CD56+ hematological neoplasms presenting in the skin: a retrospective analysis of 23 new cases and 130 cases from the literature

M. W. Bekkenk1,*, P. M. Jansen2, C. J. L. M. Meijer3 and R. Willemze1

Departments of 1 Dermatology; 2 Pathology, Leiden University Medical Center, Leiden; 3 Department of Pathology, Vrije Universiteit Medical Center, Amsterdam, The Netherlands

*Correspondence to: Dr M. Bekkenk, Department of Dermatology, B1-Q-93, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands. Tel: +31-71-526-2421; Fax: +31-71-524-8106; Email: m.w.bekkenk{at}lumc.nl


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Background: The aim of this study was to define prognostic parameters and guidelines for diagnosis and treatment for CD56+ hematological neoplasms with first presentation in the skin.

Patients and methods: The study group included 153 cases (23 new and 130 from the literature). According to the World Health Organization classification, the group included 15 nasal and 38 nasal-type natural killer (NK)/T-cell lymphomas, 63 blastic NK-cell lymphomas, 14 cutaneous CD30+ lymphoproliferations, 10 cases of myeloid leukemia, six cases of subcutaneous panniculitis-like T-cell lymphoma (SCPLTCL) and seven peripheral T-cell lymphomas, unspecified.

Results: In general, these CD56+ hematological neoplasms had a poor prognosis, with only 27% of patients alive after a median follow-up of 12 months. The median survival was 13 months. Nasal and nasal-type NK/T-cell lymphomas and CD56+ SCPLTCL had the worst prognosis, with a median survival of 5, 6 and 5 months, respectively. Only nasal-type NK/T-cell lymphomas presenting with only skin lesions had a somewhat better prognosis (median survival 27 months). In blastic NK-cell lymphomas (median survival 14 months), age ≤40 years, aggressive treatment with acute leukemia protocols and high TdT expression were associated with a more favorable prognosis. Striking similarities in histology, immunophenotype, clinical presentation and clinical behavior were found between blastic NK-cell lymphomas and CD56+ myeloid leukemias.

Conclusions: CD56+ hematological neoplasms presenting in the skin have a poor prognosis, except for primary cutaneous CD30+ lymphoproliferations. The striking similarities between blastic NK-cell lymphomas and CD56+ myeloid leukemias presenting in the skin provide a rationale to treat these patients with more aggressive regimens, rather than with CHOP(-like) regimens and radiotherapy, which have proven to be inadequate therapies for this neoplasm.

Key words: CD56, classification, cutaneous NK/T-cell lymphoma, leukemia, prognosis, treatment


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The availability of monoclonal antibodies against CD56 that are active in paraffin sections has resulted in the recognition of a subgroup of cutaneous lymphomas derived from natural killer (NK) cells or NK-like T-cells. CD56 or neuronal-cell adhesion molecule (NCAM) is a 175–185 KDa surface glycoprotein involved in cell adhesion and migration, which is expressed on all NK cells and subsets of CD4+ and CD8+ T-cells and monocytes [1Go].

These CD56+ T/NK-cell lymphomas preferentially present at extranodal sites, not uncommonly the skin, and almost without exception have an extremely poor prognosis. In the World Health Organization (WHO) classification, several groups of CD56+ lymphomas are distinguished, including extranodal NK/T-cell lymphoma (nasal/nasal type) and blastic NK-cell lymphoma [2Go]. Nasal NK/T-cell lymphoma, previously designated lethal midline granuloma, typically presents in the nasopharynx, and often shows an angiocentric growth pattern with prominent necrosis and vascular destruction. The neoplastic cells express cytotoxic proteins (TIA-1, granzyme B and perforin) in most cases. These lymphomas have been reported most frequently in Asia and Central and South America, and are almost without exception strongly associated with Epstein–Barr virus (EBV). Lymphomas with similar histology and phenotype arising at other extranodal sites are designated extranodal NK/T-cell lymphoma, nasal type.

Blastic NK-cell lymphoma most commonly presents in the skin, with or without concurrent extracutaneous disease. Histologically, these lymphomas are characterized by a diffuse monotonous infiltrate of medium-sized cells resembling lymphoblasts or myeloblasts. Characteristically, the neoplastic cells are positive for CD4 and CD56, and in some cases for TdT, but do not express surface CD3 and cytotoxic proteins, and are not associated with EBV. Recent studies showing that these blastic NK cell lymphomas express IL-3R (CD123) and an immunophenotype highly similar to that of plasmacytoid dendritic cells [3Go] argue against a NK-cell origin, which implies that the term blastic ‘NK-cell lymphoma’ is inappropriate. CD4+, CD56+ hematodermic neoplasm [4Go] and CD4+ CD56+ malignancy [5Go] have been suggested as more appropriate terms. Notwithstanding, herein the World Health Organization (WHO) term blastic NK-cell lymphoma will further be used.

In addition to these three groups, cases of myeloid leukemia presenting primarily in the skin as well as other well-defined types of cutaneous T-cell lymphoma (CTCL), such as primary cutaneous CD30+ (anaplastic) large T-cell lymphomas, subcutaneous panniculitis-like T-cell lymphoma (SCPLTCL) and CD8+ epidermotropic CTCL, may show co-expression of CD56 in a proportion of cases.

These CD56+ (cutaneous) hematological neoplasms are an extremely difficult group, for both pathologists and clinicians. For pathologists, correct classification of these lymphomas is difficult, expensive and time-consuming, since it requires application of several complimentary techniques such as extensive phenotyping, EBV analysis and T-cell receptor (TCR) gene rearrangement studies. It is therefore important to know whether such a detailed analysis is useful from a therapeutic or prognostic point of view. Clinicians are confronted with an aggressive clinical behavior and a fatal outcome often within 1 year after diagnosis, and may consider more intensive therapies, as in acute leukemias, as initial therapy. Whether patients presenting with only localized skin lesions have a somewhat better prognosis and require another therapeutic approach is uncertain. It is also not known whether coexpression of CD56 in well-defined types of CTCL is associated with a less favorable prognosis, and should be treated more aggressively.

