New long-acting bromocriptine (Parlodel MR and Parlodel LAR) in the treatment of pituitary tumours with hyperprolactinemia.

In order to assess the efficacy and tolerability of new long acting bromocriptine: Parlodel MR (oral form) and Parlodel LAR (injectable form suitable for repeatable administration) 40 patients (29 women and 11 men) with pituitary tumours with hyperprolactinemia (PRL 70 micrograms/l) were investigated in a double blind study. Patients were divided into 2 groups of 20. In the first group Parlodel R or Parlodel MR in equivalent doses was given, the other group was administered Parlodel R or Parlodel LAR. During the next 6 months 20 patients were treated with Parlodel MR and the other 20 with Parlodel LAR. In all patients pituitary and peripheral hormones, CT scan and visual fields were examined before and after 28 days of bromocriptine treatment. During the next six months 20 patients were treated with Parlodel MR while the other 20 with Parlodel LAR. Serum PRL fell in all patients and values in the normal range were obtained in 36 patients. In 30 out of 35 patients with signs of pituitary tumour in CT scan, a significant tumour shrinkage was observed. Most patients achieved considerable clinical improvement: disappearance of galactorrhoea, resumed menses in women, increased potency in men. There were no difference in efficacy in Parlodel R, Parlodel MR and Parlodel LAR, but in the case of Parlodel LAR the least number of side effects was found. Treatment with long acting bromocriptine-Parlodel MR and LAR of patients with pituitary tumours with hyperprolactinemia is an efficacious, safe and better tolerated method than Parlodel R treatment.     

Bromocriptine in management of large pituitary tumours.

Bromocriptine has an accepted place in the management of small pituitary tumours that secrete either prolactin or growth hormone. The treatment of large tumours with extrasellar extensions is more difficult, however: though surgery is the standard treatment, it is often unsuccessful in returning excessive hormone secretion to normal and may cause hypopituitarism. A prospective trial was undertaken to assess the frequency with which changes in pituitary function and size of large tumours occurs. Nineteen patients were studied before and during treatment with bromocriptine (7.5 to 60 ml/day) for three to 22 months, using contrast radiology and a detailed assessment of pituitary function. Eighteen patients had hyperprolactinaemia and two of these also had raised concentrations of growth hormones; one patient had an apparently non-functioning tumour. In 12 patients (63%) tumour size decreased with bromocriptine and no tumour enlarged. Nine patients had visual-field defects, which improved in seven, becoming normal in five. Pituitary function improved in nine patients (47%) becoming entirely normal in three. Bromocriptine should be the treatment of choice in patients with large pituitary tumours with extrasellar extensions, provided close supervision is maintained.     

Comparison between a slow-release oral preparation of bromocriptine and regular bromocriptine in patients with hyperprolactinemia: a double blind, double dummy study.

The efficacy and tolerability of a slow-release preparation of bromocriptine (Parlodel SRO) were compared to those of conventional bromocriptine (Parlodel R) in a double blind, double dummy study of 12 hyperprolactinemic women (plasma PRL 81.3 +/- 4.73, ng/ml mean +/- SEM). For 2 weeks, the patients received 2.5 mg b.i.d. Parlodel R or 5 mg once daily Parlodel SRO; for the following 2 weeks, the dose of the drugs was doubled. The patients were then treated, in an open study, with 2.5-10 mg daily Parlodel SRO for 6 months. Both preparations caused a prompt and sharp PRL fall. Hormone levels remained inhibited over the whole month of observation with both preparations. Daily PRL profiles were very close with either drug although morning PRl levels were slightly higher during Parlodel SRO than during Parlodel R administration. Doubling the doses of the two drugs did not result in further significant lowering of PRL values. During the 6-month study with Parlodel SRO, plasma PRL further decreased and normalized in 11 of 12 patients. Clinical improvement occurred in the majority of cases. Tolerability of Parlodel SRO appeared to be better, though without statistically significant differences, than that of Parlodel R. Side effects were less important with the former compound in their number, severity and duration. In conclusion, thanks to its favourable pharmacological profile, Parlodel SRO appears to be a valuable alternative to regular bromocriptine in the management of hyperprolactinemia.     

Effects of bromocriptine on pituitary tumour size.

