic study of 21 cases. J Cutan Pathol. 2013;40:236–247. Massone C, Lozzi GP, Egberts F, et al. The protean spectrum of nonHodgkin lymphomas with prominent involvement of subcutaneous fat. J Cutan Pathol. 2006;33:418–425. Cerroni L, Smolle J, Soyer HP, et al. Immunophenotyping of cutaneous lymphoid infiltrates in frozen and paraffin-embedded tissue sections: a comparative study. J Am Acad Dermatol. 1990;22:405–413. Beljaards RC, Kaudewitz P, Berti E, et al. Primary cutaneous CD30positive large cell lymphoma: definition of a new type of cutaneous lymphoma with a favorable prognosis. A European Multicenter Study of 47 patients. Cancer. 1993;71:2097–2104. Beljaards RC, Meijer CJ, Scheffer E, et al. Prognostic significance of CD30 (Ki-1/Ber-H2) expression in primary cutaneous large-cell lymphomas of T-cell origin. A clinicopathologic and immunohistochemical study in 20 patients. Am J Pathol. 1989;135:1169–1178. de Bruin PC, Beljaards RC, van Heerde P, et al. Differences in clinical behaviour and immunophenotype between primary cutaneous and primary nodal anaplastic large cell lymphoma of T-cell or null cell phenotype. Histopathology. 1993;23:127–135. Kikuchi A, Sakuraoka K, Kurihara S, et al. CD8+ cutaneous anaplastic large-cell lymphoma: report of two cases with immunophenotyping, T-cell-receptor gene rearrangement and electron microscopic studies. Br J Dermatol. 1992;126:404–408. Fukunaga M, Masaki T, Ichihashi M, et al. CD8-positive primary cutaneous anaplastic large cell lymphoma with a fair prognosis. Acta Derm Venereol. 2002;82:312–314.
18. Berti E, Gianotti R, Alessi E. Primary anaplastic large cell lymphoma of the skin. Dermatologica. 1989;178:225–227. 19. Shimizu Y, Tanae K, Takahashi N, et al. Primary cutaneous anaplastic large-cell lymphoma presenting with hemophagocytic syndrome: a case report and review of the literature. Leuk Res. 2010;34:263–266. 20. Boudova L, Kazakov DV, Jindra P, et al. Primary cutaneous histiocyte and neutrophil-rich CD30+ and CD56+ anaplastic large-cell lymphoma with prominent angioinvasion and nerve involvement in the forehead and scalp of an immunocompetent woman. J Cutan Pathol. 2006;33:584–589. 21. Delsol G, Falini B, Müller-Hermelink HK. Anaplastic large cell lymphoma (ALCL), ALK positive. In: Swerdlow SH, Campo E, Harris NE, et al. eds. WHO Classification of Tumours of Haemotopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008:312–316. 22. Akiyama C, Shibagaki N, Yasaka N, et al. Primary cutaneous CD30(Ki1)-positive lymphoma of non-T, non-B origin. Dermatology. 1995;190: 238–241. 23. Stein H, Foss HD, Dürkop H, et al. CD30+ anaplastic large cell lymphoma: a review of its histopathologic, genetic, and clinical features. Blood. 2000;96:3681–3695. 24. Booken N, Goerdt S, Klemke CD. Clinical spectrum of primary cutaneous CD30-positive anaplastic large cell lymphoma: an analysis of the Mannheim Cutaneous Lymphoma Registry. J Dtsch Dermatol Ges. 2012;10:331–339. 25. Kong YY, Dai B, Kong JC, et al. 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. 26. Paulli M, Berti E, Rosso R, et al. CD30/Ki-1-positive lymphoproliferative disorders of the skin—clinicopathologic correlation and statistical analysis of 86 cases: a multicentric study from the European Organization for Research and Treatment of Cancer Cutaneous Lymphoma Project Group. J Clin Oncol. 1995;13:1343–1354. 27. Berti E, Tomasini D, Vermeer MH, et al. Primary cutaneous CD8-positive epidermotropic cytotoxic T cell lymphomas. A distinct clinicopathological entity with an aggressive clinical behavior. Am J Pathol. 1999; 155:483–492. 28. Massone C, Chott A, Metze D, et al. Subcutaneous, blastic natural killer (NK), NK/T-cell, and other cytotoxic lymphomas of the skin: a morphologic, immunophenotypic, and molecular study of 50 patients. Am J Surg Pathol. 2004;28:719–735.
