MALIGNANT HISTIOCYTOSIS

LANGERHANS CELL HISTIOCYTOSIS

0889-8588/98 $8.00

+ .OO

MALIGNANT HISTIOCYTOSIS Histologic, Cytochemical, Chromosomal, and Molecular Data with a Nosologic Discussion Jean Gogusev, MD, and Christian Nezelof, MD

For several years, class I11 histiocytosis has been a confusing disorder, but there have been recent contributions to the subject, provided by modern techniques in cellular and molecular biology, regarding the stage of differentiation and the cell lineage involved. According to the degree of differentiation pathway from the monoblastic stem cells, the blood monocyte to the peripheral histiocyte and macrophage, malignant transformations of these lineages include acute monocytic leukemia, chronic myelomonocytic leukemia, and malignant histiocytosis (MH). Because leukemia and lymphoma are regarded as neoplasms of lymphoid cells, MH likewise is thought to represent a malignant transformation of the reticuloendothelial system (RES).5,14, R,86 Because the initial concept of RES has evolved considerably from that of a web of fixed, phagocytic cells to that of a network of bone marrow-derived, highly pleomorphic, mobile and fixed cells involved in several basic immunologic functions, the understanding and identification of these malignant changes have also evolved and given birth to various interpretations and denominations such as the mononucleur phugocyte Although it has been seriously challenged on the basis of certain immunocytochemical data, the MH concept recently received strong support from accurate chromosomal and molecular data, and it can be placed among the large and heterogeneous group of the anaplastic large cell lymphomas (ALCLs).l6,60,61, 63, Although there is general agreement on the clinical features of MH, there is still much debate as to the histiocytic or lymphoid nature of the proliferative cells, which has given rise to a number of confusing terms for this malignant entity. In addition to the equivocal innate pleomorphism of the cells of the histiocyte/ macrophage lineage, the controversy is mainly fueled by the fact that numerous Work supported by a grant from the Histiocytosis Association of America.

From the Institut National de la Sant6 et de la Recherche Medicale, Unit 90, HBpital Necker Enfants Malades (JG); the Department of Pathology, Faculte de Medecine de Paris (CN); and HBpital Saint Vincent de Paul (CN), Paris, France

HEMATOLOGY/ONCOLOGY CLINICS OF NORTH AMERICA VOLUME 12 NUMBER 2 * APRIL 1998

445

446

GOGUSEV & NEZELOF

studies have dealt with heterogeneous material, most often selected retrospectively on the basis of either some unique histochemical data (Ki-1 or CD30 positivity) or a characteristic chromosomal abnormality (the t(2;5) associated with 5q35bp).%, In addition, most of these investigations have included in a haphazard way childhood and adult cases, in spite of the fact that the clinical course is definitely not the same in children and adults, and that the disease affects preferentially a young population (Table 1). For these reasons and the fact that this issue is devoted to histiocytic disorders, the analysis of the morphologic and molecular characteristics is under the general heading of MH and deals exclusively with pediatric cases. The nosologic discussion is left to the end of the article. HISTOPATHOLOGY

MH is characterized by the proliferation of large (from 10 to 20 pm), atypical, clear, "histiocyte-like" cells according to Rappaport" and Cazal.I4Better seen on smears, these cells have a basophilic, finely or grossly vacuolated cytoplasm. The nucleus is large and irregular and exhibits prominent and dense nucleoli and a thick nuclear membrane (Fig. 1). Mitotic figures, sometimes multipolar, are frequent and provide a crucial element for distinguishing this condition from a reactive process. Although the cytoplasm frequently contains cell debris, erythrophagocytic figures are exceedingly rare. When present, the phagocytizing cells are, in most cases, indigenous normal macrophages. Accumulation of inflammatory cells is frequent and, among them, activated macrophages and lymphocytes are usually present and tend to confuse the histopathologic picture52,53, 74, 96 (Fig. 2). Areas of necrosis are also observed. In certain localizations, such as bones, these additional features can make the diagnosis difficult. The lymph nodes are rapidly preferentially involved.'" A very valuable morphologic characteristic is the sinusoidal distribution of neoplastic cell infilTable 1. MILESTONES IN MALIGNANT HlSTlOCYTOSlS (MH) 1924 Aschoff: 1939 Robb Smith: 1966 Rappaport: 1974 Epstein, Kaplan: 1981 Stein: 1982 Rousseau-Merck: 1985 Stein, Mason: 1985

Weiss:

Morgan: Soulie: 1989 Morgan: 1990 Mason: 1990 Barbey, Gogusev: 1992 Nezelof: 1994 Morris: 1995 Shiota: 1996 Miscellaneous: 1986

Reticuloendothelialsystem Histiocytic medullary reticulosis Malignant histiocytosis (MH) MH cell lines (SU-DHL-1, SUP-M2) Ki-1 or CD30 positivity of Reed-Sternberg cells MH cell line (DEL) Anaplastic large cell lymphoma (ALCL)-Reed-Sternberg cells and MH cells are derived from activated lymphoid cells Frequent immunoglobulin and T cell-receptor rearrangements in histiocytic neoplasms 5q35 bp in MH (SU-DHL-1, SupM2) with t(2;5) 5q35 bp in MH with t(5;6) Expression of c-fms in MH cell lines CD30 positivity in 5q35 ALCLs Macrophage differentiation of t(5;6)(q35;p21) cell line The 5q35 bp repesents the hallmark of MH The t(2:5) is associated with NPM/ALK fusion gene Polyclonal antibody against p80 chimeric protein Genotypic heterogeneity of ALCLs Distinction between NPM/ALK+ and NPM/ALK- neoplasms

NPM/ALK = nucleophosmin/anaplastic lymphoma kinase.

MALIGNANT HISTIOCYTOSIS

447

Figure 1. MH-derived DEL cell line. Morphology of DEL cells on cytospin smears (passage 91). MH DEL cells are large with a basophilic vacuolated cytoplasm showing numerous cytoplasmic protrusions. The nucleus is provided with numerous and large nucleoli (Giemsa, original magnification x 1250).

tration through the lymph nodes. As was noted in many early descriptions and well acknowledged by the term histiocytic medullary reticulosis, initially coined by Scott and Robb Smith, the neoplastic cells seem to proceed through the sinus network of the lymph node, sometimes sparing the follicles and the general architecture of the lymph nodeffi, (Fig. 3). However, when the proliferation is diffuse and massive and outgrows its capsule, or if it involves tissues such as skin, bone, or soft tissues, this useful topographic pattern is missing and makes the diagnosis so difficult that it often necessitates repeat biopsies to obtain additional In advanced lesions, some fibrotic changes can be seen that raise the question of a diagnosis of Hodgkin’s disease. Recently, it has been acknowledged that the histologic picture could be different from that of classical ALCL, and it could take the atypical appearance of a large cell, immunoblastic, or even follicular type lymph0rna.9~ The visceral involvement appears to be patchy and frequently nodular, and it calls into question in some circumstances the value of a blind liver or bone marrow biopsy.= In most cases, bone marrow smears yield little or show only occasional isolated and atypical cells, the recognition of which is difficult. In these situations, definite diagnosis requires additional immunocytochemical, chromosomal, and molecular data. CYTOCHEMISTRY