In an attempt to answer some of these questions and to better define these different subgroups, we evaluated the clinicopathological and immunophenotypical data of a large group of CD56+ hematological neoplasms presenting in the skin, including 23 new cases and 130 cases from the literature.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients
The two main selection criteria for this study were CD56 expression by the neoplastic T/NK-cells or myeloid cells, and the presence of skin lesions at the time of first diagnosis. Twenty-three patients meeting both criteria were found in the registry of the Dutch Cutaneous Lymphoma Group. Fourteen patients had been diagnosed in the last 2 years, when CD56 staining had been routinely applied. Retrospective analysis of a large group of CTCL other than mycosis fungoides and cases of leukemia cutis (presenting with skin lesions) yielded another nine CD56+ cases. Sixteen of 23 patients had only skin lesions at the time of diagnosis, as assessed with routine staging procedures [6Go], whereas seven patients had presented with concurrent cutaneous and extracutaneous disease. Additional immunostainings for CD2, CD3, CD4, CD5, CD7, CD8, CD30, CD68 (clone KP-1), TdT, TIA-1, granzyme B and myeloperoxidase and in situ hybridization for EBER were performed, when appropriate and possible. The clinical and histological data were retrieved and re-evaluated, and all cases were classified according to the criteria of the WHO classification [2Go].

Review of literature
Our literature search focused on reports of CD56+ hematological neoplasms with (i) skin lesions at presentation, (ii) a follow-up of at least 12 months unless prior death to lymphoma, and (iii) sufficient immunophenotypical data to allow categorization according to the WHO classification [2Go].

A Medline survey using ‘skin’ and ‘CD56’ in the PubMed search engine including papers up to 1 December 2002 revealed 87 English papers containing patients with a cutaneous CD56-positive lymphoma/leukemia. Fifty-seven papers included CD56+ hematological neoplasms with skin lesions at first presentation, which had been designated variously as nasal(-type) NK/T-cell lymphoma, NK/T-cell lymphoma, angiocentric lymphoma, blastic NK-cell lymphoma, lymphoblastic NK-cell lymphoma, NK-cell leukemia/lymphoma, CD4+/CD56+ (‘hematodermic’) lymphoma, CD30+/CD56+ lymphoma, myeloid leukemia, granulocytic sarcoma, cytotoxic T-cell lymphoma, SCPLTCL, lymphomatoid papulosis and (primary cutaneous) CD30+ (anaplastic) lymphoma [3Go, 7Go–62Go].

A total of 130 cases met all three criteria mentioned above. Using the WHO classification [2Go], these cases from literature included: 14 nasal NK/T-cell lymphomas {references: [8Go, 15Go, 24Go] (two cases), [28Go, 34Go] (two cases), [35Go, 37Go, 53Go, 56Go] (two cases), [60Go]}, 36 extranodal NK/T-cell lymphomas, nasal type {references: [7Go, 9Go] (two cases), [16Go] (three cases), [14Go, 18Go, 26Go] (three cases), [30Go, 34Go] (two cases), [35Go] (two cases), [38Go, 41Go] (four cases), [44Go] (three cases), [45Go] (two cases), [46Go] (three cases), [49Go, 53Go] (two cases), [55Go, 56Go, 57Go, 62Go]}, 52 blastic NK-cell lymphomas {references: [3Go] (13 cases), [9Go] (three cases), [13Go, 19Go, 20Go, 21Go] (15 cases), [27Go, 30Go, 33Go] (three cases), [32Go, 33Go, 35Go] (two cases), [41Go] (two cases), [45Go] (three cases), [50Go, 58Go, 61Go] (two cases)}, 12 cutaneous CD30+ lymphoproliferations {references: [10Go, 16Go, 23Go, 31Go, 43Go, 45Go, 46Go] (four cases), [48Go, 56Go]}, seven cases of acute myeloid leukemia (AML) (references: [9Go, 22Go, 31Go, 42Go, 43Go, 44Go, 50Go]), five cases of SCPLTCL {references: [25Go, 39Go] (two cases), [51Go] (two cases)}, and four peripheral T-cell lymphoma, unspecified (all with co-expression of CD56; references: [41Go, 47Go, 56Go, 59Go]).

Statistical analysis
Comparison between different subgroups of patients were performed using the {chi}2 of Fisher's exact test for categorical variables (cross-table) or the Mann–Whitney test for continuous variables. Survival duration was calculated from time of diagnosis to date of death or censoring. Overall survival rates were estimated using the method of Kaplan and Meier [63Go]. For overall survival analysis, deaths were taken into account whatever the cause; disease-specific analysis was not performed because only one case died from an unrelated cause.

Prognostic factors in the different subgroups were evaluated by overall survival univariate and multivariate analyses using a Cox proportional hazards model. Factors significant at the 0.1 level in univariate analysis were included in a stepwise regression multivariate analysis.

The following parameters were analyzed: clinical prognostic factors including gender, age, extent of cutaneous, and extracutaneous disease and therapy; histological factors including cell size, angiocentricity, expression of cytotoxic proteins and expression of several immunohistochemical markers, in particular CD30 and TdT, expression of EBV and the presence of clonal T-cell rearrangement, if performed or described.

For statistical analysis the software package for social sciences (SPSS 10.0; SPSS, Chicago, IL, USA) was used.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The clinical, histological and immunophenotypical data for our 23 new cases are summarized in Tables 1 and 2GoGo. Cumulative data on the whole group of 153 cases are summarized in Tables 3 and 4GoGo, and described below.


