Bertram Schnitzer

CD117 (KIT receptor) expression in Merkel cell carcinoma

The KIT receptor tyrosine kinase (CD117 antigen) is found in a variety of normal tissue cell types and in several malignant tumors, including acute myeloid leukemia. We recently encountered two tumors initially suspected as acute myeloid leukemia cutis and expressing CD117 that showed punctate positivity for cytokeratin 20 diagnostic for Merkel cell carcinoma. We evaluated 20 additional cases of MCC and found that 21 of 22 tumors (95%) expressed CD117. Intensity of CD117 expression did not appear to correlate with aggressive behavior. While the function of the KIT receptor in MCC is not known, its pathogenic role in other malignant neoplasms suggests the possibility of a similar role in MCC.

Hodgkin's disease in children

Relatively few studies of Hodgkins disease (H.D.) in children have used the histologic criteria of the Rye classification. In our study, 55 cases of H.D. in children of 15 years and younger were histologically classified according to the Rye scheme. The cases were divided into 3 groups: 1. all patients, 2. those in their first decade of life, and 3. those aged 11 through 15. The majority of children were boys (62%). Nodular sclerosis was the most frequent histologic type seen in all children (49%) and in the third group (57%), whereas no preponderant histologic type was seen in the younger age group. Most cases of lymphocytic predominance were clinical Stage I, and the bulk of the long‐term survivors had this histologic type. Although seen in all clinical stages at the time of diagnostic biopsy, most cases of nodular sclerosis were Stage II. Prognosis of patients with nodular sclerosis was considerably worse than that of children with lymphocytic predominance. The results of our study are compared to those of other investigators.

Characterization of NF-κB expression in Hodgkin's disease: Inhibition of constitutively expressed NF-κB results in spontaneous caspase-independent apoptosis in Hodgkin and Reed-Sternberg cells

Although the neoplastic cells of classical Hodgkin’s disease (CHD) demonstrate high levels of constitutively active nuclear NF-κB, the precise physiologic and clinical significance of NF-κB expression is currently undefined. Expression of active NF-κB p65(Rel A) was evaluated in patient samples of CHD and nodular lymphocyte predominance Hodgkin’s disease. The action of the chemical NF-κB inhibitors gliotoxin and MG132 and the effect of NF-κB inhibition utilizing an adenovirus vector carrying a dominant-negative IκBα mutant (Ad5IκB) were then demonstrated in CHD cell lines (L428, KMH2, and HS445). Hodgkin and Reed-Sternberg (HRS) cells from all patient and cell line specimens showed strong immunopositivity for active p65(Rel A). Expression was also seen in lymphocytic/histiocytic cells from all cases of nodular lymphocyte predominance Hodgkin’s disease. After chemical NF-κB inhibition, p65(Rel A) was significantly reduced in nuclear extracts from cultured HRS cells as revealed by electrophoretic mobility shift assays. Furthermore, chemical NF-κB inhibition resulted in time- and concentration-dependent apoptosis in HRS cells. With the exception of MG132-induced apoptosis in HS445, apoptosis by chemical NF-κB inhibition was not significantly altered by preincubation with various caspase inhibitors (z-DQMD-FMK, z-DEVD-FMK, z-VAD-FMK, z-VEID-FMK, and z-IETD-FMK). Regardless of the chemical inhibitor used, no significant change in caspase-3 functional activity was found in CHD cell lines. HRS cells infected with Ad5IκB also showed a marked increase in spontaneous apoptosis compared with wild type adenovirus-infected and control cells. Overall, the inhibition of active NF-κB in HRS cells resulting in spontaneous caspase-independent apoptosis demonstrates a critical role for NF-κB in HRS cell survival and resistance to apoptosis.

