A standardised multicentre procedure for plasma gonadotrophin radioimmunoassay

Vorozole results in greater oestrogen suppression than formestane in postmenopausal women and when added to goserelin in premenopausal women with advanced breast cancer

The high potency and selectivity of new aromatase inhibitors has translated to greater efficacy and improved tolerability in comparison with established second-line hormonal agents for advanced breast cancer in phase III clinical trials. Two pharmacological studies are reported which assess the use of one of these inhibitors, vorozole, in combination or comparison with well-established methods of oestrogen deprivation in pre and postmenopausal patients. When combined with the gonadotrophin-releasing hormone agonist (GnRHa) goserelin in 10 premenopausal patients, vorozole markedly enhanced the suppression of serum levels of oestrone, oestradiol and, oestrone sulphate beyond that achieved by goserelin alone (by a mean 74%, 83%, and 89%, respectively). The combination was well-tolerated and had no significant effects on androgen levels. Vorozole was compared with formestane in 13 postmenopausal women and serum oestrone, oestradiol, and oestrone sulphate levels were suppressed by 47%, 30%, and 70%, respectively, more by vorozole than by the steroidal aromatase inhibitor. Again the tolerability was excellent. The plasma oestrogen levels in the postmenopausal patients on vorozole were lower than in the premenopausal patients on goserelin plus vorozole, indicating that ovarian oestrogen synthesis may be relatively resistant to aromatase inhibition, even during GnRHa treatment. Thus, in both pre and postmenopausal patients substantially greater suppression of oestrogen can be achieved by vorozole compared with alternative approaches. Existing clinical-pharmacological correlates suggest that these increases in pharmacological effectiveness may result in enhanced clinical effectiveness.

Effects of quinagolide (CV 205-502), a selective D2-agonist, on vascular reactivity in patients with a prolactin-secreting adenoma

BACKGROUND

Quinagolide (CV 205 502) is a dopamine D2-receptor agonist which has proved effective in the treatment of prolactinomas, reducing both serum PRL and tumour size. Some of its D2-receptor effects are mediated via alpha-adrenoceptors, which have a major influence on the control of vascular tone. The aim of this study was to examine the influence of quinagolide on in-vivo dorsal hand vein vascular responses to noradrenaline in patients with a prolactinoma.

DESIGN AND PATIENTS

Seven female patients with prolactinomas (age 37 (28-46) years), intolerant of bromocriptine, were studied before and after 3 months treatment with quinagolide (0.75-1.5 mg/day). Patients were otherwise disease free, were taking no other medication, and had been on no other medication (including bromocriptine) for at least 3 months prior to enrollment into the study.

MEASUREMENTS

Vascular responses to locally infused noradrenaline were measured in dorsal hand veins using an established technique. PRL, oestradiol, FSH, LH, blood pressure and body mass index were also measured before and after 3 months treatment.

RESULTS

Quinagolide significantly reduced PRL in all 7 patients (1795 (696-4680) (mean (range)) vs 488 (290-868) mU/l, P = 0.001), with no effect on the other parameters, including mean arterial pressure (88 (2) vs 87 (4) mmHg, P = 0.6). Vascular reactivity to noradrenaline was significantly increased after 3 months therapy: log10 dose estimated to cause 50% vasoconstriction (ED50) 1.37 (0.12) vs 0.85 (0.12) ng/min (P = 0.003; a lower ED50 indicates less noradrenaline is required to constrict the vein by 50%).

CONCLUSIONS

Vasoconstrictor responses to noradrenaline were increased in all patients after 3 months treatment with quinagolide. Peripheral veins carry alpha-adrenoceptors analogous to those of systemic resistance vessels. If this increased vasoconstrictor response in patients with prolactinomas was occurring in hypophyseal vessels, it would lead to reduced tumour blood supply. Quinagolide may therefore reduce tumour blood flow, which may be one factor responsible for its effectiveness in these patients.

Isolated adrenocorticotropin deficiency presenting as primary infertility

A 31 year old female presented with primary infertility and gave a two year history of amenorrhea without symptoms or signs of endocrine dysfunction. Examination was normal and investigation showed low oestradiol and progesterone levels with decreased LH pulsatility. The cortisol responses were impaired following hypoglycaemic stress and a short synacthen test, but the cortisol response to a prolonged synacthen test was normal. An inadequate ACTH response to CRF testing confirmed the diagnosis of isolated ACTH deficiency. Hydrocortisone therapy was followed by an ovulatory menstrual cycle. Amenorrhea again ensued following the reduction of the steroid dose and normal menses resumed on normal steroid replacement therapy. Six hourly gonadotrophin pulsatility showed a significant increase in both pulse amplitude and mean LH and FSH levels following steroid treatment. Isolated ACTH deficiency is a rare but treatable cause of hypogonadism and infertility, and this case gives further insight on the role of cortisol on the hypothalamo-pituitary gonadal axis.

Inhibin as a marker for ovarian cancer

Inhibin is a polypeptide hormone produced by the granulosa cells of the ovary, and is present in body fluids as dimers of various sizes each comprising an alpha- and beta-subunit. Free forms of the alpha-subunit also circulate, and the presently available radioimmunoassay (Monash assay) cannot distinguish these from biologically active dimeric inhibin. Recently we described a new two-site enzyme immunoassay able for the first time to measure the levels of dimeric inhibin throughout the human menstrual cycle. The sensitivity limit of this assay is 2 pg ml-1 in human serum with cross-reactivity against activin of 0.05%. The normal range of inhibin in post-menopausal women is < 5 pg ml-1, in pre-menopausal women 2-80 pg ml-1 (2-10 pg ml-1 in the follicular phase, 40-80 pg ml-1 in the luteal phase). This assay was used to determine inhibin levels in sera from 15 (five pre-menopausal and ten post-menopausal) patients with granulosa cell tumours of the ovary. It was raised in a pre-menopausal patient preoperatively (261 pg ml-1), in six post-menopausal patients (32, 43, 54, 66, 24 and 58 pg ml-1) and one pre-menopausal patient with recurrent tumour, (237 pg ml-1), all confirmed clinically. Inhibin was normal in six patients in remission. Oestradiol levels were normal in all patients. Serial levels of inhibin predicted recurrence before overt clinical relapse in two patients. In 29 patients with malignant epithelial ovarian tumours inhibin levels were modestly elevated in nine and normal in the rest. Three patients with endometrioid histology, two with undifferentiated tumours, three with mucinous adenocarcinoma and one with clear cell carcinoma had elevated inhibin levels. Functional inhibin is secreted by all granulosa cell tumours of the ovary studied and can be used as a tumour marker to determine response to therapy and predict recurrence and is superior to oestradiol. A more detailed analysis of the levels of inhibin, and its subunits in epithelial ovarian cancer is needed to identify the molecular forms of the immunoreactive material before optimised assays can be applied to this more common tumour.

