Lymphoma phenotyping in formalin-fixed and paraffin wax-embedded tissues: II. Profiles of reactivity in the various tumour types

'Specific' cutaneous infiltrate of B-cell chronic lymphocytic leukemia at the site of a florid herpes simplex infection

BACKGROUND

Specific cutaneous infiltrates in patients with leukemia generally carry a grim prognosis. However, non-neoplastic skin diseases may be associated with recruitment of normal and neoplastic leukocytes circulating in the peripheral blood. In those instances, neoplastic cells may be detected in skin lesions without an adverse effect on prognosis.

METHODS

In a patient with B-cell chronic lymphocytic leukemia, a specific infiltrate developed at the site of a florid herpes simplex infection. Clinically, the lesion presented itself as an ulcerated tumor.

RESULTS

Histopathologically, the lesion was characterized by a dense, diffuse infiltrate of small hyperchromatic lymphocytes throughout the entire dermis. Lymphocytes showed an aberrant CD20(+)/CD43(+)/CD5(+) phenotype of neoplastic B cells, and monoclonal rearrangement of immunoglobulin gamma genes could be demonstrated by polymerase chain reaction. Although criteria for leukemia cutis were fulfilled, the patient did well.

CONCLUSIONS

The cutaneous infiltrate of neoplastic cells seemed to be part of a physiologic response to the antigenic stimulus, rather than indicating an exacerbation of leukemia. Ziemer M, Bornkessel A, Hahnfeld S, Weyers W. 'Specific' cutaneous infiltrate of B-cell chronic lymphocytic leukemia at the site of a florid herpes simplex infection.

Leucocyte-specific protein (LSP1) in malignant lymphoma and Hodgkin's disease

Biopsies from 319 haematopoietic neoplasms were immunostained for intracellular leucocyte-specific protein 1 (LSP1) to assess its distribution and to compare its diagnostic value with that of CD45 (leucocyte common antigen: LCA). Most small B-cell neoplasms expressed LSP1, but one third of diffuse large B-cell lymphomas (DLBCL) were LSP1 negative. Among the cases with DLBCL (76 samples) tested for both LSP1 and CD45, one fifth expressed only CD45, but five samples were LSP1-positive and negative for CD45. The latter pattern was also seen in four of nine myelomas. Five out of 14 T-lymphoblastic lymphomas co-expressed LSP1 and CD45, and three cases were LSP1 negative and CD45-positive. Most peripheral T-cell lymphomas co-expressed LSP1 and CD45. All anaplastic lymphoma kinase (ALK)-negative lymphomas of anaplastic large cell morphology (T and null phenotype) expressed LSP1 although the percentage of LSP1-positive tumour cells was variable, however, only seven out of 30 cases with ALK-positive lymphoma were LSP1 positive. LSP1 was expressed on Reed-Sternberg cells in 60 out of 66 cases with classic Hodgkin's disease but neoplastic cells were usually negative in lymphocyte predominant Hodgkin's. This study confirms the wide expression of LSP1 within haematopoietic neoplasms and its diagnostic value for a minority of lymphoid tumours that have lost CD45 expression. Furthermore, the strong expression of LSP1 in classic Hodgkin's disease, contrasting with its heterogeneous expression in ALK-negative anaplastic lymphomas, may help to distinguish the latter lymphomas from patients with tumour cell-rich Hodgkin's disease.

Apoptosis-related genes and proteins in Hodgkin's disease

During recent years it has become increasingly evident that L&H cells in nodular lymphocytic predominance (LP) Hodgkin's disease (HD) and Hodgkin and Reed-Sternberg (H-RS) cells in approximately half the cases of classical HD originate from B-lymphocytes, and that H-RS cells in most of the remaining cases of classical HD express a null phenotype. The pathogenesis of HD is unknown. An association with Epstein-Barr virus (EBV) has been suggested and there are also indications that genes involved in programmed cell death (apoptosis) may be implicated. In this study, the expression of four apoptosis-related proteins (bcl-2, bcl-x, bax and p53) in 53 cases of HD was examined and the data were correlated with the genotype, the EBV status and the phenotype (B, T or null) of the neoplastic cells. The H-RS cells expressed a B-cell phenotype in 3/3 cases of nodular LP and in 19/ 50 (38%) cases of classical HD. The remaining cases showed a null-cell phenotype in 29/50 (58%) and a T-cell phenotype in 2/50 (4%). EBV was more often positive in B (14/19, 74%) than in null (7/29, 24%) type HD. The H-RS cells were bcl-2-positive in 19/53 (36%), bcl-x-positive in 17/53 (32%), bax-positive in 1/53, and p53-positive in 41/53 (77%) cases. No relationship was found between bcl-2 expression and EBV status, or between bcl-2 and bcl-x expression. A t(14;18) translocation was seen in 2 of 34 cases. P53 point mutations were not detected. Our findings indicate that nodular LP and classical HD originate from B-cells in a high proportion of cases. They also suggest a role for bcl-2, bcl-x and p53 in tumorigenesis. The pathogenesis is not known at this stage.

