Search

We defined some standardised criteria for classifying incident cases of liver cancer into either Primary Liver Cancer (PLC) or Unspecified Liver Cancer (ULC), on the basis of the diagnostic procedures performed. A pilot hospital-based study (98 cases) was carried out in Verona, northern Italy, in order to assess the feasibility of the method. The same protocol was subsequently applied in a population-based study (349 cases) in Bresica, northern Italy. The percentage of cases with histological verification was 38.7 and 41.8%, respectively, with a wide variation among different hospitals. The percentage of cases we attributed to the PLC category was 78.6% in the hospital-based study and 78.8% in the population-based study. No differences in the proportion of cases attributed to PLC were found according to patients' age, sex or hospital of admission. Repeatability of the method was assessed through a cross-panel review of 198 cases, with a 91.9% interobserver agreement. Implications of this method are discussed and some suggestions for cancer registration and future research are proposed.

Six cases of mediastinal large B-cell lymphoma (MLCL) with sclerosis were analyzed for the presence and patterns of c-myc and bcl-2 loci rearrangements, and for the presence of Epstein-Barr virus DNA sequences by Southern blot hybridization, c-myc gene alterations were found in three of six cases. Two cases showed the presence of mutations or small rearrangements at the 3' end of the first exon. The c-myc gene abnormalities found in these two cases are similar to those observed in the translocation 8;14 of the endemic Burkitt's lymphomas or in its variants t(2;8) and t(8;22). A third case showed a major rearrangement of c-myc gene, with truncation within its first intron, similar to those observed in sporadic Burkitt's and in acquired immunodeficiency-associated lymphomas. None of the cases displayed bcl-2 gene rearrangements or contained viral sequences. Our data suggest a possible role for a translocation-mediated c-myc activation in the pathogenesis of MLCL. Conversely, bcl-2 gene and Epstein-Barr virus do not appear to be involved in the pathogenesis of these peculiar lymphomas. The association between c-myc structural modifications and MLCL also seems to be of relevance in light of the peculiar tendency of this tumor to involve unusual extranodal site (eg, kidney), reminiscent of the spreading attitude of Burkitt's limphomas.

From 1977 to 1990, 37 patients with Stage III or IV invasive thymoma (20 men and 17 women; median age, 40 years of age) were referred for chemotherapy to the Padova Medical Oncology Department. All patients initially received the same regimen (50 mg/m2 of cisplatin and 40 mg/m2 of doxorubicin intravenously (IV) on day 1, 0.6 mg/m2 of vincristine IV on day 3, and 700 mg/m2 of cyclophosphamide IV on day 4 [ADOC]), recycling at monthly intervals. No life-threatening side effects were noted. The overall clinical response rate (complete response plus partial response) was 91.8%, with 43% complete remissions. Median duration of response and survival were 12 months (range, 2 to 96+ months) and 15 months (range, 5 to 96+ months), respectively. Seven of the 16 complete remissions were pathologically confirmed at subsequent thoracotomy. Other chemotherapy combinations and radiation therapy have been applied as second-line treatment, achieving only minimal responses. In the opinion of the authors, such chemotherapy deserves evaluation for adjuvant and neo-adjuvant treatment of invasive (and/or inoperable) thymoma due to the high complete response rate and overall response rate.

A simple method was developed for extraction of DNA from stored slides of Papanicolaou stained cervical smears. HPV type 16 DNA was detected by DNA hybridization in 7/10 samples from known HPV-16 positive patients, whereas the 10 samples from negative patients were also negative with our assay. Using our method for obtaining data from stained cervical smears, it will be possible to design and perform historical cohort studies on stored material.