Because the neoplastic cells are extremely immature, usually necrotic, and frequently intermingled with locally activated lymphoid and histiocytic cells,

448

GOGUSEV & NEZELOF

Figure 2. Lymph node biopsy from a 7-year-old boy. The MH cells intermingled with numerous lymphoid cells and macrophages in this biopsy make identification difficult. They appear as large, poorly limited cells provided with a blurred nucleus (hematoxylin-eosin, original magnification x 900).

the interpretation of the cytochemical data is sometimes difficult. As anticipated, the MH cells react positively with acid phosphatase, nonspecific esterase, and most lysosome-associated enzymes.13,63 The results of the immunostaining are more characteristic, exhibiting most typically a CD30 positivity as revealed by Ki-1 or BerH2 specific antibodies (Fig. 4). CD30 positivity is a prerequisite for any MH diagnosis?z,90, 93 Also highly suggestive, albeit not constantly present, is the unexpected positivity for the epithelial membrane antigen (EMA) and the CD25 (IL-2 r e ~ e p t o r ) . " , Such ~ , ~ ~CD30, CD25, and EMA triple positivity provides a unique and diagnostic immunocytochemical profile for this hematopoietic neoplasm. Additionally, the MH cells express strongly the MHC class I1 and more weakly the transferrin receptor CD71 molecule.2, 57, 63 Although B-cell reactivity is not usually recorded, the immunocytochemical investigations detecting T-cell antigen or antigens, associated with histiocyte/macrophage cells, frequently yield inconstant and conflicting results that have largely contributed to denying the histiocytic nature of MH and fueling the nosologic debate (see later discussion). MALIGNANT HlSTlOCYTOSlS PERMANENT CELL LINES In view of certain conflicting data regarding the true nature of the proliferative cells, cell culture appeared to be a reliable method for selection of the cell in question and study of its behavior and its capacity for differentiation, as well as of the nature of the biologic products released. To date, only a few MH-

MALIGNANT HISTIOCYTOSIS

449

Figure 3. Microscopic picture of the MH proliferation involving the subcapsular sinus and the interfollicularspaces. The follicles remain easily identifiable. Note the presence of some multinucleated cells. Lymph node biopsy of a 15-year-old girl (hematoxylin-eosin, original magnification x 450).

Figure 4. Note the CD30 positivity of MH cells infiltrating the subcapsular sinus and mimicking secondaty deposits. Lymph node biopsy from a 7-year-old boy (Ber-H2 immunostaining, original magnification x 250).

450

GOGUSEV & NEZELOF

derived permanent cell lines are available. Most of them have been isolated from pleural effusions in children with a disseminated disease identified as MH by their pediatricians (Table 2). The SU-DHL-1 cell line (for Stanford University diffuse histiocytic lymphoma) was isolated in 1974 by Epstein and Kaplan.30 Subsequently, SUPM2,31,56 DEL,78,89 and Karpas cell lines3zwere isolated and made possible comparative studies and additional investigations. Remarkably representative of the original cell, they all react positively with CD30, CD25, EMA, and HLADR antigens.57,a, CHROMOSOMAL ABNORMALITIES: THE t(2;5) AND THE 5q35 BREAKPOINT

Following the first report by Morgan et a1%in 1986, chromosomal investigations performed on several MH cell lines revealed the occurrence of a constant breakpoint located on the long arm of chromosome 5 (5q35bp). Most often, this abnormality has been found associated with a second breakpoint involving the short arm of chromosome 2 and a reciprocal translocation t(2;5). Moreover, translocations involving various regions of chromosomes 1, 3, and 6 were also reported, underscoring the value of the 5q35b~.5~,The permanent occurrence of this breakpoint focused attention on the genetic material located in this chromosomal area. One of the possible candidate genes mapping to the altered band 5q35 and able to contribute to this disorder was the FLT4 receptor tyrosine kinase gene, but its involvement has been ruled out by Armstrong et al.* More recently, using several MH-related cell lines including SU-DHL-1, SUPMZ, and its own cases with t(2;5) as material of investigation, Morris et alS8 made a molecular analysis of the chromosomal region and demonstrated that the translocation specifically involved two genes: the NPM located on 5q35 and the ALK gene located on 2p23, resulting in a NPM/ALK fusion gene. The

Table 2. CELL LINES WITH 5q35bp: SOURCE, GENOTYPE AND PHENOTYPIC EXPRESSION ~

SU-DHL-1 Sexlage (yrs) Source, pleural effusion Translocation NPMIALK CD30 + EMA HLADR CD68 CD71 CD3 CD5 B-cell marker Gene rearrangement Transcripts

MI10

+ t(2;5) + +

SUPM2 F15

+ +

+ t(2;5) + + + + + +

TCRp R1 -

TCRp R2 low

DEL MI12

+ t(5;6) + + + + +

KarDas 299 MI25 LN t(2;5)

+ + + +

lgJH R1 -

‘Not documented in the original publication. LN, lymph node; R1 (mono-allelic) or R2 (bi-allelic) rearrangement following EcoRl or Hindlll digestion.