7. 8.
9. 10. 11. 12. 13.
14.
15.
16.
17.
Ó 2014 Lippincott Williams & Wilkins
www.amjdermatopathology.com |
157
Phenotypic Variability in Primary Cutaneous Anaplastic Large T-cell Lymphoma: A Study on 35 Patients Cesare Massone, MD and Lorenzo Cerroni, MD
Abstract: Primary cutaneous anaplastic large T-cell lymphoma (pcALCL) is a well-defined entity characterized by neoplastic cells expressing CD30, CD2, CD3, CD4, and CD5. Cases with different phenotype have been reported, including variable loss of CD2, CD3, and CD5, and expression of cytotoxic phenotype (CD8+) and/or of cytotoxic proteins. Aberrant phenotypes represent a diagnostic pitfall and may be the cause of misdiagnoses. We reviewed 35 cases of pcALCL (M:F = 19:16; mean age, 50.8 years; range, 14–92 years), to better characterize the immunophenotypic spectrum of the disease. Twelve cases (34%) had a T-helper phenotype (CD4+/CD82), and TIA-1 was positive in 5 of 8 stained cases. Six cases (18%) had a T-cytotoxic phenotype (CD42/CD8+) and were also positive for TIA-1. Positivity for both CD4 and CD8 was observed in 7 cases (20%), 4 of which were stained for TIA-1 and found to be positive, whereas both CD4 and CD8 were negative in 9 cases (26%, only 1/8 tested cases being TIA-1 positive). CD2 was positive in 21 of 27 tested cases (78%), CD3 in 21 of 34 cases (62%), and CD5 in 15 of 31 cases (48%). Interestingly, 11 cases (31%) showed a profoundly aberrant phenotype lacking simultaneously several T-cell markers. Our data allow a better characterization of pcALCL with aberrant phenotypes, showing the remarkable variability in expression of different markers. Key Words: primary cutaneous anaplastic large T-cell lymphoma, CD30+ cutaneous lymphoproliferative disorders, cutaneous T-cell lymphoma, immunophenotype (Am J Dermatopathol 2014;36:153–157)
INTRODUCTION Primary cutaneous anaplastic large T-cell lymphoma (pcALCL) is a well-defined entity with prognostic and biological differences from the nodal counterpart. It is usually characterized by large cells with anaplastic, pleomorphic, or immunoblastic morphology, but several histopathologic variants have been described.1–3 Neutrophils (“inflammatory-type” pattern), lymphocytes (“lymphohistiocytic” pattern), or even eosinophils can predominate in the infiltrate, and cases with small- to medium-sized cells (“small-cell” pattern) or signetring cell morphology have also been reported.4,5 From the Research Unit Dermatopathology, Department of Dermatology, Medical University of Graz, Austria. The authors declare no conflicts of interest. Reprints: Lorenzo Cerroni, MD, Department of Dermatology, Medical University of Graz, Auenbruggerplatz 8, Graz A-8036, Austria (e-mail: [email protected]). © 2014 Lippincott Williams & Wilkins
Am J Dermatopathol Volume 36, Number 2, February 2014
Neoplastic cells in conventional cases of pcALCL express CD30 and CD2, CD3, CD4, and CD5 (T-helper phenotype), and are negative for CD8, CD56, and cytotoxic proteins.1–3 Cases with different phenotype have been reported, including variable loss of the T-cell markers CD2, CD3, and CD5, and expression of cytotoxic phenotype (CD8+) and/or of cytotoxic proteins (granzyme B, TIA-1, perforin).1–3,6–10 Aberrant phenotypes represent a diagnostic pitfall and may be the cause of misdiagnoses. Although pcALCL is traditionally considered as a lymphoma of T-helper lymphocytes, the exact proportion of cases with aberrant phenotype is unknown. In this study, we analyzed the clinicopathologic and immunophenotypic features of 35 patients with pcALCL, to better characterize the immunophenotypic spectrum of this cutaneous lymphoma.