Thirty-two patients with stage III or IV invasive thymoma (14 women and 18 men; median age, 40 years) were treated at the Padua Medical Oncology Department from 1977 to 1988. All patients received the following chemotherapy in 4-day courses: 50 mg/m2 of cisplatin intravenously (IV) and 40 mg/m2 of doxorubicin IV on day 1; 0.6 mg/m2 of vincristine IV on day 3; and 700 mg/m2 of cyclophosphamide IV on day 4 (ADOC). The courses were repeated every 3 weeks, and toxic effects were tolerable. The radiologically defined overall clinical response rate (complete plus partial response) was 91% with 47% clinical complete remissions; median time to progression was 11 months (range, 0 to 96) and the median estimated (Kaplan-Meier) progression-free interval was 22 months. Five of the 15 clinical complete remissions were pathologically confirmed at thoracotomy. We believe the ADOC regimen qualifies for adjuvant and preoperative treatment of invasive thymoma due to the high complete response and overall response rates.

In this study monoclonal antibodies which recognize lymphoid-associated antigens on paraffin sections (LN1, MB2, L26, MT2, UCHL1) have been evaluated to assess their usefulness in the distinction between reactive and neoplastic lesions of lymphoid follicles. Thirty-three follicular lymphoma samples and 36 reactive samples (lymph nodes and tonsils) were analyzed. MT2 appeared as the most valuable immunophenotypic marker as emerged from a comprehensive quantitative evaluation of 2329 reactive follicles and 2288 neoplastic follicles performed on MT2 immunostained sections. MT2-positive follicles were found in all lymphoma samples but one. Overall 1908 of 2288 neoplastic follicles were judged as positive whereas no follicles with comparable strong MT2 immunoreactivity could be found in non neoplastic samples. These latter showed weak MT2 positivity only in about 10% (224/2329) of reactive follicles. This study confirms that MT2 follicular positivity can be considered a reliable marker of follicular neoplasia, although negative results ought to be considered with caution. The detection of centrofollicular cells outside the germinal centers, which is considered a reliable criterion of follicular neoplasia, was highly improved by LN1 immunostaining. On the other hand pan-B antibodies such as L26 and MB2 were less informative because of the large number of B-lymphocytes observed in interfollicular areas of nonneoplastic samples.

Human thymomas are epithelial neoplasms frequently associated with an exuberant lymphoid component. This mixture of epithelial cells and lymphocytes closely mimicks the organization of normal thymic cortex. However, it is not known whether thymocytes in thymoma express the T-cell receptor (TCR) for the antigen. We have analyzed the molecular configuration of TCR genes and their phenotypic expression in eight thymomas. In all we detected polyclonal rearrangements of TCR genes and cytoplasmic expression of TCR molecules in most thymocytes, thus indicating that rearranged TCR genes in thymomas are functioning genes. In addition, these findings suggest that the epithelial component of thymomas, even if neoplastic, is still capable of directing thymocyte differentiation.

Eight cases of mediastinal non-lymphoblastic large-cell lymphoma have been studied by histopathological and immunohistochemical methods. A common clinical, morphological and immunophenotypic pattern was identified. Six of eight cases proved to be of B-cell origin by the expression of B-associated antigens detected by specific monoclonal antibodies. Cells of large size with nuclei of varying morphology and a peculiar type of fine compartmentalizing fibrosis were observed in all specimens. Clinically the disease was characterized by the young age of the patients, primary mediastinal involvement, aggressive behaviour and spread to unusual sites (kidneys in four cases). This seems to be a hitherto unrecognized entity in the field of non-Hodgkin's lymphomas, often misdiagnosed because of location and a morphology uncommon for B-cell malignancies. Immunohistochemical analysis on frozen tissue sections appears to be mandatory for a correct diagnosis. Nevertheless, this type of lymphoma could be suspected also on the basis of its peculiar clinicopathological characteristics.