Age related changes in the pituitary-testicular axis in normal men; lower serum testosterone results from decreased bioactive LH drive

OBJECTIVE

The mechanism underlying the slight hypoandrogenism associated with normal ageing remains unclear. We have therefore examined age related changes in the activity of the pituitary-testicular axis in healthy normal males.

DESIGN

Random blood samples were obtained from 219 normal men whose ages ranged from 20 to 79 years. At the time of the study, none of the men had received treatment or had ever had any endocrine dysfunction diagnosed.

MEASUREMENTS

Luteinizing hormone was measured in the subjects' plasma using a commercially available immunoradiometric assay (IRMA, Serono Maiaclone) and a fully validated in-vitro bioassay. Testosterone and FSH were measured using standard radioimmunoassays (RIA) whilst sex hormones binding globulin was assayed by an IRMA.

RESULTS

Levels of total testosterone (total-T) and bioactive LH fell with age (r = -0.231 and -0.189 respectively) by 5.9 nmol/l and 2.3 IU/l respectively between grouped patients aged 20-39 years (Group A) and 60-79 years (Group C). In contrast, immunoreactive LH remained unchanged. The LH B:I ratio also fell with age (P < 0.0001) being 5.0 +/- 0.3 (group A) and 3.3 +/- 0.2 (group C), representing a fall of 33%. Since immunoreactivity remained constant, this fall primarily represented the decline in LH bioactivity. Bioactive, but not immunoreactive LH correlated to total-T (P = 0.009, n = 209) and the total-T:LH ratios fell by over 30% between groups A and C using the IRMA, but remained unchanged by bioassay.

CONCLUSIONS

There is an underlying decline in both total-testosterone and free-testosterone index, and bioactive LH levels with advancing age, suggestive of a hypothalamo-pituitary defect which leads to lower bioactive LH levels which in turn are responsible for the diminished gonadal steroidogenesis. Elucidation of the mechanism underlying this slight decline in hypothalamopituitary testicular activity is complicated by differences between the data obtained by immunoassay or bioassay.

Testosterone levels during systemic and inhaled corticosteroid therapy

Testosterone has importance both as a sex hormone and as an anabolic steroid promoting bone formation. Osteoporosis is associated with both hypogonadism and corticosteroid therapy. Testosterone levels are reduced by long term prednisolone treatment. Although high dose inhaled corticosteroid therapy may cause a variety of systemic effects including adrenal suppression, dermal thinning and a reduction in total bone calcium, its effect on testosterone levels is not known. Testosterone, luteinizing hormone, follicle stimulating hormone and sex hormone binding globulin were therefore measured in 35 male patients with respiratory disease attending an outpatient clinic (median age 58, range 21-75 years). They were grouped according to steroid therapy and compared with 19 age matched controls. Mean (SD) testosterone levels were 33% lower in 12 men on long term oral prednisolone [14.5 (6.0) nmol 1-1] than in controls [21.7 (6.3) nmol 1-1], but were not significantly reduced in 10 patients on low dose inhaled beclomethasone [200-800 micrograms day-1: 19.7 (3.7)] nor in 13 men taking high dose inhaled beclomethasone [1500-2,250 micrograms day-1: 17.9 (5.6)]. Levels of luteinizing hormone, follicle stimulating hormone and sex hormone binding globulin were similar in all four groups. These cross sectional data confirm that long term systemic corticosteroid therapy reduces testosterone levels. However, testosterone was reduced by only 18% (NS) by long term inhaled corticosteroids. Other mechanisms to explain the disordered bone metabolism should now be explored.

Endocrine changes with the aromatase inhibitor fadrozole hydrochloride in breast cancer

Fadrozole hydrochloride is a potent aromatase inhibitor with proven clinical effectiveness. However, its optimal dose and its effects on serum aldosterone levels/electrolyte balance have been disputed. To resolve these issues, a double-blind randomised endocrine study of three doses of fadrozole hydrochloride [0.5 mg twice daily (bd); 1.0 mg bd; 2.0 mg bd] was conducted in 80 (68 evaluable) postmenopausal patients with advanced breast cancer over a period of 3 months. There were substantial falls in the serum levels of oestradiol, oestrone and oestrone sulphate. For oestrone only, there was a significant effect of dose (on-treatment means: 0.5 mg, 38.0 pmol/l; 1.0 mg, 25.0 pmol/l; 2.0 mg, 23.9 pmol/l). All oestrogens showed a similar pattern in relation to time, with the 3-month mean being higher than those at 1 and 2 months, and this was significant for oestradiol (P = 0.012). There was an indication that complete suppression of oestradiol and oestrone was not maintained throughout the 12-h dosing period, but the data and its interpretation are complicated by a minor diurnal rhythm in these parameters. There were significant increases in 17-hydroxyprogesterone and androstenedione which may be due to a block of 11 beta-hydroxylase. There was a statistically non-significant fall in aldosterone levels (P = 0.06) during treatment (median pretreatment, 446 pmol/l; median decrease, 125 pmol/l). However, the concurrent significant fall in the plasma sodium: potassium ratio indicated that changes in aldosterone secretion did occur. None of these effects on adrenal pathways was of a degree which is likely to have clinically relevant consequences. It is concluded that fadrozole hydrochloride achieves near maximal suppression of oestrogens at 1 mg bd, and that its effects on aldosterone synthesis are unlikely to be of clinical significance.

Elevated serum immunoreactive inhibin levels in peripubertal boys with chronic renal failure. Cooperative Study Group on Pubertal Development in Chronic Renal Failure (CSPCRF)

OBJECTIVE

Boys with chronic renal failure have delayed progress through puberty and have raised gonadotrophin and low testosterone levels indicative of disturbed hypothalamo-pituitary-testicular function. Most studies into the mechanisms underlying the dysfunction have concentrated on the LH-Leydig cell interaction. However, it is now possible to probe the FSH-Sertoli cell axis by measuring plasma immunoreactive inhibin, which is a marker of Sertoli cell function. This study investigated the FSH-Sertoli cell (immunoreactive inhibin) axis in boys with chronic renal failure on conservative and dialysis treatment as they progressed through puberty. The effect of renal transplantation in chronic renal failure was also investigated.

