Immunohistochemical Examination of Routinely Processed Bone Marrow Biopsies

Path. Res. Pract. 188, 707-713 (1992)

Immunohistochemical Examination of Routinely Processed Bone Marrow Biopsies M. Werner, V. Kaloutsi, K. Walter, T. Buhr, J. Bernhards and A. Georgii Pathologisches Institut der Medizinischen Hochschule Hannover, FRG

SUMMARY

Immunohistochemistry was perfonned on paraffin sections of 169 bone marrow biopsies fixed in a buffered methanol-formalin solution and decalcified with EDTA. The biopsies included specimens with normal hematopoiesis, and specimens that were affected by various hematological dis orders as well as some metastatic carcinomas. The results demonstrate that a wide spectrum of antigens was preserved in routinely processed bone marrow biopsies, even after long-term fixation up to 12 days. Markers for granulopoietic cells were lysozyme, elastase, DAKO-M 1, and MT 1. Megakaryopoiesis was stained with glycoprotein lIla, von Willebrand factor, and Ulex europaeus agglutinin (UEA), and erythropoiesis with LN 1. Normal lymphocytes as well as lymphoma cells of all non-Hodgkin's lymphomas tested were positive for leukocyte common antigen (LCA), and at variable degree, for MB 1,4 KB 5, LN 1, LN 2, UCHL 1,01" MT 1. Reed-Sternbergand Hodgkin's cells in Hodgkin's lymphomas were reactive with Ber-H 2, LN 2 and Dako-M 1. In plasma cell dis orders, staining for immunoglobulin light chains gave best results. Metastatic carcinomas showed predominantly staining with EMA, and KL 1. A selected panel of specific cell markers is proposed, which proved to be helpful in routine bone marrow diagnosis in most cases.

Introduction Histopathological examination of bone marrow biopsies is weil established for the diagnosis of hematological disorders 3, 11, 12. Phenotyping of hematopoietic ceHs in histological sections of bone marrow biopsies, however, is limited by the sensitivity of most leucocyte antigens to the procedures necessary for tissue processing, e.g. formalin fixation, decalcification and paraffin embedding l8 . Recently, new mono clon al antibodies became available which bind to formalin-resistant epitopes on leukocyte antigens 8, 24, 26, 32. A number of antibodies were found to be also reactive in decalcified bone marrow biopsies 2,9,21, and several authors suggested- that these ceH markers may be useful for routine bone marrow diagnosis 9, 14, 31. In most immunohistochemical studies, however, different condi© 1992 by Gustav Fischer Verlag, Stuttgart

tions of fixation, decalcification and embedding have been used that are known to influence the antigenicity of the tissues to a varying degree 2, 9,19,21. We tested the reactivity of aseries of monoclonal antibodies against leukocyte antigens and epithelial ceHs in routinely processed bone marrow biopsies. The samples were fixed in a buffered methanol-formalin solution which was developed in our laboratory, decalcified with EDTA and embedded in paraffin. The specimens included biopsies with normal hematopoiesis and biopsies affected by variqus hematological disorders and metastatic tumors. Aim of the study was to test both value and reliability of immunohistochemical analyses of routine bone marrow biopsies. Furthermore, a clear panel of distinct ceH markers should be established that would be useful in bone marrow diagnosis. 0344·0338192/0188·0707$3.50/0

708 . M. Werner et al.

Material and Methods

Tissue specimens Yamshidi biopsies were obtained from the posterior iliac crest of 169 patients. The specimens were immediatcly immersed in Hannover's solution? which consisted of 1,667 ml neutral buf-

T able 1. Histopathological diagnoses of thc bone marrow biopsics examined Histopathological diagnosis

N

Normal hematopoiesis Reactive plasmocytosis or lymphocytosis Acute mycloid leukemia Chronic myeloid leukemia Polycythemia vera Thrombocythemia Chronic megakaryocytic-granulocytic myelosis Myelodysplastic syndrome Acute lymphoblastic leukemia Chronic Iymphocytic leukemia Non-Hodgkin's lymphoma Hodgkin's lymphoma Plasmocytoma Metastatic carcinomas Total

15 17 13 9 4

5 7

13 2 12 20 4 34 14 169

fered formalin (12%), 3,200 ml methanol (64%), 133 ml phosphate buffer and 6.65 g glucose. The fixation time varied from 18 h up to 12 days (median 3.2 days) depending on the time required for mailing the specimens to the Hannover Institute of Pathology. The biopsies were decalcified in neutral buffered EDTA for 48-72 hand embedded in paraffin. Five-micron-thick sections were cut and stained with Giemsa and hcmatoxylin and eosin for histopathological examination. There were 15 specimens with normal hematopoiesis, and 17 biopsies with reactive plasmocytosis or lymphocytosis. The histopathological diagnosis of 137 specimens involved by various hematological disorders or metastatic carcinoma are shown in Table 1.

