Alpha-1-Antitrypsin and Lysozyme

Transcription

1 Alpha--Antitrypsin and Lysozyme Their Limited Significance in Fibrohistiocytic Tumors YLERMI SOINI, M.D. AND MARKKU MIETTINEN, M.D. A wide range of tumors were immuriohistochemically analyzed for alpha--antitrypsin (AAT) and lysozyme in order to evaluate their specificity as histiocytic markers and their significance in the diagnostic and histogenetic evaluation of fibrohistiocytic tumors. Besides histiocytic lesions, AAT immunoreactivity was commonly found in different types of carcinomas and sarcomas, and strong immunoreactivity was found in carcinoid tumors, malignant melanomas, and schwannomas, which, however, had negative results for lysozyme. The AAT immunoreactivity could be abolished with the absorption of the antibody with purified AAT also in nonhistiocytic tumors. The neoplastic pleomorphic cells in malignant fibrous histiocytomas (MFHs)' usually had strongly positive results for AAT, whereas only entrapped histocytes had positive results for lysozyme and for two monoclonal antibodies to histomonocytic cells. The results show that AAT has a relatively low specificity as a Histiocytic marker, and one should be careful in concluding the histiocytic nature of tumors, such as MFHs, based on AAT immunostaining. It seems also questionable whether AAT can be used as a diagnostic marker for MFH. The reason for the widespread AAT immunoreactivity in various tumors may be that AAT is taken up from serum to various types of nonhistiocytic tumor cells. (Key words: Alpha- -antitrypsin; Lysozyme) Am J Clin Pathol 989;9:55-5 ALPHA- -ANTITRYPSIN (AAT) is a glycoprotein (molecular weight,,000-5,000 daltons) produced and secreted to blood plasma by liver cells. ' 8 Synthesis of AAT has also been shown in monocytes and lymphocytes. AAT is a plasma protease inhibitor whose main function is to inactivate neutrophil elastase,8 by forming a : complex with it., Lysozyme (molecular weight,000 daltons) is a bacteriolytic enzyme; which hydrolyzes N- acetylglucosaminyl-n-acetylmuramic acid linkages in the cell wall of susceptible bacteria. 0, ' Both substances are present in serum and many other body fluids. ' ' Lysozyme has been shown to be synthesized by histiomonocytic cells. ' ' 5 Both AAT and lysozyme have been used as immunohistochemical markers for histiocytic cells and to test and prove the histiocytic origin of tumors, such as benign and malignant fibrous histiocytomas (MFHs). ' 9 The presence of AAT immunoreactivity in MFHs has been Received June 9, 988; received revised manuscript and accepted for publication September, 988. Address reprint requests to Dr. Miettinen: Department of Pathology and Cell Biology, Thomas Jefferson University, Pavilion 0, 05 Walnut Street, Philadelphia, Pennsylvania 90. Department of Pathology, University of Helsinki, Helsinki, Finland thought to signify the histiocytic nature of the neoplastic cells. 9 AAT has been regarded as a sensitive marker for histiocytic tumors, whereas lysozyme has been considered to be a less sensitive marker. 9 Both AAT and lysozyme immunoreactivity have been shown in several clearly nonhistiocytic tumors, for instance, in gastric and gallbladder adenocarcinomas, ' pleomorphic adenomas of the salivary gland, and hepatocellular carcinomas. 9,0 The purpose of this study was to clarify the value of AAT and lysozyme as histiocytic markers and to explore the significance of AAT and lysozyme immunoreactivity in malignant fibrous histiocytomas. We believed this necessary, because demonstration of AAT immunoreactivity has been widely thought to prove the histiocytic origin of tumors, yet materials used in many investigations have been restricted only to selected tumor groups. We analyzed a large number of tumors for these two markers, using optimized immunostaining conditions for formaldehydefixed and paraffin-embedded tissues, including a protease (pepsin) pretreatment of the sections for both markers. Study Material Materials and Methods A total of 0 specimens were analyzed. This included malignant and benign conditions and selected normal tissues. All tissues had been fixed in neutral formalin and embedded in paraffin. The specific type diagnosis of most sarcomas was based on ultrastructural and immunohistochemical verification by appropriate diagnostic features, especially in the case of leiomyosarcomas, rhabdomyosarcomas, and synovial sarcomas. Antibodies and Immunostaining Polyclonal rabbit antihuman AAT and lysozyme antisera (Dakopatts, Copenhagen) were used in the immunostaining. The immunostaining was performed by the use of the immunoperoxidase technique via the avidinbiotin amplification system as described earlier.' 5 We also studied the effect of protease (pepsin) pretreatment in the Downloaded from on April 08 55