Bone marrow trephine biopsies from 17 patients with B-chronic lymphocytic leukemia (B-CLL) were studied by immunohistologic techniques in order to investigate the cellular phenotypes of both neoplastic (B-lymphoid) and reactive (T-lymphoid) infiltrates. For this purpose, several heteroantisera and monoclonal antibodies against human Ig isotypes, HLA-DR antigens, and T-cell subpopulations were used in immunofluorescence. The findings were analyzed in relationship to the histologic pattern of involvement, as well as to the immunologic data of cell suspensions from peripheral blood. In all cases, the dominant lymphoid population within the bone marrow infiltrates showed identical phenotypic characteristics of B-CLL cells from the blood (HLA-DR+, mu +, most frequently delta +, kappa +, or lambda +, and weakly RFA-1+). The infiltration by these malignant B cells was diffuse in 5 cases and nodular plus interstitial in 12. The number of T cells (UCHT1+, RFA-1+, mu) was variable (5%-25%) in the different samples, but the values were high when compared to the proportion of T cells in normal bone marrow and in the blood of most patients studied. Furthermore, a clear predominance of T cells exhibiting the inducer phenotype (Leu-3+) was observed in all bone marrow samples, which is in contrast with the findings from peripheral blood, where T cells with the suppressor/cytotoxic phenotype (Leu-2+) were dominant. These data suggest a different blood and tissue distribution of inducer and suppressor/cytotoxic cells in B-CLL, which may have important pathophysiologic significance.

The histologic organization of lymphoid and nonlymphoid (epithelial and interdigitating) cells in a thymoma has been compared to that of the normal thymus. Enzyme and immunohistochemical methods were applied, using both conventional antisera (to cytokeratin) and monoclonal antibodies (to epithelial cells, HLA-DR and lymphoid subsets). Throughout the tumor, the epithelial cells shared phenotypical similarities with the epithelial cells of thymic medulla (RFD-4 positive, cytokeratin strongly positive, and HLA-DR essentially negative). On the other hand, the lymphoid cells were heterogeneous in phenotype and distribution, and "mimicked" the distribution seen in the normal infant thymus. Immature thymocytes of cortical type (TdT+, OKT6+, OKT3-) were predominant in the areas with moderate lymphocytic infiltration (ML). Mature T-lymphocytes (TdT-, OKT6-, OKT3+) were found mainly in areas with scanty lymphocytes (SL) together with an additional population of HLA-DR positive interdigitating and HLA-DR+, OKT6+ Langerhans'-type cells. These findings indicate that in thymoma tissue, the lymphoid elements of cortical type are apparently surrounded by an inappropriate (medullary) epithelium.

In this study we provide evidence that MB1, a newly developed monoclonal antibody which reacts with B lymphocytes and a proportion of T cells and monocytes, can be successfully used for the direct immunohistochemical identification of osteoclasts on paraffin-embedded surgical specimens. The antigen(s) recognized by MB1 is present at high density in the cytoplasm of osteoclasts of fetal bone and in the multinucleated cells of human giant cell tumour of bone (osteoclastoma), but is weakly expressed or absent in the giant cells of granulomas. MB1 is thus proposed as a new immunohistochemical marker for osteoclasts on paraffin-embedded material.

The Southern blot hybridization technique has been applied to study the configuration of immunoglobulin and T-cell receptor genes in 6 cases of the so called mediastinal large cell lymphoma with sclerosis. This lymphoma has been recently recognized as a separate entity among non-Hodgkin lymphomas mainly affecting young adult patients. The B-cell origin of this neoplasm was suggested by means of immunohistochemical analysis. However, the immunophenotypical B-cell related markers used do not always exhibit lineage fidelity. The Southern blot analysis demonstrated the presence of unique heavy and k-light chain immunoglobulin gene rearrangements, establishing genotypically their B-cell origin.

Five cases of Rosai-Dorfman histiocytosis (RDH) (also called Sinus Histiocytosis with Massive Lymphoadenopathy; SHML) have been studied by immunohistochemical methods with heteroantisera and monoclonal antibodies. One case was also studied by Southern blot hybridization analysis with DNA probes specific for T cell receptor beta chain and immunoglobulin heavy chain. Immunophenotyping of large histiocytes, characteristic of RDH, evidenced the presence of S-100 protein and the absence of CD1 and other markers usually found in histiocytes and macrophages. DNA hybridization study showed the absence of clonal T or B lymphoid populations.