The clinical significance of CD10 antigen expression in diffuse large B-cell lymphoma

The clinical significance and prognostic value of CD10 in de novo diffuse large B-cell lymphoma (DLBCL) is largely unknown. We retrospectively studied 19 men and 9 women based on the following criteria: (1) DLBCL with no evidence of concomitant or antecedent follicular lymphoma; (2) available flow cytometric immunophenotyping data, including CD10 status; (3) older than 15 years; (4) specific exclusion of high-grade, Burkitt-like lymphoma; and (5) exclusion of primary cutaneous DLBCL. When available, clinical data at diagnosis, including components of the international prognostic index, were reviewed. Eleven cases were CD10+, and 17 were CD10-. There was no significant difference between the CD10+ and CD10- groups in age, sex, stage, performance status, extranodal involvement, or serum lactate dehydrogenase levels at diagnosis. However, in the 26 cases for which follow-up data were available, the CD10+ group displayed a shorter overall survival than the CD10- group (8 vs 30 months). Although the clinical findings at diagnosis are similar in CD10+ and CD10- DLBCL, CD10 expression is associated with shortened overall survival. Therefore, our data suggest CD10 expression may have prognostic importance in adults with de novo DLBCL.

Characterization of the Interleukin-1β-Converting Enzyme/Ced-3-Family Protease, Caspase-3/CPP32, in Hodgkin's Disease: Lack of Caspase-3 Expression in Nodular Lymphocyte Predominance Hodgkin's Disease

Apoptosis (programmed cell death) serves an important role in the normal morphogenesis, immunoregulation, and homeostatic mechanisms in both normal and neoplastic cells. Caspase-3/CPP32, a member of the ICE/Ced-3-family of cysteine proteases, is an important downstream mediator of several complex proteolytic cascades that result in apoptosis in both hematopoietic and nonhematopoietic cells. Previous studies have demonstrated that caspase-3 is commonly expressed in classical Hodgkins disease (CHD); however, the biological significance of its expression in Hodgkins disease is unknown. In this report, the expression of caspase-3 in nodular lymphocyte predominance Hodgkins disease (NLPHD) was evaluated by immunohistochemistry; in addition, we investigated the role of caspase-3 in CD95 (Fas)-mediated apoptosis in three CHD cell lines. Formalin-fixed, paraffin-embedded tissue sections from 11 cases of NLPHD were immunostained for caspase-3 using a polyclonal rabbit antibody that detects both the 32-kd zymogen and the 20-kd active subunit of the caspase-3 protease. Only 1/11 cases of NLPHD demonstrated caspase-3 immunopositivity in lymphocytic/histiocytic cells. Caspase-3 expression was also evaluated in three CHD cell lines, HS445, L428, and KMH2. Whereas caspase-3 expression was detected in HS445 and L428 cell lines, no expression was found in KMH2 cells by immunohistochemical staining. Treatment of HS445 and L428 cell lines for 72 hours with agonistic CD95 monoclonal antibody induced marked apoptosis that was significantly inhibited by pretreatment with the caspase-3 inhibitor, DEVD-FMK, as determined by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling assay and flow cytometric analysis of 7-amino-actinomycin D staining. In addition, a significant increase in caspase-3 activity as determined by an enzyme colorimetric assay was detected in HS445 and L428 cells after 48 hours of CD95 stimulation. In marked contrast, treatment of caspase-3-deficient KMH2 cells with anti-CD95 mAb did not demonstrate an increase in caspase-3 activity or induce apoptosis. These data demonstrate caspase-3 is important for CD95-mediated apoptosis in CHD cell lines. In addition, the majority of NLPHD cases examined in this study failed to express detectable levels of caspase-3, suggesting these tumor cells may be resistant to apoptotic stimuli dependent on caspase-3 activity. Furthermore, these data suggest the differential expression of caspase-3 noted between NLPHD and CHD may provide additional evidence that each is a unique disease entity.

Ki-1 lymphomas in children

Three male children, ages 8, 11, and 14 at presentation, with the recently described Ki‐1 lymphomas are reported. All three had lymph node involvement. The lymphoma was classified as immunoblastic in two children, and mixed small and large cell in the third child. In histologic terms, sinusoid, paracortical, and diffuse lymph node involvement by lymphoma was evident in each case. Both cases of immunoblastic lymphoma were T11+, T10+, T9+, HLA/Dr+, Tac+, Ki‐1+, LCA+, and EMA+, while the lage neoplastic cells of the other case were T11+, Ki‐1+, and LCA+. In all three cases DNA analysis of immunoglobulin heavy and light chain genes as well as T‐cell receptor beta‐ and gamma‐chain genes showed only germline patterns. The patients were treated with multi‐agent chemotherapy. Two are in complete remission at 13 and 16 months, while the third failed to achieve remission and is alive with disease 12 months after a diagnosis had been established.

