The effects of bromocriptine, thyrotropin-releasing hormone, and gonadotropin-releasing hormone on hormone secretion by gonadotropin-secreting pituitary adenomas in vivo and in vitro

Updating the Landscape for Functioning Gonadotroph Tumors

Functioning gonadotroph adenomas (FGAs) are rare tumors, as the overwhelming majority of gonadotroph tumors are clinically silent. Literature is based on case reports and small case series. Gonadotroph tumors are poorly differentiated and produce and secrete hormones inefficiently, but in exceptional cases, they cause clinical syndromes due to hypersecretion of intact gonadotropins. The clinical spectrum of endocrine dysfunction includes an exaggerated response of ovaries characterized as ovarian hyperstimulation syndrome (OHSS) in premenopausal females and adolescent girls, testicular enlargement in males, and isosexual precocious puberty in children. Transsphenoidal surgery and removal of tumor reduces hormonal hypersecretion, improves endocrine dysfunction, and provides tissue for further analysis. Medical therapies (somatostatin analogues, dopamine agonists, GnRH agonists/antagonists) are partially or totally ineffective in many cases, especially with respect to antitumor effect. This review aims to update recent literature on these rare functioning tumors and highlight their therapeutic management.

Gonadotrope Tumors

Gonadotrope tumors arise from the gonadotropes of the adenohypophysis. These cells rarely give rise to hyperplasia, usually only in the setting of long-standing premature gonadal failure. In contrast, gonadotrope tumors represent one of the most frequent types of pituitary tumors. Despite their relatively common occurrence, the pathogenesis of gonadotrope tumors remains unknown. Effective nonsurgical therapies remain out of reach. We review the pituitary gonadotrope from the morphologic and functional perspectives to better understand its involvement as the cell of origin of a frequent type of pituitary tumor.

Follistatin, induced by thyrotropin-releasing hormone (TRH), plays no role in prolactin expression but affects gonadotropin FSHbeta expression as a paracrine factor in pituitary somatolactotroph GH3 cells

Follistatin regulates FSHbeta gene expression by binding to and bioneutralizing activin effects. In this study, we found that thyrotropin-releasing hormone (TRH) increased follistatin gene expression in pituitary somatolactotroph GH3 cells. Treatment of GH3 with 100 nM TRH significantly increased follistatin mRNA expression as determined by real time PCR. TRH-induced follistatin expression was significantly abrogated in the presence of MEK inhibitor, U0126. Overexpression of constitutive active MEKK in GH3 cells dramatically increased follistatin expressions. Transfection of GH3 cells with follistatin siRNA reduced endogenous follistatin mRNA expression, but failed to modulate prolactin promoter activity. Prolactin mRNA levels were not affected by increasing the dose of follistatin, and TRH-induced prolactin promoter activity was not modulated in the presence of follistatin. In other experiments using pituitary gonadotroph LbetaT2 cells, activin increased FSHbeta promoter activity and mRNA expression, and follistatin completely inhibited this activin-increased FSHbeta gene expression. Treatment of GH3 cells with activin reduced the basal activity of prolactin promoter and follistatin prevented this effect. GH3 cells were co-cultured with LbetaT2 cells, which had been transfected with FSHbeta promoter-linked luciferase vectors and treated with activin in the presence of TRH. Activin-induced FSHbeta promoter activity was completely inhibited in the presence of TRH. In addition to that, FSHbeta mRNA was not detected from LbetaT2 cells which were co-cultured with GH3 cells. Our current results suggest the possibility that TRH increases follistatin gene expression in prolactin-producing cells in association with ERK pathways. Somatolactotroph-derived follistatin affects gonadotrophs by countering activin-induced FSHbeta gene expression in a paracrine fashion.

Postoperative treatment of clinically nonfunctioning pituitary adenomas with dopamine agonists decreases tumour remnant growth

OBJECTIVE

There is no consensus as to the optimal postoperative treatment of patients with clinically nonfunctioning pituitary adenomas (NFPA) in whom total tumour removal has not been achieved. In this study we assessed whether dopamine agonist (DA) treatment can prevent postoperative remnant enlargement in NFPA.

DESIGN AND METHODS

Thirty-three patients (25 men/8 women; mean age, 61.7 +/- 11.2 years; mean follow-up, 40.6 +/- 4.8 months) were treated with DA, and their outcome was compared to that of 47 untreated patients (33 men/14 women; mean age, 59 +/- 2 years; mean follow-up, 42.9 +/- 4.2 months).