View this table:
[in this window]
[in a new window]
 
Table 1 Clinical data of 23 new patients

 

View this table:
[in this window]
[in a new window]
 
Table 2 Histological and immunophenotypical data of 23 new patients

 

View this table:
[in this window]
[in a new window]
 
Table 3 Relative frequency of expression of various antibodies for different types of CD56+ lymphoma/leukemia, numbers in parentheses denote the number of cases tested

 

View this table:
[in this window]
[in a new window]
 
Table 4 Cumulative data for 153 patients with CD56 + lymphomas/leukemias presenting in the skin

 
Total group (n=153)
The total group consisted of 89 male and 56 female patients (male:female ratio 1.6:1), while no information on gender was available in eight cases. The median age was 56 years (range 2–94). Sixty-seven of 153 (44%) cases had no evidence of extracutaneous disease at presentation. Treatment consisted mostly of multiagent chemotherapy (69%). A complete remission was reached in 47% of the cases, but 89% relapsed, usually after only a short time. The prognosis was poor, with a median survival of 12 months [95% confidence interval (CI) 9–15], a 2-year survival of 34% and a 5-year survival of 12% (see Table 4Go). The two largest groups were the blastic NK-cell lymphomas (n=63), and the nasal (n=15) and nasal-type (n=38) NK/T-cell lymphomas (n=53).

Nasal NK/T-cell lymphoma (n=15)
This group consisted of 10 males and five females with a median age of 58 years (range 19–90). All cases presented with lesions in the nasopharyngeal area combined with skin lesions either on the trunk and/or extremities (14 cases) or a tumor on the nose (one case), whereas other sites were affected infrequently (see Table 4Go). Initial treatment, mostly multiagent chemotherapy (73%), resulted in a complete remission in only three cases. All cases died of lymphoma. This was the group with the worst prognosis; the median survival was only 5 months, and 2-year survival was not reached (Table 4Go). Two cases were of T-cell origin as proven by TCR and CD3 expression on frozen tissue sections.

Nasal-type NK/T-cell lymphoma (n=38)
This group included 19 males, 15 females and four cases with unknown gender. The median age at diagnosis was 50 years (range 16–79). Fifteen of 38 cases presented with only skin lesions. Treatment mostly consisted of multiagent chemotherapy (n=26; 68%) and resulted in a complete remission in 10 of 40 cases. Only three cases did not relapse, and 29 of 38 (76%) died of lymphoma. The prognosis was poor, with a median survival of only 6 months (95% CI 2–10) and a 2-year survival of 28% (see Table 4Go). Four cases were of T-cell origin as proven by TCR analysis.

Univariant analysis showed that only presentation with skin-limited disease (P=0.006) and presentation with localized as opposed to multifocal skin lesions (P=0.05) were associated with a significantly better prognosis. In multivariant analysis, only presentation with skin-limited disease was significantly related to better survival (P=0.03; relative risk 2.7; 95% CI 1.1–6.4). The median survival of patients presenting with skin-limited disease was 27 months (95% CI 9–45) compared with 4 months (95% CI 3–5) for patients presenting with both cutaneous and extracutaneous disease (see Figure 1Go).



View larger version (9K):
[in this window]
[in a new window]
 
Figure 1 Overall survival curves of different groups of nasal(-type) natural killer (NK)/T-cell lymphoma (n=53) presenting in the skin. Dashed line, nasal-type NK/T (skin lesions only, n=15); solid line, nasal-type NK/T (skin and extracutaneous localizations, n=23); dotted line, nasal NK/T (n=15).

 
Comparison between nasal and nasal-type NK/T-cell lymphoma
When comparing these groups of nasal and nasal-type NK/T-cell lymphomas, there were no differences in expression of T-cell related antigens (CD2, CD3, CD4, CD5, CD7 and CD8), CD30 expression, clonality and histology/angiocentricity (see Table 3Go). EBV expression was present in 100% and 76% of the nasal and nasal-type lymphomas, respectively (P=0.05). After exclusion of 17 nasal-type NK/T-cell lymphomas presenting with only skin lesions (median survival 27 months; 95% CI 12–42), no difference in survival was found between 15 nasal cases (median survival 5 months; 95% CI 3–7) and 23 nasal-type cases (median survival 4 months; 95% CI 3–5) (see Figure 1Go).

Blastic NK-cell lymphoma (n=63)
This group consisted of 41 male and 20 female cases (male:female ratio 2:1); two cases were of unknown gender. The median age at diagnosis was 67 years (range 8–89); 13 cases were <40 years old. A strikingly high percentage of patients had bone marrow involvement: 46% at presentation and 72% during the course of the disease (see Table 3Go). Initial treatment consisted of multiagent chemotherapy in the majority of patients (70%). Most patients had received CHOP or CHOP-like regimens (56%), whereas only eight patients had been treated more aggressively with schemes used in acute leukemias followed by autologous or allogeneic bone marrow or stem-cell transplantation. Six of these eight patients are alive, and although the follow-up is mostly short (median 25 months; range 6–40), they proved to have a significant better prognosis when compared with all other cases (P=0.01) and compared with cases that received adriamycin-containing chemotherapy (P=0.04). The prognosis of these lymphomas was poor, with a median survival of 14 months, and 2- and 5-year overall survivals of 33% and 6%, respectively (see Table 4Go). Patients presenting with only skin lesions (median survival 21 months; 95% CI 14–28) had a better prognosis than patients presenting with both cutaneous and extracutaneous disease (median survival 12 months; 95% CI 10–14) (see Figure 2Go).



View larger version (10K):
[in this window]
[in a new window]
 
Figure 2 Overall survival curves of different groups of blastic natural killer (NK)-cell lymphoma (n=63) presenting in the skin. Solid line, blastic NK (skin lesions only, n=27); dashed line, blastic NK (skin and extracutaneous localizations, n=36).

 
Univariant analysis showed that the following parameters were associated with a better prognosis: age ≤40 years (P<0.0001), presentation with only skin lesions (P=0.03), initial treatment with transplantation-directed regimens (P=0.01) and TdT expression by >50% of the neoplastic cells (P=0.03). Multivariant analysis showed that young age (≤40 years) and TdT expression of >50% of the neoplastic cells were both independent prognostic variables (relative risk 18.3; 95% CI 4.2–80.8; and relative risk 8.4; 95% CI 2.4–30) respectively.