CD30 in normal and neoplastic cells

In 1982 Stein and coworkers identified a new molecule, CD30 (Ki-1), which is expressed by Reed-Sternberg (RS) cells of Hodgkin's Disease (HD) (1). Although CD30 is not a specific RS cell marker, its characterization has assumed an important role not only in the differential diagnosis of HD, but also in the identification of a morphologically and clinically distinct type of large cell lymphoma, now designated as anaplastic large cell lymphoma (ALCL) (2). The cloning of human and murine CD30 and the utilization of genetically manipulated animal models have rapidly expanded our knowledge on its physiological role in lymphoid development and differentiation. The goal of this review is to present an overview of this rapidly evolving field and discuss the role of CD30 in normal and neoplastic lymphoid cells.

How to differentiate between T-cell-rich B-cell lymphoma and lymphocyte-predominant Hodgkin's disease. Evidence for the value of MB1 and 4KB5 immunostaining

Striking morphological similarities exist between T-cell-rich B-cell lymphoma and lymphocyte-predominant Hodgkin's disease (Hodgkin's paragranuloma), making the distinction between them extremely difficult. Immunohistochemistry provides a means of overcoming this difficulty. Immunostaining with UCHL1, L26, MB1, and 4KB5 was performed on five T-cell-rich B-cell lymphomas and 11 Hodgkin's paragranulomas (7/11 nodular, 4/11 diffuse). L26 stained the tumour cells not only of T-cell-rich B-cell lymphomas, but also of L+H Hodgkin's disease. In contrast, MB1 as well as 4KB5 identified all of the neoplastic cells in 3/5 T-cell-rich B-cell lymphomas, but did not react with the L+H cells in 8/11 Hodgkin's paragranulomas. Some overlap of staining patterns became apparent in the remaining cases, with 2/5 T-cell-rich B-cell lymphomas showing the MB1+/4KB5+ phenotype in a tumor cell subset only. Similarly, in 3/11 Hodgkin's paragranulomas, some MB1/4KB5-positive L+H cells occurred in addition to MB1/4KB5-negative L+H cells. These cases, nevertheless, could be distinguished from one another by the numbers of MB1/4KB5-positive background lymphocytes, which were scanty or absent in T-cell-rich B-cell lymphomas and more numerous in Hodgkin's paragranulomas.

Immunohistochemistry in diagnostic dermatopathology

Immunopathology continues to be important in diagnostic dermatopathology. Immunopathology is an invaluable tool for assessing the tissue of origin or direction of differentiation of cells. In some cases this can result in a more precise diagnosis. This article reviews the role of immunopathology in determining the biologic behavior of hematolymphoid infiltrates. It explores the methodology of immunoperoxidase, discusses the most commonly used antibody reagents, and presents a series of diagnostic dilemmas in which immunopathology can be useful. In each case a strategy is established that maximizes the likelihood of making a definitive diagnosis.

Fallibility of transthoracic needle biopsy of anterior mediastinal masses

Percutaneous transthoracic core needle biopsy has been advocated as a highly accurate technique for the diagnosis of anterior mediastinal masses. A patient is described with a large anterior mediastinal mass in whom the diagnosis of mediastinal carcinoid tumour was made by transthoracic core needle biopsy. At definitive surgical resection the tumour proved to be a B cell lymphoma. This case illustrates one of the important limitations of needle biopsy with its potential for sampling error.