RESULTS

Tumour mass decreased or remained stable in 18/20 patients in whom DA treatment was initiated upon detection of residual tumour on postoperative MRI (group I). In 13 subjects (group II), DA therapy was started when tumour remnant growth became evident during the course of routine follow-up. Tumour growth stabilized or decreased in 8/13 (61.5%) of these patients. In contrast, tumour size remained stable in only 38.3% (18/47) of the untreated subjects (P < 0.0001 for comparisons among the three groups) and increased in the remaining 29 patients. Tumour enlargement free mean survival time was 103.7 +/- 8.8 months (CI 86.3-121) for group I, 43.9 +/- 9.6 months (CI 25.2-62.8) for group II and 36.7 +/- 3.8 (CI 29.2-44.2) for the control group (P = 0.0017). Treatment vs. control hazard ratio for tumour enlargement was 0.135 for group I (P = 0.007, 95% CI 0.032-0.577) and 0.892 for group II (P = 0.817; 95% CI 0.34-2.34).

CONCLUSIONS

Dopamine agonist therapy is associated with a decreased prevalence of residual tumour enlargement in patients with nonfunctioning pituitary adenomas, particularly when treatment is instituted before tumour remnant growth is detected.

Expression of thyrotropin-releasing hormone messenger RNA in human pituitary adenomas with follicle-stimulating hormone immunoreactivity

OBJECTIVE

To assess the correlation between thyrotropin-releasing hormone (TRH) messenger RNA (mRNA) and the immunoreactive type of human pituitary adenomas.

METHODS

Twenty-eight patients (14 to 73 years old) who had pituitary adenomas (18 nonfunctioning adenomas, 8 growth hormone-secreting adenomas, and 2 prolactinomas) underwent surgical treatment. Pituitaries removed at autopsy from four patients without evidence of pituitary disease were used as controls. Fragments of pituitary adenomas were processed for TRH mRNA by in situ hybridization (radioactive and nonradioactive) and for TRH peptide and anterior pituitary hormones (b-thyrotropin, b-follicle-stimulating hormone [FSH], bluteinizing hormone [LH], prolactin, and growth hormone) by immunohistochemistry with use of the avidinbiotin technique. Quantitative immunohistochemical studies were performed by using image analysis software. The signal was considered positive when more than 5% of the cells were stained.

RESULTS

Cells expressing TRH mRNA were detected in 22 of 28 pituitary adenomas--15 of 18 nonfunctioning pituitary adenomas, 5 of 8 growth hormone-secreting adenomas, and both prolactinomas. TRH peptide was revealed in only 10 adenomas, all expressing TRH mRNA as well. All but one nonfunctioning adenoma expressing TRH mRNA in more than 5% of the cells were b-FSH immunoreactive (15 of 16 cases; P<0.005, c 2 test), whereas only 6 of 16 nonfunctioning adenomas exhibited both b-thyrotropin and TRH mRNA and only 5 of 16 were positive for both b-LH and TRH mRNA.

CONCLUSION

These results confirm previous data demonstrating the presence of TRH mRNA and TRH peptide in human pituitary tumor cells. We further showed that the presence of TRH mRNA is significantly correlated with FSH immunoreactive gonadotropinomas. The release of FSH after an intravenous TRH test only in gonadotropinomas, together with local production of TRH, suggests a role for TRH in pathogenesis.

Pituitary tumor diagnosis and treatment

Over the past few years, significant contributions have been made to the understanding, diagnosis, and treatment of pituitary tumors. This article reviews recent advances in the areas of biology, diagnostic imaging, medical diagnosis and treatment, surgical results and technique, and adjuvant therapy in the form of radiotherapy and radiosurgery. Of particular note are the roles of endoscopy, intraoperative magnetic resonance imaging, radiosurgery, and radiation for nonfunction tumors, the diagnosis of Cushing's disease, the management of "incidentalomas," and new medication therapies.