CD56+ myeloid leukemia first presenting in the skin (n=10)
This group included seven males and three females with a median age of 54 years (range 2–94). Seven of 10 cases presented simultaneously with skin lesions and bone marrow and/or blood involvement. The other three cases presented with only skin lesions, but developed frank leukemia after 3–12 months. According to the WHO classification, all cases were classified into the broad category of AML not otherwise categorized. When further subdivided, three cases of acute monoblastic and monocytic leukemia (FAB M5a/b) and three cases of acute myelomonocytic leukemia (FAB M4) could be distinguished. Four cases presented with only skin lesions, and would be designated as myeloid sarcoma; after the involvement of the bone marrow all four of these cases could be classified as acute monoblastic and monocytic leukemia (FAB M5). Four cases had been published as blastic NK-cell lymphoma, but were reclassified as AML because of the expression of CD33 and other myeloid markers. Initial treatment, including CHOP-like courses in three cases and AML schemes in four cases, resulted in complete remissions in six of 10 cases. The prognosis was poor, with a median survival of 6 months and a 2-year survival of 20% (see Table 4Go).

Comparison between blastic NK cutaneous CD56+ myeloid leukemia
In view of recent reports that blastic NK and a proportion of myeloid leukemias presenting in the skin are possibly derived from precursor plasmacytoid dendritic cells type 2 [3Go, 64Go–66Go], we compared the 63 cases diagnosed as blastic NK with the 10 CD56+ myeloid leukemias. Striking similarities between both groups were noted in histology and immunophenotype, including expression of CD34, CD68 and TdT (but with the exception of CD33, myeloperoxidase and lysozyme, used as exclusion criteria for blastic NK). Clinical presentation, including the percentage of bone marrow involvement at presentation (46% versus 60%), response to therapy and prognosis were also very similar (see Table 4Go and Figure 3Go).



View larger version (10K):
[in this window]
[in a new window]
 
Figure 3 Overall survival curves comparing blastic natural killer (NK)-cell lymphoma and myeloid leukemia presenting with skin localizations (n=73). Solid line, blastic NK (n=63); dashed line, CD56+ myeloid leukemia presenting on the skin (n=10).

 
Cutaneous CD30+ lymphoproliferations with CD56 expression (n=14)
This group consisted of five males, eight females and one case with unknown gender. The median age was 45 years (range 4–89). The group included 10 primary cutaneous CD30+ large cell lymphomas, including six cases designated previously CD56+ NK/NK-like T-cell lymphoma with expression of CD30 antigen, three cases of LyP and one case of anaplastic (ALK-positive) CD30+ lymphoma presenting with lesions in the skin, bone, lymph nodes and parotis.

Initial therapy consisted of multiagent chemotherapy in six of 14 cases, whereas five cases were not treated at all. Three cases died, two within 2 months due to side-effects of chemotherapy and one of unrelated disease. The median survival was not reached; the 2- and 5-year survivals were 75% and 56%, respectively.

Subcutaneous panniculitis-like T-cell lymphoma (n=6)
This group consisted of three males and three females with a median age of 50 years (range 12–56). Two cases presented with only skin lesions, whereas three cases had concurrent involvement of lymph nodes. Chemotherapy resulted in a complete remission in only one of six patients, and all but one died of lymphoma. The median survival was only 5 months (95% CI 1–9); the 2-year survival was 33% (see Table 4Go). The only patient still alive had an alpha/beta phenotype, whereas the five patients who died had a gamma/delta phenotype.

Peripheral T-cell lymphoma not otherwise specified (n=7)
This group consisted of two males, four females and one case with unknown gender. The median age was 67 years (range 28–75). All cases presented with only skin lesions. This was a heterogeneous group, including four epidermotropic CD8+ CTCL [67Go], two CD30– large primary cutaneous T-cell lymphomas [6Go] and one large granular NK/T-cell leukemia with skin localizations. Initial treatment resulted in a complete remission in four cases, but all but one patient relapsed. Four of seven patients died of lymphoma. The median survival was 66 months (95% CI 32–100); the 2- and 5-year survivals were 83% and 56%, respectively.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In the present study, 153 CD56+ hematological neoplasms, first presenting in the skin and with a minimum follow-up of 12 months or prior death to lymphoma, were evaluated. This group included 23 new cases and 130 cases retrieved from the literature. The goals of this study were to assess whether the different groups of CD56+ lymphomas recognized in the WHO classification differ in clinical behavior, to define prognostic parameters and to deduct guidelines for diagnosis and treatment. In general, these CD56+ hematological neoplasms had a poor prognosis, with a median survival of 13 months. After a median follow-up of 12 months, only 27% of patients were alive, with or without disease. Survival curves of the main groups of cutaneous CD56+ hematological neoplasms are presented in Figure 4Go. The differences in clinical behavior and prognosis underscore that it is important to make a distinction between the different groups, and justify a detailed analysis of these cases with different techniques. Considering the often difficult differential diagnosis and rarity of these cases, referring to an institute specialized in (cutaneous) lymphoma is advisable.



View larger version (11K):
[in this window]
[in a new window]
 
Figure 4 Overall survival curves of the main groups of CD56+ lymphomas/leukemias presenting in the skin (n=140). Dashed line, cutaneous CD30+ lymphoproliferation (n=14); solid line, blastic natural killer (NK)/CD56+ myeloid leukemia (n=73); dotted line, nasal(-type) NK/T (n=53).

 
Blastic NK-cell lymphomas were by far the largest group (63 of 153 cases; 40%). In recent literature, such cases have been designated variously as blastic or blastoid NK-cell lymphoma and (haematodermic) CD4+/CD56+ leukemia/lymphoma [2Go, 3Go, 21Go, 26Go]. These patients had a poor prognosis, with a median survival of only 14 months. CHOP(-like) courses may result in initial complete remissions in these patients, but quick relapses unresponsive to further chemotherapy are characteristic of these tumors. Multivariate analysis revealed that age ≤40 years and high expression of TdT by the neoplastic cells were independently associated with a better prognosis. The results of this study and others [5Go, 61Go] confirm that CHOP-like regimens and radiotherapy are inadequate therapies for this type of neoplasm. More aggressive therapies directed to alleogeneic stem-cell transplantations are probably superior. Young patients treated with aggressive regimens used in acute leukemias had a better prognosis than patients treated with CHOP(-like) regimens [5Go, 9Go, 21Go, 32Go, 33Go, 50Go, 61Go]. Further studies are necessary to confirm these findings and to detemine whether more aggressive therapies are also superior in older patients.