MATERIALS AND METHODS From a total of 114 cases of pcALCL documented in the lymphoma database of the Research Unit of Dermatopathology, Department of Dermatology, Medical University of Graz (Graz, Austria), 35 cases were selected for the study. The other 79 cases were excluded because of lack of complete data or of available material for further stainings. Patients with history of mycosis fungoides (MF) or lymphomatoid papulosis (LyP) and “borderline cases” of CD30+ cutaneous lymphoproliferative disorder (cases in which a definitive distinction between pcALCL and LyP could not be made) and cases of cutaneous Hodgkin disease have been excluded from the study. Diagnosis of pcALCL has been made according to the WHO classification of lymphomas.1,2 Primary skin involvement was defined as the presence of cutaneous lymphoma without nodal and/or visceral involvement after complete staging procedures.1,2 Details of some of the cases have been published previously.4,11
Histology
Biopsy specimens were fixed in 10% buffered formalin and subsequently embedded in paraffin. Sections were stained with hematoxylin–eosin for routine histopathologic evaluation. All cases and stainings have been reviewed independently by both the authors.
Immunohistology Detailed immunophenotypic analyses were performed on routinely fixed paraffin-embedded tissue sections according to a previously described 3-step immunoperoxidase method. Microwave enhancement was used for all the antibodies. Second and third antibodies were obtained from Dako (Dakopatts; Glostrup, Denmark). The first antibody were from www.amjdermatopathology.com |
153
Massone and Cerroni
Am J Dermatopathol Volume 36, Number 2, February 2014
Novocastra-Leica Microsystems (Wetzlar, Germany; CD2, CD3, CD4, CD5), Dako (Dakopatts, CD8, CD30, ALK-1), and Immunotech (Prague, Czech Republic; TIA-1).12
phenotype (CD42/CD8+) and were also positive for TIA-1. Positivity for both CD4 and CD8 was observed in 7 cases (20%), 4 of which were stained for TIA-1 and found to be positive. In 9 cases, both CD4 and CD8 were negative (26%), with only 1 of 8 cases stained for TIA-1 being positive in this group. Analysis of the phenotypic pattern in 24 cases with complete stainings revealed many cases with profound aberrations. In addition, it is striking that no case showed a “conventional” T-helper phenotype without loss of pan–T-cell markers and without expression of cytotoxic proteins. Five cases were positive for CD2 only and negative for all other T-cell markers and for cytotoxic proteins (Fig. 1). Four cases lacked expression of all tested pan–T-cell markers and were positive only for CD8 and TIA-1 (2 cases) (Fig. 2), CD4 (1 case), or CD4, CD8, and TIA-1 (1 case).
RESULTS Patients were 19 males and 16 females. The median age was 59 years (mean age: 50.8 years; range 14–92 years). All cases revealed strong expression of CD30 and were negative for ALK-1. No case was associated with HIV infection or iatrogenic immunosuppression. Immunohistochemical data are reported in Table 1. CD2 was positive in 21 of 27 (78%), CD3 in 21 of 34 (62%), and CD5 in 15 of 31 (48%) tested cases. Twelve cases (34%) had a T-helper phenotype (CD4+/CD82), and TIA-1 was positive in 5 of 8 stained cases. Six cases (18%) had a T-cytotoxic
TABLE 1. Clinical Data and Immunophenotype of Our Series of Patients Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Sex
Age
CD2
CD3
CD5
CD4
CD8
TIA-1
M F M M M F F M F F M F F M M M F F M M F M M M M M M F F F F M M F F
82 58 70 80 92 54 14 60 59 40 78 62 16 67 58 48 69 46 43 18 57 68 83 55 60 72 82 61 64 82 77 66 14 71 35
+ + + + 2 2 2 + + ND 2 + + + 2 + ND + 2 ND ND + + ND ND + + + ND + + + ND + +
2 2 2 2 2 2 2 2 2 + + + 2 2 2 + + + + + + + + + + + + + ND + + 2 + + +
2 2 2 2 2 2 2 2 + + + + 2 2 2 + + 2 + ND ND 2 + ND + + 2 + + + + + ND 2 2
2 2 2 2 + 2 2 2 2 2 2 2 2 2 + 2 2 + + + + + + + + + + + + + + + + 2 +
2 2 2 2 2 + + 2 2 2 2 2 + + + + ND 2 + + + 2 2 2 2 2 + + 2 2 + 2 2 + 2
2 2 2 2 2 + + 2 ND 2 2 + + + + + ND 2 + ND ND 2 + ND + + + + ND + ND ND ND + +
F, female; M, male; ND, not done.