Endocrine inactive and gonadotroph adenomas: diagnosis and management

Endocrine inactive pituitary adenomas represent about one quarter of all pituitary tumors. By immunocytochemistry, most of these tumors are positive for intact gonadotropins and/or their subunits. Clinical presentation is usually secondary to mass effect symptoms, such as visual disturbances, headache, and hypopituitarism. Differential diagnosis is usually accomplished by neuroradiologic studies, even though in selected cases positron emission tomography and/or single photon emission tomography may aid to distinguish pituitary adenomas from other endocrine inactive lesions, such as meningiomas and craniopharyngiomas. Surgical management is usually considered the first choice treatment for patients with endocrine inactive pituitary adenomas because it is very effective in ameliorating symptoms of chiasmal compression and headache. Radical removal of the tumor, however, is difficult to obtain because of the frequent invasiveness into the cavernous sinus. Radiation therapy diminishes the likelihood of tumor recurrence, especially in patients with demonstrable tumor remnants after surgery. Medical therapy with dopaminergic drugs, somatostatin analogs, or gonadotropin-releasing hormone agonists or antagonists causes mild reduction of tumor size in few patients and, therefore, seems to be of limited value in the therapeutic management of patients with endocrine inactive pituitary adenomas.

Dopamine D2 receptor gene expression in human adenohypophysial adenomas

The inhibitory effects of dopamine on adenohypophysial cells are mediated via dopamine subtype 2 receptor (D2R). Dopamine agonists inhibit hormone release and induce tumor shrinkage in most prolactin-secreting adenomas, whereas in other adenoma types such effects are sporadic. We investigated D2R gene expression by in situ hybridization (ISH) and immunocytochemistry in different types of pituitary adenomas. By ISH, a variable D2R signal was detected in 79 of 89 cases: 4 of 6 densely granulated and 8 of 8 sparsely granulated somatotroph, 4 of 4 mammosomatotroph, 7 of 7 mixed somatotroph-lactotroph, 4 of 4 acidophil stem cell, 16 of 16 sparsely granulated lactotroph, 11 of 16 corticotroph (functioning and silent), 3 of 4 silent subtype 3, 5 of 5 thyrotroph, 5 of 6 gonadotroph, 5 of 6 null cell, and 7 of 7 oncocytic adenomas. By immunocytochemistry, D2R protein was localized in cytoplasm and nuclei of 60 of 62 adenomas. In lactotroph adenomas, long-acting bromocriptine (BEC-LAR) induced a major increase in D2R mRNA, which was not accompanied by increased D2R immunoreactivity, suggesting mRNA stabilization. In conclusion, D2R gene is expressed in the majority of pituitary adenomas representing all tumor types. The significance of nuclear localization of D2R protein remains to be clarified.

Cabergoline modulation of alpha-subunits and FSH secretion in a gonadotroph adenoma

Most non-functioning pituitary adenomas respond poorly to medical therapy. We describe the case of a 62-year-old man who presented with clinical features of an invasive macroadenoma. Baseline hormonal evaluation revealed increased FSH and alpha-subunit (alpha-SU) levels. Transsphenoidal exeresis followed by radiotherapy (RT) was performed. Almost all neoplastic cells were intensely immunoreactive for alpha-SU. On PCR analysis, specific amplification products were observed for somatostatin 2, 3 and 5 receptors as well as for both short and long isoforms of the dopamine D2 receptor. In vitro, alpha-SU and FSH were released into the medium by adenoma cells and increased after TRH stimulation. After surgery, alpha-SU and FSH levels were still elevated. Short-term slow-release lanreotide treatment did not modify either alpha-SU or FSH levels. Cabergoline was started and a fast and long-lasting decrease in alpha-SU and, to a lesser extent, in FSH was observed. The tumor remnant was unmodified on magnetic resonance imaging 3 years after surgery and RT. This case report shows that the in vitro expression of somatostatin receptors may not be directly associated to the in vivo response of alpha-SU and FSH to lanreotide, probably because of a functional uncoupling of the receptors. Cabergoline should be considered as an effective therapy for hormonal, and perhaps proliferative, control of gonadotroph adenoma remnants before the effects of RT are fully effective.

Pituitary lactotroph adenomas develop after prolonged lactotroph hyperplasia in dopamine D2 receptor-deficient mice