Immunohistochemistry The primary antibodies and other cell markers, the respective antigens, spccifications and sources are listed in Table 2. Paraffin sections (5 [tm) were deparaffined in xylol and rehydrated in graded ethanols. Pronase treatment was performed on seetions to be stained with glycoprotein lIla (GPlIla), and von Willebrand factor (vWf); trypsin treatment was done prior to incubation with elastase, lysozyme, Ber-H 2 and Ulex europaeus agglutinin (UEA). The reactivity of monoclonal antibodies was demonstrated by the alkaline phosphatase anti-alkali ne phosphatasc (APAAP) method4 . Unspecific binding was blocked by incubation with rabbit serum (DIANOVA) for 30 min at room temperature (RT). Then, primary antibodies were incubated for 1 h at RT. Rabbit anti-mouse immunoglobulins (DAKOPATTS) were addcd to the sections for 1 hat RT, folIowcd by incubation with APAAP for 30 min at RT (DAKOPATTS). The last 2 steps were repeated twice for the antibodies GPlIla and LN 1. For the detection of

Table 2. Specifications and sources of thc antibodies used in this study Antibody

Sourcc

Reactivity

Refercnce

Elastase Lysozyme DAKO-M 1 (CD ]5)'" GPlIla vWf Glycophorin A LCA (CD 45) 4 KB 5 (CD 45R) MB 1 (CD 45 RA) L 26 (CD 20) LN] (CDw 75) LN 2 (CD 74)

DAKOPATTS DAKOPATTS DAKOPATTS DAKOPATTS DAKOPATTS DIANOVA DAKOPATTS DAKOPATTS BIOTEST DAKOPATTS BIOTEST BIOTEST

Kramps et al. 1984 Tabilio et al. 1982 Stein et al. 1982 Gatter et al. 1988 Naiem et aI. 1982 Andersen et al. ]979 Warnke et aI. 1983 Dalchau and Fabre 1981 Poppema et al. 1987 Nortonetal.1987a Epstein et al. 1984 Epstein et al. 1984

MT 1 (CD 43) UCHL 1 (CD 45 RO) kappa, lambda, IgG, IgM, IgA Ber-H 2 (CD 30) EMA CK 1 KL 1 UEA NASDCE

DAKOPATTS DAKOPATTS DAKOPATTS

Granulopoietic cells Myelomonocytic cells Neutrophils, myelocytes, Reed-Sternberg-cells Platelets, megakaryocytes, megakaryoblasts Megakaryocytes, cndothelium Erythroid cells Lymphoid cells B-Iymphocytes B-lymphocytes, mature T-lymphocytes B-Iymphocytes B-Iymphocytes, crythrocytcs B-Iymphocytes, interdigitating cells; subregion of MHCII T-Iymphocytes, myeloid cells T-lymphocytes, granulocytes Plasma cclls

H. SteinIBerlin DAKOPATTS DAKOPATTS PRO GEN SIGMA SIGMA

Reed-Stcrnberg-ceIls, activated lymphoid ceIIs Epithelial cells Epithelial cells Epithelial cells Endothelium, megakaryocytes Granulopoietic ccHs

Schwarting et al. 1989 CordeIl et al. 1985 Moll et al. 1982

" CD = cluster of differentiation.

Poppema et al. 1987 Norton et al. 1986

Holthofer et al. 1982 Ledcr 1964

Immunohistochemistry in Bone Marrow Diagnosis . 709 polyclonal antibodies biotinylated horse anti-mouse immunoglobulins and avidin-biotinylated alkaline phosphatase complexes (VECT ASTAIN) were used according to the instructions of the manufacturers. All sections were developed using New Fuchsin as substrate. The peroxidase labcIed lectin Ulex europaeus agglutinin (UEA) was addcd to the sections for 2 h at RT and rcactivity was visualized by diaminobenzidine. Endogenous peroxidase was previously blockcd with 0.5% H2Ü2. Additionally, a commercialIy available kit (SIGMA) was used for thc development of the enzyme naphthol-AS-D-chloroacetate esterase (NASDCE). All preparations were counterstained with hematoxylin and mountcd with Kayser's glycerin.