Comparison between blastic NK-cell lymphomas and CD56+ myeloid leukemias presenting in the skin showed striking similarities in histology, immunophenotype, clinical presentation, including the percentage of bone marrow involvement at presentation (46% compared with 60%), and prognosis. In addition, in a recent workshop of the EORTC Cutaneous Lymphoma Study Group on cutaneous CD56+ lymphomas, the great similarities between both groups were noted (Muche et al., unpublished data). These observations suggest a possible relationship between both conditions. In addition, the development of AML in patients with blastic NK-cell lymphoma has been reported [31Go].

Recent studies demonstrated that blastic NK-cell lymphomas express markers typical of plasmacytoid dendritic cells, in particular the IL-3R alpha chain (CD123) and the lymphoid protooncogene TCL1 [64Go]. Expression of CD123 has also been reported in a proportion of myeloid leukemias presenting in the skin [3Go, 65Go]. These findings argue against a NK-cell origin and suggest that the term ‘blastic NK-cell lymphoma’ is inappropriate. CD4+, CD56+ hematodermic neoplasm [4Go] and CD4+ CD56+ malignancy [5Go] have been suggested as more appropriate terms for this condition.

The overlapping clinical, histological and immunophenotypical features of blastic NK-cell lymphoma and CD56+ myeloid leukemias presenting in the skin provide also a rationale to treat these patients with more aggressive regimens used in myeloid leukemias, rather than with convential CHOP(-like) regimens, which is in line with the results of recent studies [5Go, 61Go].

Nasal and nasal-type NK/T-cell lymphomas together represented the second largest group in this study (55 of 157 cases; 35%). In cases with a true T-cell phenotype as defined by expression of surface CD3 and/or the presence of a clonal TCR gene rearrangement, which express cytotoxic proteins, but no EBV, it is an arbitrary decision whether a diagnosis of nasal type NK/T-cell lymphoma or peripheral T-cell lymphoma NOS should be made. In this study, such EBV-negative cases were classified as peripheral T-cell lymphomas, not otherwise specified. Angiocentricity proved not to be very helpful in distinguishing the two groups, because this feature was observed in all groups investigated in this study (see Table 3Go). Nasal and nasal-type NK/T-cell lymphomas showed many similarities (see Table 4Go). No difference in survival was found between nasal NK/T-cell lymphomas presenting with concurrent cutaneous disease and nasal type NK/T-cell lymphomas presenting with cutaneous and extracutaneous disease (median survival 5 and 4 months, respectively). Nasal-type NK/T-cell lymphomas presenting with skin-limited disease had a somewhat better prognosis, which is consistent with the results of previous studies in primary cutaneous angiocentric lymphomas [68Go] (see Figure 1Go). The poor survival data in these nasal(-type) NK/T-cell lymphomas also indicate that current standard regimens—generally doxorubicin-based chemotherapy—are insufficient in these cases, and that other approaches should be investigated.

Apart from the entities discussed above, other types of CTCL can occasionally show co-expression of CD56. Within the group of CD56+ SCPLTCL, five of six cases appeared to have a gamma/delta phenotype. These gamma/delta-positive cases are often not confined to the subcutis, but may extend into the dermis [69Go]. In such cases differentiation between SCPLTCL and nasal-type NK/T-cell lymphomas extending into the subcutis may be difficult [15Go]. It is noteworthy that both groups have a similar poor prognosis, and do not respond sufficiently to CHOP(-like) therapy. The only patient with a CD56+ SCPLTCL still alive in this study had an alpha/beta T-cell phenotype. This observation is consistent with the results of other studies indicating that SCPLTCL with an alpha/beta phenotype have a much better prognosis than cases with a gamma/delta phenotype, and that the two groups should be considered separately [51Go, 70Go].

Previous studies in systemic CD30+ anaplastic large-cell lymphoma suggested that expression of CD56 is associated with a less favorable prognosis [71Go]. The favorable prognosis of the patients with a primary cutaneous CD30+ lymphoma and LyP included in this study suggests that CD56 expression does not affect clinical behavior and outcome in the group of primary cutaneous CD30+ lymphoproliferations. Similarly, the four cases with an epidermotropic CD8+ CTCL with co-expression of CD56 did not differ clinically from CD56– cases, as described previously [67Go].

Classification of lymphomas that are defined primarily by phenotype, such as these CD30+ lymphomas and CD8+ CTCL, and which at the same time express CD56, may be difficult. This problem is well illustrated by the eight CD30+/CD56+ lymphomas in the study by Mraz-Gernhard et al. [41Go]. In that study, on NK/NK-like T-cell lymphoma primarily defined by the presence of CD56 expression, it was found that cases which co-express CD30 had a much better prognosis than CD30– cases. Seven of eight cases are alive and well after initial treatment, and did not relapse. It seems that these patients have a clinical presentation and course similar to that observed in primary cutaneous CD30 + CTCL [10Go, 72Go]. Although EBV-expression was noted in three out of eight cases, one wonders why at least some of the cases were not simply classified as primary cutaneous CD30+ CTCL with co-expression of CD56. Therefore, studies on larger numbers of patients with co-expression of CD30 and CD56 are necessary. If it is confirmed that primary cutaneous CD30+ CTCL with co-expression of CD56 do not have a separate clinical behavior from CD56– cases, cases with localized disease can be better treated with radiotherapy than with multiagent chemotherapy.