Tuberoinfundibular dopamine tonically inhibits PRL expression and secretion from the pituitary gland by the activation of dopamine D2 receptors (D2R) localized on lactotrophs. Mutant female mice that lack D2Rs have persistent hyperprolactinemia but also develop extensive hyperplasia of pituitary lactotrophs and peliosis of the adenohypophysis at 9 to 12 months of age, while age-matched male D2R-deficient mice have no morphologic adenohypophysial lesion. We now report that both female and male D2R-deficient mice 17 to 20 months of age develop pituitary lactotroph adenomas. Of 12 aged female mice examined, all developed monohormonal PRL-immunoreactive neoplasms that had a characteristic juxtanuclear Golgi pattern of PRL staining and loss of the reticulin fiber network. Several of these adenomas were 50-fold larger than normal glands with marked suprasellar extension and invasion of brain but no gross evidence of distant metastases. They also had striking peliosis that was more marked than the lesion seen in the hyperplastic pituitaries of the younger females. These findings demonstrate that a chronic loss of neurohormonal dopamine inhibition promotes the hyperplasia-neoplasia sequence in adenohypophysial lactotrophs. Our results are analogous to previous data indicating that protracted stimulation of adenohypophysial cells by hormones or growth factors results in proliferation with initial hyperplasia followed by the development of neoplasia. Six aged male D2R-deficient mice had slightly enlarged anterior pituitaries similar in size to normal female glands. However, each case exhibited multifocal, microscopic lactotroph adenomas with strong nuclear immunoreactivity for estrogen receptors and Pit-1 transcription factor. The unexpected development of adenomas in males without preexisting or concomitant hyperplasia suggests that prolonged loss of dopamine inhibition may also cause neoplasia by distinct cellular mechanisms in male and female animals.

Successful Long-Term Management of a Gonadotroph Adenoma with Bromocriptine

OBJECTIVE

To report the successful 10-year management of a gonadotroph macroadenoma with bromocriptine and review the management of gonadotroph adenomas with bromocriptine.

METHODS

We present a case and review the pertinent literature. The effectiveness of bromocriptine in the management of gonadotroph adenomas is evaluated.

RESULTS

A 62-year-old man was found to have a pituitary tumor after seeking medical assistance because of a 6-month history of headaches and blurred vision. He had decreased visual acuity and bitemporal field defects. Serum follicle-stimulating hormone (FSH) levels were increased, whereas serum luteinizing hormone and total testosterone levels were normal. Treatment with bromocriptine resulted in a decrease in serum FSH levels, complete resolution of his symptoms, and considerable improvement in his visual acuity and visual field defects. Treatment with only bromocriptine for 10 years resulted in maintenance of normal serum FSH levels and no recurrence of symptoms.

CONCLUSION

In the management of a gonadotroph adenoma, we recommend consideration of a therapeutic trial of bromocriptine. In cases that are refractory to this therapy, surgical treatment or external pituitary irradiation could then be used.

Serum FSH bioactivity and inhibin levels in patients with gonadotropin secreting and nonfunctioning pituitary adenomas

It has been reported that serum FSH bioactivity and inhibin levels can be used as markers of the presence of true gonadotropin-secreting pituitary adenoma (Gn-oma). To verify this hypothesis, we have investigated the bioactivity of FSH and serum inhibin alpha-alpha and alpha-beta A levels in a series of patients with either Gn-oma or nonfunctioning pituitary adenoma (NFPA). Nine patients with Gn-oma (6 men and 3 women), 21 with NFPA (9 men and 12 women) and 30 normal subjects were included in the study. We studied FSH biological activity (FSH-B) by using Sertoli cell aromatase bioassay (SAB) and alpha-alpha and alpha-beta A inhibin levels by two noncompetitive immunometric assays (IEMA). In male patients with Gn-oma, serum immunoreactive FSH (FSH-I) and FSH-B levels ranged from 5.1 to 35.5 U/L and from 8.3 to 48 U/L, respectively, FSH B/I ratio being elevated in 2 (2.5 and 4.1; normal male range: 0.3-1.5), while female patients with Gn-oma had serum FSH-I and FSH-B levels ranging from 43.2 to 162 U/L and from 41.2 to 112.8 U/l, respectively, with a normal FSH B/I ratio. In male patients with NFPA, FSH-I and FSH-B levels ranged from 2.7 to 10.7 U/l and from 2.4 to 11.4 U/l while in females they ranged from 3.4 to 67.9 and from 4.6 to 60.8 U/l, respectively. FSH B/I ratio was elevated in 1 male (3.3) and normal in the remaining patients with NFPA. Serum alpha-alpha inhibin levels were normal or low in patients with Gn-oma and NFPA, while alpha-beta A inhibin concentrations were slightly elevated in 1 of 6 postmenopausal women (0.9; normal range < 0.7 U/ml). The present study confirms and extends previous reports indicating that male patients with Gn-oma may secrete FSH molecules with increased bioactivity. However, this abnormality was also observed in one male patient with NFPA. Moreover, the measurement of inhibin levels does not appear to be a reliable in vivo marker of pituitary tumors of gonadotroph origin, as it was normal or low in almost all patients with either Gn-oma or NFPA.