MALIGNANT HISTIOCYTOSIS

451

NPM (for nucleophosmin) is a gene already known to encode a nucleolar phosphoprotein ensuring the shuttle transport of molecular components from the nucleolus to the cytoplasm and thus contributing to the assembly of ribosomal protein^.^ Apparently, the NPM gene is transcribed in numerous categories of cells, its transcription being associated with mitotic activity, increasing in G1 and decreasing in G2 phase of the cell cycle. The ALK (for anaplastic lymphoma kinase) is structurally mapped to 2p23. Its full gene size and the size of its transcript are not yet precisely determined. As with many oncogenes, the ALK gene codes for a membrane receptor with similarity to the insulin receptor subfamily of kinases. The ALK gene is strongly expressed in chorionic, fetal tissues, embryonic tumors, and rhabdomyosarcoma-derived cell lines. Surprisingly, spontaneous expression of ALK specific transcripts was not observed in hematopoietic and lymphoid cell lines nor in spleen, thymus, and peripheral blood leukocytes. Today, the 5‘NPM, 3’ALK, and NPM/ALK junction oligonucleotides described by Morris et aF8 are commercially available for reverse transcription-polymerase chain reaction (RT-PCR) analysis. The PCR analysis may increase the positive results by the use of long template specific for the NPM/ALK gene.Io3The NPM/ALK fusion gene encodes an 80-kd hyperphosphorylated protein. A rabbit polyclonal antibody against this p80 protein has been developed by a Japanese team, making it easier and quicker to detect the fusion protein.88More recently, a monoclonal antibody (ALK-1) recognizing a formalinresistant epitope present in NPM/ALK protein has been developed by English researcher^.^^ Although the reliability of these immunostainings has yet to be confirmed, the first reports indicate a satisfactory concordance between the results obtained using p80 antibody and those using the RT-PCR.51,88 In this respect, both NPM/ALK gene and positive p80 staining have been found in all cases to carry a t(2;5) translocation. Interestingly, they have also been found in variant translocations involving either chromosome 1 or chromosome 6, as in the case of the DEL cell line (ref. 11, our results, and Fig. 5). These findings indicate that the neoplastic transformation is not exclusively restricted to a unique translocation, a situation already known in Burkitt’s lymphoma, Ewing’s sarcoma, and promyelocytic leukemia. One explanation for this unexpected molecular finding would be the occurrence of either a cryptic 2p23 translocation or the activation of another gene encoding a tyrosine kinase receptor, which is indeed a common activity among several categories of oncogenes. From these data, it is determined that the 5q35bp is the most frequent, if not a constant, chromosomal abnormality/, 64, 75 Present in all MH-derived cells, and frequently discovered either cytogenetically or through its p80 expression, this 5q35bp can currently be considered to be characteristic of this condition and should be integrated as an essential precondition in the diagnosis of CD30positive neoplasms. Furthermore, tissue culture conditions make it possible to do additional investigations such as molecular analysis and drug-induced differentiation, which are able to provide valuable insights into the nature of these cells. Some of these investigations have been performed by our group using the t(5;6)(q35;p21)DEL cell line. When submitted to opsonized latex particles or previously opsonized red blood cells, these in vitro proliferative cells can develop an evident phagocytic activity. This is further associated with a reduction of nitro blue tetrazolium (NBT), revealing a respiratory burst cascade, a property usually restricted to myelomonocytic lineages.6 The SU-DHL-1 and DEL cell line, both derived from MH, express the c-fms proto-oncogene, a product highly homologous to M-CSF-R, the macrophage 39, 57 (Fig. 6). Expression of M-CSF is restricted to fetal growth factor receptor37* and embryonic tissues and to histiocyte/macrophage cells.19,71,84 Although

452

GOGUSEV & NEZELOF

B Figure 5. RT-PCR analysis of malignant histiocytosis (DEL), histiocytic lymphoma (SUDHL-l), U937 and promyelocytic leukemia (HL-60) derived cell lines for expression of the NPM/ALK fusion transcript. Total RNA was extracted from the cell lines and, after reverse transcription, was amplified for the chimeric NPM-ALK cDNA (A). After transfer and sequential hybridizationwith NPM-ALK junction-specificoligonucleotidewas end-terminally labeled with y-32P dATP and exposed to radiographic film (6). Control = internal negative control for the RT-PCR reaction. M = size marker. Note presence of NPM/ALK transcript in DEL and SU-DHL-1 cells but its absence in U937 and HL-60 cell lines.

MALIGNANT HISTIOCYTOSIS

I

28 S

-

453

TPA

M-CSF

18S-

28 S

-

-

c fms

18S-

18S-

-

I3 actin

Figure 6. Expression of CSF-1(M-CSF) and its receptor c-fms (M-CSR) mRNA in phorboldiester (TPA) treated DEL cells. DEL cells were cultured in both the absence (DEL) and presence of 33 nM TPA, and harvested at various times, as indicated. RNA was isolated and subjected to Northern blot analysis with =P dCTP labeled CSF-1, c-fms and pactin probes.

mapped to 5q33.34,*, 73, 79, 87 close to the 5q35bp, the c-fms (M-CSF-R) gene has 57 When these cells are induced to differentinot been found to be ate, the transcription of the c-fms is downregulated in parallel with that of cmyc and c-myb, whereas the expression and release of its ligand, the M-CSF, is increased, revealing the existence of a specific regulatory program, usually restricted to the cells of the histiocyte/macrophage lineage.38,39 Among various cytokines involved in the inflammatory process, tumor necrosis factor TNF-a and IL-1 are regarded as crucial molecules, the main source being monocytes, macrophages, and dendritic cells. Interestly, TNF-a is constitutively expressed by DEL cells, and the differentiation action of phorbol diester (TPA) results in a threefold increase of TNF-(r at both mRNA and protein levels. Comparatively, whereas spontaneous expression of IL-1 is not observed in DEL cells, the stimulation by TPA induces IL-1 expression within 3 hours.4O DIFFERENTIAL DIAGNOSIS

Because MH cells express CD30 antigen, as do Reed-Sternberg cells, MH has sometimes been regarded as a disseminated variety (Stage IV) of Hodgkin’s

454

GOGUSEV & NEZELOF

di~ease.3~ This assumption has recently been supported by molecular studiesthat is, the presence of NPM/ALK chimeric transcript in 11 of 13 patients with Hodgkin's These results, however, have not been confirmed by most of the investigators using the NPM/ALK probe elaborated by Morris et al.", 25, 27, 44, 50, 51, 97, lo3 Of note, whereas Hodgkin's disease is almost exclusively a nodal disease, MH is usually characterized by multiple and extranodal localizations. The leukemias involving the monocytic/histiocytic lineage are today classified as 5A or 5B. Acute myeloid leukemias, because they always imply, to varying degrees, the presence of a myeloblastic proliferation, are thought to reflect the malignant transformation of a stem cell common to both lineages. Some of these myelomonocytic leukemias represent the end stage of a myelodysplastic syndrome, sometimes exhibiting a monosomy 7.=*98 Moreover, these hemopathies are characterized by a widespread dissemination of monoblastic/ monocytic cells involving the bone marrow and the blood in the early stage. Certain acute lymphophagocytic syndromes may mimic the morphologic manifestations of MH and cause diagnostic problems,4', because they are also characterized by a disseminated cell proliferation and activation with frequent atypical cytologic features of the macrophage system. These difficulties do not concern the familial form but essentially the viral (most often Epstein-Barr virus) and drug-associated forms, often revealing an underlying immunodeficiency state15with overproduction of monokines.12 Although convincingly documented in adults, the real existence of localized histiocytic lymphoma or interdigitating reticulum cell sarcoma in children remains highly q~estionable.9~ In fact, most of the recorded cases seem to represent a monomorphic form of malignant fibrous histiocytoma.I8 The differential diagnosis should also include some unusual and poorly defined proliferative disorders such as lymphomatoid granulomatosis, in which the skin or lung manifestations are prominent.59Also perplexing is the erythrophagocytic T-cell lymphoma, a condition reported to resemble MH that also occurs in a young p ~ p u l a t i o nThis . ~ ~ condition once again raises the nosologic debate over the border between T-cell lymphomas and the histiocytic reactive or neoplastic proliferations. NOSOLOGIC DISCUSSION