2 5 SOINI AND MIETTINEN Table. Carcinomas Alpha- -Antitrypsin Lysozyme A.J.C.P.-May 989 Mammary ductal carcinoma Adenocarcinoma of the gastrointestinal tract Adenocarcinoma of the lung Adenocarcinoma of kidney Adenocarcinoma of prostate Adenocarcinoma of the thyroid gland Adenocarcinoma of suprarenal gland Thyroid medullary carcinoma Squamous carcinoma Carcinoid tumor, small bowel Other carcinomas Total Total immunostaining. For the immunostaining, the specimens were deparaffinized in xylene and rehydrated. The endogenous peroxidase was blocked with 0.5% HO in water for 5 minutes. Before the primary antibodies, the sections were incubated in normal swine serum (:0) for 5 minutes. The primary antibodies (:,000, for both markers) were incubated overnight at + C and were sequentially followed by biotinylated rabbit-antimouse IgG antiserum (Dakopatts, :00, for 0 minutes at room temperature) and by a complex of avidin and biotinylated horseradish peroxidase (Dakopatts, both diluted :50 for 0 minutes at room temperature). Careful rinses with several changes of phosphate-buffered saline (PBS) were done between each stage of the procedure. The color was developed with -amino-9-ethyl-carbazole, dissolved in dimethylformamide, and diluted in 0.05 M sodium acetate buffer, ph 5.0 (final concentration 0. mg/ml), in the presence of 0.0% hydrogen peroxide for 0 minutes at room temperature. The slides were mounted in an aqueous mounting medium. The results were evaluated quantitatively and divided into four groups (0% = -; <0% = +; 0-50% = ++; and 0-00% = +++) according to the number of positive tumor cells (Tables -). In all specimens, tissue macrophages and neutrophils were used as internal positive controls. No immunostaining for AAT and lysozyme was obtained when the primary antibody was omitted. Additionally, frozen sections of five cases of MFH werefixedwith acetone (0 minutes at room temperature) and immunostained with two monoclonal antibodies to histiomonocytic cells, KB 90 and EBM " (Dakopatts, Copenhagen; dilution of antibodies :50 and :0, respectively). The immunostaining was done as above except that rabbit antimouse immunoglobulins (Dakopatts, :00) were used as the secondary antibody. Absorption Experiments With some AAT-positive sarcoid granulomas and malignant melanomas, schwannomas; and malignant fibrous histiocytomas, an absorption test for AAT was conducted. Table. Lymphomas Alpha- -Antitrypsin Lysozyme Total Malignant histiocytosis True histiocytic lymphoma Immunoblastic lymphoma Plasmocytoma Diffuse centrocytic lymphoma Diffuse centroblastic lymphoma Large cell anaplastic lymphoma (Ki- -lymphoma) Morbus Hodgkih, nodular sclerosis Morbus Hodgkin, mixed cellularity Total Downloaded from on April 08

3 Vol.9 -No. 5 ALPHA--ANTITRYPSIN AND LYSOZYME Table. Mesenchymal Tumors 5 Malignant fibrous histiocytoma Synovial sarcoma Rhabdomyosarcoma Hemangiopericytoma Leiomyosarcoma Fibrosarcoma Dermatofibrosarcoma protuberans Osteosarcoma Myxoid chondrosarcoma Malignant melanoma Malignant schwannoma Dermatofibroma Schwannoma Chondroma Leiomyoma Leiomyoblastoma Giant cell tumor Pigmented nevus Juvenile xanthogranuloma Alpha--Antitrypsin Total Total 5 Lysozyme Twenty micrograms of purified human AAT was incubated with ml of anti-aat (:,000) before the immunostaining. In order to explore the possible cross-reactivity between AAT and alpha--antichymotrypsin (AACT), polyclonal antibodies to AACT (Dakopatts, Copenhagen) were absorbed with AAT. About 0 sarcomas and carcinomas were immunostained for AACT (:,000) without and with absorption with AAT. Twenty micrograms of purified human AAT was incubated with ml of anti-aact (:,000). The procedures of immunostaining in these absorption tests were as described above. Results Optimization of the Immunstainings Before the immunostaining of the tumor series, 0 specimens, including sarcoid granulomas, sinus