Morphological and phenotypic features in pediatric large cell lymphoma and their correlation with ALK expression and the t(2;5)(p23;q35) translocation.

Anaplastic large cell lymphoma (ALCL) was proposed as a clinicopathologic entity over 14 years ago, but has been somewhat controversial due to the variability of its defining features and variable occurrence in different age-groups. To evaluate this entity in a pediatric population, 36 cases of childhood large cell lymphoma were evaluated for abnormalities of the anaplastic lymphoma kinase (ALK) gene that has been associated with ALCL morphology and immunophenotype. ALK abnormalities were evaluated by assay for the t(2;5)(p23;q35) translocation by RT-PCR and/or expression of NPM-ALK fusion protein by immunohistochemistry. Results showed 17 patients to have evidence of ALK gene expression. All of these children (mean age, 9.3 years) had tumors that were of T-cell phenotype (with the exception of a single case of null phenotype) and that expressed CD30. In contrast, 19 children with no evidence of ALK expression were older (mean, 12.7 years), and the majority (12/19) had tumors of B-cell phenotype. CD30 was also diffusely expressed in 8 of these 19 tumors. The difference in mean age between the two groups was statistically significant (P = 0.015). In three cases tested for both ALK and the t(2;5), ALK protein was detected in the absence of the t(2;5) translocation but no cases showed the reverse pattern, consistent with ALK fusion to genes other than NPM or activation of the ALK gene by another mechanism. These findings provide further support that ALK-positive ALCL is a distinct pathologic entity among pediatric large cell lymphomas primarily characterized by expression of T-cell markers, CD30, and EMA, and by a younger mean age.

Anaplastic large-cell lymphomas of B-cell phenotype are anaplastic lymphoma kinase (ALK) negative and belong to the spectrum of diffuse large B-cell lymphomas.

There is controversy in the literature as to whether anaplastic large-cell lymphoma of B-cell phenotype is related to the t(2;5)-positive T- or 'null' cell lymphoma of the same morphology. We report a study of 24 lymphomas with morphological features of anaplastic large-cell lymphoma which expressed one or more B-cell markers and lacked T-lineage markers. Clinical features were more in keeping with large B-cell lymphoma than with classical t(2;5)-positive anaplastic large-cell lymphoma, and immunostaining for anaplastic lymphoma kinase (ALK) protein provided no evidence for the (2;5) translocation (or one of its variants). The staining patterns for CD20 and CD79 were typical of diffuse large B-cell lymphoma, CD30 expression was variable, and most cases (15/22) lacked epithelial membrane antigen (EMA). These findings support the view that 'B-cell anaplastic large-cell lymphoma' is unrelated to t(2;5)-positive (ALK-positive) lymphoma, and that it represents a morphological pattern occasionally encountered among diffuse large B-cell lymphomas. By the same reasoning, most tumours diagnosed as 'ALK-negative anaplastic large-cell lymphoma of T-cell or null phenotype' probably belong to the spectrum of peripheral T-cell lymphomas.

Detection of t(2;5) in anaplastic large cell lymphoma: comparison of immunohistochemical studies, FISH, and RT-PCR in paraffin-embedded tissue.

Anaplastic large cell lymphoma (ALCL) is associated with the t(2;5)(p23;q35) translocation involving the anaplastic lymphoma kinase gene (ALK) and the nucleophosmin gene (NPM), which result in expression of a novel fusion protein, NPM-ALK (p80). Clinicopathologic studies have shown that ALK expression in ALCL is associated with improved 5-year survival rates when compared with ALCL lacking ALK expression. This study used paraffin-embedded tissue to compare interphase fluorescence in situ hybridization (FISH) and reverse transcriptase-polymerase chain reaction (RT-PCR) for the detection of t(2;5) with immunohistochemical analysis for the detection of ALK protein expression in 27 patients with CD30-positive ALCLs. ALK protein expression was detected with ALK1 antibody in 14 of the 27 patients. The neoplastic cells in 13 of these 14 lymphomas reacted with the p80NPM/ALK antibody. FISH, using a two-color ALK DNA probe, correlated 100% with the immunohistochemical results: a translocation involving the ALK gene was detected in all 14 lymphomas that reacted with anti-ALK1. RT-PCR, performed on 21 lymphomas, detected NPM-ALK mRNA in five of the lymphomas, all of which reacted with anti-ALK1 and showed ALK gene rearrangement by FISH. Lymphomas showing ALK1 reactivity occurred in a younger patient population (median age, 19.5 years) and were associated with improved 5-year survival rates (84%), as compared with lymphomas lacking ALK1 reactivity (median age, 68.0 years; 5-year survival rate, 35%; p = 0.008). We conclude that immunohistochemical studies, using antibody ALK1. and FISH for ALK gene rearrangement are equally effective for identifying patients with ALCL who have a favorable clinical outcome.