Results Normal Hematopoiesis

Granulopoietic cells were stained by monoclonal antibodies lysozyme, elastase, Dako-M 1 and MT 1. Both elastase and lysozyme showed broad reactivity with mature and immature granulopoiesis. The staining was cytoplasmic, diffuse or granular. Staining for NASDCE was comparable to elastase with intense staining of immature and weaker staining of more mature cells. Monocytoid forms were positive for lysozyme. Dako-M 1 marked granulocytes and band forms; immature cells were only occasionally positive. MT 1 reacted with the plasma membrane of all granulopoietic cells including precursors and mature cells. In the majority of the specimens examined, anti-Ieukocyte comll10n antigen (LCA) labeled dissemina ted myelocytes and more differentiated cells, and UCHL 1 stained few granulocytes and band forms. Strong staining of cell membranes and cytoplasm was obtained with LN 1 in erythrocytes, normoblasts and erythroblasts. In proerythroblasts, staining was less intense. In few cases Ber-H 2 reacted with proerythroblasts and erythroblasts but displayed weak staining with normoblasts and was negative in erythrocytes. With glycophorin A, background staining was relatively high in most preparations, and erythroid cells were only weakly positive. In few cases, erythropoietic cells were positive with MT 1. Nearly all megakaryocytes in each section were stained by GPIIIa, vWf and UEA exhibiting strong granular or diffuse cytoplasmic staining. In some cells, a coarse paranuclear staining was noticed. MT 1 and Ber-H 2 rcacted with the cytoplasm or cell membrane of individual megakaryocytes in most sections. Plasma cells dispersed in the bone marrow were detected by antibodies against kappa and lambda light chains, and IgM, IgG or IgA heavy chains. Individual plasma cells were

stained by Ber-H 2, LN 1, or EMA. Reactivity with the cell membrane of bone marrow lymphocytes was observed with LCA, MB 1, 4 KB 5, LN 1, LN 2, MT 1 and UCHL 1. Staining with L 26 was weak or absent. Myeloproliferative Disorders and Myelodysplastic Syndromes

The different hematopoietic ceillines in myeloproliferative disorders and myelodysplastic syndromes (MDS) displayed staining patterns consistent with normal hematopoiesis (Fig. 1a-c, Table 3). In 13 ca ses of aeute myeloid leukemia, blastic cells were positive with elastase, lysozyme and NASDCE, but negative with LCA, and antibodies against lymphocyte subsets. Atypically localized my eIoblasts in MDS could be identified by elastase, lysozyme and NASDCE. LN 1 stained erythroblasts, which were increased in some cases of MDS (Fig. 1c). Dysplastic megakaryopoiesis in MDS as weil as micromegakaryocytes in chronic myelogenous leukemia or pleomorphic large megakaryocytes as seen in thrombocythemia, polycythemia vera and chronic megakaryocytic-granulocytic myelosis were strongly stained by GPIIIa, vWf and UEA (Table 3). Megakaryoblasts were positive with GPIIIa and vWf (Fig. 1 b). As also observed in normal bone marrow MT 1 showed broad reactivity with mature and immature granulopiesis including myeloblasts. Both LCA and UCHL 1 were merely positive in few mature granulopoietic cells (Table 3). Antibodies against B-lymphocytes stained only scattered lymphoid cells. Lymphoproliferative and Plasma Cell Disorders

The majority of the lymphoma cells in all nonHodgkin's lymphomas (NHL) tested were stained by LCA (Fig. le). Moreover, at least 2 additional markers against B- or T-cells were positive (Table 4, Fig. lf). In 5 B-cell lymphomas, however, UCHL 1 and MT 1 stained up to 10% of the lymphoid cells. Histological features of reactive lymphocytosis by the presence of smalllymphoid cells within the bone marrow were noticed in 5 patients. Those patients had no evidence of lymphoma and exhibited an equal admixture of B- and T-Iymphocytes in immunohistochemistry. Two aeute lymphatic leukemias were positive with LCA and antibodies against lymphocyte subsets. The Sternberg-Reed and Hodgkin's cells in Hodgkin's Imyphomas showed staining of the cell membrane or cytoplasm, either diffusely or as a paranuclear spot, with Ber-H 2 (4/4), LN 2 (3/4) and Dako-M 1 (2/4). In all 34 plasmocytomas, the neoplastic plasma cell population

Table 3. Results of immunostaining in chronic myeloproliferative disorders

Granulopoiesis

Elast.