Detection of immunoglobulin heavy chain gene rearrangement by polymerase chain reaction in chronic active gastritis associated with Helicobacter pylori

Chronic active gastritis associated with Helicobacter pylori (CAG-Hp) has been linked to the pathogenesis of gastric B-cell lymphomas of mucosa-associated lymphoid tissue (MALT). To determine whether monoclonal lymphoid populations are present in CAG-Hp and histological predictors of monoclonality exist, the authors examined 46 endoscopic biopsies from 41 patients with CAG-Hp. The authors scored gastric biopsies for the presence of lymphoepithelial lesions (LELs), intensity of lymphoid infiltrate, presence of lymphoid aggregates and germinal centers, coexpression of CD43 (Leu 22) on B cells, and cytoplasmic immunoglobulin light chain restriction in formalin-fixed, paraffin embedded tissues. DNA extracts from these routinely processed tissues were analyzed for immunoglobulin heavy chain (IgH) gene rearrangement by polymerase chain reaction (PCR). Histological features, immunophenotype, gene rearrangement status, and clinical information were correlated. Six of the 46 biopsies (13%) from six of 41 patients (15%) showed a monoclonal PCR pattern. Monoclonal PCR patterns correlated with the presence of LELs (P<.015) but not with intensity of lymphoid infiltrate, presence of germinal centers, or presence of lymphoid aggregates. LELs correlated with germinal centers (P<.003) and intensity of infiltrate (P<.0001). None of the cases showed cytoplasmic light chain restriction nor coexpression of CD43 on B cells by immunohistochemistry. Clinical follow-up was available in all six patients whose gastric biopsies had a monoclonal PCR pattern (median, 58 months; range, 1 to 66 months) and 33 of the 35 patients with biopsies showing a polyclonal PCR pattern (median, 41 months; range 0.1 to 63 months). No patient developed gastric lymphoma. Monoclonality of lymphoid cells was detected by IgH PCR in 13% of patients with CAG-Hp. Although the authors cannot exclude the possibility that some patients with monoclonal gastric lymphoid infiltrates may eventually develop overt lymphoma, no histological, immunophenotypic, nor clinical evidence of lymphoma was noted at presentation or on clinical follow-up. Given the high incidence of CAG-Hp in the general population and the relatively low incidence of gastric MALT lymphoma, clinicopathologic correlation is needed when interpreting tests for clonality in this setting. The presence of clonal IgH gene rearrangement in CAG-Hp supports the hypothesis that H pylori is involved in the pathogenesis of low grade gastric B-cell MALT lymphomas.

Acute biphenotypic leukaemia: Immunophenotypic and cytogenetic analysis

Summary. The incidence of acute biphenotypic leukaemia has ranged from less than 1% to almost 50% in various reports in the literature. This wide variability may be attributed to a number of reasons including lack of consistent diagnostic criteria, use of various panels of antibodies, and the failure to recognize the lack of lineage specificity of some of the antibodies used. The morphology, cytochemistry, immunophenotype and cytogenetics of acute biphenotypic leukaemias from our institution were studied. The diagnostic criteria took into consideration the morphology of the analysed cells, light scatter characteristics, and evaluation of antibody fluorescence histograms in determining whether the aberrant marker expression was arising from leukaemic blasts or differentiated bone marrow elements. Fifty‐two of 746 cases (7%) fulfilled our criteria for acute biphenotypic leukaemias. These included 30 cases of acute lymphoblastic leukaemia (ALL) expressing myeloid antigens, 21 cases of acute myelogenous leukaemia (AML) expressing lymphoid markers, and one case of ALL expressing both B‐ and T‐cell associated antigens. The acute biphenotypic leukaemia cases consisted of four major immunophenotypic subgroups: CD2± AML (11), CD19± AML (8), CD13 and/or CD33± ALL (24), CD11b± ALL (5) and others (4). Chromosomal analysis was carried out in 42/52 of the acute biphenotypic leukaemia cases; a clonal abnormality was found in 31 of these 42 cases. This study highlights the problems encountered in the diagnosis of acute biphenotypic leukaemia, some of which may be reponsible for the wide variation in the reported incidence of this leukaemia. We suggest that the use of strict, uniform diagnostic criteria may help in establishing a more consistent approach towards diagnosis of this leukaemic entity. We also suggest that biphenotypic leukaemia is comprised of biologically different groups of leukaemia based on immunophenotypic and cytogenetic findings.