histiocytosis, lymph node toxoplasmosis, and normal alveolar macrophages, were analyzed in order to optimize the immunostaining for histiocytes and explore the significance of protease treatment. The best results were achieved by use of a dilution of :,000 for both antibodies. A mild protease pretreatment (0.05% pepsin [,500 U/g, Merck, Darmstadt] in HC, ph.8, at C for 0 minutes) was found to intensify the immunostaining and increase the number of positive histiocytes for both markers. Therefore, such a treatment was used for the whole material. Histiocytic Markers in Tumors Of all neoplasms examined, 0% (08 of 8) had positive results for AAT and only % ( of 8) for lysozyme. The results are shown in Tables -. Of all carcinomas, half ( of 58) had positive results for AAT but only a few ( of 58) for lysozyme. Immunoreactivity for AAT in carcinomas was variably present in different kinds of adenocarcinomas, squamous cell carcinomas, and neuroendocrine carcinomas. Especially carcinoid tumors often had strongly positive results (Fig. ). Lysozyme immunoreactivity was found only in two undifferentiated carcinomas (thyroid gland and gallbladder) and one adenocarcinoma (colon). Two cases of malignant histiocytosis and one ultrastructurally confirmed true histiocytic lymphoma had positive results for both markers. In most lymphomas only reactive histiocytes had positive results. Of malignant fibrous histiocytomas, most cases showed AAT-positive large neoplastic cells, 9 of of the cases in significant numbers (Fig. ). In some cases of MFH, the extracellular matrix was also diffusely positive for AAT but not for lysozyme. In contrast, the immunostaining pattern obtained with the antibody to lysozyme was different. Eight cases showed only obviously reactive small nonneoplastic histiocytes (Fig. ), whereas of cases showed occasional possibly neoplastic cells. In their immunohistochemical profiles, the pleomorphic and myxoid variants of MFH were similar. Immunostaining with two monoclonal antibodies to histiomonocytic cells (KB 90, EBM ) showed labeling only in the small obviously entrapped histiomonocytic cells and not in the large neoplastic cells (Fig. ). Dermatofibromas also had strongly positive results for AAT, and there was only negative result of cases. AAT immunoreactivity could be observed both in the histiocyte-like and in the fibroblast- Downloaded from on April 08

4 58 A.J.C.P.-May 989 SOINI AND MIETTINEN r upr Downloaded from on April 08 i^j^

5 vol. 9i. No. 5 ALPHA--ANTITRYPSIN AND LYSOZYME 59 FIG. (upper, left). The neoplastic islands of carcinoid tumor have strongly positive results for alpha--antitrypsin. Light counterstain with hematoxylin (X00). FIG. (upper, right). The tumor cells of malignant fibrous histiocytoma have positive results for alpha--antitrypsin. Light counterstain with hematoxylin (X00). FIG. (center, left). The neoplastic cells of malignant fibrous histiocytoma have negative results for lysozyme; only nonneoplastic histiocytes or granulocytes have positive results. Light counterstain with hematoxylin (X00). FIG. (center, right). Immunostaining for KB90 monoclonal antibody to histiomonocytic cells has negative results in tumor cells of malignant fibrous histiocytoma, but some entrapped histiocytes have positive results. Light counterstain with hematoxylin (X00). FIG. 5 (lower, left). Melanoma cells have positive results for alpha--antitrypsin. Light counterstain with hematoxylin (X00). FlG. (lower, right). The spindle cells in a schwannoma have positive results for alpha--antitrypsin. Light counterstain with hematoxylin (X00). < like tumor cells. Many times the dermatofibromas also stained diffusely, which made it difficult to detect the cytoplasmic immunoreactivity in individual cells. However, the surrounding dermis always had negative results, which mitigates against the possibility that the immunostaining would result from nonspecific background staining. About half of dermatofibromas immunostained for lysozyme, the positivity being mainly in the histiocyte-like cells, but occasional positive spindle-shaped cells could also be observed. Of all sarcomas, 55% had positive results for AAT and % for lysozyme. Other sarcomas besides MFH that showed significant immunoreactivity for AAT in the neoplastic cells includedfiveof seven leiomyosarcomas, three of three