Amhives of
Arch Gynecol Obstet (1991) 249:173-177
Gynecology and Obstetrics
© Springer-Verlag1991
Immunohistochemical differentiation between ovarian granulosa cell tumors and ovarian carcinomas G. Gitsch, P. Kohlberger, E. Hanzal, H. K61bl, and G. Breitenecker Second Department of Obstetrics and Gynecology, University of Vienna, Spitalgasse 23, A-1090 Vienna, Austria Received January 24, 1991/Accepted June 13, 1991
Summary. Differential diagnosis is a major problem in histopathology of ovarian tumors. Difficulties may arise if the tumor is a poorly differentiated carcinoma or a granulosa cell tumor of the sarcomatoid type. It was the aim of the present study to evaluate the usefulness of immunohistochemistry in differentiating between granulosa cell tumors of the ovary and ovarian carcinomas. We investigated 56 ovarian malignancies (13 granulosa cell tumors, 17 serous, 14 mucinous and 12 poorly differentiated carcinomas) and performed immunohistochemical detection of Vimentin, Keratin, CA125, CA19-9, CEA, $100 and Ber-EP4. Expression of Vimentin was highest and expression of Keratin was lowest in granulosa cell tumors in contrast to carcinomas. CAi25 and CA19-9 were not expressed in granulosa cell tumors, whereas the detection rate in carcinomas (except for CA125 in mucinous carcinomas) was high. CEA, S100 and Ber-EP4 do not seem to be useful markers in differential diagnosis. A marker profile of Vimentin, Keratin, CA125 and CA19-9 allows a quite strict differentiation between poorly differentiated ovarian carcinomas and granulosa cell tumors of the ovary. Key words: Immunohistochemistry - Granulosa cell tumors - Ovarian carcinoma
Introduction Histological differentiation between ovarian granulosa cell tumors and poorly differentiated ovarian carcinomas may be difficult (Scully 1984). Immunohistochemical techniques have therefore been introduced in the diagnosis and study of ovarian neoplasms. Since the development of the hybridoma technology by K6hler and Milstein in 1975 numerous monoclonal antibodies have become available. They react with specific epitopes of tumor cells and thereby act as Offprint requests to: Dr. G. Gitsch (address see above)
174
G. Gitsch et al.
tumor markers (Bast et al. 1981). CA125, CA19-9 and CEA are well established tumor markers for ovarian carcinomas (Davis et al. 1986; Neunteufel and Breitenecker 1989). So far only a limited number of studies report on the diagnostic use of intermediate filament proteins such as Vimentin and Keratin in ovarian tumors (Bonazzi de Bogetto et al. 1983, Kurmann et al. 1984). S100 antigen which is specific for ependymal cells and Ber-EP4 which shows broad reactivity with human epithelial tissues might possibly be useful in differentiating between ovarian granulosa cell carcinomas and other ovarian malignancies (Lauriola et al. 1984; Moore 1965). The aim of the present study was to evaluate the usefulness of the immunohistochemical markers mentioned above.