The main diagnostic problem remains the distinction between MH and lymphoma, and more precisely ALCL. As already stated, it is more a nosologic than a differential discussion. The opinion that MH belongs to the ALCL group is based on the following arguments (Table 3): (1)the CD30 positivity, considered as a reliable marker of Reed-Sternberg cells; (2) the frequent expression of lymphoid markers such as CD3, CD2, CD4, CD7, CD8, CD43, and CD45R02', 55,97; and (3) the frequent underlying existence of rearrangements of the genes encoding the T-cell receptor p or the variable domain of the heavy chain gene.55.65,95 From these immunocytochemical data, it was concluded that MH, regarded as a malignant histiocytic proliferation, did not exist and should be regarded as CD30 + ALCL of T-cell origin.24,49, M, lo7 Moreover, reassessment of cases previously recorded as MH supported this new histogenesisP, lo5,lffi today widely accepted in the modern classifications of lymphomas. Grouped together, these arguments are impressive; however, taken individually, none of them is conclusive, and it is possible to appeal against this rigid concept and put forward the following arguments:

MALIGNANT HISTIOCYTOSIS

455

Table 3. 5q35 bp (NPM/ALK) MALIGNANCIES: MH VERSUS ALCL Arguments for Histiocytic Origin

Arguments for Lymphoid Origin

Class adherence lrnmunodependent phagocytosis Reduction of NBT Absence of expression of CD3 Strong expression of MHC class I1 Frequent expression of CD68, CD71 Constitutive expression and modulation of: c-fms and M-CSF Production of TNF-a and IL-l*

CD30 positivity Occasional expression of lymphoid markers: CD3, CD2, CD4, CD5, CD25 Frequent rearrangementof genes encoding T-cell receptor (TCR-p) and Ig(JH)

___

~

*TPA induced. MH = malignant histiocytosis; ALCL = anaplastic large cell lymphoma; NBT = nitro blue tetrazolium; M-CSF = macrophage colony-stimulating factor: TNF = tumor necrosis factor; IL = interleukin.

(1)Most of the previously mentioned studies have been performed in cases selected only on the basis of their CD30 positivity; a number of them lacked adequate clinical data and represented a mix-up of adult and child cases. (2) Today, it is clearly established that the CD30 positivity is not restricted to the lymphoid lineage and represents an activation-associated marker, belonging to the nerve growth factor/tumor necrosis factor receptor family, which can be also expressed in different situations including macrophage activation by various stimuli.’,2, 68 (3) Among T-cell immunocytochemical markers, only the CD3 molecule can be regarded as specific to the T-cell lineage. Indeed, the CD2, CD4, CD5, CD25, and CD4ORO can be expressed by several categories of hematopoietic cells, including B cells and histiocytic and dendritic cells. Thus, a critical review and a re-reading of data tables of previous publications reveal that the CD3 positivity of the “CD30 cells” is in fact very uncommon. Restricted to the under 20-yearold population, CD3 positivity is absent in most of the published series.s,36, 81 In this regard, the CD3 marker molecule was not documented in the series recorded by EbrahimZ7and Wellman,Io3and was found in only one of seven cases reported by in 2 of 17 cases recorded by Agnarsson? and in 3 of 3 cases in the series of Oshima.66On the other hand, none of the CD30+ cell lines with 5q35 bp (see Table 2 ) express a CD3 positivity, nor any presumed other T- or B-cell marker. Furthermore, several publications point out that the histochemical data are sometimes unclear, and that the proliferative cells frequently react either poorly or with several conflicting antigens. Also, cells of myelocyte/monocyte/ histiocyte origin can aberrantly express T- or B-cell markersj6 Conversely, although frequently not investigated, several studies mention clearly a CD68 positivity of these cells, that is, in two of two cases in the study of Carbone et all3 and in four of seven cases in the series of Mason et al.55On the whole, the results of these immunocytochemical studies in the under 20-year-old population are equivocal and lead one to consider cautiously such CD30+ neoplasms as malignancies of ”unclear lineage.”36,lol, lo2 (4) The results of genotypic investigations, in general poorly documented, are more convincing but still conflicting. For example, in a series of 30 CD30+ lymphomas (adults and children combined), O’Connor et a16 demonstrated that 16 of them were of T-cell origin, six were of B-cell type, and eight remained unclassified. Restricted to the under 20-year-old population, the reported data appear similar. For instance, in a series of 13 cases of ALCLs associated to

456

GOGUSEV & NEZELOF

5q35bp, Mason et a155reported a rearrangement of the TCRP gene in four cases, rearrangement of IglH in three, and a germline configuration of both these gene markers in three cases. Few cases are indeed homogeneous and characterized by both a CD3 positivity and a biallelic TCRP rearrangement obtained by the use of two restriction enzymes and documented by the expression of a double transcript.66,75 Among the CD30+ cell lines, only SUPM2 cells showed the presence of two rearranged bands of the TCRP gene following EcoRI digestion with a very low transcript level, the others revealing either a single rearranged TCRp band with no transcript or an IgJH-rearranged band with no transcripts (see Table 2). Sometimes the data are perplexing, because they point out the coexistence of a rearrangement for both TCRP and IgJH in cells expressing CD68.&Conversely, MH in adults, exhibiting rearrangement of the heavy chain gene and evidence of monocyte/macrophage lineage, has also been Indeed, discrepancies between the genotype and phenotype have been reported and interpreted as the presence of a basic heterogeneity within the neoplastic processzo,8o or the possible involvement of a multipotential stem cell.'" Nevertheless, it is well established today that a definite chromosomal abnormality is not necessarily restricted to a cell lineage, and it may occasionally involve close or even not close hematopoietic lineages. The tllq23 associated mixed lineage leukemia (MLL) appearing either as MLL or LLL represents a typical example of this situation.8z,91More perplexing, NPM/ALK fusion in RNA expression has been recently mentioned in few B lymphocytes of healthy donors.%It is likely that the extensive use of RT-PCR and especially p80 immunostainings will increase the frequency of these unexpected results and prevent most investigators from basing any neoplastic classification on the sole criterion of a given chromosomal abnormality. It is clear, however, that the identification of t(2;5) with a 5q35bp aberration has provided new data and added a new dimension to the MH nosologic discussion." 26, -* Of critical importance is the fact that only a limited number of CD30 + neoplasms present these chromosomal abnormalities. According to recent data (Table 4), the percentage of t(2;5) varies from 15% to 73%. Interestingly enough, as could be expected, the percentage of this cytogenetic abnormality is higher in young populations and in CD30 + disorders of the null type.lo6 Moreover, Weisenburger et ap7 and Sandlund et als3have recently reported that the t(2;5) cytogenetic abnormality is not exclusively related to the CD30 + ALCL proliferations. They also can be seen in non-ALCLs such as diffuse large cell, immunoblastic, and even follicular mixed cell type lymphomas. From these data, it is becoming evident that the designation of CD30 neoplasms covers diferent genomic conditions and therefore should be regarded useful as a primary diagnostic approach but insufficient or inadequate to define a neoplastic entity. It also