dermatofibrosarcoma protuberans, and one of two myxoid chondrosarcoma. Two of five synovial sarcomas showed some AAT-positive epithelial cells. Six of seven malignant melanomas had positive results for AAT, three of them strongly (Fig. 5). Lysozyme immunoreactivity was observed in scattered tumor cells only in one melanoma and one osteosarcoma among the malignant mesenchymal tumors. Of benign mesenchymal tumors, all schwannomas showed numerous AAT-positive spindle cells (Fig. ). The immunoreactivity for AAT, observed in sarcoid granulomas, melanomas, schwannomas, and malignant fibrous histiocytomas, could be abolished with the absorption of the antibody with purified alpha--antitrypsin. Of the 0 carcinomas and sarcomas stained for AACT, 5% had positive results. After absorption with purified AAT, also AACT immunostaining was almost completely lost. None of the tumors appeared to have positive findings, and only weak positivity could be observed in some histiocytes and neutrophils. Discussion In this study, we investigated the specificity of alpha- -antitrypsin (AAT) and lysozyme as markers for histiocytic lesions with a special attention to malignant and benign fibrous histiocytomas. The value of protease pretreatment (pepsin, trypsin, or pronase) is well known in immunohistochemistry of formalin-fixed and paraffinembedded material for many antigens. 5 We could observe a clear quantitative and qualitative intensification of immunostaining of histiocytes in various lesions with both markers when protease pretreatment was used. Protease treatment has also been used in immunostaining for AAT by some authors. 9, However, in many works on lysozyme, protease treatment has not been used. 0,, This can possibly explain differences in the immunostaining results between various authors. The use of unoptimal immunostaining conditions may also lead to a too-optimistic view on the specificity of immunostaining, because only strong immunoreactivity is detected, such as the immunostaining in histiocytes. In malignant fibrous histiocytoma (MFH), strong immunoreactivity for AAT was found in clearly neoplastic cells, but the immunostaining for lysozyme was almost exclusively found in smaller virtually entrapped histiocytic cells, similar to the immunostaining results with two monoclonal antibodies to histiomonocytic cells. This is in accordance with some 8 but not all studies describing the immunoreactivity of MFH with monoclonal antibodies for histiomonocytic cells in MFH. Our results suggest that the prominent AAT immunoreactivity in MFH may not result from the histiocytic nature of MFH. Furthermore, the fact that the immunostaining for lysozyme in MFH was similar to that obtained with monoclonal antibodies to histiomonocytic cells might suggest that this actually reflects the specificity of lysozyme in comparison with AAT and not its relative insensitivity. Prominent AAT immunostaining in many carcinomas, melanomas, and so on shows that histiocytic differentiation is not the only context in which AAT immunoreactivity can be found in tumors. In fact, the presence of AAT in malignant melanomas, schwannomas, carcinoid tumors, and other neoplasms must have another explanation. The widespread immunoreactivity for AAT might be explained in many different ways. First, AAT immu- Downloaded from on April 08

6 50 SOINI AND MIETTINEN A.J.C.P. May 989 noreactivity might result from nonspecific binding of the antibody. However, in the absorption test conducted for AAT, the immunoreactivity could be abolished after the antibody was incubated with the antigen before the immunostaining, which clearly speaks against this possibility. Secondly, the antibody used was polyclonal, and a crossreaction with some other antigens distributed in tumor cells might be possible. This issue is further complicated by the fact that there is a sequence homology between AAT, AACT, antithrombin III, chicken ovalbumin, and between the corresponding cdnas.,8, The amino acid sequence homology between AAT and AACT is %, 8 between AAT and antithrombin III %, 8 and between AAT and chicken ovalbumin 5%. We tested the hypothesis of cross-reactivity by staining a number of carcinomas and sarcomas with polyclonal anti-aact antibody by conducting an absorption test with AAT as antigen. The fact that AACT immunoreactivity was mainly abolished