Material and methods Formalin-fixed and paraffin-embedded sections of 56 malignant ovarian tumors were studied. 13 of them were granulosa cell tumors, 12 poorly differentiated ovarian carcinomas, 17 serous and 14 mucinous ovarian carcinomas. Based on an orientating pilot study and the literature we decided to study the following antigens and markers: Vimentin (monoclonal mouse anti swine Vimentin, DAKO), Keratin (monoclonal mouse anti Keratin, Boehringer Mannheim), CEA (monoclonal mouse, Biomedia), CA125 (monoclonal mouse, Biomedia), CA19-9 (monoclonal mouse, Biomedia), $100 (monoclonal rabbit anti cow, DAKO) and Ber-EP4 (monoclonal mouse anti human epithelial antigen, DAKO). We used two detection systems of immunohistochemical staining technologies for the following antigens and markers: Vectastain ABC kit by Vector Labs. for detection of CEA, CA125 and CA19-9, StrAviGen High Performance kit by Bio Genex Labs. for detection of Vimentin, Keratin, Ber-EP4 and $100. The paraffin sections were deparaffinized and hydrated through xylenes and graded alcohols, rinsed in distilled water and then air dried. If quenching of endogenous peroxidase activity was required we incubated the sections for 30 min in 0.3% H202 in methanol and then washed in 10 mM Phosphate Buffered Saline (PBS) pH 7.6 and 0.1% Triton X100. The need for enzyme predigestion of formalin-fixed tissue was only necessary for Keratin, S100 and Ber-EP4 (1 mg/ml Trypsin in 0.1% calcium chloride, adjusted to pH 7.8 with 0.1 N sodium hydroxide; incubated at 37°C for 20-30 min). The sections were incubated for 30 min with primary antiserum diluted in buffer and washed for 5 min in buffer. Using the immunoperoxidase method sections were incubated with diluted biotinylated anti-immunoglobulin for 30 min, washed and then incubated in the ABC reagent, which produces biotirdavidin/biotin-peroxidase complexes. The sections were washed again, incubated in peroxidase substrate solution (0.25 mg/ml AEC in 100 mM sodium acetate, pH 5.2) and then washed in water. The sections were finally counterstained with Mayer's Hematoxylin and mounted in Kaiser's Glycerin. Using the streptavidin-alkaline phosphatase method, sections were incubated for 30 min with biotinylated anti-immunoglobulins diluted in PBS, washed and incubated now with enzyme-labeled streptavidin in PBS. After washing the slides were incubated with alkaline phosphatase substrate solution for 10 min (Chromogen: Tablets of Fast Red TR salt, Substrate: naphthol AS-MX phosphate in Tris-HC1 buffer). Finally the slides were washed again, counterstained and mounted. Positive Control: The positive control slide was prepared from tissue known to contain the antigen. Negative Control: The negative control slide was prepared from the same tissue block as the specimen. However, instead of using a primary antibody we used a normal, non-immune serum supernatant from the same source as the primary antibody. Histochemical judgement: Strong and/or widespread staining was judged as positive, week or focal staining was judged as negative.
Immunohistochemistry: Granulosa cell tumors and ovarian carcinomas
175
Table 1. Incidence of seven markers in four types of ovarian tumor Histological ovarian tumor
Serous (n = 17) Mucinous (n = 14) Poorly diff (n = 12) Granulosa cell (n = 13)
Markers Vimentin Keratin CEA
CA125 CA19-9 S-100
Ber-EP4
2 0 2 10
14 0 8 0
5 5 0 0
12 11 9 2
5 11 4 7
8 12 5 0
3 0 8 6
Results Immunohistochemical reactions for Vimentin was most commonly found in granulosa cell tumors while Keratin was found in poorly differentiated ovarian carcinomas. Serous and mucinous carcinomas had a similar staining rate to poorly differentiated carcinomas. Granulosa cell tumors showed a high expression rate for C E A , but none for Ca125 and CA19-9. CA19-9 was detected in tissue of serous, mucinous and poorly differentiated carcinomas, whereas only serous and poorly differentiated tumors expressed CA125. Poorly differentiated carcinomas and granulosa cell tumors showed a similar behaviour concerning expression of $100 and Ber-EP4. Ber-EP4 was detectable in serous and mucinous carcinomas, whereas we found no expression of S100 in mucinous carcinomas and the detection rate was low in serous carcinomas (Table 1).
