 |
|||||
|
|||||
CD4+ CD56+ blastic tumor of the skin: cytogenetic observations and further evidence of an origin from plasmocytoid dendritic cells |
|||||
|
European Journal of Dermatology. Volume 14, Number 5, 317-22, September-October 2004, Investigative report |
|||||
 Summary
|
|||||
Author(s) : C. HALLERMANN, P. MIDDEL, F. GRIESINGER, B. GUNAWAN, H.P. BERTSCH, C. NEUMANN , Department of Dermatology, University Hospital Goettingen, Germany, Department of Pathology, University Hospital Goettingen, Germany, Department of Hematology, University Hospital Goettingen, Germany, Christian Hallermann, Fax: (+49)-551-396841. E-mail: Challermann@gmx.de. | |||||
Summary : We here report on a case of a blastic tumor, recently described to belong to a new entity sharing phenotypic similarities with blood derived plasmocytoid dendritic cells and formerly regarded as belonging to the group of natural killer cell lymphomas. Besides immunophenotypic characteristics such as the absence of T- cell markers and almost complete absence of markers of the myeloid lineage, these tumors express CD4, CD56 and CD123, the receptor for interleukin-3. Moreover, using the comparative genomic hybridisation technique, CGH, we demonstrate a gain of chromosome 7q, 22 and a loss of chromosome 3p and 13q. Since this type of hematologic disorder often shows its primary manifestation in the skin and often runs a rapidly fatal course, it is important to distinguish this from other types of primary cutaneous lymphomas. | |||||
Keywords : CD4, CD56, CGH, plasmocytoid dendritic cells | |||||
 Pictures |
|||||
ARTICLE
Auteur(s) :, C. HALLERMANN1,*, P. MIDDEL2, F. GRIESINGER3, B. GUNAWAN2, H.P. BERTSCH1, C. NEUMANN1 1Department of Dermatology, University Hospital
Goettingen, Germany accepté le 17 Mai 2004 IntroductionIn recent years several articles have described a new group of CD4+CD56+ blastic tumors (BT) with a high frequency of skin involvement. In a considerable number of patients the skin is the first organ to become clinically involved [1-5]. The disease consistently runs an aggressive course. Most patients die within the first year after the diagnosis was established.CD4+ CD56+ BT has been classified by the new WHO classification as blastic natural killer (NK)-cell lymphoma [4]. However, the origin of the malignant cells which are devoid of T-, B-, NK- as well as almost any myeloid lineage markers was a matter of debate until recently. Consistently, the malignancy is of germline type for IgH and TCR rearrangement. In contrast to classical NK-cell lymphoma, it is not associated with EBV-infection [3]. Recently, it was suggested that CD4+ CD56+ BT originate from a distinct type of dendritic blood derived cell, the plasmocytoid dendritic cell (PC-DC). Chaperot et al. [7] demonstrated functional similarities between PC-DC and leukemic CD4+ CD56+ BT cells, both lacking markers of NK- and myeloid cells. Moreover, CD123, the IL3 receptor alpha chain, which is a characteristic feature of PC-DC was demonstrated to be expressed by malignant CD4+ CD56+ BT cells, further supporting this relationship [5]. Since the disease most often primarily manifests itself in the skin, the dermatologist has to be familiar with its diagnosis. For an accurate diagnosis, besides clinical criteria, morphological, immunophenotypic and molecular features have to be taken into account. Here we present another case with typical clinical and morphological features as well as results from molecular and cytogenetic investigations of the patient’s tumor. MethodsStaining and ImmunohistochemistryA tumor biopsy was fixed in 4% buffered formalin, embedded and cut into 2-5 μm sections. Besides hematoxylin/eosin and Giemsa staining, immunohistochemistry was performed using standard immunoperoxidase techniques. For immunohistochemistry the sections were dewaxed in xylol (Merck, Darmstadt, Germany), rehydrated in serial dilutions of ethanol, and stained