Cryostat sections from frozen samples of human bone marrow trephine biopsies were used for immunohistochemical analysis. Immunoperoxidase and immunofluorescence techniques as well as enzyme histochemistry were applied to these sections in order to characterize bone marrow infiltrates in lymphoproliferative diseases. Various conventional antisera and monoclonal antibodies and several enzymes were able to identify different normal and malignant cell populations. This technology provides a new tool to be used in combination with classical morphology for studying bone marrow involvement during lymphoid malignancies.

Alpha-Naphthyl acetate esterase (ANAE) and fluoride resistant alpha-naphthyl acetate esterase (FRANAE) have been compared as histochemical methods to identify T lymphocytes in sections of normal and pathological human lymphoid tissues. In addition, the FRANAE method was combined with alkaline phosphatase (ALP) in order to simultaneously evaluate the relationship between T lymphocytes and fibroblastic reticular cells (ALP) positive). The "dot like" esterase positivity of T lymphocyte was better evaluated by using FRANAE when compared to ANAE because of fluoride inhibitor of the strong esterase activity of dendritic cells and most macrophages. The combined ALP-FRANAE method clearly demonstrated a large number of fibroblastic reticular cells within the T-areas in various normal and pathological tissues such as hyperplastic lymph nodes and especially in the lymph nodes and spleens from patients with Hodgkin's disease.

The morphological and phenotypical features of multicellular complexes formed by follicular dendritic cells and lymphocytes (FDC-LC) isolated from human hyperplastic tonsils and adenoids are described. FDC-LC obtained with this procedure were morphologically and immuno-phenotypically heterogeneous. In one type of FDC-LC, probably obtained from germinal centers, the lymphocytes exhibited ultrastructural features of centroblasts and centrocytes. In a second type, likely derived from follicular mantles, the enclosed lymphocytes were small in size and characterized by a condensed chromatin pattern. Similar heterogeneity was observed by immuno-phenotypical analysis, which revealed a prevalence of IgD+, CD3-, MT2+ small lymphocytes in a high proportion of FDC-LC. Both types of FDC-LC contained desmoplakin immunoreactivity in a typical punctate pattern corresponding to intercellular junctions when tested with a specific antibody. These findings confirm the importance of FDC in maintaining the follicular structure and also suggest that the different zones forming lymphoid follicles (mantle zone and germinal center) are formed by lymphocytes gathered in single "domains" by cytoplasmic processes of FDC. These domains have strong resistance to mechanical stress, such as that used in isolation procedures. FDC-LC have also been maintained as organized multicellular clusters for short periods (more than 48 h) in agarose gel cultures.

Bone marrow aspirates (311) from patients with various haematologic disorders were examined with both smear preparations and histologic sections. Adequate specimens for comparison were obtained in 77.4%. A wide discrepancy concerning the cellularity was noted between the two methods. The results confirm previous studies and indicate that histologic examination of bone marrow aspirates is particularly helpful in lymphoproliferative disorders. It is concluded that sectioning of aspirated marrow is an important procedure and should be included in the routine examination of bone marrow.