In a prospective study designed to assess the influence of bromocriptine on pituitary tumour size 12 patients with pituitary tumours, eight of whom had suprasellar extensions, were treated for three months with 20 mg of bromocriptine daily after a gradual increase to this dose. The group comprised eight women and four men, five with prolactin-secreting adenomas, four with acromegaly, two with functionless adenomas, and one with Nelson''s syndrome. All five patients with prolactin-secreting adenomas showed a reduction in pituitary tumour size as assessed by computerised tomography and metrizamide cisternography accompanied by a fall in prolactin concentrations and clinical and biochemical improvement in their hypopituitarism. One patient in this group had a visual-field defect before treatment, and this resolved. There was no radiological evidence of reduction in tumour size in the remaining seven patients, though this might refect the fairly short duration of treatment, particularly in view of the ancillary evidence of clinical, biochemical, and visual-field improvement in some of the patients. These results emphasise the potential value of bromocriptine in treating patients with large prolactinomas or recurrences of such tumours after previous chiasmal decompression and conventional external megavoltage irradiation on the pituitary.     

Acute administration of bromocriptine abolishes the hyperprolactinemic response induced by submaximal exercise in man.

The effective control of hypophysial prolactin (PRL) secretion with a pharmacological agent is a prerequisite for the investigation of the role of hyperprolactinemia observed during exercise. Using bromocriptine, a potent inhibitor of PRL secretion, this study established the proper experimental conditions whereby any significant increase in plasma PRL level can be prevented and basal circulating levels maintained during physical exercise. On three occasions at weekly intervals, 15 male adults, separated into two groups, exercised on an ergocycle (40 min at 65% VO2max) either 1 or 3 h after ingesting either placebo or 1.25 or 2.50 mg of bromocriptine mesylate (Parlodel; Sandoz Canada Inc., Dorval, Qué.). Under all conditions, the plasma PRL elevation observed during exercise after placebo was prevented by the administration of bromocriptine. Resting plasma PRL levels were maintained when exercise was performed 1 h after bromocriptine ingestion, but were significantly reduced when exercise was performed 3 h after administration of either bromocriptine dosages. Considering the primary and secondary effects observed, 1.25 mg of bromocriptine administered 1 h before exercise provides suitable experimental conditions to investigate the role of the increase in plasma PRL during physical exercise.     

Leukaemia inhibitory factor and interleukin 6 inhibit secretion of prolactin and growth hormone by rat pituitary MtT/SM cells.

The rat pituitary cell line, MtT/SM, has the characteristics of somatomammotrophs. The cells secrete both prolactin (PRL) and growth hormone (GH). We examined the effects of cytokines such as leukaemia inhibitory factor (LIF), interleukin 6 (IL-6), oncostatin M and interleukin 11 on the secretion of these hormones by the cells. These cytokines stimulate proliferation of the cells and inhibit the secretion of PRL by 70-80% and that of GH by 50%. They induce tyrosine phosphorylation of STAT3 in the cells. The cells containing PRL or GH decreased at 48 h after treatment of the cells with LIF or IL-6. These results suggest that the LIF/IL-6 family of cytokines inhibits the functions of mammotrophs and somatotrophs in the pituitary gland.     

Differences in prolactin receptor (PRLR) in mouse and human fallopian tubes: evidence for multiple regulatory mechanisms controlling PRLR isoform expression in mice

The anterior pituitary-derived hormone prolactin (PRL) signals through the PRL receptor (PRLR) and is important for female reproductive function in mammals. In contrast to the extensive studies of PRLR expression and regulation in human and mouse ovary and uterus, the mechanisms controlling the regulation of PRLR isoform expression in the fallopian tube are poorly understood. Because dynamic interaction of hormonal signaling in gonadal tissue and the pituitary or in gonadal tissues themselves in mammals suggests endocrine or paracrine regulation of PRLR expression, we questioned whether differential regulation of PRLR isoforms by PRL ovarian-derived estrogen (E(2)) and progesterone (P(4)) exists in the fallopian tube and pituitary of prepubertal female mice. Western blot analysis showed distinct molecular separation of PRLR isoforms in mouse and human fallopian tubes, and cellular localization was found in mouse and human tubal epithelia but not in mouse tubal smooth muscle cells. These data support the concept of an isoform- and cell type-specific expression of PRLR in human and mouse fallopian tubes. Moreover, expression of the long form of PRLR decreased after PRL treatment and increased after blockage of endogenous PRL secretion by bromocriptine (an inhibitor of PRL secretion) in a time-dependent manner in mouse fallopian tube. The opposite regulation was observed in the pituitary. Treatment with exogenous E(2) or P(4) led to changes in PRLR expression in the fallopian tube similar to those of PRL treatment. However, E(2) and P(4) did not affect PRLR expression in the pituitary. Estrogen had no effect on the long form of PRLR expression, whereas P(4) regulated the long form of PRLR in the fallopian tube, as did PRL. Taken together, the data from our comparative study provide evidence that PRLR can be regulated by an interplay of two different mechanisms, PRL or ovarian steroid hormones independently or in combination in a tissue-specific manner. Furthermore, we found that ovarian steroid hormones selectively suppress the expression of PRLR isoforms in mouse fallopian tubes. These findings may contribute to our understanding of the mechanisms controlling PRLR isoform expression in the fallopian tube (in addition to ovary and uterus), with implications for female reproduction.     