with antibodies to CD4 (clone 1F6, Novocastra, Newcastle upon Tyne, UK), CD56 (NKH-1, Biogenex San Ramo, USA), CD123 (clone 6H6, eBioscience, San Diego, USA) BDCA-2 (clone AC144, Miltenyi Biotec, Bergisch-Gladbach, Germany); CD1a (clone 010), Tia1 (clone 2G9A10F5, both Immunotech, Wildwood, Canada); CD2 (clone MT910), CD3 (clone F7.2.38), CD5 (clone CD5/54/F6 ), CD7 (clone DK24,), CD8 (clone C8/144B), CD19 (clone HD37), CD20 (clone L26), CD30 (clone Ber-H2), CD79a ( clone HM57), CD68 (clone PG-M1), CD43 (clone DF-T1), LCA (clone 2BM+PD7/26), S100 (clone S-100), Granzyme B (clone GrB-7), EBV (clone CS1/4), CD15 (clone C3D1), bcl-2 (clone 124)(all DAKO, Glostrup, Denmark); CD34 (clone My10, Becton-Dickinson, Heidelberg, Germany); KiM1P (clone KiM1P, Kiel, Germany), CD117(Code A4502, Dako), TCR alpha/beta-FITC (clone 8A3, Acris-antibodies, Hiddenhausen, Germany), TCR gamma/delta (clone Immuno/510, Acris-antibodies), The NexES IHC immunostaining device (Ventana Medical Systems, Strasbourg, France) was used in combination with an alkaline phosphatase anti-alkaline phosphatase technique (APAAP) and Fast Red as substrate according to the manufacturer’s device. The biopsy was also stained with myeloperoxidase, chloracetatesterase and lysozyme using standard techniques.Analysis of the T-cell receptor gamma chain (TCR gamma) rearrangementThe analysis of the TCR-gamma rearrangement revealed germline configuration. DNA was isolated from cryosections and TCR-gamma PCR was performed according to the protocol described by Kneba et al. 1994. The protocol allows the amplification of nearly all possible V-gamma/J-gamma recombinations [8]. For the fluorescence-based analysis of TCRgamma I and II fragmnets an ABI 310 DNA Analyser (Perkin-Elmer, Germany) was used. In addition, an internal size standard GENESCANTM, TAMRA (Perkin-Elmer, Germany) was added to the mixture. After the run, peak patterns were detected with GENESCAN software.Three-colour flow cytometryBone marrow or peripheral blood of the patient was subjected to immunophenotyping at five instances during the time course of the disease. Preparation of cells and immunophenotyping was performed as previously described [9] using 3 color staining with a FACScan (Becton Dickinson). Data were analysed using the Paint-a-Gate Programme.Comparative genomic hybridisationExtraction of DNA was performed from 20 μm thick tissue sections (n=10) of a snap frozen skin biopsy using an tissue-DNA extraction kit according to the manufacturer’s instructions (Qiagen, Hilden, Germany). Comparative genomic hybridisation (CGH) was performed as previously described [10] and was slightly modified [11]. For labeling, tumour DNA was nick-translated using biotin-16-dUTPs (Roche, Mannheim, Germany). Digoxigenin-11-dUTP-labeled normal reference DNA (Roche, Mannheim, Germany) was used for co-hybridisation to slides with metaphases from blood of a female healthy donor. After 3 days of hybridisation at 37°C, post-hybridisation washes were performed and biotin-labeled tumour DNA was detected using avidin-fluorescein isothiocyanate (FITC). Reference DNA was detected by anti-digoxigenin-rhodamin. For counterstaining 4,6-diamidino-2-phenylindole (DAPI) was used. Digital image acquisition and image analysis (Applied Imaging, Newcastle, UK) were performed as described elsewhere [10].ResultsClinical findingsAt the age of 75 an otherwise healthy woman presented with rapidly evolving skin lesions of 6 weeks duration. B-symptoms were absent. Predominantly on the trunk there were disseminated reddish, non scaling plaques and small palpable nodules located deeply in the skin (( Fig. 1 )). Within a few weeks the nodules enlarged and achieved a bluish and partly purpuric aspect. A complete staging showed no evidence for an extracutaneous manifestation of the lymphoma. Specifically, ultrasound of lymph nodes and internal organs, computerised tomography of chest and abdomen, bone marrow