Detection of normal and chimeric nucleophosmin in human cells.

In anaplastic large-cell lymphoma (ALCL), the (2;5) chromosomal translocation creates a fusion gene encoding the 80-kD NPM-ALK hybrid protein. This report describes three new monoclonal antibodies, two of which recognize, by Western blotting, the N-terminal portion of NPM present in the NPM-ALK fusion protein and also in two other NPM fusion proteins (NPM-RARalpha and NPM-MLF1). The third antibody recognizes the C-terminal portion (deleted in NPM-ALK) and reacts only with wild-type NPM. The three antibodies immunostain wild-type NPM (in paraffin-embedded normal tissue samples) in cell nuclei and in the cytoplasm of mitotic cells. Cerebral neurones, exceptionally, show diffuse cytoplasmic labeling. In contrast to normal tissues, the two antibodies against the N-terminal portion of NPM labeled the cytoplasm of neoplastic cells, in four ALK-positive ALCL, reflecting their reactivity with NPM-ALK fusion protein, whereas the antibody to the C-terminal NPM epitope labeled only cell nuclei. Immunocytochemical labeling with these antibodies can therefore confirm that an ALK-positive lymphoma expresses NPM-ALK (rather than a variant ALK-fusion protein) and may also provide evidence for chromosomal anomalies involving the NPM gene other than the classical (2;5) translocation.

Nucleolar localization of the nucleophosmin-anaplastic lymphoma kinase is not required for malignant transformation.

The (2;5)(p23;q35) lymphoma-associated chromosomal translocation creates a novel fusion gene that incorporates parts of the anaplastic lymphoma kinase (ALK) receptor tyrosine kinase and nucleophosmin genes. We report here that the product of this fusion gene accumulates within the nucleoli of neoplastic cells, and that previous reports of a predominantly cytoplasmic localization for the protein represent a tissue-processing artifact. However, nucleolar accumulation of nucleophosmin-ALK may not be necessary for its oncogenic action, because an ALK protein expressed in a lymphoma carrying a variant (1;2) chromosomal translocation did not accumulate in nucleoli. Furthermore, an engineered hybrid TPR-ALK protein can transform rodent fibroblasts and produce lymphomas in mice while remaining confined to the cytoplasm. We propose that the transforming action of ALK may not be reliant on its nucleolar localization, a hypothesis that may have implications for studies of other proteins involved in oncogenesis that are relocalized after the creation of fusion genes.

Systemic therapy with 3BIT, a triple combination cocktail of anti-CD19, -CD22, and -CD38-saporin immunotoxins, is curative of human B-cell lymphoma in severe combined immunodeficient mice.

We demonstrate in these preclinical studies that all severe combined immunodeficient mice injected with the human B-cell lymphoma cell line Ramos are cured when treated with a combination of anti-CD19, -CD22, and -CD38-saporin immunotoxins (ITs; termed 3BIT). Each component IT used individually did not cure the majority of animals but did significantly prolong their survival compared with PBS sham-treated controls, although the majority succumbed eventually to disease. The very significant improvement obtained with the three-IT combination 3BIT was not due to an antibody or antibody-plus-IT effect. We postulate that by targeting against these three cell surface molecules, we have effectively ensured delivery of saporin to each lymphoma cell with growth potential within the tumor, thus overcoming the problems of heterogeneity of target antigen expression that can limit the therapeutic efficacy of single-IT therapy or even two-IT combination therapy. These "proof of principle" findings have an obvious important bearing on antibody-based therapies for cancer and provide the rationale needed for the design and implementation of clinical trials with such combinations.