after the absorption of the antibody with AAT indicates that a cross-reactivity between these substances might be possible. This view is also supported by the fact that immunoreactivity for AAT and AACT many times is detected in same tumors. ' 9, Thirdly, AAT is widely distributed in body fluids, and its concentration especially increased in inflammatory responses; AAT is an acute phase protein. Therefore, the relatively widespread immunoreactivity for AAT in tumors might result from uptake of this substance from the serum into the tumor cells by some mechanism. However, lysozyme is also widely distributed in body fluids, and one would then expect that it also would be taken up by tumor cells. The concentration of lysozyme in serum is, however, much less than that of AAT (- mg/l for lysozyme and - g/l for AAT),,9 which could explain why lysozyme immunoreactivity cannot be detected as often as AAT immunoreactivity even though lysozyme also might be taken up by the neoplastic cells. Moreover, AAT synthesis has been shown only in some hepatoma cell lines in tissue culture studies as far as tumors are concerned. So it seems logical to assume that the widespread immunoreactivity for AAT in tumors would result from its widespread and high body concentration and the endocytic activity or passive uptake of tumor cells, rather than from the synthesis of AAT by these tumor cells. Similar nonspecific uptake of another protein, myoglobin, has been previously shown in some malignant tumors infiltrating and destroying muscle tissue. Such uptake may result in myoglobin immunoreactivity in nonmuscular tumor cells. 9 In view of the results and comments presented here, it seems likely that AAT immunoreactivity in (nonhistiocytic) tumor cells might merely result from their endocytotic capacity and not be related to their histiocytic differentiation, as was also suggested for AACT, another supposed histiocytic marker, which proved to be even more nonspecific than AAT in studies covering wide ranges of different neoplasms.,5 Consequently, the high AAT immunoreactivity in MFH can be explained in this way, and tissue culture studies of these cells to prove their possible AAT synthetic capacity are needed in order to rule out this explanation. As far as lysozyme is concerned, its role as a histiocytic marker is difficult to evaluate because the histogenesis of the so-calledfibrohistiocytictumors, such as MFH and dermatofibromas, still is controversial. 8,,,,8 Finally, the AAT immunoreactivity found in many nonhistiocytic tumors also seriously questions the value of AAT in the differential diagnosis and objective verification of MFH. References. Aroni K, Kittas C, Papadimitriou CS, Papacharalampous NX: An immunocytochemical study of the distribution of lysozyme, alpha- -antitrypsin and alpha- -antichymotrypsin in the normal and pathological gallbladder. Virchows Arch [Pathol Anat] 98;0: Bao J, Sifers RN, Kidd VJ, Ledley FD, Woo SLC: Molecular evolution of serpins: homologous structure of the human alpha-- antichymotrypsin and alpha--antitrypsin genes. Biochemistry 98;: Berninger R: Alpha--antitrypsin. J Med 985;:-99.. du Boulay CE: Demonstration of alpha--antitrypsin and alpha-- antichymotrypsin in fibrous histiocytomas using the immunoperoxidase technique. Am J Surg Pathol 98;: Brozman M: Immunohistochemical analysis of formaldehyde- and trypsin- or pepsin treated material. Acta Histochem 98;:5-0.. Burgdorf WHC, Duray P, Rosai J: Immunohistochemical identification of lysozyme in cutaneous lesions of alleged histiocytic nature. Am J Clin Pathol 98;5:-.. Carrell RW: Alpha--antitrypsin: molecular pathology, leucocytes and tissue damage. J Clin Invest 98;8:-. 8. Chandra T, Stackhouse R, Kidd VJ, Robson JH, Woo SLC: Sequence homology between human alpha- -antrichymotrypsin, alpha-- antitrypsin and antithrombin III. Biochemistry 98;: Eusebi V, Bondi A, Rosai J: Immunohistochemical localization of myoglobin in nonmuscular cells. Am J Surg Pathol 98;8: Fett JW, Strydom DJ, Lobb RR, Alderman EM, Vallee BL: Lysozyme: a major secretory product of a human colon carcinoma cell line. Biochemistry 985:: Franklin WA, Mason DY, Putord K, et al: Immunohistological analysis of human mononuclear phagocytes and dendritic cells by using monoclonal antibodies. Lab Invest 98:5:-5.. Gonzales B: Benign fibrous histiocytoma