Human Reproduction, Vol. 17, No. 6, 1468-1471,
June 2002
© 2002 European Society of Human Reproduction and Embryology
An unusual steroid-producing ovarian tumour: Case report
1 Department of Diabetes and Endocrinology, University College Hospitals, Mortimer Street, London W1N 8AA, 2 Centre for Neuroendocrinology, Royal Free and University College Medical School, Pond Street, London NW3 2QG, 3 Department of Histopathology, Royal Free Hospital, Pond Street, London NW3 2QG and 4 University Department of Obstetrics and Gynaecology, Royal Free and University College Medical School, Pond Street, London NW3 2QG, UK
 |
Abstract |
|---|
 Top Abstract IntroductionCase reportDiscussionReferences |
|---|
The rapid onset of virilization in a post-menopausal woman is usually the result of androgen secretion from a tumour of adrenal or ovarian origin. Androgen secreting neoplasms of the ovary are rare and usually show autonomous secretion. Rarely, these may be driven by the high levels of gonadotrophins seen in the post-menopausal state. We describe the case of a 67-year-old woman with high serum testosterone and estradiol in association with the high gonadotrophin levels usually associated with the post-menopausal state. All hormonal parameters showed a significant suppression over 12 h with administration of the GnRH antagonist, cetrorelix. This observation implies that excess hormone synthesis was of ovarian origin and was gonadotrophin driven. Localization of the tumour was not possible by conventional ultrasound or computerized tomography scanning, but was achieved by venous sampling. Complete cure was achieved by total abdominal hysterectomy and bilateral salpingo-oophorectomy, with restoration of the endocrine profile to that expected for a post-menopausal woman. Rapidly acting GnRH antagonists, such as cetrorelix, offer a safe and useful diagnostic and therapeutic option in the management of ovarian steroid-secreting tumours, which show gonadotrophin dependency.
Key words: cetrorelix/GnRH antagonist/hyperandrogenaemia/ovarian tumour/virilization
|
Introduction |
|---|
|
TopAbstractIntroductionCase reportDiscussionReferences |
|---|
Postmenopausal virilization due to an androgen producing ovarian tumour may be autonomous or gonadotrophin driven (Givens et al., 1975
; Lamberts et al., 1982; Drife et al., 1987; Pascale et al., 1994; Chico et al., 1995; Picon et al., 2000). Previous reports have shown that some tumours may be inhibited by long acting GnRH analogues. Ovarian tumours showing this characteristic include steroid cell tumours of uncertain origin, Leydig cell tumours, granulosa cell tumours, hilus cell tumours and the rare stromal luteoma. The high levels of pituitary gonadotrophins, LH and FSH associated with the menopause may be responsible for driving ovarian steroid production (Paraskevas and Scully, 1989). GnRH analogues result in an initial hyperstimulation followed by inhibition of endogenous gonadotrophin release, and the subsequent drive to the ovary. Therefore this class of drug has a role in establishing not only whether excess steroid production is of ovarian origin but also whether the tumour is gonadotrophin driven. To date the GnRH analogues used have had a slow onset of action (some may take up to 1 week before suppression of gonadotrophins occurs) and have to be administered i.m., thus limiting their use as a diagnostic tool. We describe a case of virilization in a post-menopausal woman with high androgen, estradiol and gonadotrophin levels. Radiological localization of an ovarian tumour, by ultrasound and computerized tomography (CT) proved unsuccessful. The tumour was localized by ovarian vein catheterisation and shown to be gonadotrophin dependent with the aid of a cetrorelix suppression test. Cetrorelix (Asta Medica, Germany), a decapeptide with a sequence derived from GnRH, is a rapidly acting GnRH antagonist which suppresses gonadotrophin production rapidly. In humans the terminal half-lives after i.v. and s.c. administration are 8–9 h and 24–40 h respectively. In our patient the plasma levels of testosterone, estradiol, FSH and LH all dropped within 3 h of administration, proving that the increased steroid hormone synthesis was of ovarian origin and gonadotrophin dependent.
|
Case report |
|---|
|
TopAbstractIntroductionCase reportDiscussionReferences |
|---|
A 67-year-old woman presented with an 11 month history of increasing hair growth affecting the face, hands, abdomen, breasts and back. Previously she had a normal menstrual history with menarche aged 14 years and menopause in her mid forties. She had a history of hypertension. Examination showed her to be virilized with clitoromegaly and marked facial, abdominal and chest hair.