154
| www.amjdermatopathology.com
DISCUSSION This is one of the largest studies performed on immunophenotype of pcALCL, showing that profoundly aberrant phenotypes are more common than previously reported and that a conventional T-helper phenotype is a rare exception rather than the rule. In the WHO-EORTC and WHO classifications of lymphomas pcALCL is defined as a tumor composed of large cells with anaplastic, pleomorphic, or immunoblastic morphology expressing CD30, T-cell markers, CD4, and frequently cytotoxic proteins (granzyme B, TIA-1, perforin).1–3 Some cases (,5%) have a CD8+ T-cell phenotype.1–3,10 Variable loss of T-cell markers is common, but the exact frequency is unknown. In our study, we were able to identify 4 of 35 cases (11%) lacking simultaneously all pan–T-cell markers (CD2, CD3, and CD5), and 7 more (20%) positive only for CD2 and lacking CD3 and CD5. Of these 11 cases, 5 also lacked CD4 and CD8 expression. These data show that almost onethird of the cases of pcALCL present with profound phenotypic aberrations concerning pan-T-cell markers expression, thus representing pitfalls in the histopathologic diagnosis of this lymphoma. Although several studies reported on loss of T-cell markers expression, detailed phenotypic aberrations and loss of all T-cell markers were never described. In a small series of 7 cases, Plaza et al10 found positivity for CD3 in all cases, and CD2 and CD5 were not expressed in 2 cases only (28%). De Bruin et al previously reported negativity of CD2 in 2 of 14 (14%), CD3 in 5 of 13 (38%), and CD5 in 7 of 10 (70%) cases, respectively, whereas Beljaards et al described loss of CD2 expression in 5 of 34 (15%), CD3 in 9 of 34 (26%), and CD5 in 15 of 28 (54%) cases, respectively.13–15 In our experience, a CD4 helper phenotype was observed only in 12 of 35 (34%) cases. Five out of 8 cases that were stained for TIA-1 were positive for this marker, showing that cytotoxic proteins are expressed in the majority of cases with a T-helper phenotype. These data demonstrate that a TIA-1– negative T-helper phenotype is rare in pcALCL. Six out of 34 stained cases (18%) expressed CD8 and TIA-1 only, whereas 7 cases (20%) revealed a coexpression of CD4 and CD8 (with expression of TIA-1 in 4 of 4 tested cases). To our knowledge, Ó 2014 Lippincott Williams & Wilkins
Am J Dermatopathol Volume 36, Number 2, February 2014
Phenotypic Variability in pcALCL
FIGURE 1. Case 4. A, Anaplastic lymphoid tumor with positivity for (B) CD2 and (C) CD30, and negativity for (D) CD3, (E) CD5, (F) CD4, (G) CD8, and (H) TIA-1.
only 17 cases of CD8+ pcALCLs have been previously reported in the literature.8–10,14,16–20 Our study expands on previous experience, showing that a cytotoxic phenotype is probably more common than previously thought. Interestingly, 9 cases (26%) were negative for both CD4 and CD8. The existence of a T/null-cell phenotype is well known for the nodal ALCL and represents a very rare variant,21 but a double CD4/CD8 negativity has been reported only rarely in pcALCL and we were able to find only 10 similar cases in the literature.7,9,21–26 Only 1 of our CD42/ CD82 cases expressed TIA-1, demonstrating that most of these double-negative cases do not express a cytotoxic phenotype. The synthesis of our data shows that pcALCL can present with 4 different and similarly common profiles, namely, positivity for CD4 only, for CD8 only, for both CD4 and CD8, or negativity for both markers. Interestingly, irrespective of the CD4/CD8 phenotypic profile, cytotoxic granules are expressed in the majority of tested cases (16/ 26, 62%) and were found to be absent mostly in the T/nullcell phenotypic variant. In fact, 7 of 10 TIA-1-negative cases Ó 2014 Lippincott Williams & Wilkins
were characterized by double negativity for both CD4 and CD8, suggesting that lack of cytotoxic proteins is mostly restricted to this phenotypic variant. Although it may be that these cases start at the onset as either CD4+ or CD8+ and subsequently undergo loss of marker expression, the frequent positivity of TIA-1 in both helper and cytotoxic variants may suggest that the T/null-cell phenotype starts as such from the beginning. The presence of different immunophenotypic subtypes and frequent aberrations in pcALCLs, and the possible expression of CD30 by neoplastic lymphocytes of other lymphoma types, implies that several differential diagnoses should be considered when evaluating biopsy specimens of putative pcALCL. Expression of alpha/beta T-cell receptor (bF1) excludes primary cutaneous gamma/delta T-cell lymphoma, and negativity for Epstein–Barr virus allows to exclude extranodal NK/T-cell lymphoma, nasal type. Cutaneous small– medium pleomorphic CD4+ T-cell lymphoma does not express CD30 and is not composed of large anaplastic cells. Primary cutaneous peripheral T-cell lymphoma not otherwise specified www.amjdermatopathology.com |
155
Massone and Cerroni
Am J Dermatopathol Volume 36, Number 2, February 2014
FIGURE 2. Case 7. A, Large dermal tumor with (B) anaplastic cells expressing (C) CD30, (D) CD8, and (E) TIA-1, but being negative for (F) CD2, (G) CD3, (H) CD4, and (I) CD5.