TAL-1 protein expression in vascular lesions.

The distribution of TAL-1 protein, an important vascular promoter in mice, has been examined immunohistochemically in a range of human vascular lesions and normal tissues. Formalin-fixed, paraffin-embedded vascular lesions including granulation tissue, haemangiomas, Kaposi's sarcomas, spindle cell haemangioendotheliomas, and angiosarcomas, were examined using a monoclonal antibody to recombinant TAL-1. Endothelial cells in all lesions gave positive immunostaining of variable intensity. Granulation tissue and spindle cell areas of the vascular tumours gave the strongest staining (nuclear and cytoplasmic). The better-differentiated endothelial cells within the tumours and resident well-formed vessels were less positive and some cells were in fact negative. The malignant endothelial cells in angiosarcomas showed less intense positive staining than KS cells. This study has shown TAL-1 protein expression in a range of reactive, benign, and malignant vascular lesions. Protein expression appears to be stronger in the spindle cell areas, perhaps reflecting greater expression in less-differentiated endothelial cells.

HML-1 antigen on mucosa-associated T cells, activated cells, and hairy leukemic cells is a new integrin containing the beta 7 subunit.

The monoclonal antibodies HML-1, B-ly7 and Ber-ACT8 recognize intramucosal gut T lymphocytes, activated cells, and hairy cell leukemia. The antigen on hairy cells consists of three glycoproteins (160 kappa D, 130 kappa D and 105 kappa D unreduced; 145 kappa D and 120 kappa D reduced). These peptides have biochemical features reminiscent of integrins but we have shown by immunoprecipitation that they are not known integrin subunits. We have used a newly produced antibody (BP6) to purify this molecule and shown by N-terminal sequence analysis that the smallest subunit is the product of integrin beta 7 cDNA. This molecule is thus a new member of the integrin family of leucocyte adhesion proteins. Immunoprecipitation experiments indicate that the two larger subunits are recognized by HML-1, B-ly7 and Ber-ACT8.

The IgM-associated protein mb-1 as a marker of normal and neoplastic B cells.

Recent evidence indicates that the transmembrane form of IgM on murine and human B lymphocytes is physically associated with at least two proteins, forming a disulfide-linked dimer, which may control cell surface expression of IgM and also play a role in signal transduction after Ag binding (by analogy with the TCR-associated CD3 components in T lymphocytes). We have used mAb and polyclonal antibodies against an intracytoplasmic epitope on one of these polypeptides (previously identified in murine B cells as the product of the B cell specific mb-1 gene) to study the distribution of the IgM-associated dimer in human cells. By immunocytochemical staining of normal and neoplastic B cells, we show that the human mb-1 protein appears early in B cell differentiation, probably before expression of cytoplasmic mu-chain, and persists until the plasma cell stage, where it is seen as an intracytoplasmic component. According to immunohistologic analysis of reactive lymphoid tissue and lymphoma samples, mb-1 protein is completely B cell specific. Anti-mb-1 also labels B cell areas in tissues from seven different mammalian species. Finally, the Ig-associated dimer could be isolated from human hairy-cell leukemia cells in high purity and yield by affinity chromatography using anti-mb-1 antibody. Mice immunized with this material have produced a strong polyclonal response, so that it should now be possible to prepare a panel of new mAb reactive with different epitopes on both mb-1 and on its associated polypeptide(s).

Characterization of a new monoclonal anti-Fc gamma RII antibody, AT10, and its incorporation into a bispecific F(ab')2 derivative for recruitment of cytotoxic effectors.