DESIGN

Blood was drawn at 15-minute intervals between 2000 and 0700 h from 51 boys with chronic renal failure at various stages of puberty. The samples were divided into two pools, corresponding to the hormone secretion in the first and second part of the night. Single blood samples were drawn from a group of normal boys between 0800 and 1000 h.

PATIENTS

A total of 37 normal boys and 51 boys with chronic renal failure were examined immediately before and during puberty. Of a total of 80 pulse profiles taken in chronic renal failure, 36 were from transplanted and 44 from non-transplanted uraemic subjects.

MEASUREMENTS

Immunoreactive inhibin, FSH and testosterone were measured using standard radioimmunoassays. The subjects were pooled into pubertal stages I, II/III and IV/V for analysis of hormone data.

RESULTS

Early morning levels of immunoreactive inhibin like molecules (i-Inh) rose steadily with pubertal progression for all subject groups, those for boys with chronic renal failure being significantly elevated over normal boys from pubertal stage II/III onwards. Uraemic boys had higher levels than those who had been transplanted at all pubertal stages (P < 0.05). Early morning levels of FSH were significantly higher in uraemic patients with pubertal stages IV/V compared to our normal boys. There were no differences in i-Inh levels in plasma pooled from the samples taken between 2000 and 0115 h and 0130 and 0700 h for either treatment group at any stage of puberty. Testosterone levels rose in the second part of the profile from pubertal stages II/III onwards for both treatment groups. The proportional increase of testosterone was lower by mid puberty in uraemic than in transplanted children (percentage increases of 92 +/- 29 and 569 +/- 190 respectively, mean +/- SEM). i-Inh failed to correlate with FSH at any Tanner stage or for any subject group.

CONCLUSION

Peripubertal boys with chronic renal failure have highly elevated serum immunoreactive inhibin and FSH levels which are partially reduced by renal transplantation. There was no evidence of any relationship between i-Inh and FSH secretion in either normal boys or in uraemic or transplanted boys with the exception of a positive correlation in late pubertal patients after transplantation. Finally, despite problems associated with the current immunoassay for inhibin, this assay may still prove to be a useful marker of Sertoli cell function in testicular pathology.

Hormonal treatment of pancreatic carcinoma: a phase II study of LHRH agonist goserelin plus hydrocortisone

Eighteen consecutive patients with measurable locally advanced or metastatic pancreatic adenocarcinoma were treated with goserelin (Zoladex) 3.6 mg subcutaneously every 4 weeks. Hydrocortisone 20 milligrams twice daily was commenced with the second injection of goserelin. Objective tumour response was monitored by computerised tomography of the abdomen. There was no objective remission in disease sites. Serial measurements of serum tumour markers showed no reduction in serum CA 19-9 and CA 195 concentrations. The median duration of survival of all cases was 5 months. Administration of goserelin resulted in significant reductions in oestradiol, testosterone, androstenedione in males and reductions in FSH and LH in both males and females. The addition of hydrocortisone resulted in further reductions of androstenedione and testosterone levels in males. Thus goserelin showed no anti-tumour effect, but concentrations required for direct inhibitory effects may be higher than those required to produce effects on hormone suppression.

Human chorionic gonadotrophin contributes to the bioactivity of Pergonal

OBJECTIVE

We examined batch variation in the LH-like bioactive content of Pergonal and determined whether hCG contributes to this.

DESIGN

Random selection of three batches of Pergonal, consisting of three ampoules in each batch.

MEASUREMENTS

The LH content in each ampoule was determined by radioimmunoassay (R-LH), immunoradiometric assay (I-LH) and in vitro Leydig cell bioassay (B-LH) using the urinary hMG International Standard 70/45. Human chorionic gonadotrophin was determined by immunoradiometric assay (I-hCG) using the hCG IRP 75/537. The isohormone content of each batch was examined by chromatofocussing over the range pH 4.5-7.0 and the fractions collected were assayed for LH and hCG content. The variability in potency between batches was assessed by single factor analysis of variance.

RESULTS

The gonadotrophin content of each batch (IU/ampoule, mean +/- SEM, n = 3 ampoules) was R-LH (40.9 +/- 0.5, 40.8 +/- 0.2, 39.3 +/- 0.7, P > 0.15), I-LH (39.0 +/- 1.5, 28.3 +/- 0.8, 36.9 +/- 3.3, P < 0.001), B-LH (27.3 +/- 0.3, 12.0 +/- 0.9, 19.3 +/- 0.9, P < 0.001) and I-hCG (16.4 +/- 0.7, 11.7 +/- 0.2, 10.5 +/- 0.5, P < 0.001). The chromatofocussing recoveries below pH 5.5 expressed as a percentage of the total amount of analyte eluted from the column and collated for all three batches of Pergonal were (mean % +/- SD, n = 3 batches) R-LH (58.4 +/- 4.0), I-LH (41.3 +/- 7.5), B-LH (81.4 +/- 2.8) and I-hCG (87.8 +/- 3.7).

CONCLUSIONS

There was significant batch variation in the I-LH, B-LH and I-hCG (P < 0.001) but not the R-LH (P > 0.15) content of Pergonal. More than 80% of the total B-LH recovery chromatofocussed below pH 5.5 and corresponded to the region of highest I-hCG (> 87%) and lowest I-LH (< 42%) recovery. This was highly suggestive of hCG contributing to the LH-like bioactivity of Pergonal.

Pituitary-ovarian function in women with minimal or mild endometriosis and otherwise unexplained infertility

OBJECTIVE

To determine whether abnormalities of pituitary-ovarian function are associated with minor degrees of endometriosis in women with otherwise unexplained infertility.

DESIGN

Comparison of pituitary-ovarian function in a group of 22 patients with that in a control group of 10 healthy fertile women.

PATIENTS

Twenty-two women with minimal or mild endometriosis and otherwise unexplained infertility.

MEASUREMENTS

Ovarian ultrasound and radioimmunoassay of pituitary and ovarian hormones.

RESULTS

Abnormalities, including luteinization of unruptured follicles, broad LH surges and low concentrations of progesterone during the luteal phase, were demonstrated in 82% of study cycles.