Somatostatin receptors in GH-secreting pituitary adenomas--their relationship to the response to octreotide

Twenty pituitary adenomas, surgically removed from patients suffering from acromegaly, were studied. The tumours were immunostained with anti-GH and anti-PRL antibodies and with antibodies raised against particular subtypes of somatostatin receptors (rsst1-5). Expression of rsst immunoreactivity was scored using the following scale: 0--negative reaction, 1--weak reaction, 2--moderate reaction, 3--strong reaction. In 15 patients in whom the GH response to the acute administration of 200 ug octreotide was tested the correlation between the expression of rsst and the percentage of GH drop was estimated. All the tumours were GH-immunopositive and in the majority (14/20) co-expression of PRL was also found. All the adenomas examined expressed rsst2A (20/20) and rsst5 (12/12) receptor proteins. Receptors sst2B and rsst3 were found in all but one of the tumours examined (19/20 and 11/12, respectively). None of tumours investigated presented rsst4 immunopositivity. The mixed (GH/PRL) adenomas showed a tendency to a higher expression of rsst2A + + rsst2B and a greater response to octreotide administration. A significant positive correlation was found between rsst2A + rsst2B expressions and a drop in GH after octreotide. To conclude, the GH-inhibiting effect of octreotide depends on the intensity of expression of both rsst2A and rsst2B. Both isoforms of rsst2 mediate the same biological response (inhibition of GH secretion) in GH-secreting and GH/PRL-secreting adenomas.     

Expression of hypoxia-inducible factor-1alpha (HIF-1alpha) in pituitary tumours

The present study was performed to investigate HIF-1alpha (hypoxia-inducible factor-1alpha) expression in a large number of immunohistochemically and ultrastructurally characterized surgically removed pituitary tumours. The potential relation of HIF-1alpha with outcome variables as well as the presence of HIF-1alpha expression in the tumours treated with dopamine agonists and octreotide, a long-acting somatostatin analogue was also investigated. HIF-1alpha immunoreactivity was confined to the nucleoplasm whereas the nucleoli were unconspicuous. The distribution of HIF-1alpha was evident in the tumours whereas normal adenohypophysial cells showed no HIF-1alpha staining. HIF-1alpha expression was detected not only in the tumour cells but also in endothelial cells lining the blood vessels within the tumour. ACTH producing adenomas showed the lowest level of HIF-1alpha expression whereas pituitary carcinomas and GH producing adenomas had the highest counts. The statistical study demonstrated no significant correlation between HIF-1alpha expression, patient age, gender, tumour, size, invasiveness, cell proliferation rate and vascularity. These results suggest that the behaviour of pituitary tumours does not primarily depend of HIF-1alpha expression. Our study demonstrated an increase HIF-1alpha expression in bromocriptine treated PRL producing pituitary adenomas compared with untreated tumours but no increase in octreotide treated tumours.     

Effects of prolonged treatment with diltiazem on pituitary secretion of luteinizing hormone, follicle-stimulating hormone, thyrotropin and prolactin.

In previous studies it has been observed that acute administration or short-term treatment with calcium channel blockers can influence the secretion of some pituitary hormones. In this study, we have examined the effect of the long-term administration of diltiazem on luteinizing-hormone (LH), follicle-stimulating hormone (FSH), thyrotropin (TSH) and prolactin (PRL) levels under basal conditions and after gonadotropin-releasing hormone (GnRH)/thyrotropin-releasing-hormone (TRH) stimulation in 12 subjects affected by cardiovascular diseases who were treated with diltiazem (60 mg 3 times/day per os) for more than 6 months and in 12 healthy volunteers of the same age. The basal levels of the studied hormones were similar in the two groups. In both the treated patients and the control subjects, a statistically significant increase (p < 0.01) in LH, FSH, TSH and PRL levels was observed after GnRH/TRH administration. Comparing the respective areas under the LH, FSH, TSH and PRL response curves between the two groups did not present any statistically significant difference. These findings indicate that long-term therapy with diltiazem does not alter pituitary hormone secretion.     