Fc gamma RII (CDw32) on monocytes is capable of triggering both phagocytosis and lysis of chick red blood cells (CRBC) coated with antibody of the appropriate isotype. In this report we describe the production and characterization of a mouse monoclonal IgG1 antibody specific for Fc gamma RII and compare its activity in binding studies, tissue distribution and redirected cellular cytotoxicity (RCC), with the previously identified anti-Fc gamma RII antibodies KB61 and IV.3. Immunohistochemical and flow cytometry analyses demonstrated that AT10 binds very strongly to Fc gamma RII on normal monocytes, but only weakly to that expressed on lymphocytes. This pattern does not correspond to the staining seen with either KB61 or IV.3, and appears to give an intermediate profile. The binding constant (Ka) for the Fab' fragment of AT10 was calculated at 5.3 x 10(8) M-1, four times higher than that for KB61 (1.4 x 10(8) M-1). Bispecific F(ab')2 antibodies were constructed from Fab' fragments of AT10 or KB61 thioether-linked to Fab' from an anti-CRBC monoclonal antibody. These bispecific derivatives directed monocyte cytotoxicity against CRBC as efficiently as either a monoclonal or polyclonal anti-chick erythrocyte antibody. The bispecific F(ab')2 antibodies had a distinct advantage over the conventional reagents, in that they were not blocked in the presence of human Fc gamma at 3.5 mg/ml (a concentration comparable with that provided by IgG in serum). Therefore, bispecific derivatives constructed with the high affinity anti-Fc gamma RII antibody, AT10, may be used as therapeutic reagents for targeting tumour cell lysis in vivo.

Ultrastructural localization of the CD68 macrophage-associated antigen in human blood neutrophils and monocytes.

The ultrastructural localization of the CD68 antigen, a 110-kd intracellular glycoprotein associated with myeloid cells and with monocytes/macrophages, was investigated in human neutrophil granulocytes by postembedding immunogold staining, using monoclonal antibody KP1. The antigen was found in the primary granules of neutrophils, although not all primary granules were labeled. It was absent from the plasma membrane. In monocytes, it was also detected within cytoplasmic granules, colocalized with lysozyme and myeloperoxidase. This observation confirms and completes results obtained by immunofluorescence and other light-microscopic methods. Moreover this study shows that the CD68 epitope recognized by antibody KP1 is able to resist fixation and embedment and therefore emphasizes the value of using KP1 as a marker for this macrophage-associated molecule.

Human B cells cannot be triggered to kill target cells through their Fc gamma RII or Fc epsilon RII receptors.

There has been some controversy as to whether or not B cells can kill target cells through their Fc receptors. To address this, we have examined the ability of human B cells from a variety of sources to lyse hybridoma cells with specificity for either the B cell Fc gamma RII or Fc epsilon RII using a reverse killing assay, as well as their ability to lyse opsonized chicken erythrocytes using a classic ADCC assay. Tonsil B cells, chronic lymphocytic leukemia B cells, and Epstein-Barr virus-induced B cells, even after preactivation with a cocktail of cytokines, all failed to lyse any of these targets. We conclude that Fc gamma RII and Fc epsilon RII on human B cells are not cytotoxic trigger molecules.

Expression of the bcl-2 oncogene protein is not specific for the 14;18 chromosomal translocation.

It has been reported previously that the bcl-2 protooncogene protein is detectable in neoplastic cells from cases of human lymphoma in which the 14;18 chromosomal translocation is present, but not in lymphomas that lack this chromosomal rearrangement or in normal lymphoid tissue. In the present study we confirmed, by immunohistologic labeling with polyclonal and monoclonal antibodies, that bcl-2 protein is strongly expressed in many cases of follicular lymphoma and that these neoplastic follicles differ clearly from their nonmalignant counterpart (reactive germinal centres) in which bcl-2 protein is undetectable. However we also found bcl-2 protein in normal T- and B-lymphoid cells and in a variety of lymphoproliferative disorders in which the 14;18 translocation is not present. It is therefore concluded that expression of bcl-2 protein is not a specific marker for lymphomas bearing the 14;18 chromosomal translocation and that the observations of other investigators may have reflected the inadequate sensitivity of their staining procedure.