CONCLUSION

Pituitary-ovarian dysfunction appears to be associated with minimal or mild endometriosis in women with otherwise unexplained infertility but the diversity of endocrine abnormalities demonstrated suggests that endometriosis is unlikely to be an aetiological factor in their development.

An endocrine and pharmacokinetic study of four oral doses of formestane in postmenopausal breast cancer patients

43 postmenopausal breast cancer patients were treated orally with the aromatase inhibitor formestane (4-hydroxyandrostenedione) at daily doses of 62.5, 125, 250 or 500 mg for 4 weeks followed by 250 mg daily for a further 4 weeks. For some patients, 62.5 mg did not suppress serum oestradiol levels maximally. The doses of 250 and 500 mg did not differ in their effectiveness. Oestrone levels were suppressed by all doses of formestane but no consistent changes of aldosterone, cortisol or 17-hydroxyprogesterone occurred. Serum levels of sex hormone binding globulin fell by about 15% during treatment with 250 mg formestane reflecting its minor androgenic activity. The maximum concentration and area under the curve of serum formestane levels after the first dose varied in an approximately linear manner with dose. It is concluded that formestane is an effective, specific suppressant of oestradiol levels via the oral route requiring no more than 250 mg to be given daily.

Endocrine, pharmacokinetic and clinical studies of the aromatase inhibitor 3-ethyl-3-(4-pyridyl)piperidine-2,6-dione ('pyridoglutethimide') in postmenopausal breast cancer patients

The aromatase inhibitor, 'pyridoglutethimide' (PyG), has been shown previously to suppress serum oestrogen levels in postmenopausal breast cancer patients and to achieve clinical responses at a dose of 500 mg twice daily (b.d.). This report gives the results of a detailed pharmacokinetic and endocrine study of PyG in ten patients. Four doses were tested at intervals of 2 weeks in the following order: 200 mg b.d., 400 mg b.d., 800 mg b.d., 1200 mg b.d. Concentration-time profiles of serum levels of PyG were curvilinear in all patients probably reflecting a saturation of metabolic enzymes. During repeat-dosing metabolism was enhanced approximately 2-fold. Plasma levels of oestradiol were significantly suppressed by the lowest dose of PyG. Although higher doses appeared to achieve greater suppression this was not statistically significant in this small group of patients. There were no significant effects at any dose on the serum levels of cortisol, aldosterone, luteinising hormone, follicle stimulating hormone, prolactin, sex hormone binding globulin or thyroid stimulating hormone. There was a dose-related increase in 17 alpha-hydroxyprogesterone levels and a dose-related decrease in levels of dehydroepiandrosterone sulphate (DHAS). The androgens DHA, testosterone and androstenedione also were significantly suppressed with at least one of the doses of PyG. Synacthen tests did not support these changes being a result of inhibition of 17,20 lyase. It is possible that they are due to enhanced clearance of DHAS. Two patients experienced no toxicity throughout the study, whilst a total of four patients were withdrawn because of side-effects: one at 400 mg b.d., two at 800 mg b.d., and one at 1200 mg b.d. The most frequent side-effects were nausea and lethargy. One patient showed an objective response to treatment.

Normal postnatal androgen production and action in isolated micropenis and isolated hypospadias

To try and find out if a defect in androgen biosynthesis or action could be responsible for the incomplete virilisation seen in boys with isolated hypospadias and isolated micropenis, androgen receptor binding was studied in genital skin fibroblasts established from 18 boys with isolated micropenis and 19 boys with isolated hypospadias. The production of gonadotrophins and testosterone was also measured in the boys with micropenis. There was no evidence of gonadotrophin deficiency, or of a defect in testosterone biosynthesis in the boys with micropenis, and there was no evidence of a quantitative or qualitative defect of androgen binding in either group. These isolated abnormalities may be the result of transient defects in androgen synthesis or action, or both, during a critical phase of embryogenesis.

The effect of danazol on pulsatile gonadotropin secretion in women with endometriosis

OBJECTIVE

To determine whether the effect of danazol on gonadotropin pulsatility was due to a direct effect of the drug or the suppression in the estradiol (E2) levels.

DESIGN

Prospective analysis of serial blood samples at 10-minute intervals for 5 hours on days 3, 4, or 5 of the control cycle and 2 months after the start of danazol treatment.

SETTING

Tertiary institutional outpatient care.

PATIENTS, PARTICIPANTS

Six eugonadal women with minimal endometriosis.

INTERVENTIONS

Oral administration of danazol, 200 mg three times daily.

MAIN OUTCOME MEASURES

Gonadotropin mean levels, pulse frequency, and amplitude.

RESULTS

The mean level of E2 was the same in the control cycle as that during danazol treatment (170 pmol/L). Danazol administration resulted in a 16% increase in the mean luteinizing hormone (LH) pulse amplitude (95% confidence interval [CI] 6% to 26%, P less than 0.01), associated with a 20% decrease in LH pulse frequency (95% CI -71% to +31%, P = 0.37). There was a nonsignificant increase in follicle-stimulating hormone (FSH) pulse amplitude (2%, 95% CI -9% to +14%, P = 0.68) and in FSH pulse frequency (27%, 95% CI -4% to +58%, P = 0.08). The 22% decrease in the mean LH level (95% CI -85% to +12%, P = 0.13) and the 20% decrease in the mean FSH level (95% CI -53% to +37%, P = 0.33) were also not significant.

CONCLUSIONS

The increase in LH pulse amplitude represents a direct effect of danazol on the hypothalamic-pituitary axis.

A dose-comparative endocrine-clinical study of leuprorelin in premenopausal breast cancer patients

Twelve premenopausal patients with advanced breast cancer were randomised to receive 3.75 or 7.5 mg of a slow release formulation of the luteinising hormone releasing hormone agonist leuprorelin once every 4 weeks. All patients were oestrogen receptor positive or unknown. Serum levels of gonadotrophins and oestrogens were suppressed markedly by both doses. All oestrogen values during treatment were within the postmenopausal range except for a single oestradiol level (274 pmol l-1) in one patient on the lower dose. There was no other indication that this lower dose was less effective as an oestrogen suppressant. There were two objective responders to the 3.75 mg dose and three to the 7.5 mg dose. Toxicity was confined almost entirely to hot flushes which occurred in 11/12 patients. We conclude that the slow release formulation of leuprorelin is effective in breast cancer treatment and that there is no major detriment to the use of the 3.75 rather than 7.5 mg dose.