characteristically shows loss of T-cell markers but usually does not express CD30 in the majority of neoplastic cells. Tumor-stage MF may show CD30 expression together with loss of CD2, CD3, and/or CD5. In contrast to MF, a pertinent history is missing in patients with pcALCLs, and lesions are solitary or localized rather than generalized. LyP type C presents with multiple chronic, recurrent, self-healing papulonodular lesions that show on histopathology a nodular infiltrate with anaplastic cell morphology. Although the phenotype may show similarities, the clinical history is diagnostic. Primary cutaneous aggressive epidermotropic CD8+ cytotoxic T-cell lymphoma is characterized by multiple ulcerated lesions and usually by negativity for CD30, and morphological features are different from those of pcALCL.1,3,27,28 In conclusion, our data show that many cases of pcALCL present with a phenotype that deviates from the conventional profile described in the literature. Simultaneous loss of several T-cell markers is a relatively frequent event. An accurate diagnosis is possible only with precise
156
| www.amjdermatopathology.com
morphologic and phenotypic studies and with clinicopathologic correlation. REFERENCES 1. Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas. Blood. 2005;105:3768–3785. 2. Ralfkiaer E, Willemze R, Paulli M, et al. Primary cutaneous anaplastic large cell lymphoma. In: Swerdlow SH, Campo E, Harris NE, et al. eds. WHO Classification of Tumours of Haemotopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008:300–301. 3. Cerroni L, Gatter K, Kerl H. An Illustrated Guide to Skin Lymphoma. 2nd ed. Malden, Oxford, Carlton: Blackwell Publishing; 2004. 4. Massone C, El-Shabrawi-Caelen L, Kerl H, et al. The morphologic spectrum of primary cutaneous anaplastic large T-cell lymphoma: a histopathologic study on 66 biopsy specimens from 47 patients with report of rare variants. J Cutan Pathol. 2008;35:46–53. 5. Burg G, Kempf W, Kazakov DV, et al. Pyogenic lymphoma of the skin: a peculiar variant of primary cutaneous neutrophil-rich CD30+ anaplastic large-cell lymphoma. Clinicopathological study of four cases and review of the literature. Br J Dermatol. 2003;148:580–586. 6. Kaudewitz P, Stein H, Dallenbach F, et al. Primary and secondary cutaneous Ki-1+ (CD30+) anaplastic large cell lymphomas. Morphologic,
Ó 2014 Lippincott Williams & Wilkins
Am J Dermatopathol Volume 36, Number 2, February 2014
Phenotypic Variability in pcALCL
immunohistologic, and clinical-characteristics. Am J Pathol. 1989;135: 359–367. Kummer JA, Vermeer MH, Dukers DF, et al. Most primary cutaneous CD30-positive lymphoproliferative disorders have a CD4-positive cytotoxic T-cell phenotype. J Invest Dermatol. 1997;109:636–640. Krenacs L, Wellmann A, Sorbara L, et al. Cytotoxic cell antigen expression in anaplastic large cell lymphomas of T- and null-cell type and Hodgkin’s disease: evidence for distinct cellular origin. Blood. 1997; 89:980–989. Boulland ML, Wechsler J, Bagot M, et al. Primary CD30-positive cutaneous T-cell lymphomas and lymphomatoid papulosis frequently express cytotoxic proteins. Histopathology. 2000;36:136–144. Plaza JA, Feldman AL, Magro C. Cutaneous CD30-positive lymphoproliferative disorders with CD8 expression: a clinicopathologic study of 21 cases. J Cutan Pathol. 2013;40:236–247. Massone C, Lozzi GP, Egberts F, et al. The protean spectrum of nonHodgkin lymphomas with prominent involvement of subcutaneous fat. J Cutan Pathol. 2006;33:418–425. Cerroni L, Smolle J, Soyer HP, et al. Immunophenotyping of cutaneous lymphoid infiltrates in frozen and paraffin-embedded tissue sections: a comparative study. J Am Acad Dermatol. 