Effects of prolactin on the luteinizing hormone response to gonadotropin- releasing hormone in primary pituitary cell cultures during the ovine annual reproductive cycle

In the sheep pituitary, the localization of prolactin (PRL) receptors in gonadotrophs and the existence of gonadotroph-lactotroph associations have provided morphological evidence for possible direct effects of PRL on gonadotropin secretion. Here, we investigated whether PRL can readily modify the LH response to GnRH throughout the ovine annual reproductive cycle. Cell populations were obtained from sheep pituitaries during the breeding season (BS) and the nonbreeding season (NBS), plated to monolayer cultures for 7 days, and assigned to receive one of the following treatments: 1) nil (control), 2) acute (90- min) bromocriptine (ABr), 3) chronic (7-day) bromocriptine (CBr), 4) ABr and PRL, 5) CBr and PRL, 6) PRL alone, or 7) thyrotropin-releasing hormone. Cells were treated as described above, with the aim of decreasing or increasing the concentrations of PRL in the culture, and simultaneously treated with GnRH for 90 min. The LH concentrations in the medium were then determined by RIA. GnRH stimulated LH in a dose-dependent manner during both stages of the annual reproductive cycle. During the NBS, single treatments did not significantly affect the LH response to GnRH. However, when PRL was combined with bromocriptine, either acutely or chronically, GnRH failed to stimulate LH release at all doses tested (P < 0.01). In contrast, during the BS, the LH response to GnRH was not affected by any of the experimental treatments. These results reveal no apparent effects of PRL alone, but an interaction between PRL and dopamine in the regulation of LH secretion within the pituitary gland, and a seasonal modulation of this mechanism.     

Pregnancy in patients presenting with hyperprolactinaemia.

Ninety-two pregnancies occurred in 76 hyperprolactinaemic patients treated with bromocriptine. Half conceived within three months of attempted conception. There was no evidence of an increased rate of spontaneous abortion, fetal abnormality, or multiple pregnancy; the three twin pregnancies occurred in women who were additionally treated with clomiphene and human chorionic gonadotrophin. Thirty-one patients had radiological evidence of a pituitary tumour; 14 with major radiograph changes in the pituitary fossa or serum prolactin concentrations greater than 100 ng/ml received pituitary irradiation before conception. None of the latter showed evidence of enlargement of the tumour during pregnancy. In contrast two of the four patients with similar tumours but who were not irradiated developed visual field defects, one with gross destruction of the pituitary fossa. Prophylactic treatment to limit subsequent tumour expansion during pregnancy in patients with prolactinomas is indicated, and pituitary irradiation before conception appears to be a safe and effective method to achieve this goal.     

Interactions of steroids with prolactin secretion in vitro

Estrogens prevent or diminish the sensitivity to dopamine of prolactin (PRL) secretion by cultured rat pituitary cells. Cultured tumor cells prepared from a transplantable rat PRL-secreting tumor were insensitive to dopamine and bromocriptine, while the anti-estrogen tamoxifen restored this sensitivity. Cultured normal human pituitary cells were shown to be more sensitive to dopamine, if they were preincubated with estradiol, while cultured human prolactinoma cells became insensitive to bromocriptine after they were exposed to estrogens. This sensitivity was restored, however, by tamoxifen. These results point to an important species difference between primates and rodents with regard to the normal regulation of PRL secretion.     

Abnormal dopamine sensitivity in some human prolactinomas

In most of human prolactin (PRL)-secreting adenomas, dopamine and dopamine agonists normally suppress the excessive PRL secretion. Nevertheless, a subpopulation of such patients presents a relative insensitivity to the ergot derivative bromocriptine. Six patients with a macroadenoma (n = 5) or microadenoma (n = 1) were considered resistant to bromocriptine which, at a daily dose of 15-60 mg, did not normalize high plasma PRL levels. Culture studies of these adenoma cells showed that: (1) 10(-8) M bromocriptine produced a 32 +/- 16% inhibition of PRL release versus 65 +/- 12% obtained in the same conditions with normal human pituitary cells; (2) sulpiride (10(-6) M) reversed the inhibitory effects of bromocriptine, and (3) the bacterial endotoxins, cholera toxin (10(-11) M) and pertussis toxin (250 ng/ml), respectively, produced a 45-500% increase and a total abolition of bromocriptine-induced PRL inhibition. These observations and recent data of the literature allow to discuss the possibility of receptor or postreceptor defects in such tumors.     