Initial investigations showed a total testosterone of 10.7 nmol/l (normal <2.9), androstenedione of 7.2 nmol/l (3.5–14), FSH of 34.7 U/l (37–125), LH of 19.9 IU/l (10.5–42), dihydroepiandrosterone sulphate (DHEAS) of 1.9 µmol/l (0.9–11.6) and estradiol of 392 pmol/l (post-menopausal <50). During a low dose dexamethasone suppression test, the plasma cortisol suppressed to <20 nmol/l but the total testosterone remained elevated at 15.2 nmol/l, confirming the non-adrenocorticotrophic hormone dependency of the hyperandrogenic state. A pelvic ultrasound examination noted the endometrial thickness to be 6.9 mm (usually atrophic and <5 mm in the post-menopausal state). The ovaries and uterus were otherwise normal. A CT scan of the pelvis and adrenal gland was normal, giving no clue as to the origin of the hyperandrogenism.
Adrenal and ovarian vein sampling was performed. The results are summarized in Table I and Figure 1, which clearly demonstrate an increase in testosterone and estradiol production from the right ovarian vein, consistent with the presence of a right ovarian testosterone- and estradiol-secreting tumour. We wished to ascertain whether ovarian hormone synthesis might be gonadotrophin dependent or autonomous and for the first time in the literature, report the acute effect of a GnRH antagonist using cetrorelix. Cetrorelix is a fast acting GnRH antagonist, which suppresses gonadotrophin secretion shortly after administration. Testosterone, FSH and LH were collected over a 24 h period following the administration of cetrorelix 0.5 mg at baseline and 6 h. Control samples were taken over a 24 h period prior to the test. The results are illustrated in Figure 2. These show that plasma testosterone was suppressed from 8 nmol/l to 2.3 nmol/l, along with the expected decrease in LH and FSH concentration.
|
|
|
The investigations concluded that the patient had a right ovarian tumour secreting both testosterone and estradiol. The cetrorelix suppression test suggested that the tumour was gonadotrophin dependent. No adverse effects were observed with the drug.
The patient underwent a total abdominal hysterectomy and bilateral salpingo-oophorectomy. Both ovaries were of normal size. Histology of the right ovary showed a 12 mm firm parenchymal nodule with a homogeneous pale yellow cut surface. Microscopic examination showed this to be a solid lesion, well demarcated from the surrounding stroma, composed of nests of cells with pale vesicular cytoplasm. There were no features of cellular hyperplasia. Reinke crystalloids were not identified. Occasional mitoses were noted, but no other sinister features such as nuclear atypia or haemorrhage. Immunostaining for inhibin was positive. These features are in keeping with an ovarian steroid cell tumour. No further sub-classification could be made. Examination of the endometrium showed evidence of simple cystic hyperplasia, presumably resulting from the continued estrogen exposure. Immunohistochemistry of the right ovary and endometrium did not show the presence of GnRH receptors. The left ovary was normal.
Three months post-surgery the patient was asymptomatic and repeat endocrinology had returned to normal (Table II). Of note, both LH and FSH were found to have further increased to the levels usually seen in a post-menopausal woman, demonstrating that although gonadotrophin dependent, the tumour products and inhibin were exerting a negative feedback action in the pituitary gland.
|
|
Discussion |
|---|
|
TopAbstractIntroductionCase reportDiscussionReferences |
|---|
We describe the unusual case of a post-menopausal woman with virilization secondary to a gonadotrophin driven ovarian steroid cell tumour. Of interest was the observation that post-operatively, the pituitary FSH and LH increased to the expected levels found in a post-menopausal woman indicating that there was an inhibitory negative feedback from the high levels of estradiol and testosterone. As demonstrated by venous sampling, this tumour not only secreted testosterone but also estradiol. Sixty percent of ovarian steroid cell tumours cannot be further sub-classified (Clement and Young, 2000
). Twenty percent are Leydig cell tumours and a further 20% stromal cell tumours (Hayes and Scully, 1987; Paraskevas and Scully, 1989). Leydig cell tumours are typified by the presence of cytoplasmic crytalloids of Reinke (Mandel et al., 1981). Previously, estrogenic manifestations have been reported in 60% and virilization in 12% of patients with ovarian steroid cell tumours (Paraskevas and Scully, 1989).