Received for publication October 28, 2003. Revision received February 27, 2004. Accepted for publication February 27, 2004.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
1. Griffin JD, Hercend T, Beveridge R et al. Characterization of an antigen expressed by human natural killer cells. J Immunol 1983; 130: 2947–2951.[Abstract/Free Full Text]

2. Jaffe ES, Harris NL, Stein H, Vardiman JW (eds): World Health Organization classification of tumours. Pathology and Genetics of Tumours of the Hematopoetic and Lymphoid Tissues. Lyon, France: IARC Press 2001.

3. Petrella T, Comeau MR, Maynadie M 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 2002; 26: 852–862.[CrossRef][ISI][Medline]

4. Petrella T, Dalac S, Mayandie M et al. CD4+ CD56+ cutaneous neoplasms: a distinct hematologic entity? Groupe Francais d'Etude des Lymphomes Cutanes (GFELC). Am J Surg Pathol 1999; 23: 137–146.[CrossRef][ISI][Medline]

5. Reimer P, Rüdiger T, Kraemer D et al. What is CD4+ CD56 + malignancy and how should it be treated? Bone Marrow Transplantation 2003; 32: 637–646.[CrossRef][ISI][Medline]

6. Willemze R, Kerl H, Sterry W et al. EORTC classification for primary cutaneous lymphomas: a proposal from the Cutaneous Lymphoma Study Group of the European Organization for Research and Treatment of Cancer. Blood 1997; 90: 354–371.[Abstract/Free Full Text]

7. Abe Y, Muta K, Ohshima K et al. Subcutaneous panniculitis by Epstein–Barr virus-infected natural killer (NK) cell proliferation terminating in aggressive subcutaneous NK cell lymphoma. Am J Hematol 2000; 64: 221–225.[CrossRef][ISI][Medline]

8. Ahn SJ, Jang KA, Choi JH et al. Nasal and nasal type CD56+ natural killer cell/T-cell lymphoma: a case with rapid progression to bone marrow involvement. Br J Dermatol 2000; 142: 1021–1025.[CrossRef][ISI][Medline]

9. Bayerl MG, Rakozy CK, Mohamed AN et al. Blastic natural killer cell lymphoma/leukemia: a report of seven cases. Am J Clin Pathol 2002; 117: 41–50.[CrossRef][ISI][Medline]

10. Bekkenk MW, Kluin PM, Jansen PM et al. Lymphomatoid papulosis with a natural killer-cell phenotype. Br J Dermatol 2001; 145: 318–322.[CrossRef][ISI][Medline]

11. Chan JK, Sin VC, Wong KF et al. Nonnasal lymphoma expressing the natural killer cell marker CD56: a clinicopathologic study of 49 cases of an uncommon aggressive neoplasm. Blood 1997; 89: 4501–4513.[Abstract/Free Full Text]

12. Chan JK, Tsang WY, Lau WH et al. Aggressive T/natural killer cell lymphoma presenting as testicular tumor. Cancer 1996; 77: 1198–1205.[CrossRef][ISI][Medline]

13. Chang SE, Choi HJ, Huh J et al. A case of primary cutaneous CD56+ TdT+, CD4+, blastic NK-cell lymphoma in a 19-year-old woman. Am J Dermatopathol 2002; 24: 72–75.[CrossRef][ISI][Medline]

14. Chang SE, Choi JH, Sung KJ et al. Cutaneous nasal-type CD56+ natural killer/T-cell lymphoma preceded by Epstein–Barr virus antigenaemia. Br J Dermatol 2001; 145: 668–669.[CrossRef][ISI][Medline]

15. Chang SE, Huh J, Choi JH et al. Clinicopathological features of CD56+ nasal-type T/natural killer cell lymphomas with lobular panniculitis. Br J Dermatol 2000; 142: 924–930.[CrossRef][ISI][Medline]

16. Chang SE, Park IJ, Huh J et al. CD56 expression in a case of primary cutaneous CD30+ anaplastic large cell lymphoma. Br J Dermatol 2000; 142: 766–770.[CrossRef][ISI][Medline]

17. Chang SE, Yoon GS, Huh J et al. Comparison of primary and secondary cutaneous CD56 + NK/T cell lymphomas. Appl Immunohistochem Mol Morphol 2002; 10: 163–170.[ISI][Medline]

18. Coupland SE, Foss HD, Assaf C et al. T-cell and T/natural killer-cell lymphomas involving ocular and ocular adnexal tissues: a clinicopathologic, immunohistochemical, and molecular study of seven cases. Ophthalmology 1999; 106: 2109–2120.[CrossRef][ISI][Medline]

19. Dummer R, Potoczna N, Haffner AC et al. A primary cutaneous non-T, non-B CD4+, CD56+ lymphoma. Arch Dermatol 1996; 132: 550–553.[Abstract]

20. Falcao RP, Garcia AB, Marques MG et al. Blastic CD4 NK cell leukemia/lymphoma: a distinct clinical entity. Leuk Res 2002; 26: 803–807.[CrossRef][ISI][Medline]

21. Feuillard J, Jacob MC, Valensi F et al. Clinical and biologic features of CD4(+)CD56(+) malignancies. Blood 2002; 99: 1556–1563.[Abstract/Free Full Text]

22. Ginarte M, Abalde MT, Peteiro C et al. cell leukemia/lymphoma with cutaneous involvement. Dermatology 2000; 201: 268–271.[CrossRef][ISI][Medline]

23. Gould JW, Eppes RB, Gilliam AC et al. Solitary primary cutaneous CD30+ large cell lymphoma of natural killer cell phenotype bearing the t(2;5)(p23;q35) translocation and presenting in a child. Am J Dermatopathol 2000; 22: 422–428.[ISI][Medline]

24. Hirakawa S, Kuyama M, Takahashi S et al. Nasal and nasal-type natural killer/T-cell lymphoma. J Am Acad Dermatol 1999; 40: 268–272.[ISI][Medline]

25. Jang KA, Choi JH, Sung KJ et al. Primary CD56+ nasal-type T/natural killer-cell subcutaneous panniculitic lymphoma: presentation as haemophagocytic syndrome. Br J Dermatol 1999; 141: 706–709.[CrossRef][ISI][Medline]