Specific cutaneous infiltrates of B-cell chronic lymphocytic leukemia arising at the site of herpes zoster and herpes simplex scars

BACKGROUND

Cutaneous lymphoid infiltrates at sites of herpes zoster scars in patients with B-cell chronic lymphocytic leukemia (B-CLL) often are diagnosed as benign lymphoid hyperplasia ("pseudolymphomas"). The histologic and immunophenotypic features of these lesions are not well characterized. Appearance of skin lesions in B-CLL patients is considered a poor prognostic sign.

METHOD

Eight punch biopsies from five patients (three males, two females; mean age, 66.7 years) affected by B-CLL and presenting with lesions at sites of previous herpes simplex (upper lip, one patient) or herpes zoster (trunk, four patients; forehead, one patient) infections were included in the study. Histologic examination was performed on routine sections stained with hematoxylin and eosin and Giemsa. Immunohistologic stainings were performed with a standard three-step immunoperoxidase technique on formalin fixed, paraffin embedded tissue sections.

RESULTS

Specific cutaneous infiltrates of B-CLL were diagnosed histopathologically and immunophenotypically in eight biopsies from all five patients. Clinically, patients presented with erythematous papules or plaques confined to the area of previous herpes virus eruptions. Histopathologic features in most cases were characterized by a variably dense perivascular and periadnexal infiltrate of small hyperchromatic lymphocytes throughout the entire dermis, reaching the subcutaneous fat. In one case, a dense, diffuse infiltrate involving the entire dermis was observed. A granulomatous reaction with presence of epithelioid and multinucleated giant cells was a prominent feature in four biopsies from three patients. Light areas containing large lymphoid cells with features of prolymphocytes and paraimmunoblasts (so-called "proliferation centers") could be observed only in the case characterized by a diffuse infiltrate. Immunohistology revealed an aberrant CD20+/CD43+ phenotype of neoplastic B cells, which is not found in normal B lymphocytes (CD20+/CD43-). Reactive T lymphocytes were present in all lesions and had a normal CD20-/CD43+/CD45Ro+ phenotype. At the time of this writing, four patients were alive without signs of skin disease after a mean follow-up of 58.5 months, and one patient died of B-CLL 24 months after the cutaneous eruption.

CONCLUSIONS

Specific cutaneous infiltrates of B-CLL are not uncommon at sites of herpes virus scars. The diagnosis can be confirmed by histopathologic and immunophenotypic criteria. The prognosis is better than previously reported.

T cell-rich B cell lymphoma bearing Epstein-Barr virus in tumor cells: a case of IBL-T-like lesion following Lennert's lesion

A case of T cell-rich B cell lymphoma (TCRBCL) with Epstein-Barr virus (EBV) infection in tumor cells is reported. A 50 year old male developed right cervical lymph node swelling in July 1988. Initial biopsy in April 1989 demonstrated many scattered Hodgkinoid atypical cells with Lennert's lesion. After partial remission following chemotherapy, the lymph nodes enlarged again, and a second biopsy in February 1991 showed an IBL-T-like lesion. Only a small number of Hodgkinoid atypical cells were still observed. After apparently, complete remission, the lesion soon recurred and the patient died in November 1992. Immunohistochemically the Hodgkinoid cells were positive for L26, but negative for LN2, LN3, UCHL-1, MT1, lysozyme, Ber-H2 and Leu-M1. Reactivity for immunoglobulins showed false-positive because of polyclonal staining. IgH monoclonality was detected by the polymerase chain reaction method in the first biopsied specimen, and by Southern blotting in the second biopsied snap-frozen specimen. Monoclonal TCR beta rearrangement was not detected. The Hodgkinoid atypical cells were positive for EBV-encoding RNA by in situ hybridization, and LMP-1 by immunostaining. Occasionally, EBV-bearing immunoblastic, medium sized, or small lymphocytic cells were also observed. This case indicates the possibility that EBV is related to the pathogenesis of TCRBCL.