Ovarian steroid cell tumours are usually benign, unilateral tumours characterized by a steroid cell proliferation, located within the ovarian stroma. The small size (usually <1cm) accounts for the poor visualization with ultrasound and CT scanning. In line with previous studies (Kirshner and Jacobs, 1971), selective venous sampling in our case proved to be highly effective in tumour localization with a marked testosterone and estradiol gradient being present in the right ovarian vein. This is, however, an invasive procedure with the risk of haemorrhage. This procedure is operator dependent and is also dependent on reliable anatomy.
Eight case reports of gonadotrophin dependent tumours have been previously described (Givens et al., 1975; Lamberts et al., 1982; Drife et al., 1987; Paraskevas and Scully, 1989; Pascale et al., 1994; Chico et al., 1995; Marcondes et al., 1997; Picon et al., 2000). These consisted of four Leydig cell tumours, one granulosa cell tumour, two hilus cell tumours and one stromal luteoma. This gonadotrophin dependency appears to be related to a high LH receptor expression seen in some tumours (Gutierrez et al., 1983). In five of these cases, the tumour was 1 cm in size and six cases were post-menopausal. All of these showed suppression of testosterone levels with administration of i.m. GnRH analogues over a period of 2 weeks. In our case, cetrorelix induced suppression of testosterone and gonadotrophin levels within 6 h demonstrating this approach as a safe and rapid diagnostic procedure, which could be performed in an outpatient setting.
There have also been reports of GnRH analogues reducing androgen production in ovarian tumours, where patients (in the reproductive years) had normal levels of gonadotrophins. It has been proposed that, in these cases, GnRH analogues may have a direct effect on the ovarian tumour. These effects have been observed in vitro and appear to be related to inhibition of cellular steroidogenesis by antagonism of the GnRH receptor in ovarian tumour cells (Lamberts et al., 1982; Gutierrez et al., 1983; Latouche et al., 1989). Caution is needed as androgen secretion by adrenal tumours may occasionally be under gonadotrophin control (Werk et al., 1973; Leinonen et al., 1991) as opposed to being autonomous. Such adrenal tumours are LH dependent and will be suppressed by GnRH analogues. Also, poorly differentiated ovarian androgen secreting tumours may be partially or not at all suppressed by GnRH analogues (McLellan et al., 1990). However in the case we describe it is clear that their action is to decrease gonadotrophin levels and subsequent ovarian steroid hormone synthesis.
In conclusion, cetrorelix may offer a useful diagnostic test and therapeutic option for gonadotrophin dependent ovarian androgen-producing tumours. Localization is best achieved with venous sampling, but in centres where this facility may not be available and where radiological imaging is inconclusive, a cetrorelix suppression test may point to an ovarian cause of hyperandrogenaemia. Moreover, in patients in whom surgery is contraindicated, the identification of gonadotrophin dependency may allow a medical form of therapy using longer-lasting GnRH antagonists.
|
Notes |
|---|
5 To whom correspondence should be addressed. E-mail: jwstephens{at}tinyworld.co.uk
|
References |
|---|
|
TopAbstractIntroductionCase reportDiscussionReferences |
|---|
Chico, A., Garcia, J.L., Matias-Guiu, X., Webb, S.M., Rodriguez, J., Prat, J. and Calaf, J. (1995) A gonadotrophin dependent stromal luteoma; a rare cause of post-menopausal virilization. Clin. Endocrinol., 43, 645–649.[Medline]
Clement, P.B. and Young, R.D. (2000) Germ cell tumours of the ovary. Chapter 15. In Clement, P.B. and Young, R.D. (eds) Atlas of Gynecological Surgical Pathology, Mosby, London. pp. 230–290.