Distribution of the CD68 macrophage/myeloid associated antigen.

The distribution of the pan-macrophage CD68 antigen, recognized by six different monoclonal antibodies, was examined in human blood, tissue, and cell lines using APAAP staining and Western blotting. All antibodies stained monocytes and macrophages, but labelling of neutrophils, basophils, and lymphocytes was seen with some of the reagents. In addition, the CD68 antibodies demonstrated a variety of staining patterns on some non-haemopoietic cells. The subtle differences between the reactions of the different antibodies suggested that the CD68 antigen may be heterogeneous, possibly due to differences in glycosylation. While CD68 antibodies are very useful markers of the macrophage/myeloid series, the presence of small amounts of the antigen on some lymphoid and non-haemopoietic cells means that care should be taken when using them for the diagnosis of tumours of unknown origin.

Diagnosis of myelomonocytic and macrophage neoplasms in routinely processed tissue biopsies with monoclonal antibody KP1.

A new monoclonal antibody, KP1, against the CD68 antigen, which labels macrophages and other members of the mononuclear phagocyte lineage in routinely processed tissue sections, has been used to stain a range of lymphoid, histiocytic, and myelomonocytic proliferations. All 20 neoplasms of myeloid, myelomonocytic, and presumed macrophage derivation reacted with antibody KP1. None of the 22 cases of T cell neoplasia had positive reactions. Although 14 of 41 B lineage lymphomas and leukaemias were stained by antibody KP1, staining was usually confined to small dots of reactivity, in contrast to the strong and extensive cytoplasmic staining seen in the neoplasms of myeloid and macrophage/monocyte origin. Furthermore, positive B cell neoplasms were almost all small cell proliferations, which are unlikely to be confused with myelomonocytic malignancies. It was concluded that antibody KP1 is a valuable addition to a panel of monoclonal antibodies for phenotyping lymphomas, particularly in routinely fixed tissues. It should assist the pathologist in the recognition of extramedullary presentation of leukaemia, aid in the diagnosis of suspected cases of true histiocytic neoplasia, and allow for quantitation of macrophages infiltrating lymphomas and other solid tumors.

Expression of macrophage-associated antigens in tissues involved by Langerhans' cell histiocytosis (histiocytosis X).

The expression of macrophage antigens KP1, Mac, lysozyme, and alpha-1-antichymotrypsin was investigated on routine paraffin sections from 17 cases of Langerhans' cell histiocytosis (LCH). All the major clinical forms were represented, including single lesions and monosystemic and multisystemic disease. In all the cases, a variable fraction (3-35%) of LCH cells was immunoreactive with KP1 and anti-Mac; the staining pattern was quite typical because the immunoreaction product was often confined to the perinuclear space and the Golgi area. LCH cells containing lysozyme and AACT were detected less frequently; however, in positive cases the percentage of LCH cells immunoreactive for lysozyme and AACT was in the same range as that of KP1-positive cells. On immunostained cytosmears (one case), about 10% of the CD1a-positive cell population was reactive for the macrophage antigens CD14 and PAM-1. No association was noted between the number of KP1-positive cells and the clinical form and/or anatomic site of the lesion. Phagocytic macrophages were significantly and diffusely immunoreactive with KP1 and anti-Mac and for AACT and lysozyme. Multinucleated giant cells with irregular nuclei were frequently observed; these cells were rarely S-100 positive, were consistently stained by KP1 and AACT, and were occasionally anti-Mac positive. The authors' findings suggest that antimacrophage monoclonals, in conjunction with S-100 protein, may represent a useful tool to establish the diagnosis of LCH in paraffin-embedded material.

Colocalization of elastase and myeloperoxidase in human blood and bone marrow neutrophils using a monoclonal antibody and immunogold.