Lysoz.

DakoMI

NASDCE MT 1

+

+

+

+

Megakaryopoiesis ,. scattercd granulocytes, band forms and myelocytes.

GPIIIa

vWf

UEA

+

+

+

+

LCA

UCHLI

+*

+"

710 . M. Werner et al. Table 4. Reactivity of monoclonal antibodies with non-Hodgkin's lymphomas in paraffin sections of bone marrow biopsies* Histopathol. diagnosis

N

LCA

MB 1

4KB5

LN 1

CLL

12 8 3 5 4

+ +

+ + +

+

± ±

Immunocytoma Cbcc Large cell lymphoma T-cell lymphoma

+ + +

+

+ ± ±

+

LN2

MT1

UCHL 1

±

=+=

=+=

+ + + +

=+=

+

+

* (+) all cases positive; (±) majority of the cases positive; (=+=) majority of the cases negative; (-) all cases negative. Abbreviations: CLL = chronic lymphatic leukemia ; cbcc = centroblastidcentrocytic lymphoma.

Fig. 1. Immunohistochemistry of decalcified and paraffin-embedded bone marrow specimens, APAAP-method and hematoxylin counterstain: (1 A) Positive reaction for elastase in immature granulopoiesis in a case with chronic myeloid leukemia, x 320. - (1 B) Reactivity of von Willen brand factor (vWf) with immature megakaryocytes and megakaryoblasts (arrow) in a ca se with myelodysplastic syndrome, x 480. - (1 C) Erythropoiesis is stained by antibody LN 1 in a ca se with myelodysplastic syndrome, x 480. - (1 D) Monoclonal antibody KL llabelling metastatic careinoma cells in a case with gastric cancer, x 420. - (1 E) Leukocyte common antigen (LCA) , x 180, and (1 F) 4 KB 5, x 420, in Non-Hodgkin's lymphoma.

Immunohistochemistry in Bone Marrow Diagnosis . 711

could be identified by antibodies against light and heavy chains. Staining with the latter was weak in some preparations. The majority of the plasmocytomas examined (n = 30) were positive for anti-lgG, 18 of them with kappa, and 12 of them with lambda light chain restriction. Three cases were stained by anti-lambda and anti-lgA, one case by anti-kappa and anti-lgA. Plasmocytoma could be excluded in 12 biopsies with reactive plasmocytosis by the presence of corresponding numbers ofkappa- and lambdapositive plasma cells.

Metastatic Carcinomas The metastatic carcinoma cells present in 14 bone marrow specimens were reactive with monoclonal antibodies both EMA and KL 1 (Fig. 1 d) in all cases examined including patients with carcinoma of the breast, stomach and prostate. Antibody CK 1 gave only weak staining in 4 cases. UEA stained carcinoma cells in 3 specimens. From the remaining antibodies, occasional staining of tumor cells was observed with LN 1 in 4, and LN 2, MB 1 and Ber-H 2, each in 2 biopsies. Discussion Most mono clon al antibodies used he re showed a distinct staining of bone marrow cells and allowed the identification of different celilines at all stages of differentiation. In general, our"staining results are in agreement with earlier reports 9, 14,32. Discrepant staining patterns of a few markers, e.g. glycophorin A, MT 1, UEA, CD 15, elastase, and NASDCE6,16,21,32, may be the result of varying conditions in tissue processing. Especially fixation and decalcification are known to influence the antigenicity of tissues 2,19,21. Mullink et al. (1985) described that the glycophorin A antigen was resistant to formalin fixation and several decalcification procedures including EDTA as used in our study. In our material the glycophorin A antigen may have been affected by the Hannover's fixative. Our fixative differs from that used by Mullink et al. (1985) in the addition of methanol. Interestingly, even after long-term fixation up to 12 days all other antigens tested were not affected by the buffered methanol-formalin solution. In recent immunohistochemical studies where fixatives without addition of methanol were used, the fixation time was limited to 18-24 h 9, 19,21,32. Hannover's solution, therefore, seems to be very useful for the fixation of routine bone marrow biopsies that have to be analyzed immunohistochemically and in which fixation time cannot necessarily be limited to 24 h. The value of the cell markers in routine bone marrow diagnosis has been tested on specimens involved by a variety of hematological and metastatic disorders. In our experience, a selected panel of distinct antibodies and enzymes is sufficient to support morphological diagnoses in bone marrow biopsies (Table 5). This panel includes at least two distinct markers for most of the hematopoietic celiline allowing reliable~phenotyping of the cells, even in the case of anyone's failing or weak reaction. For the identification of granulopoietic cells, elastase and lyso-