Discussion Differential diagnosis is a m a j o r problem in histopathology of ovarian tumors. Difficulties m a y arise especially if the t u m o r is either a poorly differentiated carcinoma or a granulosa cell t u m o r of the sarcomatoid type (Sculty 1984). T h e r e f o r e we investigated 56 ovarian malignancies (12 of them poorly differentiated carcinomas and 13 granulosa cell tumors to see if immunohistochemistry could be a useful tool for differential diagnosis. We used as it was several markers, namely Vimentin, Keratin, S100, C E A , CA125, CA19-9 and Ber-EP4. As 10 of our 13 granulosa cell tumors showed microfollicular or macrofollicular structures histological diagnosis was easy. The remaining three granulosa cell tumors were unilateral and caused endometrium proliferation. These three patients showed no recurrence of disease after 5, 7 and 8 years of observation. The differentiation between granulosa cell tumors and poorly differentiated ovarian carcinomas is important because of the different prognosis for these malignancies. In our study Vimentin was detected in 10 of 13 granulosa cell tumors which is 5.1 fold m o r e than in undifferentiated carcinomas. Most ovarian carcinomas expressed Keratin whereas there were only two positives for granulosa cell tumors.
176
G. Gitsch et al.
These findings agree with those of Mann et al. (1986) and Chadha and van der Kwast (1989). Our finding of Keratin in two granulosa cell tumors contradicts the findings of Chadha and van der Kwast (1986) but is in accordance with Czernobilsky et al. (1987) and Miettinen et al. (1985). Expression of Keratin in other non epithelial tumors and tissues has also been reported (Czernobilsky et al. 1987, Brown et al. 1987, Norton et al. 1987). The low expression rate of Keratin in our material may be due to the fact that we used formalin-fixed paraffin-embedded material and not frozen sections. As frozen material may not always be available the loss of sensitivity has to be accepted. On the other hand Pan-Keratin might not be as sensitive in the detection of all Keratins as single Keratin markers. The use of more specific antibodies may lead to better immunostaining results than can be obtained by application of the multispecific (Cytokeratin 1-19) antikeratin we used in our study. CEA expression was detected in every group of ovarian tumors whereas CA19-9 was detected in serous, mucinous and undifferentiated carcinomas but not in granulosa cell tumors. CA125 might be helpful in differentiating between granulosa cell tumors (no positivity) and poorly differentiated carcinomas (Positivity in 8 of 12 cases). $100 and Ber-EP4 do not seem to be helpful in the differential diagnosis. In conclusion it seems that a tumor marker combination of vimentin, Keratin, CA125 and CA19-9 allows imnmnohistochemical differentiation between granulosa cell tumors of the ovary and poorly differentiated ovarian carcinomas and this is important for prognosis.
Acknowledgements.The authors thank Mrs.
A. Steiner for excellent technical assistance. The study was supported by Bender MedSystems, Austria.
References 1. Bast RC, Fenney M, Lazarus H, Nadler LM, Colvin RB, Knapp RC (1981) Reactibility of a monoclonal antibody with human ovarian carcinoma. J Clin Invest 68:1331-1337 2. Bonazzi del Pogetto C, Virtanen I, Lehto VP, Wahlstrom T, Saksch E (1983) Expression of intermediate filaments in ovarian and uterine tumors. Int J Gynecol Pathol 1:359-366 3. Brown DC, Theaker JM, Banks PM, Garter KC, Mason DY (1987) Cytokeratin expression in smooth muscle and smooth muscle tumors. Histopathology 11:477-486 4. Chadha S, van der Kwast ThH (1989) Immunohistochemistry of ovarian granulosa cell tumors. The value of tissue specific proteins and tumour markers. Virch Arch 414:439-445 5. Czernobilsky B, Moll R, Leppien G, Schweickhart G, Franke WW (1987) Desmosomal plaque associated vimentin filaments in human ovarian granulosa cell tumors of various histologic patterns. A m J Pathol 126:476-486 6. Davis HM, Zurawski