Histiocytic sarcomas

In this study a review of malignancies classified as histiocytic in the literature is given. The available data suggest that a distinction can be made between three main categories, i.e., follicular dendritic cell (FDC) sarcomas, Langerhans' cell/interdigiting reticulum cell (LC/IRC) sarcomas and histiocytic sarcomas (HS) which are unrelated to accessory cells. With the exception of FDC sarcomas these tumours are high-grade malignancies with an aggressive course and short survival times. The recognition of FDC sarcomas and LC/IRC sarcomas rests upon the identification of accessory cell related antigens (e.g., R4/23, CD1a, S-100) and/or the demonstration of desmosomes or Birbeck granules. The diagnosis of HS which are unrelated to accessory cells is more complex. These tumours are heterogenous with respect to morphology and phenotype and can only be recognized with the use of an extensive panel of antibodies supplemented when possible by analysis of T-cell receptor--or immunoglobulin genes.

Immunophenotypic analysis of childhood Burkitt's lymphoma in the West Midlands 1957-1986

AIMS

To analyse the immunophenotype of a large number of non-endemic Burkitt's lymphomas to determine whether a B cell phenotype is consistently recognisable using formalin fixed, paraffin wax embedded archival material and a standard panel of commercially available antibodies.

METHODS

Archival material was obtained from 30 cases of childhood Burkitt's lymphoma registered with the West Midlands Regional Children's Tumour Research Group. These were analysed by a standard avidin biotin complex immunoperoxidase method using antibodies to CD45, CD43, CD30, CD20, CD15, and immunoglobulin heavy and light chains.

RESULTS

There was a high incidence of the CD45RB and CD20 immunophenotypes, with a clearly recognisable B cell lineage even in archival material. IgM was identifiable in 13 of the 23 (56.5%) cases tested. Only three of 17 (18%) cases expressed CD30. Positive membrane staining with CD45RO was observed in two (6.7%) cases.

CONCLUSIONS

A B cell lineage can be identified in Burkitt's lymphoma in formalin fixed, paraffin wax embedded material, even in archival tissue. There was a low incidence of membrane staining with CD45RO which is a potential source of diagnostic confusion.

Immunohistochemical analysis of CDw52 antigen expression in non-Hodgkin's lymphomas

AIM

To determine the antigen expression of CDw52 using Campath-1 antibodies in a series of non-Hodgkin's lymphomas (NHLs).

METHODS

Tissue sections of lymphoma were stained immunohistochemically using rat Campath-1G and humanised Campath-1H with avidin-biotin-peroxidase complex techniques. Fifty-two fresh frozen lymphomas and a further 26 paraffin wax embedded sections were studied.

RESULTS

Thirty-seven out of 41 B cell lymphomas were positive with Campath-1H in frozen sections (low grade, 24 of 24; high grade, 13 of 17) as were three out of five T cell lymphomas. Reed-Sternberg cells in six cases of Hodgkin's disease did not react. Eleven out of 16 high grade B cell lymphomas also stained positively with Campath-1G in paraffin wax sections as did five out of 10 T cell lymphomas.

CONCLUSIONS

The Campath-1 antibodies showed that CDw52 antigen expression was present in all cases of low grade B cell NHL examined. Immunohistochemical staining in high grade B cell NHL and in T cell NHL was variable. These findings may be relevant to patient selection when considering treatment with Campath-1 antibodies.

Immunohistochemistry in tumour diagnosis

Immunohistochemistry has become an everyday tool in diagnostic pathology to increase the accuracy in tumour typing. The most important recent developments include availability of antibodies to leukocyte CD-antigens to nearly comprehensive lymphoma typing in paraffin sections, use of certain leukocyte antigens in typing of solid tumours, e.g. neoplastic endothelial cells (CD31 and CD34) and neural and neuroendocrine cells (CD56, CD57). Also, antibodies selective to individual keratin polypeptides promise to be helpful in the subtyping of carcinomas. Antibodies to nuclear proliferation antigens have already proven helpful in the characterization of tumour cell proliferation. Technical optimization, using enzymatic and non-enzymatic antigen retrieval modalities, is also important.

CD43 expression in B cell lymphoma

AIMS

To determine the expression of CD43 in frozen sections in a range of B cell lymphomas.

