INAPPROPRIATE ACTH CONCENTRATIONS IN TWO PATIENTS WITH FUNCTIONING ADRENOCORTICAL CARCINOMA
A. LAW, W. M. HAGUE, J. G. DALY, J. W. HONOUR, N. TAYLOR, S. L. JEFFCOATE, R. L. HIMSWORTH AND G. F. JOPLIN
Endocrine Unit, Hammersmith Hospital; Departments of Gynaecology and Endocrinology, Chelsea Hospital for Women; Departments of Medicine and Clinical Chemistry, Northwick Park Hospital and Clinical Research Centre
(Received 27 November 1987; returned for revision 6 January 1988; finally revised 1 February 1988; accepted 29 February 1988)
SUMMARY
Two female patients with functioning adrenocortical carcinomas had plasma ACTH detectable by RIA at presentation. In both patients there was evidence for biological activity of ACTH. There was no evidence for an ectopic source of ACTH, nor for a pituitary tumour. Urinary steroid analysis showed patterns of multiple hormone secretion characteristic of adrenocortical carcinomas. The finding of detectable ACTH concentrations in a patient with Cushing’s syndrome does not exclude the presence of an adrenocortical tumour.
Functioning adrenocortical carcinomas cause about 10% of adult cases of Cushing’s syndrome (Neville & O’Hare, 1985). There is a marked female predominance (Didolkar et al., 1981). The mortality is high, the 5 year survival in three recent large series ranging from 24-31% (King & Lack, 1979; Didolkar et al., 1981; Nader et al., 1983). The tumours are usually unilateral and secrete a variety of adrenocortical steroids (Kelly et al., 1979). Excess glucocorticoid secretion induces suppression of the hypothalamic-pituitary- adrenal system, and the resulting lack of ACTH results in atrophy of the unaffected contralateral adrenal gland. Two patients are described here who both had radioimmuno- logically detectable circulating ACTH at presentation and evidence for the presence of biologically active ACTH.
METHODS
Serum or plasma was stored at -20℃ until assayed. Plasma samples for ACTH assay were taken into plastic lithium heparin tubes on ice and separated within 15 min prior to freezing.
Plasma cortisol, 11-deoxycortisol, androstenedione, testosterone, DHAS and oestra- diol were measured after extraction by our own standard radioimmunoassays. Plasma
Correspondence: Dr A. Law, Clinical Research Centre, Northwick Park Hospital, Harrow, Middlesex, UK.
aldosterone and DOC for patient 1 were measured using the methods of Cope and Loizou (1973; 1975). Plasma aldosterone for patient 2 was measured by radioimmunoassay following extraction.
Plasma N-terminal ACTH was measured by radioimmunoassay according to the methods of Rees et al. (1971). Plasma N-terminal LPH was measured by radioimmuno- assay according to the methods of Jeffcoate et al. (1978).
Urinary steroids were analysed by gas chromatography following extraction using Sep-Pak cartridges, enzyme hydrolysis of conjugates, and the formation of methyloxime- trimethylsilyl ether derivatives. The identity of all components was verified by gas chromatography-mass spectrometry (Honour et al., 1984; Bevan et al., 1986).
CASE REPORTS
Patient 1
A 19 year old girl presented with 8 months secondary amenorrhoea and 6 months rapidly increasing hirsutism. She had been previously well, with 28 day menstrual cycles since menarche at 13 years. On examination, she was masculinized with florid stigmata of Cushing’s syndrome. Her blood pressure was 170/110 mm Hg. She had increased muscle bulk, severe hirsutes, seborrhoea and acne. There was no proximal myopathy, bone tenderness, excess pigmentation, voice change, abdominal mass or clitoromegaly. Her mental state was normal.
Investigations revealed hypokalaemia (K + 3.0 mmol/1) and impaired glucose toler- ance. The haemoglobin was 18.1 g/dl. Cortisol, androgen and mineralocorticoid concentrations were elevated in both plasma and urine and did not suppress on high dose dexamethasone (8 mg/day for 48 h) (Table 1). Plasma ACTH concentrations were in the normal range or raised on all of four occasions (0800-51 ng/1, 40 ng/1; 2400 -25 ng/1, 35 ng/l) (normal at 0800 < 80 ng/1; at 2400 < 10 ng/l). Plasma LH was 9 U/1 (normal 6-16 U/1), FSH 3 U/l (normal 2-8 U/l), and SHBG low at 28 nmol DHT bound/1 (normal 40- 80 nmol DHT bound/l).