Role of prolactin and luteinizing hormone in regulating timing of implantation in the spotted skunk

The western spotted skunk exhibits an obligate delay of implantation lasting 200-220 days. The pituitary is essential for luteal activation. The corpora lutea, in turn, secrete the hormones necessary for blastocyst implantation. Two experiments were designed to determine which pituitary hormones are responsible for increasing luteal activity and induction of implantation. Forty-two pregnant skunks with delayed implanting blastocysts were treated as follows: 13 served as untreated controls, 6 received 0.5 mg prolactin (PRL) daily, and 5 received diluent beginning in January. Four received 1.5 mg bromocriptine (CB-154) daily, 3 received both CB-154 and PRL, 3 received diluent, 5 received a gonadotropin-releasing hormone agonist (GnRHa) dispensed from osmotic minipumps, and 3 received diluent dispensed from osmotic minipumps starting in April. The skunks were subjected to a natural photoperiod. Duration of preimplantation and blood levels of progesterone and luteinizing hormone were measured. PRL significantly (p less than 0.05) shortened and CB-154 significantly (p less than 0.05) prolonged the duration of preimplantation when compared to controls (148 +/- 33.6 vs. 251 +/- 3.2 vs. 199 +/- 5.1 days, respectively). PRL was able to reverse the inhibitory effect of CB-154 when both were administered simultaneously (195 +/- 4.0 vs. 251 +/- 3.2 days). GnRHa had no significant (p greater than 0.05) effect on duration of preimplantation (199 +/- 5.1 days) when compared to controls (203 +/- 3.2 days). These results indicate that PRL is the primary pituitary hormone responsible for increased luteal activity and subsequent blastocyst implantation in the spotted skunk.     

Ectopic corticotrophin-releasing-factor and growth hormone releasing factor secretion: diagnosis using human pituitary cell culture

Cell culture of human pituitary tissue has been used to diagnose a patient with Cushing's syndrome due to ectopic secretion of corticotrophin-releasing factor (CRF; case 1) and a case of acromegaly associated with ectopic secretion of a growth-hormone releasing factor (GRF; case 2). In both patients a pituitary tumour was not detected. Case 1 had a small cell carcinoma and symptoms of the ectopic ACTH syndrome, but in culture the carcinoma failed to secrete detectable ACTH. However, the culture medium used to maintain this carcinoma in vitro was found to contain a substance which stimulated ACTH secretion by human pituitary corticotrophs in cell culture. Radioimmunoassays and HPLC indicated that this substance had similar elution characteristics to human CRF and cross-reacted with antiserum to ovine CRF. Case 2 was found to have a lung tumour, the removal of which led to regression of her acromegalic symptoms. In culture, this tumour did not secrete GH, but did secrete a GRF. We conclude that the Cushing's syndrome and acromegaly, in cases 1 and 2, respectively, were due to ectopic secretion of CRF and GRF leading to hyperstimulation of the pituitary gland.     

Long-term Treatment of Galactorrhoea and Hypogonadism with Bromocriptine

Seventeen women and four men with galactorrhoea and associated hypogonadism have been treated with bromocriptine for 2 to 28 months. In 18 patients the gonadal status became normal as the galactorrhoea improved. The gonadally unresponsive patients had either pituitary tumours or premature menopause. Prolactin levels fell with treatment; withdrawal of the drug was associated with an increase in serum prolactin and a recurrence of the galactorrhoea and hypogonadism. Two patients tried to become pregnant on treatment and both succeeded. Raised prolactin levels appear to block the actions of the gonadotrophins at a gonadal level rather than prevent their synthesis or release; lowering prolactin secretion with bromocriptine allows resumption of normal gonadal function. Bromocriptine appears to be the treatment of choice for inappropriate lactation in association with hypogonadism on a long-term basis.     