METHODS

The monoclonal antibody WR14, clustered provisionally in the Fourth Leucocyte Typing Workshop as a CD43 reagent, was investigated by epitope blocking studies on formalin fixed reactive lymph node tissue, using the established CD43 antibody MT1, to validate its use as a CD43 reagent. CD43 expression was studied in 131 immunophenotypically defined B cell lymphomas, including lymphocytic lymphoma (Lc, n = 13), centrocytic lymphoma (Cc, n = 14), and a range of follicle centre cell lymphomas (FCC) including centroblastic/centrocytic follicular (CbCcF, n = 48), centroblastic diffuse (CbD, n = 39), centroblastic/centrocytic diffuse (CbCcD, n = 4), centroblastic follicular and diffuse (Cb FD, n = 3) and centroblastic/centrocytic follicular and diffuse (CbCc FD, n = 1). Nine lymphomas of mucosa associated lymphoid tissue (MALT) were also examined.

RESULTS

Epitope blocking studies showed that WR14 is a CD43 reagent that binds to an epitope identical with or close to that recognised by MT1. Eleven of 13 (84%) cases of Lc and 11 of 14 (78%) cases of Cc expressed CD43; 87 of 95 (91%) cases of FCC did not. All eight low grade lymphomas of MALT were negative. One high grade lymphoma, transformed from a low grade MALT lymphoma, was positive for CD43. The expression of CD43 by tumours of B cell lineage was associated with the expression of CD5 (p < 0.001) although either antigen could occasionally be found in the absence of the other.

CONCLUSION

CD43 reagents can be used in conjunction with CD5 antibodies for the immunophenotypic discrimination of follicle centre cell lymphomas from non-follicle centre cell lymphomas.

Monoclonal antibody OPD4 detects neoplastic T cells but does not distinguish between CD4 and CD8 neoplastic T cells in paraffin tissue sections

Monoclonal antibody (MoAb) OPD4, reported to preferentially react with benign CD4 T cells in formalin-fixed tissue sections, was examined for its reactivity with 56 T-cell neoplasms after formalin or Bouin's fixation to determine if it also preferentially detects neoplastic CD4 T cells in paraffin tissue sections. Monoclonal antibody OPD4 did not preferentially detect neoplastic CD4 T cells, since it reacted with 22 of 38 (58%) CD4-positive compared with nine of 14 (64%) CD4-negative T-cell neoplasms. However, MoAb OPD4 appears to detect neoplastic T cells in Bouin's-fixed (11 of 20 cases [55%]) about as well as in formalin-fixed (20 of 32 cases [63%]) tissues. Since MoAb OPD4 does not preferentially react with neoplastic CD4 T cells, the utility of this MoAb as a pan-T-cell marker in routinely processed tissues was also explored and compared with that of Leu-22, UCHL-1, and CD3. All four antibodies reacted with approximately the same percentage of T-cell malignancies (51% to 57%). However, examination of different clinicopathologic groups and types of fixative highlighted differences. Monoclonal antibodies OPD4 and Leu-22 reacted with 62%, while CD3 detected only 41% of formalin-fixed, postthymic T-cell neoplasms. OPD4, UCHL-1, and CD3 each reacted with 55%, but Leu-22 recognized only 45% of Bouin's-fixed, postthymic T-cell malignancies. OPD4 reacted with none, but CD3 reacted with all four T-cell lymphoblastic lymphomas. Various antibody combinations were examined to determine an optimal panel for the recognition of T-cell neoplasms in paraffin sections. The combination of MoAbs OPD4 and Leu-22 detected 86% of postthymic T-cell neoplasms in formalin-fixed tissue sections. Furthermore, MoAb OPD4 appears to be relatively specific for T-cell neoplasms, detecting 31 of 56 (55%) T-cell malignancies, while only reacting with two of 39 (5%) B-cell neoplasms. Therefore, while not preferentially reactive with neoplastic CD4 T cells, MoAb OPD4 may be useful as a pan-T-cell marker of postthymic T-cell neoplasms in routinely processed, formalin-fixed tissues, especially when used in conjunction with MoAb Leu-22.

Spontaneous remission of multilobated non-Hodgkin lymphoma

A 67-year-old previously well man was admitted with hypercalcemia after a 7-day history of fever, night sweats, and back pain. The blood showed 5% multilobated lymphoid cells. A multilobated non-Hodgkin lymphoma associated with marked reticulin fibrosis was diagnosed on a bone marrow biopsy. During the next 7 days, in the absence of specific therapy, his symptoms disappeared. Three weeks after admission his biochemical abnormalities had resolved and after 9 weeks his bone marrow examination was normal. He remains in clinical and laboratory remission 16 months after presentation.