VR, Bast RC, Klug TL (1986) Characterization of the CA125 antigen associated with human epithelial ovarian carcinomas. Cancer Res 46:6143-6148 7. K6hler G, Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specifity. Nature 256:495-497 8. Kurmann RJ, Ganjei P, Nadji M (1984) Contributions of immunohistochemistry to the diagnosis and study of ovarian neoplasms. Int J Gynecol Pathol 3:3-26 9. Lauriola L, Michetti F, Sentinelli S, Cocchia D (1984) Detection of S-100 labeled cells in nasopharyngeal carcinoma. J Clin Pathol 37:1235-1238 10. Mann WJ, Chumas J, Rosenwaks Z, Merril JA, Davenport D (1986) Elevated serum alpha fetoprotein associated with Sertoli-Leydig cell tumors. Obstet Gynecol 67:141-144
Immunohistochemistry: Granulosa cell tumors and ovarian carcinomas
177
11. Miettinen M, Walstrom T, Vitranen I, Talerman A, Astengo-Osuna C (1985) Cellular differentiation in ovarian sex cord-stromal and germ-cell tumors studied with antibodies to intermediate-filament proteins. Am J Surg Pathol 9:640-651 12. MooreBW (1965)A solubleprotein characteristic of the nervous system. Biochem Biophys Res Comm 19:739-744 13. Neunteufel W, Breitenecker G (1989) Tissue expression of CA125 in benign and malignant lesions of the ovary and fallopian tube: a comparison with CA19-9 and CEA. Gynecol Oncol 32:297-302 14. Norton AJ, Thomas JA, Isaacson PG (1987) Cytokeratin-specific monoclonal antibodies are reactive with tumors of smooth muscle derivation. An immunocytochemical and biochemical study using antibodies to intermediate filament cytosceletal proteins. Histopathology 11:487499 15. Scully RE (1984) Immunohistochemistry of ovarian tumors. In: Jose Russo (ed) Immunocytochemistry in tumor diagnosis. Boston Dordrecht Lancaster, pp 293-321 16. Young RH, Scully RE (1982) Ovarian sexcord-stromal tumors: recent progress. Int J Gynecol Pathol 1:101-123
Arch Gynecol Obstet (1991) 249:173-177
Gynecology and Obstetrics
© Springer-Verlag1991
Immunohistochemical differentiation between ovarian granulosa cell tumors and ovarian carcinomas G. Gitsch, P. Kohlberger, E. Hanzal, H. K61bl, and G. Breitenecker Second Department of Obstetrics and Gynecology, University of Vienna, Spitalgasse 23, A-1090 Vienna, Austria Received January 24, 1991/Accepted June 13, 1991
Summary. Differential diagnosis is a major problem in histopathology of ovarian tumors. Difficulties may arise if the tumor is a poorly differentiated carcinoma or a granulosa cell tumor of the sarcomatoid type. It was the aim of the present study to evaluate the usefulness of immunohistochemistry in differentiating between granulosa cell tumors of the ovary and ovarian carcinomas. We investigated 56 ovarian malignancies (13 granulosa cell tumors, 17 serous, 14 mucinous and 12 poorly differentiated carcinomas) and performed immunohistochemical detection of Vimentin, Keratin, CA125, CA19-9, CEA, $100 and Ber-EP4. Expression of Vimentin was highest and expression of Keratin was lowest in granulosa cell tumors in contrast to carcinomas. CAi25 and CA19-9 were not expressed in granulosa cell tumors, whereas the detection rate in carcinomas (except for CA125 in mucinous carcinomas) was high. CEA, S100 and Ber-EP4 do not seem to be useful markers in differential diagnosis. A marker profile of Vimentin, Keratin, CA125 and CA19-9 allows a quite strict differentiation between poorly differentiated ovarian carcinomas and granulosa cell tumors of the ovary. Key words: Immunohistochemistry - Granulosa cell tumors - Ovarian carcinoma
Introduction Histological differentiation between ovarian granulosa cell tumors and poorly differentiated ovarian carcinomas may be difficult (Scully 1984). Immunohistochemical techniques have therefore been introduced in the diagnosis and study of ovarian neoplasms. Since the development of the hybridoma technology by K6hler and Milstein in 1975 numerous monoclonal antibodies have become available. They react with specific epitopes of tumor cells and thereby act as Offprint requests to: Dr. G. Gitsch (address see above)