Four samples of thymoma obtained from patients affected by myasthenia gravis have been immunohistologically analysed on cryostat sections using a panel of antisera and monoclonal antibodies specific for antigens which define different stages of intrathymic lymphocyte differentiation and antigens specific for different types of thymic epithelial cells (cortical, medullary). When the thymoma samples were compared to age-matched normal thymuses and hyperplastic thymuses obtained from patients with myasthenia gravis some evident microenvironmental differences could be demonstrated using these reagents. In all the thymoma samples in fact the neoplastic lobules appeared as grossly enlarged cortical-type areas, formed by accumulations of T lymphocytes exhibiting the cortical immature phenotype (TdT+, T6+, etc.) within a network of putatively neoplastic epithelial cells characterized by cortical phenotype as defined by reactivity with various monoclonal antibodies (RFD4-, MR3+). These 'cortical' epithelia showed some abnormal features such as lack or irregular distribution of HLA-DR and enhanced keratin expression. Small areas of 'medullary' differentiation could be observed in 3/4 thymoma samples. In thymic hyperplasia, on the other hand, the cortical areas appeared somewhat compressed (but comparable to those observed in normal age-matched samples) by enlarged medullary areas. The expansion of medullary areas was due to the infiltration of 'peripheral' lymphoid tissue intruding through the extraparenchymal zone and forming organized B and T areas. These observations are discussed in the light of the clinical heterogeneity observed in myasthenia gravis.

Using a sensitive immunohistochemical technique and a specific monoclonal antibody (RFD-3), follicular dendritic cells (FDC) have been demonstrated in a significant proportion (6/12) of bone marrow samples from patients with nodular marrow involvement of B-cell chronic lymphocytic leukemia (B-CLL). In all cases with "packed marrow" involvement and advanced stages of disease the FDC were absent. As these accessory cells are normally present only in the lymph nodes, their presence in the bone marrow is in accord with the view that B-CLL might be the malignancy of an immature subpopulation of lymph node lymphocytes that invade the bone marrow.

Malignant tumours mortality among residents of the city of Verona (267,703 at 31-12-1980) during the year 1980 is assessed from the ISTAT individual death cards and/or municipal death certificates (especially in the case of deaths outside the municipality). The figures are broken down by site, apparatus or system (in accordance with the VIII-ICD), sex, age at death, and compared with the Italian mortality figures for 1976 (indirectly standardised) and those for the province of Verona for 1976-77 (directly standardised). Tumour mortality per site was generally higher than the Italian level. This is line with the differences noted between geographical areas and the similarity of the rates in more industrialised areas. A salient finding was that lung tumour mortality exceeded that for the province of Varese, which is among the highest in Italy entered on the Register of Tumours. Genital and breast tumours were responsible for the highest number of deaths in women, while stomach cancer was the second cause of death in both sexes. The study will be completed with an analysis of tumour deaths during 1980 and 1981 in the province of Verona as a whole.

A 45-year old male presented latero-cervical lymphoadenopathy. Biopsy revealed a malignant proliferation of immature "lymphoid" cells bearing T6 antigen and HLA-DR but negative for other lymphoid markers, suggesting a phenotype similar to Langerhans cells. The patient did not receive any therapy and six months later developed a histologically typical malignant histiocytosis, involving spleen and liver. Other reported cases of lymphoid malignancies (two bearing the T6 antigen on blast cells) preceding malignant histiocytosis were found and compared with ours. Most of these cases were characterized by the pediatric age of the patients and were presented as acute leukemias exhibiting, in at least some of them, reliable T-cell markers. Our case appears to represent, on the other hand, a blastic proliferation of precursors of both histiocytes and Langerhans dendritic cells which eventually progressed to malignant histiocytosis. The relevance of this observation in the debate on the origin of Langerhans cells and the relationships existing between macrophages and dendritic cells is discussed.

The enzyme glycyl-proline-naphthylamidase (dipeptidyl-amino-peptidase IV), (DAP-IV) has been histochemically analysed in normal and pathological specimens of different lymphoid tissues. In all the tissues examined, the enzyme appeared to be highly specific for T cells. Using chloroformacetone fixation and sections of proper thickness, DAP-IV could be successfully demonstrated in cortical thymocytes, which exhibited weak reactivity, as well as in medullary thymocytes which showed a more intense and variable reactivity. A similar reactivity was observed in T cells of peripheral blood, lymph node, and spleen. Various acute lymphoblastic and nonlymphoblastic leukemias and malignant lymphomas were studied using the histochemical method for DAP-IV. Malignant cells from all the acute leukemias and from the B-cell lymphomas were DAP-IV unreactive, while strong reactivity was observed in one case of T-CLL and one case of T-PLL. The possible relationship of DAP-IV positivity pattern with T-cell maturation and its potential use as a diagnostic tool in lymphoproliferative disorders are discussed.