Nodular sclerosing Hodgkin's disease and large cell lymphoma. Immunophenotypic characterization of a composite case

Composite lymphomas have rarely been reported in Hodgkin's disease (HD), except in the lymphocyte predominance sub-type, and immunohistochemical investigations have been performed in only few cases. We describe the histological and immunophenotypical findings in a case of composite nodular sclerosing HD and high-grade, large cell non-Hodgkin's lymphoma (NHL). In our case HD and NHL cells displayed striking morphological and immunophenotypical divergence, suggesting a lack of correlation between the two neoplasms.

Pathological variables determining the prognosis of non-Hodgkin's lymphomas

The aim of the study was to assess the role of pathological grade, cell proliferation, ploidy, immunophenotype and site in determining the prognosis of non-Hodgkin's lymphomas. Of particular interest was the relative value of grades derived from the Kiel classification as opposed to the National Cancer Institute (NCI) working formulation. The study consisted of 181 cases, treated in a relatively uniform way over an 18-month period spanning 1986. Using life table analysis, both NCI working formulation grade and Kiel grade correlated strongly with survival. However, the differences between grades were entirely due to an excess of early deaths in the high-grade and intermediate-grade categories. In patients surviving greater than 0.1 years (37 days), phenotype, site, ploidy and cell proliferation had no effect on survival. There was no evidence that intermediate-grade tumours, when subdivided into Kiel low- and high-grade types, differed in survival from tumours graded as low- or high-grade by both methods. However, NCI working, formulation high-grade tumours, especially those with a high proliferation rate, formed a group with a very high likelihood of death within 0.1 years.

Childhood lymphoma in Yorkshire

AIMS

A histopathological review of 43 cases of childhood non-Hodgkin's lymphoma (NHL) in an attempt to identify histological variables of prognostic importance.

METHOD

Each case was reclassified according to the Working Formulation and an attempt made to allocate an immunophenotype using a panel of monoclonal antibodies. Results were correlated with clinical data on site and survival.

RESULTS

Of the 43 cases, 30 were males and 13 females. There were 17 cases of lymphoblastic lymphoma, 15 cases of small non-cleaved cell lymphoma (SNCC), and four cases of large cell lymphoma. The SNCC group was subdivided into 10 cases of Burkitt's lymphoma and five cases of non-Burkitt's lymphoma. An immunophenotype was allocated in 65.1% of cases (23 B, 5 T). The SNCC cases were spread throughout the 0-16 year age range while the lymphoblastic lymphoma cases tended to occur in older children. Most mediastinal tumours were lymphoblastic lymphoma and most abdominal tumours were SNCC. Statistical analysis failed to show a significant difference in survival among histological subgroups or immunophenotypes.

CONCLUSION

No histological variables of prognostic importance were identified partly due to the great variation in treatment regimens, standard of supportive care, and prognosis over the period of the study (1972 to 1988).

Large cell, multilobated, B-cell lymphoma of the palate. A case report

A case of B-large cell non-Hodgkin's lymphoma (NHL) with multilobated nuclei arising in the palatal mucosa is described. Immunologic typing of tumor cells was crucial to determine the exact cell of origin of this lesion.

Immunohistochemical evaluation of bone marrow lymphoid nodules in chronic myeloproliferative disorders

One hundred and seventy bone marrow biopsies from patients with chronic myeloproliferative disorders (CMPDs) were evaluated for the presence of lymphoid nodules (LNs) and were immunostained using a panel of monoclonal antibodies (UCHL1, 4KB5 and L26) recognizing different lymphocyte antigens. LNs were found in 35% of cases of idiopathic thrombocythaemia, 24.6% of myelofibrosis/osteomyelosclerosis, 18.2% of polycythaemia vera 12.1% of chronic myeloid leukaemia and 19.2% of borderline cases. Varying degrees of immunohistochemical positivity for the three antibodies tested were found. LNs were always made up of variable proportions of both T- and B-lymphocytes with a prevalence of T-cells. This latter observation suggests that bone marrow LNs in CMPDs could be an expression of reactivity.