Metyrapone administration (4.5 g in 24 h) produced a paradoxical fall in urinary 17-oxogenic steroids (239 umol/day to 179 umol/day) (normal 17-55 umol/day) and 17-oxosteroids (266 umol/day to 150 umol/day) (normal 10-70 umol/day). Plasma 11- deoxycortisol increased slightly (96 nmol/l to 106 nmol/l), but plasma DOC showed a large response to metyrapone (1.89 nmol/1 to 10 nmol/l) whereas plasma aldosterone fell (515 pmol/l to 203 pmol/l). Thyroid function, calcitonin and gut hormone levels were all normal.
The 24-h urinary profile shown on gas chromatography demonstrated that most of the steroid components found in the urine of normal subjects were excreted in increased amounts. For example, total cortisol metabolite excretion was increased four-fold. The compounds that showed the greatest increase had a 30-hydroxy-5-ene configuration (i.e., compounds 3, 4, 8, 11, 12, 14, 15 and 18-see Table 2), indicating a relative deficiency of 3ß-hydroxysteroid dehydrogenase in the tumour. Although plasma 11-deoxycortisol and DOC concentrations were elevated, this was insufficient to affect their urinary metabolite excretions.
Intravenous urography showed a rotated right kidney. Abdominal computed tomo- graphy (CT) scan showed a large right adrenal tumour invading the liver and inferior vena cava. The left adrenal gland was of normal size and not atrophic. Angiography
| Plasma levels | Urine levels | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Cortisol | 11-deoxycortisol | Testosterone (nmol/l) | Androstenedione | Dehydroepiandrosterone | Oestradiol | Deoxycorticosterone | Aldosterone (pmol/l) | Urine free cortisol (nmol/24 h) | |
| (nmol/l) | (nmol/l) | (nmol/l) | (pmol/l)-SO4 | (pmol/l) | (nmol/l) | ||||
| Normal range (follicular phase) | 150-550 | 28-84 | 1.0-3.2 | 3.1-10-1 | 50-7-5 | 50-180 | 0-08-0.61 | 30-130 | <270 |
| Patient 1 Basal 0900h | 524 | 101 | 11.6 | 22-8 | 14.0 | 264 | 1.89 | 515 | 1781 |
| 2400h | 552 | 1.53 | 499 | ||||||
| Dexamethasone* | 1080 | 9.7 | 15-0 | 25-0 | 248 | 1.45 | 446 | 2100 | |
| Patient 2 Basal 0900h | 1080 | 3.7 | 15.9 | 4-0 | 302 | 468 | 3300 | ||
| 2400h | 900 | ||||||||
| Dexamethasone* | 940 | 1560 | |||||||
*Dexamethasone 8mg/day for 48hrs, with serum sample at 48h.