The authors describe a 63-year-old woman who developed a histologically distinctive malignant cutaneous neoplasm composed of large pleomorphic cells with abundant cytoplasm and multilobate, often clefted nuclei that occasionally contained small nucleoli. This neoplastic cell population metastasized to a regional lymph node already involved by a B-cell derived chronic lymphocytic leukemia expressing surface IgMk, BA-1, and OKT1. The large metastatic tumor cells lacked surface immunoglobulin, B-lymphocyte associated antigen BA-1, T-lymphocyte associated antigens OKT1 and OKT3, and the monocyte/macrophage markers lysozyme and alpha 1-antichymotrypsin. These tumor cells expressed HLA-DR antigens, adenosine triphosphatase (ATPase), OKT6, and contained S-100 protein, i.e., they expressed the phenotype peculiar to epidermal Langerhans cells. The typical clinical and histologic features of Histiocytosis X were absent. Thus, this case appears to represent a distinctive cutaneous neoplasm composed entirely of malignant cells of dendritic cell origin which, by immunophenotypic and histochemical analysis, appear to be related to epidermal Langerhans cells.

The distribution of a recently produced melanoma specific antibody (HMB-45) has been evaluated histochemically on paraffin sections in a large panel of melanocytic and non melanocytic tumours. Results have been compared with the presence of S-100 protein. HMB-45 was shown to be a highly specific antibody being present only in melanomas, junctional melanocytes and histogenetically related neoplasms such as melanocytic neuroectodermal tumour of infancy and, at low levels, on a proportion of peripheral nerve sheath tumours. The high specificity of HMB-45 antibody, coupled with the greater sensitivity of S-100, makes the combined use of these markers practical in the differential diagnosis of skin tumours and of metastatic lesions of uncertain primary site.

In this study we provide evidence that the sera of patients with hairy cell leukemia (HCL) contain a factor that can prevent the binding of a monoclonal antibody specific for interleukin-2 receptor (IL-2R) to its target. This factor corresponds to the soluble form of IL-2R (sIL-2R), as assessed by a specific enzyme-linked immunosorbent assay test, and appears to be released by neoplastic hairy cells. The serum sIL-2R levels were very high at diagnosis and significantly reduced during recombinant alpha-interferon (rIFN alpha 2) therapy. Values of sIL-2R appeared to be inversely related to the natural killer in vitro function displayed by peripheral blood mononuclear cells from the same patients. The presence of sIL-2R in the serum of patients with HCL might be involved in the impairment of cell-mediated immunity observed in these patients and could represent a valuable marker for monitoring different phases of the disease and for modulating IFN therapy.

Cryostat sections from bone marrow biopsies submerged in gum-sucrose solution and snap-frozen can be stained with various antisera to detect both the membrane and cytoplasmic antigens on normal and malignant populations in their tissue environments. This technique seems to be the method of choice to analyse the degree of BM involvement in cases of non-Hodgkin's lymphoma where conventional histology provides equivocal results. The uniform expression of only one type of light chain in B lymphoma and the analysis of the T cell infiltrate in T-CLL using monoclonal antibodies is demonstrated.