Spontaneous and agonist-induced calcium oscillations in single human nonfunctioning adenoma cells

The effects of gonadotropin-releasing hormone (GnRH) and GnRH-associated peptide (GAP) on cytosolic free calcium concentration ([Ca(2+)](i)) were investigated in 20 human nonfunctioning pituitary adenomas. We divided these tumors into three classes according to their response pattern to hypothalamic peptides. In type I adenomas (8 out of 20 adenomas), GnRH and GAP mobilized intracellular calcium ions stored in a thapsigargin (TG)-sensitive store. For the same concentration of agonist, two distinct patterns of GnRH-GAP-induced Ca(2+) mobilization were observed (1) sinusoidal oscillations, and (2) monophasic transient. The latter is followed by a protein kinase C (PKC)-dependent increase in calcium influx through L-type channels. In type II adenomas (7 out of 20 adenomas), GnRH and GAP only stimulate calcium influx through dihydropyridine-sensitive Ca(2+) channels by a PKC-dependent mechanism. TG (1 μM) did not affect [Ca(2+)](i) in these cells, suggesting that they do not possess TG-sensitive Ca(2+) pools. All the effects of GnRH and GAP were blocked by an inhibitor of phospholipase C (PLC), suggesting that they were owing to the activation of the phosphoinositide turnover. Type I and type II adenoma cells showed spontaneous Ca(2+) oscillations that were blocked by dihydropyridines and inhibition of PKC activity. GnRH and GAP had no effect on the [Ca(2+)](i) of type III adenoma cells that were also characterized by a low resting [Ca(2+)](i) and by the absence of spontaneous Ca(2+) fluctuations. K(+)-induced depolarization provoked a reduced Ca(2+) influx, whereas TG had no effect on the [Ca(2+)](i) of type III adenoma cells. The variety of [Ca(2+)](i) response patterns makes these cells a good cell model for studying calcium homeostasis in pituitary cells.

Gonadotropin-releasing hormone receptor mRNA expression by human pituitary tumors in vitro

An important question in the pathogenesis and regulation of human gonadotroph adenomas is whether heterogeneous gonadotropin responses to gonadotropin-releasing hormone (GnRH) are due to dysregulation of GnRH receptor biosynthesis and/or cell-signaling pathways. We investigated gonadotropin responsiveness to pulsatile GnRH in 13 gonadotroph adenomas. All tumors had evidence of follicle-stimulating hormone (FSH) beta and alpha subunit biosynthesis using reverse transcriptase/polymerase chain reaction (RTPCR) techniques. Four tumors significantly increased gonadotropin and/or free subunit secretion during pulsatile 10(-8) M GnRH administration. The GnRH antagonist Antide (10(-6) to 10(-8) M) blocked secretory increases in all GnRH-responsive tumors. Gonadotropin and/or free subunit secretion increased after 60 mM KCl, confirming that GnRH nonresponsiveness was not due to intracellular gonadotropin depletion. We hypothesized that GnRH nonresponsiveness in these tumors may be due to GnRH receptor (GnRH-Rc) biosynthetic defects. RTPCR analyses detected GnRH-Rc transcripts only in responsive tumors and normal human pituitary. This is the first demonstration of a cell-surface receptor biosynthetic defect in human pituitary tumors. We conclude (a) one third of gonadotroph tumors respond to pulsatile GnRH in vitro, (b) GnRH-Rc mRNA is detected in human gonadotroph adenomas and predicts GnRH responsiveness, and (c) GnRH-Rc biosynthetic defects may underlie GnRH nonresponsiveness in gonadotroph tumors.

Responses to gonadotropin releasing hormone agonist and antagonist administration in patients with gonadotroph cell adenomas

As they are clinically silent, gonadotroph cell pituitary adenomas are usually diagnosed only when pituitary enlargement causes visual impairment or hypopituitarism. In postmenopausal women presenting with pituitary tumors it can be difficult to determine whether gonadotropin hypersecretion is due to adenomatous or normal gonadotrophs prior to surgery. The usual GnRH dependency of gonadotropin secretion may be of diagnostic and therapeutic value. We therefore evaluated responses to the GnRH antagonist Nal-Glu-GnRH and to the long-acting GnRH agonist D-Trp6 (3.75 mg IM) in 9 and 4 patients with FSH- and/or alpha-subunit-secreting adenomas, respectively. Six of the 7 patients with FSH-secreting adenomas and one of the 2 patients with pure alpha subunit-secreting adenomas were studied postoperatively. In these patients postoperative FSH and/or alpha-subunit levels remained elevated and pituitary imaging by CT-scan and/or MRI disclosed tumoral residues. In the 2 remaining patients testing was performed preoperatively. A single administration of 5 mg Nal-Glu to the 7 patients with FSH-secreting adenomas produced a slight but significant fall in above-normal FSH levels from 24.4 +/- 15.4 IU/l to a nadir of 20.3 +/- 11.9 IU/l (-17%, p < 0.05) 20 h following the injection. LH levels fell markedly in the 6 patients with normal basal serum LH concentrations to those observed in hypophysectomized patients, while mean alpha-subunit levels were not modified. Alpha-subunit levels were not modified by Nal-Glu administration in the 2 patients with alpha-subunit-secreting adenomas.(ABSTRACT TRUNCATED AT 250 WORDS)