| Steroid | Case 1 | Case 2 | Median | Range |
|---|---|---|---|---|
| 1. Androsterone (5a) | 5 | 4 | 3.9 | 1-9-8.9 |
| 2. Aetiocholanolone (5฿) | 13 | 9 | 3.6 | 2-3-9.5 |
| 3. Dehydroepiandrosterone (DHA) | 76 | 0-6 | 0-4 | 0-2-1.5 |
| 4. 5-Androstene-3, 17x-diol | 13 | * | ||
| 5. 11-oxo-aetiocholanolone | 4 | 0.3 | 0.4 | 0-1-1.2 |
| 6. 11-hydroxy-androsterone | 12 | 40 | 2.4 | 0-8-5.0 |
| 7. 118-hydroxy-aetiocholanolone | 12 | 11 | 0.5 | 0-3-1.1 |
| 8. 16x-hydroxy-DHA | 26 | 5 | 1.0 | 0-6-1-5 |
| 9. Pregnanediol | 20 | 6 | 1-4 | 0-5-5-4 |
| 10. Pregnanetriol | 14 | 5 | 2.5 | 0-5-5.4 |
| 11. 5-Pregnene-3, 20x-diol | 12 | * | ||
| 12. 5-Androstene-38, 16x, 178-triol | 14 | 20 | 1.0 | 0-1-1-8 |
| 13. 58-Pregnane-3x, 16x, 20x-triol | 6 | * | ||
| 14. 16x-hydroxypregnenolone | 8 | * | ||
| 15. 5-Pregnene-38, 17x, 20x-triol | 27 | * | ||
| 16. Tetrahydrocortisone (THE) | 31 | 18 | 5.3 | 3.3-9.3 |
| 17. Tetrahydro-11-dehydrocorticosterone | 1 | 0.2 | 0-2-0.3 | |
| 18. 5-Pregnene-38, 16x, 208-triol | 21 | * | ||
| 19. 5č-Pregnane-35, 118, 20¢, 21-tetrol | 4 | * | ||
| 20. Tetrahydrocortisol (THF) | 11 | 17 | 1.8 | 1.2-3.3 |
| 21. allo-Tetrahydrocortisol (5x-THF) | 5 | 5 | 1.5 | 0.8-2.9 |
| 22. «-cortolone (20x) | 17 | 1 | 1.0 | 0.7-2.5 |
| 23. B-cortol + B-cortolone | 20 | 4 | 1.7 | 1.0-4.9 |
| 24. a-cortol (20x) | 6 | 3 | 0.5 | 0-4-0-9 |
* < 0.5 umol/day.
-Not detectable.
demonstrated a pathological tumour circulation and confirmed invasion of the liver and vena cava. Selective venous sampling was not possible from the right adrenal vein as this was occluded. The left adrenal vein had a higher cortisol concentration (828 nmol/l) than simultaneous arterial (662 nmol/l) or distal venous (607 nmol/l) samples. Chest, skull X-rays and a high resolution CT scan of the pituitary were normal and a bone scan showed no metastases.
Resection of the tumour was considered impossible, owing to the invasion of the liver and vena cava. The patient was therefore treated with a course of op’DDD, but she died 3 months later. Permission for autopsy was not given.
Patient 2
A 74 year old woman presented with a 4 month history of thirst, polyuria, tiredness, easy bruising, difficulty in climbing stairs and unsteadiness of gait. On examination, she had signs of Cushing’s syndrome with petechiae, bruising and a proximal myopathy. There was no pigmentation, hirsutes, striae or abdominal mass. The blood pressure was 190/110 mm Hg.
On investigation, random plasma glucose was 19·1 mmol/1, plasma K + 2.7 mmol/l and HCO-3 38 mmol/1. Basal plasma cortisol and urinary free cortisol values were elevated (Table 1), falling in the urine, but not in the plasma, on high dose dexamethasone (8 mg/ day over 48 h). Plasma sex steroids were elevated while plasma gonadotrophins were suppressed (LH <1 U/1, FSH 1·2 U/1). Plasma ACTH was readily detectable and within the normal range (0800-58 ng/1; 2400 h-24 ng/1) and a high concentration of plasma LPH was also found (1084 ng/1; normal <200 ng/l). Thyroid function, calcitonin and gut hormone levels were all normal. Abdominal CT scanning showed a large left adrenal tumour with a second mass below it. There was no other evidence of metastases. The right adrenal gland was normal. A high resolution CT scan of the pituitary fossa was normal.
The 24 h urinary steroid profiles on gas chromatography obtained during the initial phase of the disease (Table 2) showed an increase in absolute excretion of C21 cortisol metabolites with ratios of THE to THF, and THF to 5x-THF, characteristic of Cushing’s syndrome (Phillipou, 1982); a marked increase in 11ß-hydroxyandrosterone excretion provided evidence for the presence of an adrenal tumour. DHA excretion was not increased. Further steroid profiles obtained after the development of metastases were similar.
At operation, a tumour 6 x 6 cm was found arising from the upper aspect of the left adrenal gland and was resected, together with an enlarged para-aortic gland. The right adrenal gland was inspected and found to be of normal size.
Histological examination revealed a small normal portion of the left adrenal gland to which was attached a tumour containing large nests of tumour cells resembling adrenal
200
100 mg hydrocortisone i.v.