The effect of danazol and the LHRH agonist analogue goserelin (Zoladex) on the biological activity of luteinizing hormone in women with endometriosis

In an attempt to determine whether the suppression in oestradiol levels caused by danazol is due to an effect on the hypothalamic-pituitary axis, we compared the endocrine effects of danazol with those of the LHRH (GnRH) agonist analogue goserelin. Serum levels of immunoreactive LH (I-LH), FSH, 17 beta-oestradiol (E2) and bioactive LH (B-LH) (using a mouse Leydig cell bioassay), were measured in ten and 20 women with endometriosis treated with danazol and goserelin, respectively. I-LH was measured both by radioimmunoassay (RIA) and immunoradiometric assay (IRMA). During 6 months of treatment with 600 mg of danazol daily, mean serum E2 decreased (P less than 0.05) to levels near the upper limit of the post-menopausal range (to a mean (and 95% confidence interval of the mean) of 117 (65-169) pmol/l) whereas FSH, I-LH (both by RIA and IRMA) and B-LH levels were not significantly altered. During 6 months of treatment with monthly depot injections of 3.6 mg goserelin, mean serum E2 decreased (P less than 0.001) to well within the post-menopausal range (to 23 (18-28) pmol/l). The mean FSH, I-LH and B-LH levels also decreased (P less than 0.05) during therapy with goserelin (from 3.9 (3.1-4.7) to 2.0 (1.6-2.4) IU/l for FSH, from 5.3 (4.5-6.1) to 1.9 (1.7-2.1) IU/l for RIA-LH, from 2.9 (2.5-3.3) to less than 0.5 (less than 0.5) IU/l for IRMA-LH and from 9.1 (7.1-11.1) to 2.9 (2.6-3.2) IU/l for B-LH).(ABSTRACT TRUNCATED AT 250 WORDS)

The clinical and endocrine effects of 4-hydroxyandrostenedione alone and in combination with goserelin in premenopausal women with advanced breast cancer

The aromatase inhibitor, 4-hydroxyandrostenedione (4OHA) is an effective treatment for advanced post-menopausal breast cancer. The clinical and endocrine effects of 4OHA treatment were studied in five pre- and perimenopausal women with metastatic breast cancer. Serum oestradiol levels were not significantly reduced as a result of treatment with 500 mg of 4OHA by weekly i.m. injections and no patient had a tumour response. Four patients were subsequently treated with the luteinising hormone releasing hormone (LHRH) analogue, gosereline, and three had objective responses. The endocrine effects of combined treatment with goserelin (Zoladex), and 4OHA were studied in a further five premenopausal women. Serum oestradiol levels after treatment with goserelin alone were typical of post-menopausal women. Addition of 4OHA led to a further suppression of oestradiol to within the range observed in post-menopausal patients treated with further suppression of oestradiol to within the range observed in post-menopausal patients treated with 4OHA. Six patients whose tumours had regressed as a result of goserelin treatment and who subsequently relapsed were then given combined treatment. Four of the six experienced a second remission. We conclude that while 4OHA alone is unlikely to be a satisfactory treatment for premenopausal patients with advanced breast cancer, 4OHA in conjunction with goserelin leads to profound suppression of oestradiol. The combination of LHRH analogue and aromatase inhibitor may prove to be a superior treatment to LHRH analogue alone in these patients.

Potency and selectivity of the non-steroidal aromatase inhibitor CGS 16949A in postmenopausal breast cancer patients

A selective inhibitor of aromatase is widely sought for the treatment of postmenopausal women with breast cancer. CGS 16949A has been shown to be a highly selective, potent inhibitor of aromatase in vitro. Its potency as an oestrogen suppressant and its selectivity were examined by treating 24 postmenopausal patients with advanced breast cancer for 4 weeks with doses of 0.3, 1.0 and 2.0 mg twice daily. The study was conducted in two parts which compared the two lower doses and the two higher doses separately in a cross-over design protocol. All doses significantly suppressed serum oestradiol and oestrone levels below pretreatment levels. Cross-over analysis indicated that the 2.0 mg twice daily dose achieved significantly greater suppression of oestradiol levels than 0.1 mg twice daily but there was no significant differences between any of the doses in the suppression of oestrone. No significant effects were noted on serum levels of LH, FSH, SHBG, prolactin, testosterone, androstenedione, 17-hydroxyprogesterone or cortisol. For the four steroids this was true both for basal samples and those collected after Synacthen stimulation. However, serum aldosterone levels were significantly suppressed by 1.0 mg twice daily CGS 16949A and further suppressed by 2.0 mg twice daily. It is concluded that CGS 16949A is a potent oestrogen suppressant in postmenopausal patients but that its effect is not totally selective.

Rapid pulses of luteinizing hormone during the ovarian cycle

Luteinizing hormone has been shown to be secreted in a complex pattern of low-amplitude, high-frequency pulses superimposed on high-amplitude, low-frequency pulses. To test the hypothesis that the rapid, small pulses vary during the ovarian cycle, 16 normal women were studied on two occasions each, in the early and late follicular phases of one cycle or the luteal and following early follicular phase. Blood samples were taken every 1 min for 1 h. Statistical methods of time series analysis were applied to the data. There were significant amplitude differences between subjects at all stages of the cycle. We found no evidence to support the hypothesis that the frequency of the small rapid pulses varied during folliculogenesis. The results for the luteal phase, while less uniform, showed no consistent change. It is suggested that the rapid pulses may be related to function of the hypothalamic-pituitary axis rather than direct ovarian control.

Phase I/II study of the anti-oestrogen zindoxifene (D16726) in the treatment of advanced breast cancer. A Cancer Research Campaign Phase I/II Clinical Trials Committee study

We report a phase I/II study of the indole derivative, zindoxifene, an anti-oestrogen with intrinsic oestrogenic activity. We have treated 28 women with advanced breast cancer of whom 26 had received prior endocrine therapy. Oral zindoxifene doses ranged from 10 to 100 mg daily; doses were escalated in some patients. Twenty-five patients were assessed for response; the remaining three patients completed less than 3 weeks of treatment. There were no objective responses; disease stabilised in seven patients for up to 5 months and progressed in the remaining 18. Five patients (including three treated with tamoxifen) responded to subsequent endocrine therapy. Nausea, which was dose-limiting, affected half of the patients treated with 80 mg daily. Metabolites of zindoxifene were detectable in serum at all doses used, and sex hormone binding globulin (SHBG) levels showed a strong tendency to rise at the higher doses, indicating that zindoxifene is absorbed and has biological activity. We conclude that zindoxifene in the doses used in this study has only marginal therapeutic activity in the treatment of advanced breast cancer.