biopsy, extensive immunophenotyping of white blood cells (among others the CD56 marker) yielded normal results. So did the physical examinations including those of the throat and the nose.No antibodies to borrelia burgdorferi, HIV und CMV were detectable. Serology was without evidence for active or persistent EBV- disease. A tentative diagnosis of an unclassified primary cutaneous lymphoma, possibly of NK-type was made and extracorporal photopheresis was started. Ten weeks later, the patient developed erythrocytopenia and thrombopenia. Bone marrow biopsy now revealed an infiltration by the CD56 positive lymphoma. The patient died of progressive disease while receiving CHOP polychemotherapy. Histology, immunohistochemistry and TCR analysis of lesional skin and bloodThere was a nodular infiltrate in the dermis, sparing the epidermis. It was composed of medium sized monomorphic lymphoid cells. Large granular cells, as described in NK-cell lymphoma, were absent (( Fig. 2 a, d, e )).The tumor cells stained positive for LCA, CD4, CD43, CD56 (( Fig. 2b )), CD123 (bright), BDCA-2 (weak). They stained negative for CD1a, CD2, CD3 (( Fig. 2c )), CD5, CD7, CD8, CD19, CD20, CD30, CD34, CD68, KiM1P, CD79a, CD117, TCR alpha/beta, TCR gamma/delta, S100, Tia1, granzyme-B, bcl-2, EBV and the myeloid marker CD15. Moreover, the tumor cells stained negative for myeloperoxidase, chloroacetatesterase and lysozyme which allowed a definitive distinction from myeloic leukemia. The analysis of the TCRgamma rearrangement revealed germline configuration Flow-cytometry using three color immunophenotyping of peripheral blood and bone marrowStaining of bone marrow and peripheral blood cells at diagnosis of systemic disease revealed the following immunophenotype: cells coexpressed CD4 and high levels of CD56 (( Fig. 3 a-d )). They also were positive for CD38 and HLA-DR. No expression of the following differentiation antigens was observed: T-cell antigens CD2, CD3, CD5, CD7, CD8, TCR gamma/delta and TCR alpha/beta, as well as CD34 and CD11c.Comparative genomic hybridisationCGH analysis revealed a gain of chromosomes 7q and 22 as well as a loss of chromosome 3p21.3-26 and 13q (( Fig. 4 )).DiscussionInitially considered to be of natural killer cell origin, a new classification was proposed for CD4+CD56+CD3- lymphomas based on a study of seven cases with distinct features.3 The tumors primarily involved the skin and were composed of small blasts forming a nodular non-angiogenic infiltrate in the dermis down to the subcutis. Distinct cytoplasmic granules were absent. The immunophenotype was CD4+, CD56+, CD2-, CD3-, CD5-, CD7-,CD34-,CD38+/-,CD43+,CD68+/-, TIA-1-, granzyme B-, thus being incompatible with T-, B- and NK-cells. Configuration of the TCR- and IgH gene was germline. In contrast to true NK lymphoma, there was no evidence for EBV infection. In 6/7 patients the disease took a fatal course within 5-27 months. The features described meet well with the case presented in the present paper.More recently further studies have supported the view that cutaneous and extracutaneous CD4+CD56+, CD3- neoplasms are a distinct lymphoma entity [1, 12]. Unlike classical NK-lymphomas even though expressing the NK-marker CD56, BT cells do not express killer cell receptors [13]. Phenotypic similarities of CD4+,CD56+ BT cells with blood- derived plasmocytoid dendritic cells were described, both expressing CD45RA and CD123 (IL-3 receptor alpha chain) on their surface [7]. Upon incubation with IL-3, lymph node derived PC-DC will become matured DCs. Interestingly, similar to their neoplastic counterpart in BT, PC-DC usually do not express myeloid cell lineage markers such as CD 116, the receptor for GM-CSF [14]. More recently, the expression of the lymphoid protooncogene TCL1 was demonstrated in 83% of BT and normal lymphoid PC-DC as well, in further support of a common origin [12]. Jacob et al. [15] reviewed the literature on BT