Usefulness of thyrotropin-releasing hormone test, SMS 201-995, and bromocriptine in the diagnosis and treatment of gonadotropin-secreting pituitary adenomas

Five patients with gonadotropin-secreting pituitary adenomas were studied. The utility of gonadotropin response to TRH stimulation in the diagnosis and follow-up of these tumors was evaluated, as well as the effects of somatostatin analogue SMS 201-995 and bromocriptine on gonadotropin release. Three patients had FSH and LH secreting adenomas while the other two tumors secreted FSH and alpha-subunit. Transsphenoidal resection of the pituitary adenomas were performed in all patients. Following preoperative TRH administration (400 micrograms i.v.), marked increases were observed in FSH levels in two cases, in LH levels in three and in alpha-subunit in one. The FSH and LH responses to t.his stimulus persisted in the same patients after surgery. Following acute bromocriptine administration (5 mg orally), FSH was reduced in all cases by 19% to 46%, LH in three cases by 50-67% and alpha-subunit in one by 33%. In patient no. 5, with persistent high FSH levels in the immediate postoperative period, long-term bromocriptine treatment was administered (15 mg/d orally), resulting in normalization of FSH levels 6 months later, although the size of the tumor was not reduced. After acute SMS 201-995 administration (100 micrograms sc) FSH decreased in two cases by 38% and 76%, LH in three by 30-56% and alpha-subunit in one by 20%. We conclude that gonadotropin response to TRH stimulation is useful in the diagnosis and follow-up of patients with gonadotroph adenoma. Bromocriptine and SMS 201-995 may be effective as coadjuvant treatment following surgery and radiotherapy in these patients, although long-term studies will be necessary to confirm these proposals.

Long-term treatment with the dopamine agonist CV 205-502 of patients with a clinically non-functioning, gonadotroph, or alpha-subunit secreting pituitary adenoma

OBJECTIVE

We aimed to assess the effects of prolonged treatment with the dopamine agonist CV 205-502 on tumour volume, visual field defects, and serum gonadotrophin and alpha-subunit concentrations in patients with gonadotroph, alpha-subunit secreting, or clinically non-functioning pituitary adenomas.

DESIGN

The patients were treated with CV 205-502 in a final daily dose of 300 micrograms for at least 1 year. The patients were seen at 2 or 3-week intervals during the first 3 months of treatment, and thereafter every 1 or 2 months. Computerized tomography and Goldmann perimetry were performed before treatment and during follow-up. Blood samples were drawn before treatment and at each out-patient visit.

PATIENTS

One patient with gonadotroph, two with alpha-subunit secreting, and two with clinically non-functioning pituitary adenomas were studied.

RESULTS

Computerized tomography showed tumour shrinkage in one patient. In two other patients an improvement of visual field defects was observed. In four patients, a significant decrease in serum FSH and/or alpha-subunit concentrations occurred within the first 3 months of treatment. In the remaining patient, a significant decrease of serum FSH and alpha-subunit concentrations was found after more than 3 months of treatment.

CONCLUSIONS

In patients with clinically non-functioning, gonadotroph, or alpha-subunit secreting pituitary tumours, long-term treatment with the dopamine agonist CV 205-502 decreases serum FSH and/or alpha-subunit concentrations. This decreased secretory activity from the pituitary tumour may be accompanied by an improvement of visual field defects, or tumour shrinkage on computerized tomography. Therefore, treatment with CV 205-502 may be useful in patients with clinically non-functioning, gonadotroph, or alpha-subunit secreting pituitary tumours, who cannot be operated upon.