174
G. Gitsch et al.
tumor markers (Bast et al. 1981). CA125, CA19-9 and CEA are well established tumor markers for ovarian carcinomas (Davis et al. 1986; Neunteufel and Breitenecker 1989). So far only a limited number of studies report on the diagnostic use of intermediate filament proteins such as Vimentin and Keratin in ovarian tumors (Bonazzi de Bogetto et al. 1983, Kurmann et al. 1984). S100 antigen which is specific for ependymal cells and Ber-EP4 which shows broad reactivity with human epithelial tissues might possibly be useful in differentiating between ovarian granulosa cell carcinomas and other ovarian malignancies (Lauriola et al. 1984; Moore 1965). The aim of the present study was to evaluate the usefulness of the immunohistochemical markers mentioned above.
Material and methods Formalin-fixed and paraffin-embedded sections of 56 malignant ovarian tumors were studied. 13 of them were granulosa cell tumors, 12 poorly differentiated ovarian carcinomas, 17 serous and 14 mucinous ovarian carcinomas. Based on an orientating pilot study and the literature we decided to study the following antigens and markers: Vimentin (monoclonal mouse anti swine Vimentin, DAKO), Keratin (monoclonal mouse anti Keratin, Boehringer Mannheim), CEA (monoclonal mouse, Biomedia), CA125 (monoclonal mouse, Biomedia), CA19-9 (monoclonal mouse, Biomedia), $100 (monoclonal rabbit anti cow, DAKO) and Ber-EP4 (monoclonal mouse anti human epithelial antigen, DAKO). We used two detection systems of immunohistochemical staining technologies for the following antigens and markers: Vectastain ABC kit by Vector Labs. for detection of CEA, CA125 and CA19-9, StrAviGen High Performance kit by Bio Genex Labs. for detection of Vimentin, Keratin, Ber-EP4 and $100. The paraffin sections were deparaffinized and hydrated through xylenes and graded alcohols, rinsed in distilled water and then air dried. If quenching of endogenous peroxidase activity was required we incubated the sections for 30 min in 0.3% H202 in methanol and then washed in 10 mM Phosphate Buffered Saline (PBS) pH 7.6 and 0.1% Triton X100. The need for enzyme predigestion of formalin-fixed tissue was only necessary for Keratin, S100 and Ber-EP4 (1 mg/ml Trypsin in 0.1% calcium chloride, adjusted to pH 7.8 with 0.1 N sodium hydroxide; incubated at 37°C for 20-30 min). The sections were incubated for 30 min with primary antiserum diluted in buffer and washed for 5 min in buffer. Using the immunoperoxidase method sections were incubated with diluted biotinylated anti-immunoglobulin for 30 min, washed and then incubated in the ABC reagent, which produces biotirdavidin/biotin-peroxidase complexes. The sections were washed again, incubated in peroxidase substrate solution (0.25 mg/ml AEC in 100 mM sodium acetate, pH 5.2) and then washed in water. The sections were finally counterstained with Mayer's Hematoxylin and mounted in Kaiser's Glycerin. Using the streptavidin-alkaline phosphatase method, sections were incubated for 30 min with biotinylated anti-immunoglobulins diluted in PBS, washed and incubated now with enzyme-labeled streptavidin in PBS. After washing the slides were incubated with alkaline phosphatase substrate solution for 10 min (Chromogen: Tablets of Fast Red TR salt, Substrate: naphthol AS-MX phosphate in Tris-HC1 buffer). Finally the slides were washed again, counterstained and mounted. Positive Control: The positive control slide was prepared from tissue known to contain the antigen. Negative Control: The negative control slide was prepared from the same tissue block as the specimen. However, instead of using a primary antibody we used a normal, non-immune serum supernatant from the same source as the primary antibody. Histochemical judgement: Strong and/or widespread staining was judged as positive, week or focal staining was judged as negative.