Classification of primary gastric lymphomas according to histologic features

Histologic features of low-grade gastric lymphomas of mucosa-associated lymphoid tissue (MALT) have been extensively described, and transformation to a large cell (high-grade) lymphoma can occur. We characterize high-grade gastric lymphoma histologically in an attempt to distinguish between MALT-type and non-MALT-type lesions. We studied a series of 60 gastric lymphomas and characterized them clinically, histopathologically, and immunophenotypically. Low-grade gastric lymphomas were classified according to established criteria. High-grade lymphomas were classified in three groups based on the presence or absence of a low-grade component and lymphoepithelial lesions (LELs): 1) high-grade MALT lymphomas appearing in low-grade MALT lymphomas (LG/HG MALT lymphoma); 2) large cell lymphoma with LELs composed of large cells (high-grade LELs) but without a low-grade component (HG MALT lymphoma); and 3) diffuse large cell lymphoma without a low-grade MALT lymphoma component or LELs (DLCL). Twenty-two lymphomas were classified as low-grade MALT lymphomas, 16 as LG/HG MALT lymphomas, 10 as HG MALT lymphomas, and 12 as DLCL. B-cell immunophenotype was confirmed in all 55 cases in which immunophenotyping was performed. Low-grade LELs were seen in all low-grade MALT lymphomas, and CD20(L26) expression confirmed B-cell phenotype in the LELs in 20 of 20 cases. Clinical follow-up was available for 56 patients (range, 1-264 months; mean, 57 months). Actuarial analysis of disease-specific survival and relapse-free survival showed that clinical stage was highly statistically significant (P < 0.0001), whereas histologic type and grade approached statistical significance. Multivariate analysis showed that clinical stage was the only significant factor in relapse-free and disease-specific survival.

Lymphocytic Mastitis and Diabetic Mastopathy: A Molecular, Immunophenotypic, and Clinicopathologic Evaluation of 11 Cases

Lymphocytic mastitis and diabetic mastopathy are uncommon fibroinflammatory breast diseases. The lesions seen in these entities are unique in that the associated lymphoid infiltrates are composed of predominantly B cells. In addition, B-cell lymphoepithelial lesions, a finding commonly associated with extranodal marginal zone B-cell/mucosa-associated lymphoid tissue (MALT) lymphomas, are also often present in lymphocytic mastitis and diabetic mastopathy. Although the clinical and immunomorphologic features are well characterized, the clonality of the B-cell infiltrate and the lymphomatous potential of lymphocytic mastitis and diabetic mastopathy have not been emphasized in the literature. We evaluated 11 cases of lymphocytic mastitis/diabetic mastopathy for immunoglobulin heavy chain gene rearrangement and correlated the findings with all available clinical data. A longstanding history of Type I diabetes mellitus was present in seven patients. One nondiabetic patient had Sjogren’s syndrome, and two patients had no history of diabetes mellitus or other autoimmune disease. Clinical data were unavailable for one patient. B-cell–predominant lymphoid infiltrates were seen in all cases, and B-cell lymphoepithelial lesions were found in five. No evidence of a B-cell clone was found in any of the 11 cases by appropriately controlled immunoglobulin heavy chain gene rearrangement studies, and none of the patients developed lymphoma during follow-up intervals ranging from 2–126 months. These findings suggest that despite the presence of B-cell–predominant lymphoid infiltrates and lymphoepithelial lesions, lymphocytic mastitis and diabetic mastopathy do not appear to be associated with an increased risk for lymphoma.