Table 4. CD30+ NEOPLASMS: PERCENTAGE OF 5q35bp AND NPM/ALK FUSION GENE DETECTED BY POLYMERASE CHAIN REACTION OR p80 IMMUNOSTAINING' Ebrahim et al (1990) Bullrich et al (1994) Lopategui et al (1994) Herbst et al (1 995) Wellman et al (1995) Shiota et al* (1995) Lamant et al" (1996) Weisenburger et al (1996)

42 15 16 24 45 30 73 66

MALIGNANT HISTIOCYTOSIS

457

ensues that the results of immunocytochemical and molecular investigations previously recorded from series of CD30+ lymphoma including children and adults should be submitted to control and revision. As was previously suggested and recently agreed to by Cline,17the chromosomal abnormalities should be regarded as a hallmark of this entity. The chromosomal aberrations are numerous, and changes always occur in tissue culture. One therefore should focus, as is done in leukemia16and in embryonic tumors, on the most frequent anomaly occurring in these disorders, namely the t(2;5) and 5q35bp. Because these abnormalities seem to be associated with a satisfactory response after treatment and with a more favorable course, this new basis of definition is of more than just speculative interest.QIffiIt is obvious that these remarks lead one to ponder anew the nosologic debate surrounding 5q35bp neoplasms. The principal arguments for a lymphoid and/or a histiocytic origin are summarized in Table 3. In addition, some indirect arguments can be put forward: (1) the sinusoidal infiltration of neoplastic cells within the lymph nodes, which is not the usual manner of progression for a lymphoma; (2) the fact that the NPM/ALK transcript has been isolated by Morris from t(2;5) cell lines that do not express T-cell antigen but CD68 antigen58,88; (3) the argument that any of the three different monoclonal antibodies produced using SU-DHL-1 as a fusion cell failed to react with B or T cells in frozen tissue; one of them reacts with the nuclear membrane of histiocytes and interdigitating cells as well as with the cell membranes of neoplastic cells of true histiocytic lymphoma and MH45;(4)among the countless chromosomal abnormalities recorded in lymphoid leukemia and lymphoma, it is surprising that none of them (except the 5q-associated myelodysplasia) concerns the long arm of chromosome 516,70, 82; and (5) the fact that the ALK gene constantly implicated in t(2;5) and variant translocations, albeit expressed in many categories of cells, is not constitutively expressed in lymphoma cells.58 In summary, from these conflicting data, it appears that the histogenesis and hence the nosologic place of this neoplastic condition is not yet After a half century, Robb-Smith admits that the condition he described initially remains a "cell teaser."77 Be that as it may, all recent chromosomal and molecular investigations lead one to think that ALCL is heterogeneous group of neoplastic entities artificially grouped under the heading of certain positive immunohistochemical markers. In this context, the chromosomal data and the related molecular abnormalities seem to represent a better criterion for the identification of these disorders and a better field of investigation. In our opinion, the 5q35bp related MH and ALCL are very closely related disorders, if not the same condition.@ What denomination would be most appropriate? Should we return to the old denominations such as difuse kistiocytic lymphoma, neoplasm of uncertain or, as we suggest, the time-honored term of MH? Because the chromosomal abnormalities currently appear to be the hallmark of this neoplastic process, the term 5q35bp associated kematopoietic neoplasm can also be suggested. SUMMARY

Although myelomonoblastic leukemia is thought to originate from a malignant transformation of the stem cell of the mononuclear phagocyte system, malignant histiocytosis (MH) is classically assumed to represent a malignant change of the terminal and fixed elements of this system. Indeed, MH is charac-

458

GOGUSEV & NEZELOF

terized by the proliferation of large, clear, pleomorphic, ”histiocytic-like” HLADR and CD30+ cells resulting in a nodal and extranodal disseminated neoplasm affecting preferentially and severely children and young adults. Although there is broad agreement on the clinicopathologic presentation of this condition, there is currently quite a controversy over the T-lymphoid or histiocytic origin of the proliferative cells that results in a nosologic discussion between the anaplastic large cell lymphoma (ALCL) advocates and the MH supporters. This article has dealt mainly with this nosologic discussion and with the contributions provided by the investigations performed on MH permanent cell lines. These in vitro studies have demonstrated that the proliferation is characterized by a unique chromosomal abnormality, the 5q35bp usually associated with a t(2;5) translocation generating a fusion gene NPM/ALK and the subsequent translation of p80 protein. Although it is known that no single chromosomal abnormality is strictly restricted to a cell lineage, this 5q35bp and associated translocations seem today to represent the hallmark for this condition. In view of these chromosomal aberrations, the CD30 + ALCLs represent a heterogeneous group because 15% to 50% express the NPM/ALK fusion gene. In addition, these in vitro investigations have shown that 5q35bp proliferative cells are glass-adherent, can develop an immunodependent phagocytosis, and are able to reduce NBT and produce TNF-a.More significantly, they express constitutively the c-fms (the receptor of the macrophage growth factor) and, under TPA stimulation, are able to modulate the expression of this receptor and its ligand, as well as TNF-aand IL-1. None of these cell lines express CD3, but several express CD68 and CD71. In contrast, genomic investigations have shown the underlying existence of monoallelic and even biallelic gene rearrangements for TCRp and IgJH. In view of these discrepancies between the genomic and phenotypic features of these cells, the histogenetic debate should remain open but must take into account these new chromosomal and molecular data.

References 1. Abbondanzo SL, Sat0 N, Straus SF, et al: Acute infectious mononucleosis. CD30 (Ki1) antigen expression and histologic correlations. Am J Pathol 93:698, 1990 2. Agnarsson BA, Kadin M E Ki-1 positive large cell lymphoma. A morphologic and immunologic study of 19 cases. Am J Surg Pathol 12264, 1988 3. Andreesen R, Brugger W, Lohr GW, et al: Human macrophages can express the Hodgkin’s cell-associated antigen Ki-1 (CD30). Am J Pathol 134:187, 1989 4. Armstrong E, Kastury K, Aprelikova 0, et al: FLT4 receptor tyrosine kinase gene mapping to chromosome band 5q35 in relation to the t(2;5) t(5;6) and t(3;5) translocations. Genes Chromosom Cancer 7:144, 1993 5. Aschoff L Das Reticuloendothelial System. Ergebnisse der Inneren Medizin und Kinderheilkunde 26:1, 1924 6. Barbey S, Gogusev J, Mouly H, et al: DEL cell line. A malignant histiocytosis cell CD30+, t(5;6)-q35;p21) cell line. Int J Cancer 45546, 1990 7. Bern-Lemoine E, Brizard A, Huret JL, et al: Malignant histiocytosis: A specific t(2;5)(p23;q35) translocation? Review of the literature. Blood 721045, 1988 8. Bikfer MA, Franklin WA, Larson RA, et al: Morphology in Ki-1 (CD30) positive nonHodgkin’s lymphoma is correlated with clinical features and the presence of a unique chromosomal abnormality t(5;6)(p23;q35). Am J Pathol 14:305, 1990 9. Borer RA, Lehner CF, Eppenberger HM, et al: Major nucleolar proteins shuttle between the nucleus and cytoplasm. Cell 56:379,1989