T able 5. Selected panel of antibodies for immunohistochemical analysis of bone marrow biopsies Ceilline

Antibody

Granulopoiesis Erythropoiesis Megakaryopoiesis Lymphoid cells Plasma cells Epithelial cells

Elastase, lysozyme, NASDCE* LN 1 GPIIIa, vWf LCA, MB 1, 4 KB 5, UCHL 1 kappa, lambda EMA, KL 1

<-

Enzyme naphthol-AS-D-chloroacetate esterase.

zyme were found to be most suitable. Mature and immature granulopoietic cells can be stained by lysozyme and elastase as shown earlier 27,30, although some authors describe that elastase is negative in myeloblasts 6 ,16,32. In contrast to Crocker and Burnett (1986), we demonstrated activity of NASDCE also in myeloblasts, whereas granulocytes were only weakly positive. Staining of myeloblasts for NASDCE is highly dependent upon the fixation procedure 6 . Evaluating 15 cases of acute leukemias we found that elastase, lysozyme and NASDCE may be useful for the determination of myelomonocytic or lymphatic differentiation which is important for the prognosis and therapy of the patients 13 . Erythropoietic cells were stained by LN 1, also Ber-H 2 in apart of the sections examined. The binding of both mono clon al antibodies to erythroid cells cannot be explained, because Ber-H 2 and LN 1 were originally generated against Hodgkin's cells and lymphoid cells, respectively8,28. Possibly, this reaction may reflect a cross-reaction with other antigens that were altered by our fixative. However, in our preparations Ber-H 2 and LN 1 also stained the cell types against which both antibodies were raised. The antibodies GPIIIa and vWf were strong markers for megakaryopoiesis. We found no difference between both antibodies in the labelling of mature as weIl as immature megakaryopoietic cells including megakaryoblasts. Lymphoid cells in all NHLs were stained by LCA. In addition, at least two markers against lymphocyte subsets were positive. For the identification of lymphoid cells in the bone marrow the panel of cell markers (Table 5) should therefore include LCA, MB 1, 4 KB 5, and UCHL 1, and optionally LN 1, LN 2 and MT 1. MT 1 as weIl as UCHL 1 were positive in up to 10% of lymphoid cells in 5 cases with CLL or immunocytomas. Ir cannot be excluded that these positive cells are reactive T-lymphocytes. T 0 our opinion an unspecific binding of MT 1 and UCHL 1 to the neoplastic cells of the B-cell lymphomas 25 ,26 seems to be more likely. The discrimination of reactive lymphocytosis from neoplastic infiltration of lymphoma cells may be difficult in some cases. In addition to morphological examination, staining for B- and T-cell markers may be helpful. Reactive or neoplastic plasma cells"could be differentiated with anti-light chain antibodies in all cases examined. Metastatic carcinoma cells in the bone marrow were reliably stained by monoclonal antibodies EMA and KL 1 which may be important for the detection of micrometastases.

712 . M. Werner et al.