Immunohistochemistry: Granulosa cell tumors and ovarian carcinomas
175
Table 1. Incidence of seven markers in four types of ovarian tumor Histological ovarian tumor
Serous (n = 17) Mucinous (n = 14) Poorly diff (n = 12) Granulosa cell (n = 13)
Markers Vimentin Keratin CEA
CA125 CA19-9 S-100
Ber-EP4
2 0 2 10
14 0 8 0
5 5 0 0
12 11 9 2
5 11 4 7
8 12 5 0
3 0 8 6
Results Immunohistochemical reactions for Vimentin was most commonly found in granulosa cell tumors while Keratin was found in poorly differentiated ovarian carcinomas. Serous and mucinous carcinomas had a similar staining rate to poorly differentiated carcinomas. Granulosa cell tumors showed a high expression rate for C E A , but none for Ca125 and CA19-9. CA19-9 was detected in tissue of serous, mucinous and poorly differentiated carcinomas, whereas only serous and poorly differentiated tumors expressed CA125. Poorly differentiated carcinomas and granulosa cell tumors showed a similar behaviour concerning expression of $100 and Ber-EP4. Ber-EP4 was detectable in serous and mucinous carcinomas, whereas we found no expression of S100 in mucinous carcinomas and the detection rate was low in serous carcinomas (Table 1).
Discussion Differential diagnosis is a m a j o r problem in histopathology of ovarian tumors. Difficulties m a y arise especially if the t u m o r is either a poorly differentiated carcinoma or a granulosa cell t u m o r of the sarcomatoid type (Sculty 1984). T h e r e f o r e we investigated 56 ovarian malignancies (12 of them poorly differentiated carcinomas and 13 granulosa cell tumors to see if immunohistochemistry could be a useful tool for differential diagnosis. We used as it was several markers, namely Vimentin, Keratin, S100, C E A , CA125, CA19-9 and Ber-EP4. As 10 of our 13 granulosa cell tumors showed microfollicular or macrofollicular structures histological diagnosis was easy. The remaining three granulosa cell tumors were unilateral and caused endometrium proliferation. These three patients showed no recurrence of disease after 5, 7 and 8 years of observation. The differentiation between granulosa cell tumors and poorly differentiated ovarian carcinomas is important because of the different prognosis for these malignancies. In our study Vimentin was detected in 10 of 13 granulosa cell tumors which is 5.1 fold m o r e than in undifferentiated carcinomas. Most ovarian carcinomas expressed Keratin whereas there were only two positives for granulosa cell tumors.
176
G. Gitsch et al.
These findings agree with those of Mann et al. (1986) and Chadha and van der Kwast (1989). Our finding of Keratin in two granulosa cell tumors contradicts the findings of Chadha and van der Kwast (1986) but is in accordance with Czernobilsky et al. (1987) and Miettinen et al. (1985). Expression of Keratin in other non epithelial tumors and tissues has also been reported (Czernobilsky et al. 1987, Brown et al. 1987, Norton et al. 1987). The low expression rate of Keratin in our material may be due to the fact that we used formalin-fixed paraffin-embedded material and not frozen sections. As frozen material may not always be available the loss of sensitivity has to be accepted. On the other hand Pan-Keratin might not be as sensitive in the detection of all Keratins as single Keratin markers. The use of more specific antibodies may lead to better immunostaining results than can be obtained by application of the multispecific (Cytokeratin 1-19) antikeratin we used in our study. CEA expression was detected in every group of ovarian tumors whereas CA19-9 was detected in serous, mucinous and undifferentiated carcinomas but not in granulosa cell tumors. CA125 might be helpful in differentiating between granulosa cell tumors (no positivity) and poorly differentiated carcinomas (Positivity in 8 of 12 cases). $100 and Ber-EP4 do not seem to be helpful in the differential diagnosis. In conclusion it seems that a tumor marker combination of vimentin, Keratin, CA125 and CA19-9 allows imnmnohistochemical differentiation between granulosa cell tumors of the ovary and poorly differentiated ovarian carcinomas and this is important for prognosis.