100
Log ACTH (ng/I)
80
60
t = 82 min
40
1
2
30
20
10
-180
-150
120
-90
-60
-30
0
30
60
90
120
150
180
Time (min)
cortex with abundant eosinophilic cytoplasm and florid nuclear pleomorphism, divided by strands of fibrous tissue, and with evidence of venous invasion. The para-aortic gland showed similar histology. Electron microscopy of the tumour cells demonstrated particularly prominent smooth endoplasmic reticulum (SER) together with mitochondria having tubular cristae, suggestive of cells actively engaged in steroid biosynthesis (Fig. 1). The normal adrenal tissue had less prominent SER, but numerous clear lipid droplets and compound lysosomal bodies of the lipofuscin type, suggestive of reduced but not totally suppressed steroid synthesis (Mackay, 1969) (Figure 2). Immunocytochemistry of the tumour using the Wellcome Anti-ACTH antibody showed no stainable ACTH present.
The patient made a rapid recovery after surgery and was maintained on prednisolone 7.5 mg/day. Since prednisolone metabolites are distinguishable from endogenous cortisol metabolites, this management enabled a rise in cortisol production on recurrence of the tumour to be detected in the urinary steroid profile at an early stage.
At 3 months after operation, a hydrocortisone suppression test (100 mg i.v. as a bolus, 36 h after interrupting prednisolone) (Besser et al., 1971) demonstrated normal basal ACTH levels which were promptly shut off (Figure 3). At this time, the 24 h urinary steroid profile on gas chromatography showed no endogenous cortisol metabolites. At 9 months after operation, the urinary steroid profile showed an increase in steroid metabolites, indicative of tumour recurrence despite absence of clinical progression. At 13 months, abdominal CT scanning revealed extensive tumour recurrence in the left para- aortic area. Her clinical condition slowly deteriorated thereafter, despite treatment with metyrapone, and she died 3 years and 4 months after initial presentation.
Autopsy demonstrated widespread local invasion by tumour and metastases in liver and lung. The right adrenal gland was atrophic. The pituitary gland showed no abnormalities.
DISCUSSION
The presence of ACTH in the plasma of patients with Cushing’s syndrome is widely interpreted as excluding adrenocortical tumours (Howlett et al., 1985; Bondy, 1985; Nabarro, 1986). In this paper, we present two cases of functioning adrenocortical carcinomas in whom prevailing plasma ACTH concentrations were not suppressed, despite elevated plasma concentrations of glucocorticoids.
Various hypotheses could account for this paradoxical observation:
(1) The adrenal tumour itself secretes peptide hormones in addition to steroids.
(2) There is a concurrent alternative source of ectopic ACTH.
(3) There is a concurrent pituitary ACTH-secreting tumour.
(4) One or more of the steroids produced by the adrenal tumour block the usual negative feedback effect of cortisol on the hypothalamus and pituitary.
(5) There is an abnormal cortisol receptor in the hypothalamus or pituitary.
Adrenomedullary production of ACTH is well recognised (Forman et al., 1979; Spark et al., 1979). There was, however, no histological nor autopsy evidence for a phaeochromocytoma in Case 2. Furthermore, there was no ACTH demonstrated by immunocytochemical staining in the tumour of Case 2, which suggests that it was not secreting ACTH. Corticotrophin releasing factor (CRF) was not measured in either
patient. There is no published data on CRF production by adrenal cortical tissue or tumours (Sudat et al., 1986). However, our data do not exclude secretion of ACTH or CRF by the tumour tissue.
Both our patients were hypokalaemic at presentation-a feature commonly found in Cushing’s syndrome due to ectopic ACTH production (Howlett et al., 1985). Bevan and Burke (1984) have described the coincidence of an adrenal carcinoma causing Cushing’s syndrome with normal ACTH levels in association with an ACTH-containing bronchial neoplasm found at autopsy. Such a second tumour was not found in either of our two patients, although it cannot be excluded in Case 1.