MALIGNANT HISTIOCYTOSIS

459

10. Buchsbaum RJ, W w a r t z RS:Cellular origins of hematologic neoplasms [editorial]. N Engl J Med 322694,1990 11. Bullrich F, Morris SW, Hummel M, et al: Nucleophosmin (NPM) gene rearrangements in Ki-1 positive lymphomas. Cancer Res 54:273, 1994 12. Caniglia M, Prieur AM, Le Deist F, et a 1 Activitk des monokines au cours du syndrome d’activation histiocytaire [abstract]. Arch Fr Pediatr 47:305, 1990 13. Carbone A, Gloghini A, De Re W, et al: Histopathologic, immunophenotypic and genotypic analysis of Ki-1 anaplastic large cell lymphomas that express histiocyteassociated antigens. Cancer 66:2547, 1990 14. Cazal P:Aspects cliniques et hCmatologiques de la rkticulose maligne. Acta Haematol 765, 1952 15. Chen RL, Su IJ, Lin KH, et al: Fulminant childhood hemophagocytic syndrome mimicking histiocytic medullary reticulosis. An atypical form of Epstein-Barr virus infection. Am J Clin Pathol 96:171, 1991 16. C l i e MJ: Histiocytes and histiocytosis. Blood M.2840, 1994 17. Cline MJ: The molecular basis of leukemia. N Engl J Med 330:328, 1994 18. Cozzutto C, Bronzini E, Bandelloni R, et al: Malignant monomorphic histiocytoma in children. Cancer 48:2112, 1981 19. Datta R, Inamura K, Goldman SJ, et al: Functional expression of the macrophage colony-stimulating factor receptor in human THP-1 monocytic leukemia cells. Blood 79:904, 1992 20. Davey MP, Waldmann TA. Clonality and lymphoproliferative lesions [editorial]. N Engl J Med 315509, 1986 21. Davis E, Wilson M, Weiss L, et al: Malignant histiocytosis: A reassessment of cases previously reported in 1975 based upon paraffin section immunophenotyping studies [abstract]. Mod Pathol 3:24, 1990 22. Dekmezian R, Goodacre A, Cabanillas F: The 2;5 translocation: Is it specific for anaplastic (Ki-1) large cell lymphomas? Mod Pathol 325(A), 1990 23. Delcourt A, Raphael M, Auriol M, et al: L‘histiocytose maligne. Etude anatomique de 6 cas d’expression clinique atypique. Semaine des HBpitaux, Paris 59:1811, 1983 24. Delsol G, A1 Saati T, Gatter KC, et a1 Coexpression of epithelial membrane antigen (EMA) Ki-1 and interleukin-2 receptor by anaplastic large cell lymphomas. Diagnostic value in so-called malignant histiocytosis. Am J Pathol 13059, 1988 25. Dirks WG, Ziborski M, Jager K, et al: The (2;5)(p23;q35) translocation in cell lines derived from malignant lymphomas: Absence of t(2;5) in Hodgkin analogous cell lines. Leukemia 10142, 1996 26. Downing JR, Shurtleff SA, Zielenska M, et al: Molecular detection of the t(2;5) translocation in non Hodgkin’s lymphoma by reverse transcriptase polymerase chain reaction. Blood 85:3416, 1995 27. Ebrahim SAD, Ladanyi M, Desai SB, et a1 Immunohistochemical, molecular and cytogenetic analysis of a consecutive series of 20 peripheral T-cell lymphomas and lymphomas of uncertain lineage, including 12 Ki-1 positive lymphomas. Genes Chromosom Cancer 2:27, 1990 28. Egeler RM, Schmitz L, Sonneveld P, et al: Malignant histiocytosis: Reassessment of cases formerly classified as histiocytic neoplasms and review of the literature. Med Pediatr Oncol 25:1, 1995 29. Elmberger PG, Lozano MD, Weisenburger DD, et al: Transcripts of the npm-alk fusion gene in anaplastic large cell lymphoma, Hodgkin‘s disease and reactive lymphoid lesions. Blood 6:3517, 1995 30. Epstein AL, Kaplan HS: Biology of the human malignant lymphomas. I. Establishment in continuous cell culture and heterotransplantation of diffuse histiocytic lymphomas. Cancer 343851, 1974 31. Epstein AL, Levy R, Kim H, et al: Biology of the human malignant lymphomas. IV. Functional characterization of ten histiocytic lymphoma cell lines. Cancer 422379, 1978 32. Fischer P, Nacheva E, Mason DY, et al: A Ki-1 (CD30) positive human cell line (karpas.299) established from a high grade non Hodgkin’s lymphoma showing a 2;5 translocation and rearrangement of the T-cell receptor beta gene. Blood 72:234, 1988