and demonstrated a positive staining for the Blood Dendritic Cell Antigens (BDCA)-2 and -4 which are thought to be specific for immature blood dendritic cells, preferably of the described plasmocytoid type. Also in our case tumor cells stained weakly positive for BDCA2. While similarities between the malignant cells of BT and PC-DC are obvious, there is still some discussion about its relation to myelo-monocytic lymphoma/leukemia [4]. A significant percentage of patients with BT will develop acute or chronic myeloid leukemia [12, 16]. This discussion is closely related to the still ongoing dispute about the origin of the plasmocytoid dendritic cell itself. There is both evidence for a lymphoid and for a myeloid origin of PC-DC. Interestingly, recent data suggest the existence of an intermediate phenotype undergoing cell fate conversion from a functionally distinct and numerically predominant lymphoid to a myeloid cell type of PC [17]. BT appears to be a morphologically well defined entity which is characterised by clinical, histological and distinct immunophenotypic features with a germline TCR- and IgH chain rearrangement. Interestingly, CD4+ CD56+ blastic tumors often show primary skin involvement, e.g. at the time of diagnosis in one study 9/23 cases were without extracutaneous spread [1]. However, also in these cases bone marrow involvement was diagnosed within a few months, and since normal skin is devoid of PC-DC it is unlikely that these BT had their origin in the skin. Also in our patient bone marrow involvement became clinically apparent within three months. Thus, a complete staging procedure is always indicated. Most cases, irrespective of their response to chemotherapy, showed rapid progression [15]. In the study of Feuillard [1] with 80% of patients having developed full remission after polychemotherapy, relapses occurred after a median time of 9 months with only 25% survival after 2 years. Also in our patient rapid disease progression occurred during CHOP polychemotherapy. Presently, possibly bone marrow transplantation seems to be the best choice also for patients with disease limited to the skin as these patients will also almost inevitably experience a rapidly fatal course [1]. Optimised treatment protocols still have to be established. Until now there are only a few cytogenetic data available on cutaneous CD4+CD56+ BT. A study by the Groupe Francaise de Cytogenetique Hematologique has analyzed 21 cases with the primary leukemic onset [18]. Although present in 66% of cases, no specific clonal aberrations were identified. Deletion of chromosome 13q was most frequently found, being present in 64% of those cases with apparent chromosomal aberrations. A deletion of 13q was also seen in our case. Deletion of chromosome 13 is also frequently found in different other types of systemic lymphoma and also in myeloid neoplasms and here it appears to have no prognostic impact [19]. Two out of 4 cases with primary cutaneous manifestation of BT studied by Petrella et al. [3] by classical banding techniques, had complex chromosomal aberrations, including a loss of chromosome 13. Both cases also presented a deletion 5q which, though not specific, is very common in myelodysplastic syndrome [3, 20]. In another recent publication using CGH and micorarray analysis, 4 cases of BT among mostly NK- lymphomas were investigated [21]. Also in this study 3 out of 4 BT showed a loss of chromosome 13 with a deletion of the RB1 gene which is located on this chromosome. Deletion of RB1 has been previously demonstrated in blastic NK-lymphoma pointing to a possible functional significance also in BT. In summary, CD4+CD56+ BT mostly show complex chromosomal aberration patterns which are similar to those described in different types of lymphoid or myeloid neoplasms. Until now it is not clear whether any of the described aberrations may be either relevant for the genesis of the disease or causally