26. Kamarashev J, Burg G, Mingari MC et al. Differential expression of cytotoxic molecules and killer cell inhibitory receptors in CD8+ and CD56+ cutaneous lymphomas. Am J Pathol 2001; 158: 1593–1598.[Abstract/Free Full Text]

27. Kameoka J, Ichinohasama R, Tanaka M et al. A cutaneous agranular CD2–CD4+ CD56+ lymphoma: report of two cases and review of the literature. Am J Clin Pathol 1998; 110: 478–488.[ISI][Medline]

28. Kanavaros P, Lescs MC, Briere J et al. Nasal T-cell lymphoma: a clinicopathologic entity associated with peculiar phenotype and with Epstein–Barr virus. Blood 1993; 81: 2688–2695.[Abstract]

29. Kato N, Yasukawa K, Kimura K et al. CD2– CD4+ CD56+ hematodermic/hematolymphoid malignancy. J Am Acad Dermatol 2001; 44: 231–238.[CrossRef][ISI][Medline]

30. Kato N, Yasukawa K, Onozuka T et al. Nasal and nasal-type T/NK-cell lymphoma with cutaneous involvement. J Am Acad Dermatol 1999; 40: 850–856.[ISI][Medline]

31. Khoury JD, Medeiros LJ, Manning JT et al. CD56(+) TdT(+) blastic natural killer cell tumor of the skin: a primitive systemic malignancy related to myelomonocytic leukemia. Cancer 2002; 94: 2401–2408.[CrossRef][ISI][Medline]

32. Kimura S, Kakazu N, Kuroda J et al. Agranular CD4+ CD56+ blastic natural killer leukemia/lymphoma. Ann Hematol 2001; 80: 228–231.[CrossRef][ISI][Medline]

33. Knudsen H, Gronbaek K, thor Straten P et al. A case of lymphoblastoid natural killer (NK)-cell lymphoma: association with the NK-cell receptor complex CD94/NKG2 and TP53 intragenic deletion. Br J Dermatol 2002; 146: 148–153.[CrossRef][ISI][Medline]

34. Ko YH, Ree HJ, Kim WS et al. Clinicopathologic and genotypic study of extranodal nasal-type natural killer/T-cell lymphoma and natural killer precursor lymphoma among Koreans. Cancer 2000; 89: 2106–2116.[CrossRef][ISI][Medline]

35. Kojima H, Mukai HY, Shinagawa A et al. Clinicopathological analyses of 5 Japanese patients with CD56+ primary cutaneous lymphomas. Int J Hematol 2000; 72: 477–483.[ISI][Medline]

36. Kwong YL, Chan AC, Liang RH. Natural killer cell lymphoma/leukemia: pathology and treatment. Hematol Oncol 1997; 15: 71–79.[CrossRef][ISI][Medline]

37. Ling TC, Harris M, Craven NM. Epstein–Barr virus-positive blastoid nasal T/natural killer-cell lymphoma in a Caucasian. Br J Dermatol 2002; 146: 700–703.[CrossRef][ISI][Medline]

38. Lima M, Goncalves C, Teixeira MA et al. Aggressive natural-killer cell lymphoma presenting with skin lesions, breast nodule, suprarenal masses and life-threatening pericardial and pleural effusions. Leuk Lymphoma 2001; 42: 1385–1391.[ISI][Medline]

39. Marzano AV, Berti E, Paulli M et al. Cytophagic histiocytic panniculitis and subcutaneous panniculitis-like T-cell lymphoma: report of 7 cases. Arch Dermatol 2000; 136: 889–896.[Abstract/Free Full Text]

40. Matano S, Nakamura S, Nakamura S et al. Monomorphic agranular natural killer cell lymphoma/leukemia with no Epstein–Barr virus association. Acta Haematol 1999; 101: 206–208.[CrossRef][ISI][Medline]

41. Mraz-Gernhard S, Natkunam Y, Hoppe RT et al. Natural killer/natural killer-like T-cell lymphoma, CD56+, presenting in the skin: an increasingly recognized entity with an aggressive course. J Clin Oncol 2001; 19: 2179–2188.[Abstract/Free Full Text]

42. Mukai HY, Kojima H, Suzukawa K et al. High-dose chemotherapy with peripheral blood stem cell rescue in blastoid natural killer cell lymphoma. Leuk Lymphoma 1999; 32: 583–588.[ISI][Medline]

43. Murakami Y, Nagae S, Matsuishi E et al. A case of CD56+ cutaneous aleukaemic granulocytic sarcoma with myelodysplastic syndrome. Br J Dermatol 2000; 143: 587–590.[ISI][Medline]

44. Nakamura S, Suchi T, Koshikawa T et al. Clinicopathologic study of CD56 (NCAM)-positive angiocentric lymphoma occurring in sites other than the upper and lower respiratory tract. Am J Surg Pathol 1995; 19: 284–296.[ISI][Medline]

45. Natkunam Y, Smoller BR, Zehnder JL et al. Aggressive cutaneous NK and NK-like T-cell lymphomas: clinicopathologic, immunohistochemical, and molecular analyses of 12 cases. Am J Surg Pathol 1999; 23: 571–581.[CrossRef][ISI][Medline]

46. Natkunam Y, Warnke RA, Haghighi B et al. Co-expression of CD56 and CD30 in lymphomas with primary presentation in the skin: clinicopathologic, immunohistochemical and molecular analyses of seven cases. J Cutan Pathol 2000; 27: 392–399.[CrossRef][ISI][Medline]

47. Noguchi M, Kawano Y, Sato N et al. T-cell lymphoma of CD3+ CD4+ CD56+ granular lymphocytes with hemophagocytic syndrome. Leuk Lymphoma 1997; 26: 349–358.[ISI][Medline]

48. Paulli M, Boveri E, Rosso R et al. CD56/neural cell adhesion molecule expression in primary extranodal Ki-1/CD30+ lymphoma. Report of a pediatric case with simultaneous cutaneous and bone localizations. Am J Dermatopathol 1997; 19: 384–390.[CrossRef][ISI][Medline]