The authors have localized elastase in human blood and bone marrow neutrophils by immunoelectron microscopy using a monoclonal anti-human elastase antibody (NP 57) and compared its distribution with myeloperoxidase (MPO) and lactoferrin (LF), which mark primary and secondary neutrophil granule, respectively. Human bone marrow and blood polymorphonuclear leukocytes (PMN), either unstimulated or after phagocytosis of latex microbeads, were fixed in 4% paraformaldehyde. Ultrathin frozen sections were immunolabeled with NP 57, followed by an immunogold probe. In bone marrow granulocyte precursors elastase appeared simultaneously in the immature first granules of myeloblasts with MPO. As these granules became denser with maturation, labeling for both enzymes became weaker and sometimes negative (possibly due to masking of immunoreactivity). The ellipsoidal primary granules were strongly labeled by NP57. LF positive granules appeared later, at the myelocyte stage, and contained neither MPO nor elastase. In mature neutrophils, immunolabeling for elastase was found together with MPO in the large electron-dense primary granules and in a different granule population from the LF-positive secondary granules. Double labeling with two different-sized gold particles was used to compare the kinetics of degranulation of secondary and primary granules. The observation and the analysis of single phagosome content was made possible by this new technique. In conclusion, immunoelectron microscopy was used to show elastase in the primary granules of neutrophils, where it appears simultaneously with MPO. This technique has also allowed comparison of the kinetics of degranulation of both types of granules, and could be applied to different experimental and pathologic conditions.

Diagnosis of acute myeloid leukaemia with the use of monoclonal anti-neutrophil elastase (NP57) reactive with routinely processed biopsy samples.

The reactivity of a monoclonal anti-neutrophil elastase antibody (NP57) with routinely processed biopsy samples from various acute leukaemias has been examined and compared with that of chloroacetate esterase and CD15 (hapten X), two other myeloid cell-associated markers detectable in paraffin sections. No staining was seen with these markers in 14 cases of acute lymphoblastic leukaemia. In contrast the neoplastic cells in 27 of 37 acute myeloid leukaemias were NP57 positive. Twenty of these were also positive for chloroacetate esterase, whereas CD15 was expressed in only six cases. These results indicate that detection of elastase with monoclonal NP57 forms a useful supplement to traditional methods for the histopathological diagnosis of acute myeloid leukaemias.

Plasmacytoid monocytes (so-called plasmacytoid T cells) in granulomatous lymphadenitis.

Immunohistochemical evidence that the plasmacytoid T cell is closely related to the blood monocyte has been reported, and the term plasmacytoid monocyte has been proposed to describe this cell. The present study was undertaken to analyze the presence and distribution of plasmacytoid monocytes in human reactive lymph nodes showing epithelioid cell reactions. Numerous plasmacytoid monocytes (detected by a panel of monoclonal antibodies) were found in the majority of the lymph nodes studies, usually in close topographical association with epithelioid cells and multinucleated giant cells. The present findings suggest that plasmacytoid monocytes may give rise to epithelioid cells. This is further supported by the ultrastructural similarities between plasmacytoid monocytes and plasmacytoid epithelioid cells, a cell type that has been identified previously in granulomas and considered a direct precursor of the classical epithelioid cell.

KP1: a new monoclonal antibody that detects a monocyte/macrophage associated antigen in routinely processed tissue sections.

A new monoclonal antibody, KP1, raised against a lysosomal fraction of human lung macrophages, recognises a fixation-resistant epitope in a wide variety of tissue macrophages (such as Kupffer cells germinal centre, splenic, and lamina propria macrophages), and in granulocyte precursors. Its broad reactivity with cells of the mononuclear phagocytic lineage was established by testing on routinely processed samples of normal and reactive lymphoid tissues. Interdigitating reticulum cells were unstained or showed limited cytoplasmic staining while Langerhans' cells and follicular dendritic reticulum cells were unreactive. KP1 recognises a molecule of about 110 kilodaltons in macrophage-rich human tissue when tested by either immunoprecipitation or Western blotting (although the latter procedure also shows two additional components with molecular weights of 70 and 40 kilodaltons). KP1 should be of considerable value for studying disorders of the monocyte/macrophage system, including both reactive and neoplastic states (such as true histiocytic proliferations).