Some antibodies examined were not found to be advantageous for immunohistochemical bone marrow dia gnosis and were, therefore, not included in the panel of selected cell markers. For instance, DAKO-M 1 (CD 15) was positive mainly in mature granulopoietic cells which can easily be identified by morphology. Others32 demonstrated broad expressionof CD 15 in immature granulopoietic cells except myeloblasts by use of monoclonal antibody Leu-M 1. MT 1 was proposed as marker for granulopoiesis 32 . In addition, MT 1 stained also megakaryocytes and erythroid cells in our material. In conclusion, this study demonstrates that immunohistochemistry in routinely processed bone marrow biopsies fixed with a buffered methanol-formalin solution gives reliable results. A selected panel of 15 cell markers was established that is helpful in routine bone marrow diag!loses for most hematological disorders. Acknowledgements The authors are grateful to Fabian Kausche for reading of the manuscript, and to Angelika Hörtinger and Anja Tenfelde for excellent technical assistance.

References 1 Anderson LC, Gahmberg CG, Teerenhovi L, Vuopio P (1979) Glycophorin A as a cell surface marker of early erythroid differentiation in acute leukemia. Int J Cancer 24: 717-720 2 Athanasou NA, Quinn J, Heryet A, Woods CG, McGree JOD (1987) EHect of decalcification agents on immunoreactivity of cellular antigens. J Clin Pathol 40: 874-878 3 Bartl R, Frisch B, Burkhardt R, Jäger K, Pappenberger R, Hofman-Fezer G (1984) Lymphoproliferations in the bone marrow: Identification and evolution, classification and staging. J Clin Pathol 37: 233-254 4 CordeIl JL, Falini B, Erber WN, et al. (1984) Immunoenzymatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J Histochem Cytochem 29: 219-229 5 CordeIl J, Richardson TC, Pulfot'd KAF, Ghosh AK, Gatter KC, Heyderman E, Mason DY (1985) Production of monoclonal antibodies against human epithelial membrane antigen for use in diagnostic immunocytochemistry. Brit] Cancer 52: 347-354 6 Crocker ] and Burnett D (1986) Granulocyte markers in histopathology: A review. ] Path 150: 77-88 7 Dalchau R, Fabre JW (1981) Identification with a monodonal antibody of predominantly B Iymphocyte-specific determinant of the human leukocyte common antigen. J Exp Med 153: 753-765 8 Epstein AL, Ivlarder R], Winter JN, Fox RI (1984) Two new mono clon al antibodies (LN 1, LN 2) reactive in B 5 fixed, paraffin-embedded tissues with follicular center and mantle zone human B-Iymphocytes and derived tumors. J Immunol 133: 1028-1036 9 Erber WN, McLachlan] (1989) Use of APAAP technique on paraffin-wax-embedded bone marrow trephines. ] Clin Pathol 42: 1201-1205 III Gatter KC, CordeIl JL, Turley H, Heryet A, Kieffer N, Anstee DJ, Mason DY (1988) The immunohistological detection of platelets, megakaryocytes and thrombi in routinely processed spccimens. Histopathology 13: 257-267

