PubMed1998
We investigated 203 ovarian tumors and tumor-like lesions for inhibin expression using a monoclonal anti-inhibin alpha antibody. Inhibin was present in the tumor cells in all 14 primary adult granulosa cell tumors (4 luteinized) plus 3 metastatic, all 10 primary juvenile granulosa cell tumors plus 1 metastatic, 10 of 11 thecomas, 3 of 11 fibromas, 4 of 11 sclerosing stromal tumors, 6 of 11 Sertoli cell tumors (1 oxyphilic), 7 of 11 Sertoli-Leydig cell tumors, 1 gynandroblastoma, 10 primary ovarian sex cord tumors with annular tubules plus 2 metastatic, 8 of 9 steroid cell tumors, both pregnancy luteomas, 1 of 2 unclassified sex cord tumors, 2 of 5 gonadoblastomas, 9 of 10 female adnexal tumors of probable wolffian origin, and in the non-neoplastic stroma of many carcinomas and germ cell tumors. The tumor cells were inhibin-negative in 10 fibrosarcomas, 12 small cell carcinomas of hypercalcemic type, 24 germ cell tumors (except for a focus of inhibin-positive syncytiotrophoblast in one case), and 17 ovarian carcinomas. Two of the three inhibin-positive fibromas showed diffuse immunostaining and were associated with evidence of estrogenic activity. Among nine Sertoli-Leydig cell tumors with available clinical data, four that were more than minimally inhibin-positive were accompanied by androgenic manifestations; five inhibin-negative or only minimally positive tumors lacked such evidence. Inhibin immunostaining may be useful in the differential diagnosis of inhibin-positive sex cord tumors versus histologically similar inhibin-negative neoplasms, but inhibin negativity does not preclude a diagnosis of sex cord tumor. The unexpected, common inhibin positivity of female adnexal tumors of probable wolffian origin indicates that inhibin staining cannot be used to differentiate these tumors from Sertoli cell tumors. Inhibin immunostaining is also helpful in identifying potential steroid hormone-secreting cells in the non-neoplastic stromal component of epithelial, germ cell, and other ovarian tumors.