460

GOGUSEV & NEZELOF

33. Foucar K, Foucar E The mononuclear phagocyte and immunoregulatory effector (M.PIRE) system: Evolving concepts. Semin Diagn Pathol7:4, 1990 34. Freedman MH, Estrov Z, Chan H S Juvenile chronic myelogenous leukemia. Am J Pediatr Hematol Oncol 10261, 1988 35. Frizzera G: The distinction of Hodgkin’s disease from anaplastic large cell lymphoma. Semin Diagn Pathol9291, 1992 36. Fujimoto J, Hata JI, Ishii E, et al: Ki-1 lymphomas in childhood: Immunohistochemical analysis and the simificance of epithelial membrane antigen (EMA) as a new marker. Virchows Arch A ;12:307, 1988 37. Gogusev J, Barbey S, Nezelof C: Genotype markers and proto-oncogene analysis in the CD30 positive malignant histiocytosis with t(5;6)(q35,p21). h t J Cancer 46:106, 1990 38. Gogusev J, Barbey S, Nezelof C: Human malignant histiocytosis CD30-k DEL cell line differentiates into macrophage-like cells when treated with a phorbol diester. Cancer Res 575712, 1991 39. Gogusev J, Barbey S, Nezelof C: Modulation of c-myc, c-myb, c-fos, c-sis and c-fms proto-oncogene expression and of CSF-1 transcripts and protein by phorbol diester in human malignant histiocytosis DEL cell line with 5q35bp. Anticancer Res 13:1043, 1993 40. Gogusev J, Barbey S, Nezelof C Regulation of TNFa and IL-1 gene expression during TPA induced differentiation of malignant histiocytosis DEL cell line t(5;6)(q35;p21). Anticancer Res 1645, 1996 41. Goldberg J, Nezelof C: Lymphohistiocytosis. A multifactorial syndrome of macrophagic activation. Clinico-pathological studies of 38 cases. Hematol 011~014275,1986 42. Gonzalez-Crussi F, Mankomkanok M, Hsueh W Large-cell lymphoma. Diagnostic difficulties and case study. Am J Surg Pathol 11:59, 1987 43. Groffen J, Heisterkamp N, Spurr NK,et al: Regional assignment of the human c-fms oncogene to band 434 of chromosome 5 (HGM7). Cytogenet Cell Genet 37484,1984 44. Herbst H, Anagnostopoulos J, Heinze B, et a1 ALK gene products in anaplastic large cell lymphomas and Hodgkin’s disease. Blood 61694,1995 45. Hsu SM, Hsu PL Aberrant expression of T cell and B cell markers in myelocyte/ monocyte/histiocyte derived lymphoma and leukemia cells. Am J Pathol 134:203, 1989 46. Hsu SM, Pescovitz MD, Hsu Pt: Monoclonal antibodies against SU-DHL-1 cells stain the neoplastic cells in true histiocytic lymphoma, malignant histiocytosis and Hodgkin’s disease. Blood 68213,1986 47. Kadin ME, Kamoun M, Lamberg J: ErythrophagocyticTy lymphoma. A clinicopathologic entity resembling malignant histiocytosis. N Engl J Med 304648, 1981 48. Kamesaki H, Koya M, Miwa H, et al: Malignant histiocytosis with rearrangement of the heavy chain gene and evidence of monocyte-macrophage lineage. Cancer 621306,1988 49. Kaneko Y, Frizzera G, Edamura S, et a1 A novel translocation t(2;5)(p23;q35) in childhood phagocytic large T-cell lymphoma mimicking malignant histiocytosis. Blood 73806,1989 50. Ladanyi M, Cavalchire G, Morris SW, et a1 Reverse transcriptase polymerase chain reaction for Ki-1 anaplastic large cell lymphoma-associated t(2;5) translocation in Hod&in’s disease. Am J Pathol 1451296, 1994 51. Lamkt L, Megetto F, A1 Saati T, et a1 High incidence of t(2;5)(p23;q35) translocation in anaplastic large cell lymphoma and its lack of detection in Hodgkin’s disease. Comparison of cytogenetic analysis, reverse transcriptase-polymerase chain reaction and p80 immunostaining. Blood 87:284,1996 52. Lampert IA,Catowsky D, Bergier N: Malignant histiocytosis: A clinico-pathological study of 12 cases. Br J Haematol4065, 1978 53. Lombardi L, Carbone A, Pilotti S, et al: Malignant histiocytosis: A histological and ultrastructural study of lymph nodes in six cases. Histopathology 2315,1978 54. Lopategui JR, Sun LH, Chan JKC, et a1 Low frequency association of the t(2;5)(p23;q35) chromosomal translocation with CD30 + lymphomas from American and Asian patients. Am J Pathol 146:323,1995 55. Mason DY, Bastard C, Rimokh R, et a1 CD30 positive large cell lymphomas (Ki-1 Y~

MALIGNANT HISTIOCYTOSIS

461

lymphomas) are associated with a chromosomal translocation involving 5q35. Br J Haematol 74:161, 1990 56. Morgan R, Hecht BK, Sanberg AA, et al: Chromosome 5q35 breakpoint in malignant histiocytosis. N Engl J Med 314:1322,1986 57. Morgan R, Smith S, Hecht B, et a1 Lack of involvement of c-fms and N-myc genes by chromosomal translocation t(2;5)(p23;q35) common to malignancies with features of so-called malignant histiocytosis. Blood 732155, 1989 58. Morns SW, Kirstein MN, Valentine MB, et al: Fusion of a kinase gene, ALK to a nucleolar protein gene, NPM in non-Hodgkin's lymphoma. Science 2631281, 1994 59. Myers JL, Kurtin PJ, Katzenstein AL, et al: Lymphomatoid granulomatosis. Evidence of immunophenotypic diversity and relationship to Epstein-Barr virus infection. Am J Surg Pathol 19:1300, 1995 60. Nezelof C: Vers une nouvelle definition de l'histiocytose maligne chez l'enfant. Arch Fr Pediatr 47697, 1990 61. Nezelof C The 5q35bp chromosomal abnormality characterizes certain CD30 positive anaplastic large cell lymphomas offering a new definition of malignant histiocytosis in childhood. Nouv Rev Fr Hematol35:463, 1993 62. Nezelof C, Barbey S Histiocytosis: Nosology and pathobiology. Pediatr Pathol 341, 1985 63. Nezelof C, Barbey S, Gogusev J, et al: Malignant histiocytosis in childhood: A distinctive CD30 positive clinicopathological entity associated with a chromosomal translocation involving 5q35. Semin Diagn Pathol 9:75, 1992 64. Nezelof C, Egeler M, Bucsky P, et al: Malignant histiocytosis in childhood A disease in quest of new nosological criteria. Med Pediatr Oncol 2567, 1995 65. OConnor NT, Stein H, Gatter KC, et al: Genotypic analysis of large cell lymphomas which express the Ki-1 antigen. Histopathology 11:733, 1987 66. Ohshima K, Kikuchi M, Masuda Y, et al: Genotypic and immunophenotypic analysis of anaplastic large cell lymphoma (Ki-1 lymphoma). Pathol Res Pract 186582, 1990 67. Oreschechek K, Merz H, Hell J, et a1 Large-cell anaplastic lymphoma-specifictranslocation t(2;5)(p23;q35) in Hodgkin's disease: Indication of a common pathogenesis? Lancet 34587,1995 68. Pallesen G: The diagnostic significance of the CD30 (Ki-1) antigen. Histopathology 16409,1990 69. Pittaluga S, Pulford K, Wlodarska I, et al: ALK-1 antibody staining pattern in anaplastic large cell lymphoma (ALCL) and ALCL Hodgkin's like [abstract]. USCAP, Orlando, 1997 70. Pui C H Childhood leukemias. N Engl J Med 3321618, 1995 71. Radzun HJ, Kreipe H, Heidorn K, et a1 Modulation of c-fms proto-oncogene expression in human blood monocytes and macrophages. J Leuk Biol44:198,1988 72. Rappaport R Tumors of the hematopoietic system. In Atlas of Tumor Pathology, sect 3, fasc 8. Washington, DC,Armed Forces Institute of Pathology, 1966 73. Rettenmier CW, Sacca R, Furman WL, et al: Expression of the human c-fms protooncogene product (colony-stimulatingfactor-1receptor) on peripheral blood mononuclear cells and choriocarcinoma cell lines. J Clin Invest 773740, 1986 74. Rilke F, Carbone A, Musumeci R, et al: Malignant histiocytosis: A clinicopathologic study of 18 consecutive cases. Tumori 64:221, 1978 75. Rimokh R, Magand JP, Berger F, et a1 A translocation involving a specific breakpoint (q35) on chromosome 5 is characteristic of anaplastic large cell lymphoma (Ki-1 lymphoma). Br J Haematol 71:31, 1989 76. Risdall RJ, McKenna RW, Nesbitt ME, et al: Virus associated hemophagocytic syndrome. Cancer 44:993,1979 77. Robb-Smith AHT: Before OUT time: Half a century of histiocytic medullary reticulosis: A T-cell teaser? Histopathology 17279, 1990 78. Rousseau-Merck MF, Jaubert F, Bach MA, et al: Tumor cell line characterization of a malignant histiocytosis transplanted in nude mice. Virchows Arch A 397171, 1982 79. Roussel MF, Sherr CJ, Barker PE, et al: Molecular cloning of the c-fms locus and its assignment to human chromosome 5. J Virol48:770, 1983