Use of monoclonal antibodies for the typing of malignant lymphomas in routinely processed biopsy samples

Eight antibodies (UCHL1 (CD45RO), MT1 (CD43), MT2 (CD45R), 4KB5 (CD45R), MB1 (CD45R), MB2, L26 (CD20) and LN1 (CDw75)) have been examined for reactivity with routine specimens of normal and hyperplastic lymphoid organs (n = 6), non-Hodgkin's lymphomas (n = 62), Hodgkin's disease (n = 27) and non-lymphoid malignancies (n = 9). In normal and hyperplastic lymphoid organs, UCHL1 and MT1 stained predominantly T cells; 4KB5, MB1, MB2, L26 and LN1 stained predominantly B cells; and MT2 reacted with a subset of B and T cells. The lineage of the neoplastic cells was correctly identified in 24 of 28 (86%) peripheral T-cell lymphomas; and in 31 of 35 (88%) B-cell malignancies. In two cases of lymphocyte-predominant Hodgkin's disease, the Hodgkin's and Reed-Sternberg (H&RS) cells were 4KB5+, L26+ and/or LN1+. The H&Rs cells in nodular sclerosis and mixed cellularity Hodgkin's disease were positive with 4KB5 in 17 of 25 cases. Antibodies UCHL1, MT1, MB1, MB2, L26 and LN1 also labelled some H&RS cells, but in a much smaller proportion of the cases. In three of nine non-lymphoid neoplasms, UCHL1 and MB2 showed a staining of the neoplastic cells, but the staining was cytoplasmic rather than membrane-associated. The remaining antibodies were unreactive with the non-lymphoid malignancies. It is concluded that many non-Hodgkin's lymphomas can be typed in routine specimens, and that antibodies UCHL1, MT1, L26 and LN1 are especially useful in this respect. The antibodies do not provide a means of distinguishing between non-Hodgkin's lymphomas and Hodgkin's disease.

Immunophenotyping of Nigerian cases of non-Hodgkin's lymphomas on paraffin sections

One hundred cases of routinely fixed and processed non-Hodgkin's lymphoma from Nigeria were immunostained with a small panel of monoclonal antibodies against B-, T- and macrophage antigens. The aims of the study were to assess the suitability of stored material from a country like Nigeria for immunohistochemical examination and the ability of the antibody panel to evaluate the distribution of B- and T-cell neoplasms. Eighty-seven of the 100 cases gave interpretable immunostaining, with 75 being B-cell and 12 T-cell neoplasms. Eighty-seven of the 100 cases gave interpretable immunostaining, with 75 being B-cell and 12 T-cell neoplasms. There were no tumours of macrophage lineage. Four cases gave satisfactory staining of reactive lymphoid cells but no reactivity with malignant cells and thus were not phenotyped. The remaining nine cases gave no staining of neoplastic or reactive cells, suggesting that they were unsuitable for immunohistochemical study, presumably because of inappropriate fixation and handling. We concluded that a panel of three monoclonal antibodies is suitable for routine immunostaining of conventionally fixed and processed blocks in Third World countries and will give diagnostically useful information in approximately 95% of cases.

Sinonasal non-Hodgkin's lymphomas and Wegener's granulomatosis: a clinicopathological study

Reports of sinonasal non-Hodgkin's lymphomas, analysed with monoclonal antibodies, are scarce, and differentiation of these lymphomas from Wegener's granulomatosis can be difficult. In this study, we investigated histopathologically and immunohistologically 20 cases of non-Hodgkin's lymphoma, primary in the sinonasal region, and sinonasal biopsies from 11 patients with Wegener's granulomatosis. All T-cell lymphomas (n = 7) and plasmacytomas (n = 4) were stage I at clinical presentation, while all B-cell lymphomas (n = 9) presented at higher stages. T-cell lymphomas tended to be more frequent in the nasal cavity and paranasal sinuses; B-cell lymphomas more often presented in the nasopharynx. Remarkably, 1 B-cell lymphoma expressed MT1, and 1 T-cell lymphoma expressed L26 (CD 20). The follow-up of 2 patients with a clinical diagnosis of Wegener's granulomatosis was suggestive of non-Hodgkin's lymphoma. Retrospective immunohistochemical analysis revealed that the original histological diagnosis of non-specific inflammation had to be changed to T-cell lymphoma, pleomorphic small cell type. We conclude that a biopsy from the sinonasal region with a dense inflammatory infiltrate, consisting predominantly of T-lymphocytes, renders a diagnosis of Wegener's granulomatosis unlikely and is at least suspicious of T-cell lymphoma. Immunohistochemical analysis is warranted for this type of biopsy.