Acknowledgements.The authors thank Mrs.
A. Steiner for excellent technical assistance. The study was supported by Bender MedSystems, Austria.
References 1. Bast RC, Fenney M, Lazarus H, Nadler LM, Colvin RB, Knapp RC (1981) Reactibility of a monoclonal antibody with human ovarian carcinoma. J Clin Invest 68:1331-1337 2. Bonazzi del Pogetto C, Virtanen I, Lehto VP, Wahlstrom T, Saksch E (1983) Expression of intermediate filaments in ovarian and uterine tumors. Int J Gynecol Pathol 1:359-366 3. Brown DC, Theaker JM, Banks PM, Garter KC, Mason DY (1987) Cytokeratin expression in smooth muscle and smooth muscle tumors. Histopathology 11:477-486 4. Chadha S, van der Kwast ThH (1989) Immunohistochemistry of ovarian granulosa cell tumors. The value of tissue specific proteins and tumour markers. Virch Arch 414:439-445 5. Czernobilsky B, Moll R, Leppien G, Schweickhart G, Franke WW (1987) Desmosomal plaque associated vimentin filaments in human ovarian granulosa cell tumors of various histologic patterns. A m J Pathol 126:476-486 6. Davis HM, Zurawski VR, Bast RC, Klug TL (1986) Characterization of the CA125 antigen associated with human epithelial ovarian carcinomas. Cancer Res 46:6143-6148 7. K6hler G, Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specifity. Nature 256:495-497 8. Kurmann RJ, Ganjei P, Nadji M (1984) Contributions of immunohistochemistry to the diagnosis and study of ovarian neoplasms. Int J Gynecol Pathol 3:3-26 9. Lauriola L, Michetti F, Sentinelli S, Cocchia D (1984) Detection of S-100 labeled cells in nasopharyngeal carcinoma. J Clin Pathol 37:1235-1238 10. Mann WJ, Chumas J, Rosenwaks Z, Merril JA, Davenport D (1986) Elevated serum alpha fetoprotein associated with Sertoli-Leydig cell tumors. Obstet Gynecol 67:141-144
Immunohistochemistry: Granulosa cell tumors and ovarian carcinomas
177
11. Miettinen M, Walstrom T, Vitranen I, Talerman A, Astengo-Osuna C (1985) Cellular differentiation in ovarian sex cord-stromal and germ-cell tumors studied with antibodies to intermediate-filament proteins. Am J Surg Pathol 9:640-651 12. MooreBW (1965)A solubleprotein characteristic of the nervous system. Biochem Biophys Res Comm 19:739-744 13. Neunteufel W, Breitenecker G (1989) Tissue expression of CA125 in benign and malignant lesions of the ovary and fallopian tube: a comparison with CA19-9 and CEA. Gynecol Oncol 32:297-302 14. Norton AJ, Thomas JA, Isaacson PG (1987) Cytokeratin-specific monoclonal antibodies are reactive with tumors of smooth muscle derivation. An immunocytochemical and biochemical study using antibodies to intermediate filament cytosceletal proteins. Histopathology 11:487499 15. Scully RE (1984) Immunohistochemistry of ovarian tumors. In: Jose Russo (ed) Immunocytochemistry in tumor diagnosis. Boston Dordrecht Lancaster, pp 293-321 16. Young RH, Scully RE (1982) Ovarian sexcord-stromal tumors: recent progress. Int J Gynecol Pathol 1:101-123