Dopamine, the dopamine D2 receptor and pituitary tumours

Dopamine plays an important role in the hypothalamic-pituitary axis where its major effects are to inhibit pituitary hormone secretion and cell division. Chronic dopamine deficiency has been postulated as a cause of pituitary tumour formation and several lines of evidence exist to suggest that a functional deficiency may develop as a result of defective dopamine receptor action. The available data suggest that a number of sites in the dopamine-D2 receptor-second messenger pathways may be implicated. These abnormalities are reflected in the variety of responses to dopamine and its agonists which have been observed in pituitary tumours both in the clinical situation and in cultured cells in vitro. Whilst it seems likely that the primary defect in pituitary tumour formation lies within the pituitary itself, the role of hypothalamic factors in facilitating tumour growth remains to be explored. Further studies of the dopamine receptor and its function will be of value not only in pathophysiological studies of human pituitary adenomas, but also in the development of new pharmacological agents to treat patients with these tumours.

Recognition of gonadotroph adenomas in women

BACKGROUND

Pituitary adenomas that arise from the gonadotroph cells are being recognized with increasing frequency in men, but they are still rarely recognized in women. This rarity could be the result of an actual difference in occurrence or of greater difficulty in recognition. The tumors are usually recognized in men more than 50 years old, but elevated serum gonadotropin levels in women of that age could be produced by normal gonadotroph cells.

METHODS

Because the stimulation of follicle-stimulating hormone (FSH), luteinizing hormone (LH), and the beta subunit of LH (LH beta) by thyrotropin-releasing hormone (TRH) is a characteristic of gonadotroph adenomas in men, we administered TRH to 16 women with apparently nonsecreting pituitary macroadenomas and measured serum FSH, LH, LH beta, and the glycoprotein hormone alpha subunit every 15 minutes for 90 minutes before and 90 minutes after. The results were compared with the responses in 16 healthy women matched for age and in 10 women with macroadenomas secreting prolactin, growth hormone, or corticotropin. The tumors from 12 of the women with nonsecreting adenomas were cultured, and the secretion of FSH, LH, and LH beta in culture was determined.

RESULTS

Eleven of the 16 women with apparently nonsecreting adenomas had significant increases in serum LH beta in response to TRH, 3 had FSH responses, and 4 had LH responses. None of the 16 healthy women and none of the 10 women with secreting macroadenomas had LH beta, FSH, or LH responses to TRH. Ten of the 12 adenomas that were cultured secreted readily detectable amounts of FSH, LH, and LH beta, and their secretion in vitro correlated with the patients' responses to TRH in vivo.

CONCLUSIONS

Most apparently nonsecreting pituitary macroadenomas in women arise from gonadotroph cells. The majority of these can be recognized, even in postmenopausal women, by the serum LH beta responses to TRH, and some can be recognized by the responses of serum FSH and LH.

LH and FSH secretion and responses to GnRH and TRH in patients with clinically functionless pituitary adenomas

Serum concentrations of LH and FSH and their response to the separate administration of GnRH (100 micrograms i.v.) and TRH (200 micrograms i.v.) have been studied preoperatively in 12 patients with a clinically functionless pituitary adenoma, of whom nine (3F: 6M) were found to secrete gonadotrophins in vitro. In three patients with a gonadotrophin-secreting adenoma (GSA) the pulsatile release of LH and FSH was also assessed preoperatively. An elevated serum FSH was recorded in six of nine patients with a GSA, and was subnormal in one, whilst an elevated LH was recorded in only two and was subnormal in six. A doubling of LH occurred in only four of the nine patients after GnRH and in three of six after TRH. None of the three patients with a non-GSA was shown to have an aberrant response to GnRH or TRH. In patients with a GSA, pulsatile release of LH and FSH was usually asynchronous and neither hormone demonstrated any regular harmonic pattern. These data show that in patients with a GSA the serum FSH level is usually elevated but this is not invariable, and the LH may well be low. Pathological responses of LH are frequently found following the administration of either GnRH or TRH and these stimulation tests should be performed separately in patients presenting with a clinically 'non-functioning' pituitary tumour to assist in the preoperative diagnosis. The absence of normal LH and FSH pulsing also appears to be a feature of GS adenomas, and suggests that tumorous gonadotrophin secretion is not under physiological control by hypothalamic GnRH.