11 Georgii A, Vykoupil KF, ThieleJ (1984) Histopathology of bone marrow and clinical findings in chronic myeloproliferative disorders. In: Lennert K, Hübner K (Eds) Pathology of the Bone ivlarrow, pp. 147-169, Fischer Verlag, Stuttgart 12 Georgii A, Vykoupil K-F, Buhr Th, Choritz H, Döhler U, Kaloutsi V, Werner M (1990) Chronie myeloproliferative disorders in bone marrow biopsies. Path Res Pract 186: 3-27 13 Griffin ]D, Mayer R], Weinstein H], Rosenthai DS, Coral FS, Berverdidge RP, Schlossman SF (1983) Surface marker analysis of aeute myeloblastic leukemia: identification of differentiation-associated phenotypes. Blood 62: 557-562 14 Hall PA, Lindeman R, Butler MG, Amess JAL, D'Ardenne AJD (1987) Demonstration of lymphoid antigens in decalcified bone marrow trephines. J Clin Pathol 40: 870-873 15 Holthofer H, Virtanen I, Kariniemi AL, Hormia M, Linder E, Miettinen A (1982) Ulex europaeus I lectin as a marker for vaseular endothelium in human tissues. Lab Invest 47: 60-66 16 KrampsJA, van der Valk P, van der Sandt MM, LindemanJ, Meijer CJLM (1984) Elastase as a marker for neutrophilic myeloid cells. J Histochem Cytochem 32: 389-394 17 Leder LD (1964) Über die selektive fermentcytoehemische Darstellung von neutrophilen myeloischen Zellen und Gewcbsmastzellen im Paraffinschnitt. Klin Wochensehr 42: 553 18 Mason DY, Gatter KC (1987) The role of immunoeytochemistry in diagnostic pathology. J Clin Pathol 40: 1042-1054 19 Matthews JB, Mason Gi (1984) Influencc of decalcifying agents on immunoreactivity of formalin-fixed, paraffin-embedded tissue. Histochemical J 16: 771-787 20 Moll R, Franke 00, Schiller DL, Geiger B, Krepler R (1982) The catalog of human cytokeratins: Patterns of expression in normal epithelia, tumors and cultured eells. Cell 31: 11-24 21 Mullink H, Henzen-Logmans SC, Tadema TM, Mol lJ, Meijer CJLM (1985) Infillence of fixation and decalcification on the immunohistochemical staining of cell-specific markers in paraffin-embedded human bone biopsies. J Histochem Cytochem 33: 1103-1109 22 Naiem M, GerdesJ, Abdulaziz Z, Sunderland CA, Allington MJ, Stein H, Mason DY (1982) The value of immllnohistological screening in the production of monoclonal antibodies. J Immunol Meth 50: 145-160 23 Norton AJ, Ramsay AD, Smith SH, Berly PC, Isaacson PG (1986) Monoclonal antibody (UCHL 1) that recognizes normal and neoplastic T-eells in routinely fixed tissues. J CI in Pathol39: 399-405 24 Norton AJ, Isaaeson PG (1987a) Monoclonal antibody L 26: an antibody that is reactive with normal and neoplastic B Iymphocytes in routinely fixed and wax-embedded tissue. J Clin Pathol40: 1405-1412 25 Norton AJ and Isaacson PG (1987b) Detailed phenotypic analyses of B-celllymphoma using a panel of antibodies reactive in routinely fixed wax-embedded tissue. Am J Pathol 128: 225-240 26 Poppema S, HoHema H, Visser L, Vos H (1987) Monoclonal antibodies (MT 1, MT 2, MB 1, MB 2, MB 3) reactive with leukocyte subsets in paraffin-embedded tissue sections. Am J Pathol127: 418-429 27 Ralfkiaer E, Pulford KA, Lauritzen AF, Avnstrom S, Guldhammer B, Mason DY (1989) Diagnosis of aeute myeloid leukaemia with the use of monoclonal anti-neutrophil elastase (NP 57) reactive with routinely processed biopsy sampies. Histo,pathology 14: 637-643 28 Schwarting R, Gerdes J, Dürkopp H, Stein H (1989) Ber-H 2: A new high avidity monoclonal antibody to a formol resistant epitope of the Ki-1 (CD 30) antigen. Immunohistochemical and iml11unobioehemical characterization. Blood 74: 1678-1689

Immunohistochemistry in Bone Marrow Diagnosis . 713 29 Stein H, Uchanska-Ziegler B, Gerdes j, Ziegler A, Wernet P (1982) Hodgkin and Sternberg-Reed cells contain antigens specific to !ate ceHs of granuJopoiesis. [nt J Cancer 29: 283-290 30 Tabilio A, Falini B, Aversa F, Zuccaccia M, Cernetti C, Gerli R, Rutli 0, Grignani F, Martelli MF (1982) Intracytoplasmic lysozyme in malignant hematologic disorders: an immunoperoxidase study. Tumori 68: 417-425

31 van der Valk P, Mullink H, Huijgens PC, Tadema TM, Vos W, Mcijjer CJLM (1989) Immunohistochemistry in bone marrow diagnosis. Value of a panel of monoc!onal antibodies on routinely processed bone marrow biopsies. Am J Surg Pathol 13: 97-106 32 Warnke RA, Gatter KC, Falini B, Hildreth P, Woolston RE, Pulford K, CordeIl JL, Cohen B, Oe Wolf-Peeters C, Mason DY (1983) Diagnosis of human lymphoma with monoc!onal antileukocyte antibodics. N Engl J Med 309: 1275-1281

Received February 15, 1991 . Accepted in revised form August 28,1991

Key words: Bone marrow biopsy - Monoclonal antibodies - Hematological disorders - Hodgkin's lymphoma Cell marker Prof. Dr. med. A. Georgii, Direktor des Pathologischen Instituts der Medizinischen Hochschule Hannover, Konstanty-Gutschow-Str. 8, 0-3000 Hannover 61, FRG