Transforming growth factor-beta and activin inhibit basal secretion of prolactin in a pituitary monolayer culture system

Incubations of rat anterior pituitary cells with transforming growth factor (TGF)-beta 1 for 48 hr suppressed the secretion of basal prolactin (PRL) in a dose-dependent manner (ED50, 100 pg/ml). Activin, a gonadal hormone processing cysteine distribution similar to TGF beta, also suppressed basal PRL secretion, but it was less effective (ED50, 4 mg/ml). Treatment with TGF beta 1 significantly suppressed basal PRL secretion from the pituitary after 24 hr and up to 72 hr of incubation. TGF beta 1 also inhibited thyrotropin-releasing hormone-mediated PRL secretion and activin inhibited thyrotropin-releasing hormone-mediated PRL secretion slightly, but significantly. In addition, we also measured the secretion of growth hormone by cultured pituitary cells treated with TGF beta 1 or activin for 24 to 72 hr. TGF beta 1 and activin showed an opposite effect on growth hormone secretion; TGF beta stimulated and activin inhibited basal secretion of growth hormone. These results suggest that TGF beta 1 is a potent inhibitor of basal secretion of PRL by the pituitary, and both TGF beta 1 and activin play a multifunctional role in basal secretion of pituitary hormones.     

Inhibitory effects of cysteamine on neuroendocrine function

The action of cysteamine on anterior pituitary hormone secretion was studied in vivo using conscious, freely moving male rats and in vitro using anterior pituitary cells in monolayer culture. Administration of 500 micrograms cysteamine into the lateral cerebral ventricles of normal rats caused the complete inhibition of pulsatile GH secretion for a minimum of 6 h. This treatment also significantly decreased plasma concentrations of LH for at least 6 h in orchiectomized rat, TSH in short-term (0.5 month) thyroidectomized rats, and PRL in long-term (6 months) thyroidectomized rats. The in vivo stimulation of GH, LH, TSH and PRL with their respective releasing hormones 60 min after administration of cysteamine was not different from the response observed in rats pretreated with saline except for PRL where cysteamine pretreatment significantly inhibited the expected PRL increase. In vitro, 1 mM cysteamine decreased basal and TRH stimulated PRL release while not affecting basal or stimulated GH, LH, TSH and ACTH secretion. These data demonstrate the dramatic and wide-ranging effects of cysteamine on anterior pituitary hormone secretion. This action appears to be mediated through hypothalamic pathways for GH, LH and TSH and through a pituitary pathway for PRL.     

Predicting therapeutic response and degree of pituitary tumour shrinkage during treatment of acromegaly with octreotide LAR

BACKGROUND/AIMS: The efficacy of transsphenoidal surgery in the treatment of patients with acromegaly is largely dependent on tumour size. A reduction in pituitary tumour volume by medical therapy might therefore improve subsequent surgical cure rates. This study prospectively determined the effects of the depot somatostatin analogue octreotide LAR on pituitary tumour size, GH and IGF-I levels and clinical symptoms in a cohort of previously untreated patients with acromegaly. METHODS: Six patients newly diagnosed with acromegaly (mean age 53 years; range 42-76 years) received intramuscular octreotide LAR every 28 days for 6 months. The initial dose of LAR was 20 mg, but increased to 30 mg after the initial 3 injections if mean GH levels were >5 mU/l. Prior to commencing LAR therapy, each patient received 3 injections of subcutaneous octreotide (50, 100 and 200 mug) in a randomized order on separate days, and the serum GH response was measured. Pituitary tumour volume was calculated from MRI or computed tomography scans at baseline, then 3 and 6 months after initiation of treatment, and assessed by a 'blinded' radiologist in random order. At baseline, 4 patients had a macroadenoma and 2 patients had a microadenoma. For the latter, the whole gland volume was measured. RESULTS: Serum GH levels decreased from 29.6 +/- 19.2 mU/l (mean +/- SD) at baseline to 12.1 +/- 10.5 mU/l at 3 months and 10.4 +/- 9.3 mU/l at 6 months. Three patients achieved a mean serum GH level of <5 mU/l. In these patients, the serum GH had declined to <5 mU/l in response to a single 100 mug subcutaneous octreotide injection. Serum IGF-I levels decreased by a mean of 45 +/- 7.4%. Tumour volume decreased in all patients: mean baseline volume 2,175 mm(3) (range 660-6,998) decreasing to 1,567 mm(3) (range 360-4,522) at 3 months (p < 0.05) and 1,293 mm(3) (range 280-4,104) at 6 months (p < 0.002). The mean percentage decrease in size was 29% (range -54 to +4%) at 3 months (p < 0.02) and 47% (range 21-97%) at 6 months (p < 0.002). There was no statistically significant correlation between GH response and tumour shrinkage. CONCLUSIONS: A single test dose of subcutaneous octreotide may be useful in predicting the subsequent efficacy of octreotide LAR. Octreotide LAR results in significant shrinkage of pituitary tumours of newly diagnosed patients with acromegaly. Whether its administration to such patients for 6-12 months can improve the efficacy of subsequent transsphenoidal surgery will require further study.