Confounding factors in the interpretation of gonadotropin and gonadotropin-subunit release from cultured human pituitary adenomas

Culture data of 31 human pituitary non-functioning adenomas and effects of cross-reactivity and in vitro culturing conditions on immunoreactivity of gonadotropins and subunits were investigated. Using immunoradiometric assays for FSH and LH and radioimmunoassays for alpha-subunit and LH beta-subunit crossreactivities were reduced to a minimum. Repeated freezing and thawing had no effect on immunoreactivity of hormones and subunits tested. Incubation at 37 degrees C did not affect the immunoreactivity of purified subunit preparations and no recombination of alpha-subunit and LH beta into intact LH could be demonstrated after coincubation of the subunits. FSH immunoreactivity in culture media from 3 pituitary tumors was not affected by incubation at 37 degrees C. LH from a purified preparation and LH in culture media from 3 pituitary adenomas showed a rapid decrease of LH immunoreactivity when left at 37 degrees C. Concomitant with decreasing LH levels at 37 degrees C, a rise in the concentration of alpha-subunit occurred. A direct correlation between gain in alpha-subunit and loss of LH was found. LH beta levels remained stable while LH decreased. This observation may be attributed to an increase in LH beta levels which is compensated by the loss of LH, which has a relatively high crossreactivity in the LH beta immunoassay. LH, FSH, alpha-subunit, LH beta or a combination of these glycoproteins could be demonstrated in 26 out of 31 cultured tumours from patients operated upon because of a clinically non-functioning adenoma. In none of the media of 15 adenomas in which both alpha-subunit and LH were detected, could alpha-subunit levels have been caused by dissociation of LH at 37 degrees C. In two cases, measured LH levels could have been caused by crossreactivity of alpha-subunit and FSH. It was concluded that: (1) in research of non-functioning pituitary adenomas data on gonadotropin and gonadotropin-subunit secretion may suffer from bias caused by crossreactivity; (2) that dissociation of LH into subunits at 37 degrees C is relatively unimportant in in vitro research of non-functioning adenomas; (3) that virtually all non-functioning pituitary adenomas contain or release gonadotropins and/or subunits.

In vitro studies of human pituitary adenomas

The application of morphologic and tissue culture techniques to the study of human pituitary adenomas allows further investigation of structure-function correlations. Using these methods, we have documented morphometric differences between densely and sparsely granulated somatotroph adenomas but the release of growth hormone in vitro and responses to adenohypophysial hormones/drugs do not correlate with tumor type. The morphologic and functional alterations in somatotroph adenomas exposed to SMS 201-995 appear to be reversible in vitro. Incubation of lactotroph adenomas with bromocriptine for 3 days directly reduces tumor cell size, cytoplasmic volume and cytoplasmic volume densities of endoplasmic reticulum and Golgi regions; these changes are similar to the effects of longterm bromocriptine therapy in vivo. Tissue culture studies of gonadotroph adenomas of men confirm that gonadotropin-releasing hormone (GnRH) stimulates gonadotropin release by tumor cells and yields morphologic evidence of increased hormone synthesis whereas these tumors have variable sensitivity to gonadal steroids; structural changes in tumor cells correlate with hormone release after stimulation, suggesting that morphologic parameters may reflect the hormonal milieu of these adenomas. Null cell adenomas and oncocytomas release small quantities of glycoprotein hormones, predominantly gonadotropins in vitro and there are no functional differences between these 2 tumor types; gonadotropin release responds to GnRH stimulation and, paradoxically, to other adenohypophysiotropic hormones, but such stimulation does not result in secretion of other adenohypophysial hormones by these tumors.

Immunohistochemical studies of human FSH producing pituitary adenomas

Ten FSH producing pituitary adenomas were studied immunohistochemically. 9 cases were in males, and 7 showed elevated serum FSH levels. Immunohistochemically, all cases showed the presence of alpha-subunit and FSH-beta subunits in many tumour cells. These two subunits were frequently colocalized in the same cells. However, the expression of LH-beta subunit was extremely low (1 of 10 cases exhibiting occasional LH-beta positive tumour cells), although it has been reported that FSH-beta and LH-beta subunits are colocalized in the same cells of the normal adult pituitary gland. Immunoelectron microscopically, alpha-subunits and FSH-beta were present in the secretory granules and suggested the co-release of subunits or secretion of combined form of FSH. In 7 cases, TSH-beta was positive, and in some cases, TSH-beta was colocalized in the same tumour cells which contained alpha-subunit and FSH-beta subunit. A few cases also demonstrated immunoreactivity for PRL and ACTH. Our immunohistochemical studies suggest that FSH adenomas are multihormonal and that there is abnormal gene expression in FSH cells with loss of LH-beta appearance and co-expression of TSH-beta.