Extending the duration of the pill-free interval during combined oral contraception

The effect of increasing the duration of the pill-free interval from 7 to 9 and 11 days during the first 4 months of combined oral contraceptive therapy was studied by hormonal, ovarian follicle and mucus changes. Increasing the pill-free interval allowed more follicular development and more estradiol production. In some cases, pre-ovulatory values were achieved although ovulation did not occur. No differences could be detected between different types of combined oral contraceptive, including monophasic, triphasic, levonorgestrel-containing and gestodene-containing. The variation between individuals in the group of 28 was far greater than any variation in dose.

Comparison of the pharmacokinetics and pharmacodynamics of unformulated and formulated 4-hydroxyandrostenedione taken orally by healthy men

A study of the aromatase inhibitor 4-hydroxy-androstenedione (4-OHA) was conducted in normal healthy men to compare the oral administration of two preparations of the drug: an unformulated, micronized powder and a formulated microcrystalline material (CGP 32349). The formulated material achieved a significantly higher mean peak concentration (88% greater than that obtained using the unformulated powder) and a higher mean AUC (not significant). The median time to peak was 1.5 h for both preparations and the elimination rate constants were similar (0.31 for micronized 4-OHA and 0.36 h-1 for formulated 4-OHA). Plasma concentrations of 4-OHA in this group were markedly lower than those previously observed in postmenopausal breast cancer patients. Significant biological activity was demonstrated with the formulated material in its suppression of plasma oestradiol levels, whereas no significant suppression was obtained using the micronized powder. An increase in androgen levels was observed that may have been due to competitive inhibition of enzymes involved in metabolic clearance of androgens and/or to decreased feedback inhibition of gonadotrophin secretion by oestradiol.

Testicular enlargement and elevated serum inhibin concentrations occur in patients with pituitary macroadenomas secreting follicle stimulating hormone

We studied four male patients with pituitary macroadenomas. Before treatment all had high serum FSH concentrations, but LH and testosterone were normal or subnormal; all patients were found to have large testes. All had had normal sexual function, and three patients had fathered children. After pituitary surgery there were decreases in serum gonadotrophins and testosterone, which were accompanied by decreases in testicular volumes. hCG stimulation tests in two patients showed normal responses of testosterone and oestradiol, confirming normal Leydig cell function. Inhibin levels were increased in two patients studied when FSH levels were high, suggesting a defect in gonadal-pituitary feedback control. Later, as FSH concentrations decreased to normal, so did inhibin levels. Histology showed that increased testicular size was due to increased lengths of seminiferous tubules. The association of pituitary macroadenomas, large testes and increased serum inhibin has not been reported previously. Assessment of testicular size in patients with raised serum FSH is important, since enlarged testes suggest the likely pathogenesis is that of a pituitary gonadotrophinoma, rather than primary gonadal failure. Increased inhibin levels may then confirm this, and be a biochemical marker for these tumours.

Use of luteinising hormone-releasing hormone agonist (leuprorelin) in advanced post-menopausal breast cancer: clinical and endocrine effects

Fifteen post-menopausal patients with advanced breast cancer were treated with the LH-RH agonist leuprorelin (D-leu6-des-gly10-Gn-RH-ethylamide) given in a dosage of 7.5 mg as a monthly subcutaneous depot injection, to assess the clinical activity and endocrine response to treatment. None of the 15 patients showed an objective response to treatment, although four patients had stable disease for at least 6 months. No toxicity was demonstrated. Endocrine effects after 4 weeks' treatment were as follows: mean levels of serum gonadotrophins fell to 10% of their pretreatment values; there were no significant changes in the levels of prolactin on treatment; there was a significant decrease in the levels of serum testosterone in 12 out of 14 patients; there were no significant changes in the levels of oestradiol, androstenedione and oestrone. The lowering of serum testosterone suggests that androgens in post-menopausal women may be partly produced by the ovaries, stimulated by LH and FSH. This fall in testosterone may explain why some post-menopausal breast cancer patients in other studies have been reported to respond to treatment with LH-RH agonists, as it would decrease the substrate for the peripheral synthesis of oestrogens.

Programmed gamete intrafallopian transfer (GIFT)

Thirty-three patients with unexplained infertility underwent a total of 42 programmed superovulation cycles in a gamete intrafallopian transfer (GIFT) programme. The date of oocyte retrieval was decided in advance and the cycle preceding oocyte collection was modified with norethisterone from mid-cycle until 14 days before the scheduled laparoscopy. This was followed by a fixed schedule superovulation regimen. Serum oestradiol, progesterone and luteinizing hormone were monitored and the data analysed retrospectively. A single ultrasound scan was performed on the day of laparoscopy to exclude ovulation. Thirty-eight GIFT procedures were performed, resulting in 11 (29%) clinical pregnancies of which four twin and four singleton pregnancies are continuing. There was a significant correlation between the oestradiol response pattern and the maturity of oocytes retrieved, the fertilization rate of supernumerary oocytes and the pregnancy rate. Programmed cycles may be conveniently combined with GIFT, and basic endocrinological monitoring can be used to identify cycles with a poor prognosis before laparoscopy.

Effects of 180 micrograms and 250 micrograms norgestimate on pituitary-ovarian function and cervical mucus

A double-blind cross-over study of 16 healthy women was carried out to evaluate the effects of norgestimate, a new progestogen, on pituitary ovarian function and cervical mucus. Treatment from cycle day 7 to 16 with 180 micrograms and 250 micrograms norgestimate led to suppression of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) levels, while progesterone (P) remained suppressed in 28 of 32 cycles studied, indicating inhibition of ovulation. Lack of ovulation in 30 cycles was associated in all but 3 with functional follicles, 2 of which luteinized. With 250 micrograms norgestimate, estradiol (E2) concentrations reached levels similar to those achieved during control follicular phase, but significantly higher concentrations of E2 were achieved with 180 micrograms norgestimate. Cervical mucus score was significantly depressed in all but 5 cycles (3 norgestimate 180 micrograms and 2 norgestimate 250 micrograms cycles). In conclusion, both dosages of norgestimate show good suppression of the hypothalamic-pituitary axis and cervical mucus.