related to its aggressive behaviour. AcknowledgementsThe authors thank Prof. Dr. H. H. Kreipe, Director of the Department of Pathology, Hannover Medical University for support with immunohistological stainings and Dr. Thilo Schlott, Department of Pathology, Göttingen Medical University for the analysis of the TCR-gamma rearrangement.References1 Feuillard , Jacob , Valensi , et-al. Clinical and biological features of CD4+CD56+ malignancies Blood 99 2002 1556-15632 Dummer , Potoczna , Haffner , et-al. A primary cutaneous non-T, non-B CD4+, CD56+ lymphoma Arch Dermatol 132 1996 550-553 3 Petrella , Dalac , Maynadie , et-al. CD4+CD56+ cutaneous neoplasms : A distinct hematological entity? Am J Surg Pathol 23 1999 137-146 4 Penven , Macro , Salaun , et-al. Skin manifestations in CD4+, CD56+ malignancies Eur J Dermatol 13 2003 161-165 5 Petrella , Comeau , Maynadie , et-al. Agranular Cd4+CD56+ hematodermic neoplasm’ (Blastic NK-Cell Lymphoma) originates from a population of Cd56+ precursor cells related to plasmacytoid monocytes Am J Surg Pathol 26 2002 852-862 6 Jaffe , Harris , Stein , et-al. World Health Organization Classification of tumours. Pathology and genetics of tumours of haematopoetic and lymphoid tissues ARRAY(0x24dba0)ARRAY(0x24dbc4) 7 Chaperot , Bendriss , Manches , et-al. Identification of a leukemic counterpart of the plasmacytoid dendritic cells Blood 97 2001 3210-3217 8 Kneba , Bolz , Linke , et-al. Characterization of clone-specific rearranged T-cell receptor gamma-chain genes in lymphomas and leucemias by the polymerase chain reaction and DNA sequencing Blood 84 1994 574-581 9 Reinhardt , Langebrake , Creutzig , et-al. Minimal residual disease in acute myeloid leukemia in children-standardization and evaluation of immunophenotyping in the AML-BFM-98 study Klin Padiatr 214 2002 179-187 10 Kallioniemi , Kallioniemi , Sudar , et-al. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors Science 258 1992 818-821 11 Hallermann , Kaune , Siebert , et-al. Chromosomal aberration patterns differ in subtypes of primary cutaneous B-cell lymphomas J Invest Dermatol 2004 in press 12 Herling , Teitell , Shen , et-al. TCL1 Expression in plasmocytoid dendritic cells (CD2) and related CD4CD56 blastic tumors of skin Blood 101 2003 5007-5009 13 Bagot , Bouloc , Charue , et-al. Do primary cutaneous non-T non-B CD4+CD56+ lymphomas belong to the myelo-monocytic lineage ? J Invest Dermatol 111 1998 1242-1244 14 Grouard , Rissoan , Filgueira , et-al. The enigmatic plasmocytoid T-cells develop into dendritic cells with interleucin (IL)-3 and CD40 ligand J Exp Med 185 1997 1101-1111 15 Jacob , Chaperot , Mossuz , et-al. CD4+CD56+ lineage negative malignancies : a new entity developed from malignant early plasmocytoid dendritic cells Haematologica 88 2003 941-955 16 Khoury , Medeiros , Manning , et-al. CD56(+)TdT(+) blastic natural killer cell tumor of the skin: a primitive systemic malignancy related to myelomonocytic leucemia Cancer 94 2002 2401-2408 17 Comeau , Van der Vuurst de Vries , Maliszewski , et-al. CD123 bright plasmocytoid predendritic cells : Progenitors undergoing cell fate conversion? J Immunol 169 2002 75-83 18 Leroux , Mugneret , Callanan , et-al. CD4+CD56+DC 2 acute leukemia is characterized by recurrent clonal chromosomal changes affecting 6 major targets : a study of 21 cases by the Groupe Français de Cytogenetique Hematologique Blood 99 2002 4154-4159 19 Chaganti , Nanjangud , Schmidt , Teruya-Feldstein Recurrent chromosomal abnormalities in Non-Hodgkin’s lymphoma : Biologic and clinical significance Semin Hematol 37 2000 396-411 20 Boultwood , Lewis , Wainscoat The 5q-syndrome Blood 84 1994 3253-3260 21 Mao , Onadim , Price , et-al. Genomic alterations in blastic natural killer/extranodal natural killer-like T cell lymphoma with cutaneous involvement J Invest Dermatol 121 2003 618-627 |
|||||
Version PDF
Copyright © 2007 John Libbey Eurotext - Tous droits réservés