A case of angiofollicular lymph node hyperplasia (Castleman's disease) characterized by monotypic (IgG+, lambda+) plasmacytosis is described. Fresh tissue was available and a thorough immunohistochemical analysis of lymphoid and non-lymphoid cells was performed on cryostat sections. Although lymphoid follicles were numerous and exhibited some abnormal features they did not appear part of the monoclonal cell proliferation. Follicular lymphocytes were mixtures of Kappa+ and lambda+ cells. Vessels penetrating within these abnormal follicles expressed reduced levels of FVIII and Leu-M5 antigens and exhibited thicker layer of collagen type IV. The analysis of T-cell subsets showed a normal (3:1) T4/T8 ratio. This case extends to the mixed variant of hyaline-vascular Castleman's disease, the neoplastic potential previously associated to the plasma cell variant of the disease.

Enzyme histochemical and immunohistological (immuno-fluorescence and -peroxidase) techniques have been routinely used for investigating over 70 normal and pathological bone marrow samples. This recently standardized diagnostic procedure is very quick and can be performed in a few hours. In 6 cases the clinical diagnosis of leukaemia/lymphoma has become apparent only after the immunohistological analysis of the bone marrow. In 6 other cases the information about the staging of B cell malignancies was superior in the frozen biopsies to the paraffin embedded preparations. Amongst many other features the monoclonality of B CLL/lymphomas, the special features of B CLL infiltrates (RFA-1+, Leu-1+, HLA-DR+, SmIg+), follicular lymphoma deposits (containing follicular dendritic cells) and non-T, non-B acute lymphoblastic leukaemic blasts (terminal transferase+, HLA-DR+) as well as the sometimes conspicuous presence of infiltrating normal T cells could be clearly and reproducibly demonstrated.

So far, enzyme histochemical examination has been applied, with few exceptions, either to tissue imprints or to cryostat sections of freshly collected samples. This procedure is not easily applicable in routine histopathologic examination. In this study, a simplified tissue embedding procedure is presented which can be performed using an automatic tissue changer. The paraffin embedded samples can be used for both conventional histopathologic examination and for demonstrating enzymes in sections. The enzymes studied (alkaline phosphatase, alpha-naphthyl acetate esterase, acid phosphatase, tartrate resistant acid phosphatase, ATPase, peroxidase, and chloroacetate esterase) gave comparable results in formalin-fixed cryostat sections and paraffin sections in both normal and pathologic lymphoid samples. The only exception was ATPase, which could not be demonstrated on paraffin-embedded material. The technic described has broad application in the analysis of lymphoid diseases.

A recently developed procedure, that has been shown to be suitable for detailed immunohistological analysis, has been used to prepare cryostat sections of bone marrow to investigate whether enzyme-histochemical techniques are also feasible on such material. A selected group of enzymes, some of which are inhibited or destroyed in paraffin- or plastic-embedded samples, have been demonstrated. The morphological details obtained were satisfactory in the preparations. The enzymes were dipeptidyl(amino)peptidase IV (for T lymphocytes); tartrate-resistant acid phosphatase (for hairy cell leukaemia); acid phosphatase and non-specific esterase (for macrophages and monocytes); ATPase and 5'nucleotidase (for B lymphocytes); and peroxidase or chloroacetate esterase (for granulocytic cells). In these preparations strong enzyme activities were shown. In adjacent sections the immunological analysis of membrane markers could also be performed contributing to a comprehensive study of the normal and malignant bone marrow cells.

Eight cases of human thymoma have been analyzed on cryostat sections with the monoclonal antibody Ki67, which reacts with cells in the proliferative phases of the cell cycle. The aim was to assess the proportion of proliferating thymocytes among lymphoid cells in the thymoma samples. In all cases a large number of cells (mean, 58.75%; range, 35-80%), recognized as thymocytes by morphology and lack of cytokeratin expression in a combined immunohistochemical assay, exhibited nuclear Ki67 staining. These findings differ from the reactivity pattern observed in age-matched nonneoplastic thymuses where lower growth activity of cortical thymocytes was observed (15-20% Ki67+ cells). Intensive thymocyte proliferation in thymomas may represent one of the factors which lead to autoimmunity in myasthenia gravis and thymomas.