A puzzling ovarian tumour: pregnancy luteoma with diffuse endometriosis.

BACKGROUND: Pregnancy luteoma is a distinctive non-neoplastic hormone dependent lesion arising in pregnancy and mimicking an ovarian tumour. Fewer than 200 cases have been described in the English-language literature. Its clinical and morphological features are characteristic and must be considered in order to prevent diagnostic misinterpretation. To the best of our knowledge the association of pregnancy luteoma with endometriosis has not been reported in literature to date. CASE REPORT: A 30-year-old pregnant woman with no particular past medical history, consulted her gynaecologist at 17 weeks gestation for routine check-up. The patient was asymptomatic and did not show any signs of virilization. Ultrasonography disclosed a left adnexal heterogeneous mass measuring 7 cm in diameter with intramural vegetations. The right ovary was unremarkable. The patient underwent salpingo-oophorectomy considering the imaging findings were suspicious for malignancy. Histologically, the lesion was constituted of large sheets of luteinized polygonal cells with abundant eosinophilic cytoplasm and small round nuclei devoid of atypia and mitotic figures. In addition, there were several ectopic endometrial glands surrounded by abundant decidualized or edematous stroma. Immunohistochemically, these glands were immunoreactive for cytokeratin 7. The final pathological diagnosis was pregnancy luteoma associated with diffuse endometriosis. CONCLUSIONS: Because of its relative rarity, pregnancy luteoma is likely to be clinically misinterpreted and overtreated, as in the present case.

Development of an experimental ovarian tumor: immunocytochemical analysis.

OBJECTIVE: The aim of the present work was to study whether immunocytochemical parameters present in the normal ovary were altered after tumor development under high gonadotropin levels. METHODS: Ovarian tumors (luteoma): castrated female rats had an ovary grafted into the spleen; tumors were left to develop for 1, 2, 3 or 7 months. The presence of apoptotic cells (TUNEL method) and the expression of proliferating cell nuclear antigen (PCNA), gap junction protein (Cx43), steroidogenic acute regulatory protein (StAR), aromatase and synaptosome-associated protein of 25 kDa (SNAP-25) were determined by immunocytochemistry. Some of these findings were confirmed by RT-PCR (Cx43, StAR, SNAP-25). Inhibin subunit mRNAs were investigated by Northern blot. RESULTS: PCNA staining of tumors was mainly found in granulosa cells of transforming follicles and was absent from luteinized follicles. A nearly complete absence of apoptosis was observed. Cx43 was mainly found in follicles, while it was very weakly expressed or absent in luteinized follicles. StAR protein expression, indicating active steroidogenesis, was demonstrated only in luteinized follicles and in thecal cells, but was absent from granulosa cells. Aromatase immunoreactivity was very intense in granulosa and also present in luteal cells. Membrane-associated and cytoplasmic SNAP-25 immunostaining was determined in patches of endocrine cells in the follicles, as well as in the luteinized follicles. The expression of mRNAs for Cx43, StAR and SNAP-25 (RT-PCR) and inhibin subunits (Northern blots) were confirmed in 1-, 3- and 7-month-old tumors. CONCLUSIONS: These results indicated that luteoma most likely develop from unruptured follicles by hypertrophy and proliferation of follicular cells. Circulating gonadotropins seem to play a fundamental role in maintaining the expression of proteins typically expressed in normal ovary, while avoiding apoptosis in this tissue.

Dimeric inhibin concentrations in mares with granulosa-theca cell tumors.

OBJECTIVE: To determine whether concentrations of dimeric inhibin (CaCA) are greater in plasma and tumor fluid from mares with granulosa-theca cell tumors (GTCT), compared with concentrations in plasma and equine follicular fluid (eFF) from control mares. ANIMALS: 6 mares with GTCT and 12 clinically normal mares. PROCEDURE: The alphabetaA immunoradiometric assay used 2 antibodies, one against each subunit of inhibin (alpha and betaA subunits). Tumor tissue, tumor fluid, and a single blood sample were collected at the time of surgical removal of the GTCT. A single blood sample was collected from 7 control mares during various stages of the estrous cycle. Five other control mares were ovariectomized when their ovaries contained growing follicles of 25 to 35 mm in diameter. A blood sample and eFF from the largest follicle were collected at the time of ovariectomy. RESULTS: Mares with GTCT had significantly greater plasma concentrations of betabetaA (mean +/- SEM, 0.86 +/- 0.53 ng of recombinant human-alphabetaA/ml), compared with control mares (0.14+/-0.02 ng/ml). Concentrations of alphabetaA in tumor fluid and eFF were similar. Concentrations of alphabetaA were significantly lower after ovariectomy. CONCLUSIONS AND CLINICAL RELEVANCE: Dimeric inhibin concentration was higher in plasma from mares with GTCT than in plasma from control mares. Increased granulosa cell mass and loss of mechanisms regulating alphabetaA release in mares with GTCT likely accounted for the increase in plasma concentrations. Measurement of alphabetaA concentrations may be useful for identifying mares with GTCT.

Pregnancy luteoma with granulosa cell proliferation: an unusual hyperplastic lesion arising in pregnancy and mimicking an ovarian neoplasia.

A pregnancy luteoma (PL) was incidentally found at a term cesarean section in a 27-year-old black woman without any endocrine abnormality. The lesion involved only the left ovary; it had a nodular and focal pseudoalveolar growth pattern and was associated with areas of tubular sertoliform component, consistent with granulosa cell proliferation. Immunohistochemistry revealed a diffuse positivity to Inhibin A, CD99, cytokeratin and vimentin. The ultrastructure was typical of steroid-producing cells. PL is a tumor-like lesion arising in pregnant women and often misdiagnosed as a neoplastic lesion; awareness of this rare entity and its differential diagnoses may avoid unnecessary surgery in young patients.

Lipid cell (steroid cell) tumor of the ovary: immunophenotype with analysis of potential pitfall due to endogenous biotin-like activity.

Twenty-eight lipid cell (steroid cell) tumors of the ovary were studied by immunohistochemistry using an avidin-biotin complex detection system; 75% of tumors were vimentin positive, 46% were positive for cytokeratin (CAM5.2 antibody), 37% were positive with the cytokeratin cocktail AE1/AE3 and CK1, and 29% were positive for smooth muscle alpha-actin. Three tumors were positive for CD68 (KP-1), a histiocyte marker, and each of the following markers was positive in two cases: desmin, epithelial membrane antigen, neuron-specific enolase, and S-100 protein. All tumors tested were negative for chromogranin A, CD15 (Leu-M1), myoglobin, neurofilament protein, alpha-fetoprotein, carcinoembryonic antigen, and melanoma-associated antigen (HMB-45 antibody). Immunoreactivity for cytokeratins was usually focal, paranuclear, and globoid, while reactivity for actin and vimentin was diffuse and cytoplasmic. Based on these findings, melanomas and some carcinomas should be distinguishable from lipid cell tumors. However, the immunohistochemical profiles of smooth-muscle tumors, other gonadal stromal tumors (granulosa cell tumors, thecomas), and hepatocellular, renal cell, and adrenocortical carcinomas overlap with that of lipid cell tumors, and therefore these tumors may not be distinguishable from lipid cell tumors using this technique. In 10 cases (36%), negative controls exhibited weak to moderate nonspecific cytoplasmic staining. Evidence obtained using a biotin blocking kit, and a monoclonal antibody against biotin, suggests endogenous biotin-like reactivity as the source of the nonspecific staining.