A randomized comparative study of purified follicle stimulating hormone and human menopausal gonadotropin after pituitary desensitization with Buserelin for superovulation and in vitro fertilization

Twenty patients entered a randomized, crossover study of purified follicle-stimulating hormone (pure-FSH) or human menopausal gonadotropin (hMG) superovulation, 2 ampules per day after pituitary desensitization with the luteinizing hormone-releasing hormone (LH-RH) analogue Buserelin (D-Ser tBu6 LH-RH 1-9 ethylamide) nasal spray. There were no cycles cancelled. Six patients conceived (five on pure-FSH, one on hMG). There were 24.2 +/- 2.5 (mean +/- standard error of the mean [SEM]) ampules of pure-FSH and 24.3 +/- 3.6 ampules of hMG stimulation required. There were similar numbers of preoperation follicles: 6.9 +/- 1.0 on hMG and 6.6 +/- 1.1 on pure-FSH, of oocytes collected; 8.5 +/- 1.4 on hMG and 5.8 +/- 1.4 on pure-FSH, and of pre-embryos achieved; 5.1 +/- 0.9 on hMG and 3.4 +/- 1.0 on pure-FSH; on either treatment. The fertilization rate on hMG was 60% and on pure-FSH was 55%. Pre-embryo transfer rates were 3.2 +/- 0.3 in the hMG group and 2.7 +/- 0.4 in the pure-FSH group. There were no differences in serum FSH, LH, estradiol, or progesterone levels between the hMG and pure-FSH groups. Mean +/- SEM luteal phase length was 10.6 +/- 0.4 days in the nonpregnant cycles.

Ovarian follicular development in oral contraceptive cycles

The cyclic changes in serum hormone concentrations, ovarian follicular development, uterine size, and endometrial appearance have been observed in 58 spontaneous ovulatory cycles and compared with 22 initial cycles of oral contraceptive (OC) therapy. A far greater inhibitory effect on the ovary was achieved by starting contraception on day 1 of the cycle rather than on day 5. The relevance of these findings with regard to OC failures is discussed.

Assays for follicle stimulating hormone and luteinising hormone: guidelines for the provision of a clinical biochemistry service

The measurement of serum follicle stimulating hormone (FSH) and luteinising hormone (LH), together with the appropriate sex steroid, is of great value in the investigation of delayed and precocious puberty, hypogonadism, subfertility, polycystic ovarian disease and hypothalamic-pituitary disorders. Dynamic function testing of the hypothalamic-pituitary-gonadal axis should be restricted to a few defined situations. Sequential LH measurements, either in serum or in urine, may be used to time ovulation during artificial insemination or in vitro fertilisation programmes. No special precautions are necessary when sampling for FSH and LH measurement; serum is preferred to plasma and should be stored frozen before assay. Aliquots of timed urine specimens of known volume should be stored frozen without preservative. Gonadotrophin results should be available within 2-3 weeks; laboratories unable to meet this schedule are advised to send their samples to a Regional Centre for assay. Reagents for the radioimmunoassay of FSH and LH are readily available, and standard techniques have been developed for their use. Laboratories using 'in-house' methods should pay particular attention to the matrix used for preparing standard solutions, the purification of radioligands and the optimisation of the separation system. Low cost matched reagents of proven performance are available in kit form from the Chelsea Hospital for Women; several commercial kits are also available, although few are widely used in the UK. The overall performance of laboratories in the UK External Quality Assessment Scheme (EQAS) for FSH and LH has remained steady for several years. Of the 130 participants, only about 15% in each scheme have 'good' performance (cumulative bias less than 10%, plus cumulative variability of bias less than 10%), whilst a similar proportion have 'unacceptable' performance (cumulative bias greater than 20% and/or cumulative variability of bias greater than 25%). The remaining 70% of laboratories have 'adequate' performance but are at risk of producing results that are clinically misleading. Within any one method group, the performance of FSH and LH assays are closely related. Optimal assay performance depends upon sensible laboratory management to ensure skilled operators, a regular programme of reagent/kit renewal, comprehensive internal and external quality assessment, and attention to detail in all aspects of gonadotrophin assay. The working range of each individual assay should be defined and no absolute result reported from outside this range.(ABSTRACT TRUNCATED AT 400 WORDS)

Hypergonadotropic amenorrhea--etiology and outcome in 93 young women

Ninety-three cases of amenorrheic women aged 40 years or less in whom serum FSH concentrations were found to be greater than 20 units/l have been reviewed. The etiology and diagnosis of this uncommon but serious condition and the potential for spontaneous recovery of ovarian follicular activity (17.1%) are discussed. Management is reviewed in the light of ultrasonic, karyotypic and immunological findings.

Bacterial contamination of QC specimens as a cause of artefactually poor performance in the UK EQAS

During 1984-85, users of a solid-phase radioimmunoassay kit for LH and FSH had problems of both variability and bias in the assay of some EQAS specimens despite adequate internal QC. The cause has been identified as contamination of these specimens with Pseudomonas fluorescens.

Growth hormone secreting pituitary adenomas are heterogeneous in cell culture and commonly secrete glycoprotein hormone alpha-subunit

Cell culture methods were used to assess whether human pituitary adenomas secreting GH and associated with clinical acromegaly also secreted the structurally unrelated glycoprotein hormone alpha-subunit. Thirty-two tumours, together with peri-adenomatous tissue from two of them and three normal pituitaries were studied. Anterior pituitary hormones were measured by radioimmunoassay and included PRL, TSH, LH, FSH, and ACTH, as well as GH and alpha-subunit. Normal pituitary tissues secreted all hormones assayed. All 32 tumours secreted GH ranging from 241 to 5556 ng/2 X 10(5) cells/24 h and 12 (37.5%) secreted alpha-subunit in amounts which could not be accounted for by cross-reaction of other hormones or contamination by normal pituitary tissue, and which ranged from 10.3 to 73.5 ng/2 X 10(5) cells/24 h. Ten other tumours also secreted alpha-subunit but in very small amounts, not exceeding 1.8 ng/2 X 10(5) cells/24 h. PRL was secreted from 21 tumours (66%), and small amounts of other hormones, chiefly LH and TSH, were occasionally secreted from tumours. These cell culture studies would suggest that pituitary adenomas causing acromegaly are hormonally heterogeneous and that PRL and glycoprotein alpha-subunit are commonly detected in addition to GH.