49. Radonich MA, Lazova R, Bolognia J. Cutaneous natural killer/T-cell lymphoma. J Am Acad Dermatol 2002; 46: 451–456.[CrossRef][ISI][Medline]

50. Rakozy CK, Mohamed AN, Vo TD et al. CD56+/CD4+ lymphomas and leukemias are morphologically, immunophenotypically, cytogenetically, and clinically diverse. Am J Clin Pathol 2001; 116: 168–176.[CrossRef][ISI][Medline]

51. Salhany KE, Macon WR, Choi JK et al. Subcutaneous panniculitis-like T-cell lymphoma: clinicopathologic, immunophenotypic, and genotypic analysis of alpha/beta and gamma/delta subtypes. Am J Surg Pathol 1998; 22: 881–893.[CrossRef][ISI][Medline]

52. Sanda T, Lida S, Ito M et al. Successful treatment of nasal T-cell lymphoma with a combination of local irradiation and high-dose chemotherapy. Int J Hematol 2002; 75: 195–200.[ISI][Medline]

53. Savoia P, Fierro MT, Novelli M et al. CD56-positive cutaneous lymphoma: a poorly recognized entity in the spectrum of primary cutaneous disease. Br J Dermatol 1997; 137: 966–971.[CrossRef][ISI][Medline]

54. Siu LL, Chan JK, Kwong YL. Natural killer cell malignancies: clinicopathologic and molecular features. Histol Histopathol 2002; 17: 539–554.[ISI][Medline]

55. Takahashi N, Miura I, Chubachi A et al. A clinicopathological study of 20 patients with T/natural killer (NK)-cell lymphoma-associated hemophagocytic syndrome with special reference to nasal and nasal-type NK/T-cell lymphoma. Int J Hematol 2001; 74: 303–308.[ISI][Medline]

56. Takeshita M, Akamatsu M, Ohshima K et al. Angiocentric immunoproliferative lesions of the skin show lobular panniculitis and are mainly disorders of large granular lymphocytes. Hum Pathol 1995; 26: 1321–1328.[ISI][Medline]

57. Tao J, Shelat SG, Jaffe ES et al. Aggressive Epstein–Barr virus-associated, CD8+, CD30+, CD56+, surface CD3-, natural killer (NK)-like cytotoxic T-cell lymphoma. Am J Surg Pathol 2002; 26: 111–118.[CrossRef][ISI][Medline]

58. Uchiyama N, Ito K, Kawai K et al. CD2–, CD4+, CD56+ agranular natural killer cell lymphoma of the skin. Am J Dermatopathol 1998; 20: 513–517.[CrossRef][ISI][Medline]

59. Yamashita Y, Tsuzuki T, Nakayama A et al. A case of natural killer/T cell lymphoma of the subcutis resembling subcutaneous panniculitis-like T cell lymphoma. Pathol Int 1999; 49: 241–246.[CrossRef][ISI][Medline]

60. Yamazaki M, Kakuta M, Takimoto R et al. Nasal natural killer cell lymphoma presenting as lethal midline granuloma. Int J Dermatol 2000; 39: 931–934.[CrossRef][ISI][Medline]

61. Yamada O, Ichikawa M, Okamota T et al. Killer T-cell induction in patients with blastic natural killer cell lymphoma/leukemia: implications for successful treatment and possible therapeutic strategies. Br J Haematol 2001; 113: 153–160.[CrossRef][ISI][Medline]

62. Amo Y, Ogata F, Yonemoto K et al. CD56-positive cutaneous lymphoma with multicentric Castleman's disease-like systemic manifestations. J Dermatol 2001; 28: 746–752.[ISI][Medline]

63. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 157–181.

64. Herling M, Teitell MA, Shen RR et al. TCL1 expression in plasmacytoid dendritic cells (DC2s) and the related CD4+ CD56+ blastic tumors of skin. Blood 2003; 101: 5007–5009.[Abstract/Free Full Text]

65. Munoz L, Nomdedeu JF, Lopez O et al. Interleukine-3 receptor alpha chain (CD123) is widely expressed in hematologic malignancies. Haematologica 2001; 86: 1261–1269.[ISI][Medline]

66. Chaperot L, Bendriss N, Maches O et al. Identification of a leukemic counterpart of the plasmacytoid dendritic cells. Blood 2001; 97: 3210–3217.[Abstract/Free Full Text]

67. 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.[Abstract/Free Full Text]

68. Wechsler J, Willemze R, van der Brule A et al. Differences in Epstein–Barr virus expression between primary and secondary cutaneous angiocentric lymphomas. French Study Group of Cutaneous Lymphomas. Arch Dermatol 1998; 134: 479–484.[Abstract/Free Full Text]

69. Santucci M, Pimpinelli N, Massi D et al. Cytotoxic/natural killer cell cutaneous lymphomas. Report of EORTC Cutaneous Lymphoma Task Force Workshop. Cancer 2003; 97: 610–627.[CrossRef][ISI][Medline]

70. Hoque SR, Child FJ, Whittaker SJ et al. Subcutaneous panniculitis-like T-cell lymphoma: a clinicopathological, immunophenotypic and molecular analysis of six patients. Br J Dermatol 2003; 148: 516–525.[ISI][Medline]

71. Suzuki R, Kagami Y, Takeuchi K et al. Prognostic significance of CD56 expression for ALK-positive and ALK-negative anaplastic large cell lymphoma of T/null cell phenotype. Blood 2000; 96: 2993–3000.[Abstract/Free Full Text]

72. Bekkenk MW, Geelen FA, van Voorst Vader PC et al. Primary and secondary cutaneous CD30(+) lymphoproliferative disorders: a report from the Dutch Cutaneous Lymphoma Group on the long-term follow-up data of 219 patients and guidelines for diagnosis and treatment. Blood 2000; 95: 3653–3661.[Abstract/Free Full Text]