What's new in primary central nervous system lymphomas?

Primary central nervous system lymphomas (CNSL) are uncommon neoplasms accounting for about 1% of primary brain tumors. Patients with congenital or acquired immunodeficiencies including AIDS patients and transplant recipients represent the main high-risk population for CNSL occurrence. An important point emerging from the literature is that CNSL incidence has dramatically increased during the last years not only in HIV infected patients by virtue of the AIDS epidemic spread, but also for unclear reasons in immunologically normal persons. Although c-myc oncogene activation and Epstein-Barr virus infection are considered to play a role in CNSL development, the peculiar tendency of these lymphomas to occur and remain inside the CNS is not well understood and may involve putative CNS binding molecules carried by lymphocytes. The clinical presentation is characterized by a great variety of neurological disorders. Radiological features consist of hyperdense homogeneous deposits within the subcortical white matter with a pattern of marked enhancement after injection of contrast material. The tumor masses are usually ill-defined and multicentric. Although all cytological types can be observed, the most common types belong to the high-grade category of non-Hodgkin's lymphoma. Monoclonal antibodies reactive with formalin-fixed, paraffin-embedded sections can be used in conjunction with stereotactic needle biopsy to provide accurate immunological characterization of CNSL. The large majority of CNSL is of B-cell origin but T-cell lymphomas seem at the present time less exceptional than previously thought. Although radiotherapy and chemotherapy can increase length of survival, the prognosis of CNS remains dramatically poor, the shortest survival being observed in AIDS patients.

Primary lymphoma of the liver: clinical and pathological features of 10 patients

Nine out of 10 patients with primary lymphoma of the liver presented in a manner that did not suggest a tumour. The initial diagnoses were chronic active hepatitis in three cases and "granulomatous cholangitis", inflammatory pseudotumour, and anaplastic carcinoma in one case each. Moreover, extensive haemorrhagic necrosis in three cases initially suggested the Budd-Chiari syndrome. All the tumours were diffuse non-Hodgkin's lymphomas like the 50 cases reported previously, but they differed from most of these in that nine were of T cell phenotype. Five were pleomorphic small T cell, two T zone, and two T lymphoblastic lymphomas: only one was centrocytic and of B cell lineage. This report extends the range of clinical manifestations (diffuse hepatomegaly without a tumour), histological appearances (resemblance to chronic inflammatory or vascular liver diseases) and phenotype (of T cell lineage) of primary lymphoma of the liver: these features seemed to be related in this series. Recognition is important as prognosis remains favourable in appropriately treated cases. Although the appearances of the liver biopsy specimens may be difficult to interpret, the destructiveness of the infiltrate is an important clue to the diagnosis.

Diagnosis of primary cerebral lymphoma with particular reference to CT-guided stereotactic biopsy

In establishing the histological diagnosis of primary cerebral lymphoma, stereotactic brain tumour biopsy is the method of choice as the mainstay of therapy is radiation and chemotherapy. This study describes the histopathology and diagnostic immunohistochemistry of 54 primary brain lymphomas in a mainly non-AIDS population. The stereotactic biopsies were performed using the Leksell CT stereotactic frame and a spiral needle which procured about 10-mm-long tissue cylinders. Usually, three successive biopsy cylinders were taken along the target trajectory. Histological examination revealed the prevalence of high-grade non-Hodgkin's lymphoma of the polymorphous centroblastic type. The series did not include any low-grade lymphomas or T-cell lymphomas. L-26 immunohistochemistry resulted in a positive staining of the blasts, thus confirming the B-cell origin of primary brain lymphomas. Small reactive T-lymphocytes and monohistiocytic cells were also found within and at the periphery of the lymphomas and in areas of degeneration. In the biopsies of nine patients, who had shown significant reduction of the lesions on the CT scans, after corticosteroid medication, regressive tissue changes were predominant and consisted of T-lymphocytes, macrophages, and occasionally bizarre reactive astrocytes.