of the skin. An immunohistochemical analysis of 0 cases. Pathol Res Pract 985; 80: Gordon S, Todd J, Cohn ZA: In vitro synthesis and secretion of lysozyme by mononuclear phagocytes. J Exp Med 9; 9:8-8.. Haneberg B, et al: In vitro release of lysozyme from monocytes and granulocytes. J Leukocyte Biol 98;5: Hsu SM, Raine L, Fanger H: Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem 98;9: Huang SN: Application of immunofluorescent staining on paraffin sections improved by trypsin digestion. Lab Invest 9;5: Isaacson P, Jones DB, Millward-Sadler GH, Judd MA, Payne S: Alpha--antitrypsin in human macrophages. J Clin Pathol 98;: Kanitakis J, Schmitt D, Thivolet J: Immunohistologic study of eel- Downloaded from on April 08

7 Vol. 9 -No. 5 ALPHA--ANTITRYPSIN AND LYSOZYME 5 lular populations of histiocytofibromas ("dermatofibromas"). J Cutan Pathol 98;: Kindblom L-G, Jacobsen G, Jacobsen M: Immunohistochemical investigations of tumors of supposedfibroblastic-histiocyticorigin. Hum Pathol 98;: Klockars M, Reitamo S: Tissue distribution of lysozyme in man. J Histochem Cytochem 95;: Lappin D, Hamilton AD, Morrison L, Aref M, Whaley K: Synthesis of complement components (C, C, B and CI-inhibitor) and lysozyme by human monocytes and macrophages. J Clin Lab Immunol 98;0: Leader M, Patel J, Collins M, Henry K: Anti-alpha--antichymotrypsin staining of 9 sarcomas, 8 carcinomas and malignant melanomas. Its lack of specificity as a tumour marker. Am J Surg Pathol 98;:-9.. Long GL, Chandra T, Woo SLC, Davie EW, Kurachi K: Complete sequence of the cdna for human alpha--antitrypsin and the gene for the S variant. Biochemistry 98;: Mason DY, Taylor CR: The distribution of muramidase (lysozome) in human tissues. J Clin Pathol 95;8:-. 5. McClelland D, van Furth R: In vitro synthesis of lysozyme by human and mouse tissues and leucocytes. Immunology 95,8:099.. Mihatsch-Konz B, Schaumburg-Lever G, Lever WF: Ultrastructure of dermatofibroma. Arch Dermatol Res 9;:8-9.. Miyauchi J, Sasadaira H, Watanabe K, Watanabe Y: Ultrastructural immunocytochemical localization of lysozyme in human monocytes and macrophages. Cell Tissue Res 985;: Mornex JF, Chytil-Weir A, Martinet Y, Courtney M, LeCocq JP, Crystal RG: Expression of the alpha--antitrypsin gene in mononuclear phagocytes of normal and alpha--antitrypsin-deficient individuals. J Clin Invest 98;: Okushin H, Yamada G, Nagashima H: Immunohistochemical study of fibrbnectin, lysozyme and alpha-fetoprotein (AP) in human hepatocellular carcinoma. Gastroenterol Jpn 98;: Ordonez NG, Manning JT: Comparison of alpha--antitrypsin and alpha--antichymotrypsin in hepatocellular carcinoma: an immunoperoxidase study. Am J Gastroenterol 98;9: Roholl P, KJeijne J, van Buoken C, van der Putte S, van Unnik J: A study to analyze the origin of tumour cells in malignant fibrous histiocytomas. A multiparametric characterization. Cancer 985;5: Roholl P, KJeyne J, Elbers H, van der Vegt M, Albus-Lutter CH, Van Unnik J: Characterization of tumour cells in malignant fibrous histiocytomas and other soft tissue tumours in comparison with malignant histiocytes. Immunohistochemical study on paraffin sections. J Pathol 985;: Roholl P, Kleyne J, Pijpers H, van Unnik J: Comparative immunohistochemical investigation of markers for malignant histiocytes. Hum Pathol 985;:-.. Sehested M, Barfoed C, Krogdahl A, Bretlau P: Immunohistochemical investigation of lysozyme, lactoferrin, alpha--antitrypsin, alpha--antichymotrypsin and ferritin in parotid gland tumours. J Oral Pathol 985;: Soini Y, Miettinen M: Widespread immunoreactivity for alpha-lantichymotrypsin in different types of tumors. Am J Clin Pathol 988:90:-.. Strauchen JA, Dimitriu-Bona A: Malignant fibrous histiocytoma. Expression of monocyte/macrophage differentiation antigens detected with monoclonal antibodies. Am J Pathol 98; : Wittekind C, Wachner R, Henke W, von KJeist SI Localization of CEA, HCG, lysozyme, alpha--antitrypsin and alpha--antichymotrypsin in gastric cancer and prognosis. Virchows Arch [Pathol Anat] 98;09: Wood GS, Beckstead JH, Turner RR, Hendrickson MR, Kempson RL, Warnke RA: Malignant fibrous histiocytoma tumor cells resemble fibroblasts. Am J Surg Pathol 98;0: Zucker S, Hanes DJ, Vogler WR, Eanes RZ: Plasma muramidase: a study of methods and clinical applications. J Lab Clin Med 90;5:8-9. Downloaded from on April 08