Drife, J.O., O'Malley, P.O. and Rosenthal, F.D. (1987) Long-term suppression of a testosterone-producing tumour by oestrogen/progesterone therapy. Clin. Endocrinol., 27, 121–124.[Medline]
Givens, J.R., Andersen, R.N., Wiser, W.L., Donelson, A.J. and Coleman, S.A. (1975) A testosterone secreting, gonadotrophin-responsive pure thecoma and polycystic ovarian disease. J. Clin. Endocrinol. Metab., 42, 845–853.
Gutierrez, S., Bermudez, M., Zylber, S., Tein, C., Naredo, E., Illanes, L. and Scaglia, H.E. (1983) LH receptor in testosterone producing ovarian tumour. Eur. J. Gynecol. Oncol., 3, 182–191.
Hayes, M.C. and Scully, R.E. (1987) Stromal luteoma of the ovary: A clinicopathological analysis of 25 cases. Int. J. Gynecol. Pathol., 6, 313–321.[ISI][Medline]
Kirshner, M.A. and Jacobs, J.B. (1971) Combined ovarian and adrenal vein catheterisation to determine the sites of androgen overproduction in hiruite women. J. Clin. Endocrinol., 33, 199–209.[ISI][Medline]
Lamberts, S.W.J., Timmers, J.M., Oosterom, R., Verleun, T., Rommerts, F.G. and De Jong, F.H. (1982) Testosterone secretion by cultured arrhenoblastoma cells: Suppression by luteinizing hormone releasing hormone agonist. J. Clin. Endocrinol., 54, 450–454.[Abstract]
Latouche, J., Crumeyrolle-Arias, N. and Jordan, D. (1989) GnRH receptors in human granulosa cells; anatomical localization and characterisation by autoradiographic study. Endocrinology, 125, 1739–1744.[Abstract]
Leinonen, P., Ranta, T., Siegberg, R., Pelkonen, R., Heikkila, P. and Kahri, A. (1991) Testosterone-secreting virilising adrenal adenoma with human chorionic gonadotrophin receptors and 21-hydroxylase deficiency. Clin. Endocrinol., 34, 31–35.[Medline]
Mandel, F.P., Voet, R.L., Weiland, A.J. and Judd, H.I. (1981) Steroid secretion by masculising and `feminising' hilus cell tumours. J. Clin. Endocrinol. Metab., 52, 779–784.[Abstract]
Marcondes, J.A.M., Nery, M., Mendoca, B.B., Hayashida, S.A.Y., Halbe, H.W., Carvalho, F.M., Wajchenberg, B.L. (1997) A virilizing tumor of the ovary associated with stromal hyperplasia under gonadotrophin control. J. Endocrinol. Investig., 20, 685–689.[ISI][Medline]
McLellan, A.R., Mowat, A., Cordiner, J., Beastall, G.H., Wallace, A.M., Connell, J.M.C. and Davies, D.L. (1990) Hilus cell pathology and hirutism. Clin. Endocrinol., 32, 203–212.[Medline]
Paraskevas, M. and Scully, R.E. (1989) Hilus cell tumour of the ovary. Int. J. Gynaecol. Pathol., 8, 299–309.[ISI][Medline]
Pascale, M.M., Pugeat, M., Roberts, M., Rousset, H., Dechaus, H., Dutriex-Berger, N. and Tourniaire, J. (1994) Androgen suppressive effect of GnRH agonist in ovarian hyperthecosis and virilizing tumours. Clin. Endocrinol., 41, 571–576.[Medline]
Picon, M.J., Lara, J.I., Sarasa, J.L., Recases, J.D., Clouet, R., Gonzalo, M.A. and Rovira, A. (2000) Use of long–acting gonadotrophin-releasing hormone analogue in a postmenopausal woman with hyperandrogenism due to a hilus call tumour. Eur. J. Endocrinol., 142, 619–622.[Abstract]
Werk, E.E., Sholiton, L.J. and Kalejs, L. (1973) Testosterone secreting adrenal adenoma under gonadotrophin control. New Eng. J. Med., 289, 767–769.[ISI][Medline]
Submitted on December 12, 2001; accepted on February 11, 2002.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. de Boer, M. de Man, K. de Bruyn, and A. van Sorge Cetrorelix suppression test to assess the source of androgen overproduction in postmenopausal hirsutism Eur. J. Endocrinol., September 1, 2006; 155(3): 391 - 393. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||