462

GOGUSEV & NEZELOF

80. Rovigatti U, Mirro J, Kitchingman G, et al: Heavy chain immunoglobulin gene rearrangement in acute nonlymphocytic leukemia. Blood 63:1023, 1984 81. Rubie H, Gladieff L, Robert A, et al: Childhood anaplastic large cell lymphoma Ki1/CD30: Clinicopathologic features of 19 cases. Med Pediatr Oncol22:155, 1994 82. Russell N H Biology of acute leukemia. Lancet 349:118, 1997 83. Sandlund JT, Pui CH, Roberts WM, et al: Clinico-pathological features and outcome of children with large-cell lymphoma and t(2;5)(p23;q35). Blood 84:2467, 1994 84. Sariban E, Mitchell T, Kufe D Expression of the c-fms proto-oncogene during human monocytic differentiation. Nature 31664, 1985 85. Sarris AH, Luthra R, Waasdorp M, et al: Genomic DNA PCR defines unique t(2;5) breakpoints in anaplastic large cell lymphoma (ALCL) [abstract]. Blood 143b, 1997 86. Scott RB, Robb-Smith AHT: Histiocflc medullary reticulosis. Lancet 2:194, 1939 87. Sherr CJ, Rettenmier CW, Sacca R, et a1 The c-fms proto-oncogene product is related to the receptor for the mononuclear phagocyte growth factor, CSF-1. Cell 41:665,1985 88. Shiota M, Nakamura S, Ichinohasama R, et a 1 Anaplastic large cell lymphomas expressing the novel chimeric protein p80 NPM/ALK. A distinct clinicopathologic entity. Blood 86:1954, 1995 89. Soulik J, Rousseau-Merck MF, Mouly H, et al: Cytogenetic study of malignant histiocytosis transplanted in nude mice: Presence of translocation between chromosomes 5 and 6 and a unique marker (13q+). Virchows Arch B 50339, 1986 90. Stein H, Mason DY, Gerdes J, et al: The expression of the Hodgkin's disease associated antigen Ki-1 in reactive and neoplastic lymphoid tissue: Evidence that Reed-Stemberg cells and histiocytic malignancies are derived from activated lymphoid cells. Blood 66:848,1985 91. Thirman MJ, Gill HJ, Bumett RC, et al: Rearrangement of the MLL gene in acute lymphoblastic and acute myeloid leukemias with llq23 chromosomal translocation. N Engl J Med 329909, 1993 92. Trtimper L, Daus H, Bonin FU,et a1 NPM/ALK fusion mRNA expression occurs in peripheral blood by B lymphocytes of healthy donors as evidenced by reverse transcriptase polymerase chain reaction (RT-PCR). In Programs and Abstracts of the 38th Annual Meeting of the American Society of Hematology, Orlando, 1996, p 225 93. Tumer RR, Wood GS, Beckstead JH, et al: Histiocytic malignancies: Morphologic, immunologic and enzymatic heterogeneity. Am J Surg Pathol 8485,1984 94. Van den Oord JJ, de Wolf-Peters C, de Vos R, et al: Sarcoma arising from interdigitating reticulum cells: Report of a case, studied with light and electron microscopy and enzyme and immuno-histochemistry. Histopathology 10509,1986 95. Vannier JP, Bastard C, Rossi A, et a 1 Chromosomal t(2;5) and hematological malignancies. Pediatr Hematol Oncol4177, 1987 96. Wamke RA, Kim H, D o r h a n RF: Malignant histiocytosis (histiocytic medullary reticulosis) I. Clinicopathologic study of 29 cases. Cancer 35:215, 1975 97. Weisenburger DD, Cordon BG, Vose JM, et a 1 Occurrence of the t(2;5)(p23;q35) in non Hodgkin's lymphoma. Blood 87:3860,1996 98. Weisgerber G, Schaison G, Chavelet F, et al: Les leuckmies myklo-monocytaires de l'enfant. Arch Fr Pediatr 29:11, 1972 99. Weiss LM, Trela MJ, Cleary ML, et a1 Frequent immunoglobulin and T cell receptor gene rearrangements in histiocytic neoplasms. Am J Pathol 121:369, 1985 100. Weiss LM, Azzi R, D o r h a n RF, et al: Sinusoidal hematolymphoid malignancy: Malignant histiocytosis presenting as atypical sinusoidal proliferation. A study of nine cases. Cancer 981681, 1986 101. Weiss LM, Picker LJ, Copenhauer CM, et al: Large-cell hematolymphoid neoplasms of uncertain lineage. Hum Pathol 19:967, 1988 102. Weiss LM, Lopategui JR, Sun LH, et a1 Absence of the t(2;5) in Hodgkin's disease. Blood 85:2845, 1995 103. Wellmann A, Otsuki T, Vogelbmch M, et al: Analysis of the t(2;5)(p23;q35) translocation by reverse transcription polymerase chain reaction in CD30+ anaplastic large cell lymphomas in other non Hodgkin's lymphomas of T cell phenotype and in Hodgkin's disease. Blood 862321, 1995 104. Wellman A, Fest T, Otsuki T, et a1 Detection of NPM/ALK DNA rearrangement in

MALIGNANT HISTIOCYTOSIS

463

CD30 positive classical ALCL with long template PCR. Pathol Res Pract 192:388(A), 1996 105. Wilson MS, Weiss LM, Gatter KC, et al: Malignant histiocytosis. A reassessment of cases previously reported in 1975, based on paraffin section immunophenotyping studies. Cancer 66:530, 1990 106. Wu CD, Greiner TC, Vose JM, et a 1 Anaplastic large cell lymphoma in adults: A clinicopathologic study of 31 cases [abstract]. USCAP, Orlando, 1997 107. Xerri L, Horschowski N, Payan MJ, et al: Genotypic analysis in large cell lymphomas expressing a restricted set of differentiation antigens. Pathol Res Pract 186:317, 1990

Address reprint requests to Jean Gogusev, MD Charge de Recherche B l'LNSERM U 90 HBpital Necker Enfants Malades 149, rue de 9 v r e 75743 PARIS Cedex 15 France