1990;22:405–413. Beljaards RC, Kaudewitz P, Berti E, et al. Primary cutaneous CD30positive large cell lymphoma: definition of a new type of cutaneous lymphoma with a favorable prognosis. A European Multicenter Study of 47 patients. Cancer. 1993;71:2097–2104. Beljaards RC, Meijer CJ, Scheffer E, et al. Prognostic significance of CD30 (Ki-1/Ber-H2) expression in primary cutaneous large-cell lymphomas of T-cell origin. A clinicopathologic and immunohistochemical study in 20 patients. Am J Pathol. 1989;135:1169–1178. de Bruin PC, Beljaards RC, van Heerde P, et al. Differences in clinical behaviour and immunophenotype between primary cutaneous and primary nodal anaplastic large cell lymphoma of T-cell or null cell phenotype. Histopathology. 1993;23:127–135. Kikuchi A, Sakuraoka K, Kurihara S, et al. CD8+ cutaneous anaplastic large-cell lymphoma: report of two cases with immunophenotyping, T-cell-receptor gene rearrangement and electron microscopic studies. Br J Dermatol. 1992;126:404–408. Fukunaga M, Masaki T, Ichihashi M, et al. CD8-positive primary cutaneous anaplastic large cell lymphoma with a fair prognosis. Acta Derm Venereol. 2002;82:312–314.
18. Berti E, Gianotti R, Alessi E. Primary anaplastic large cell lymphoma of the skin. Dermatologica. 1989;178:225–227. 19. Shimizu Y, Tanae K, Takahashi N, et al. Primary cutaneous anaplastic large-cell lymphoma presenting with hemophagocytic syndrome: a case report and review of the literature. Leuk Res. 2010;34:263–266. 20. Boudova L, Kazakov DV, Jindra P, et al. Primary cutaneous histiocyte and neutrophil-rich CD30+ and CD56+ anaplastic large-cell lymphoma with prominent angioinvasion and nerve involvement in the forehead and scalp of an immunocompetent woman. J Cutan Pathol. 2006;33:584–589. 21. Delsol G, Falini B, Müller-Hermelink HK. Anaplastic large cell lymphoma (ALCL), ALK positive. In: Swerdlow SH, Campo E, Harris NE, et al. eds. WHO Classification of Tumours of Haemotopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008:312–316. 22. Akiyama C, Shibagaki N, Yasaka N, et al. Primary cutaneous CD30(Ki1)-positive lymphoma of non-T, non-B origin. Dermatology. 1995;190: 238–241. 23. Stein H, Foss HD, Dürkop H, et al. CD30+ anaplastic large cell lymphoma: a review of its histopathologic, genetic, and clinical features. Blood. 2000;96:3681–3695. 24. Booken N, Goerdt S, Klemke CD. Clinical spectrum of primary cutaneous CD30-positive anaplastic large cell lymphoma: an analysis of the Mannheim Cutaneous Lymphoma Registry. J Dtsch Dermatol Ges. 2012;10:331–339. 25. Kong YY, Dai B, Kong JC, et al. 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. 26. Paulli M, Berti E, Rosso R, et al. CD30/Ki-1-positive lymphoproliferative disorders of the skin—clinicopathologic correlation and statistical analysis of 86 cases: a multicentric study from the European Organization for Research and Treatment of Cancer Cutaneous Lymphoma Project Group. J Clin Oncol. 1995;13:1343–1354. 27. Berti E, Tomasini D, Vermeer MH, et al. Primary cutaneous CD8-positive epidermotropic cytotoxic T cell lymphomas. A distinct clinicopathological entity with an aggressive clinical behavior. Am J Pathol. 1999; 155:483–492. 28. Massone C, Chott A, Metze D, et al. Subcutaneous, blastic natural killer (NK), NK/T-cell, and other cytotoxic lymphomas of the skin: a morphologic, immunophenotypic, and molecular study of 50 patients. Am J Surg Pathol. 2004;28:719–735.
7. 8.
9. 10. 11. 12. 13.
14.
15.
16.
17.
Ó 2014 Lippincott Williams & Wilkins
www.amjdermatopathology.com |
157