Studies of gonadotrophin pulsatility and pelvic ultrasound examinations distinguish between isolated premature thelarche and central precocious puberty

We have studied the pulsatile secretion of gonadotrophins at night and made ovarian ultrasound examinations in three girls with central precocious puberty and three with isolated premature thelarche. The three girls with precocious puberty had well-defined pulsatile secretion of LH and FSH with LH predominating, as would be expected in normal puberty. Pulsatile secretion of gonadotrophins was also seen in girls with premature thelarche but the pattern was reversed. In girls with precocious puberty, large "multicystic" ovaries and large uteri were seen on ultrasound examination, whereas girls with isolated premature thelarche had small uteri and ovaries with less than four "cysts" up to 15 mm in diameter. These data provide the key to understanding the aetiology of isolated premature thelarche.

Gonadotrophin control of follicular development

In order to investigate the endocrine requirements for induction of follicle maturation, ovulation and luteal function we have measured plasma LH and FSH concentrations in 109 cycles induced by treatment with LHRH in 25 women with amenorrhoea of diverse aetiology. The major clinical and endocrine subgroupings were polycystic ovarian disease (PCO), Kallmann's syndrome, weight related amenorrhoea and hyperprolactinaemia. By analysis of variance of the LH and FSH concentrations obtained three times per week in the follicular phase in 77 ovulatory cycles (including 24 conception cycles), we found that in women with PCO the mean LH concentration was 19.5 IU/l. In hyperprolactinaemia and secondary amenorrhoea of non-specific aetiology the mean LH concentration was 11.6 IU/l and in patients with hypogonadotrophic hypogonadism and weight-related amenorrhoea it was 7.2IU/l. These mean LH concentrations were significantly different from each other (P less than 0.001). We speculate that the high LH concentrations in the follicular phase may impair the final stages of oocyte maturation and so contribute to the infertility of these patients. FSH concentrations in the follicular phase were lower in patients with hypogonadotrophic hypogonadism and weight-related amenorrhoea than in other groups. Although the differences were statistically significant (P less than 0.001) the difference was small (mean FSH in hypogonadotrophic hypogonadism and weight-related amenorrhoea 4.3 IU/l vs 6.3 IU/l in the others) and of uncertain biological significance. During the luteal phase, no differences between any of the groups in LH and FSH concentrations were found.

Cryopreservation of semen in men with testicular tumour or Hodgkin's disease: results of artificial insemination of their partners

After cryopreservation of semen from men with testicular tumours or Hodgkin's disease the success rate of artificial insemination of their partners was analysed. The cumulative probability of pregnancy at 6 months was 45%. The chance of pregnancy was greater when timing of ovulation was predicted with measurements of urinary luteinising hormone. Sperm density and sperm motility were also important in predicting the likelihood of pregnancy.

Improved performance of plasma gonadotrophin assays using common reagents and assay protocols: evidence from the UK External Quality Assessment Scheme

Radioimmunoassay kits prepared in the Chelsea Hospital for Women for follicle stimulating hormone (FSH) and luteinising hormone (LH) have been used in 26 UK laboratories for over 2 years. Data from the UK External Quality Assessment Schemes for FSH and LH have been used to provide an independent assessment of the performance of these kits over a 12-month period. For both analytes, users of the kits were found to have: a low variability of the bias, implying good within-laboratory, between-assay precision; a lower between-laboratory, within-sample geometric coefficient of variation than users of 'own' methods, implying better between-laboratory agreement; method bias that did not differ significantly from laboratories using 'own' method protocols. This survey indicates that a non-commercial organisation can produce immunoassay kits that improve the quality of FSH and LH assays generally available.

Inhibition of 3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) activity in first- and second-trimester human pregnancy and the luteal phase using Epostane

The effects of a competitive inhibitor of 3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) (Epostane, Sterling Winthrop, Guildford, England) on serum progesterone (P), estradiol (E2), and cortisol have been studied in three groups of pregnant women awaiting termination of pregnancy (5 to 8 weeks, 8 to 12 weeks, and 12 to 18 weeks of pregnancy) and 15 women in the luteal phase of the menstrual cycle. A single-dose randomized double-blind study was performed, each woman receiving a placebo, 50 mg of Epostane, or 100 mg of Epostane. In the pregnant group, there was a significant decline in the serum P concentration after both 50 mg and 100 mg of Epostane. The percentage fall increased with both drug dosage and advancing gestation. A similar fall in serum E2 was observed. Both of these effects were temporary. In the luteal phase group, a significant decline in serum P was observed after 100 mg of Epostane, but the serum E2 was not significantly different from the pretreatment concentration. Serum cortisol did not differ significantly from control values. These findings suggest that Epostane is an effective inhibitor of placental and ovarian 3 beta-HSD, which may have a role as an interceptive agent.

Endocrine and clinical consequences of combination tamoxifen-aminoglutethimide in postmenopausal breast cancer

By analogy with combination chemotherapy, endocrine agents with different mechanisms of action have been combined in the treatment of patients with advanced breast cancer. The clinical use of tamoxifen+aminoglutethimide+hydrocortisone showed no clinical benefit over the individual use of tamoxifen or aminoglutethimide+hydrocortisone. The endocrine changes occurring in postmenopausal patients as a consequence of their treatment with tamoxifen+aminoglutethimide+hydrocortisone have been examined. Suppression of gonadotrophin and oestrogen levels and increased levels of sex hormone binding globulin were observed. These changes might be expected to be of benefit in the treatment of advanced breast cancer, and do not explain the lack of clinical benefit in combining the treatments. Non-responders to this combination therapy had higher levels of oestrone and dehydroepiandrosterone sulphate whilst on treatment than responders, confirming previous observations in patients treated with aminoglutethimide+hydrocortisone.

Endocrine effects of danazol in the treatment of endometriosis

The effects of danazol therapy on circulating gonadotrophins, oestradiol and sex hormone-binding globulin (SHBG) binding capacity have been studied in eight patients with endometriosis. There was a significant rise in luteinizing hormone (LH) in the first week of treatment with a fall in oestradiol levels. There was a marked fall in SHBG in the first week which continued up to 4 weeks. It is concluded that danazol is not 'anti-gonadotrophic' as frequently claimed and that its therapeutic actions, side effects and the endocrine changes could be explained by an effect on SHBG production. The rationale of its use in other conditions requires reconsideration.