A concomitant pituitary lesion was sought in both cases. Anderson et al. (1978) described a patient with Cushing’s syndrome, a virilizing adrenal carcinoma and consistently detectable ACTH in whom the contralateral adrenal showed nodular hyperplasia at the time of surgery and in whom autopsy revealed a 5 mm pituitary basophil adenoma. High resolution CT scanning of the pituitary of both our patients, however, together with dexamethasone suppression testing, and, in Case 2, autopsy failed to reveal any evidence of such a lesion. Moreover, the patient described by Anderson et al. (1978) had a long history of preceding hypertension, leucocytosis, exophthalmos and pigmentation although the diagnosis of Cushing’s syndrome was not established prior to the rapid onset of the virilisation. Hamwi et al. (1957) described a woman with untreated long-standing probable 21-hydroxylase deficiency in whom an adrenocortical carcinoma was found. Neither of our two patients, however, had a significant past history prior to presentation.
Both our patients showed high basal plasma levels of adrenal steroids as well as high basal urinary excretion of steroid metabolites. It is possible that atypical circulating steroids from the carcinoma interfere with the normal cortisol feedback mechanisms, allowing persistent ACTH secretion by the pituitary (Taylor et al., 1986). Neither 30-hydroxy-5-ene steroids nor any of the other unusually increased metabolites are close structural analogues of cortisol, and so are unlikely to have such effects.
After surgery in patient 2, the administration of hydrocortisone rapidly suppressed plasma ACTH levels. This suggests that the normal feedback responses were operating after tumour resection and the removal of excessive and aberrant steroid and/or peptide production. The response of plasma ACTH to hydrocortisone is evidence against the presence of an abnormal cortisol receptor in either the hypothalamus or pituitary, although this is necessarily an indirect method of assessing receptor integrity.
There is indirect evidence that the ACTH measured by radio-immunoassay in both of our patients was also biologically active. In case 1, the contralateral adrenal was not atrophic on high resolution CT scanning, while its venous effluent had a higher concentration of cortisol than simultaneously taken mixed venous or arterial samples. The contralateral adrenal gland of Case 2 was inspected at operation by an experienced adrenal surgeon who stated that it was not atrophic, while the ipsilateral adrenal removed with the tumour had normal histology and features of tissue active in steroid biosynthesis on electron microscopy. We suggest that the normal circulating concentrations of ACTH were responsible for maintaining the integrity of structure and function. The subsequent finding at post mortem of right adrenal atrophy might suggest a change in the biological activity of ACTH following removal of the primary tumour, although no ACTH measurements were made, and the patient had been receiving steroid replacement therapy
and, in later months, metyrapone.
In conclusion, we emphasize that the finding of normal plasma ACTH concentrations in Cushing’s syndrome may not exclude the diagnosis of an adrenocortical tumour.
ACKNOWLEDGEMENTS
We are indebted to Dr M. J. O’Hare for immunocytochemistry, Dr J. Crow for electron microscopy, Professor D. J. Allison for angiographic investigations and CT scans of patient 1, Mr S. Loizou for aldosterone and DOC measurements of patient 1, Professor L. Rees for the ACTH and LPH measurements of both patients.
REFERENCES
ANDERSON, D.C., CHILD, D.F., SUTCLIFFE, C.H., BUCKLEY, C.H., DAVIES, D. & LONGSON, D. (1978) Cushing’s syndrome, nodular adrenal hyperplasia and virilizing carcinoma. Clinical Endocrinology, 9, 1-14.
BESSER, G.M., CULLEN, D.R., IRVINE, W.J., RATCLIFFE, J.G. & LANDON, J. (1971) Immunoreactive corticotrophin levels in adrenocortical insufficiency. British Medical Journal, i, 374-376.
BEVAN, J.S. & BURKE, C.W. (1984) Cushing’s syndrome due to primary adrenal disease with circulating ACTH. Abstract 84, British Endocrine Societies Meeting, Edinburgh.
BEVAN, B.R., SAVVAS, M., JENKINS, J.M., BAKER, K., PENNINGTON, G.W. & TAYLOR, N.F. (1986) Abnormal steroid excretion in gestational trophoblastic disease complicated by ovarian thecalutein cysts. Journal of Clinical Pathology, 39, 627-634.
BONDY, P.K. (1985) Disorders of the adrenal cortex. In Williams Textbook of Endocrinology (eds J.D. Wilson & D.W. Foster), pp. 816-890, W.B. Saunders, Philadelphia.
COPE, C.L., & LOIZOU, S. (1973) Simplification of urinary aldosterone measurements by radioimmunoassay. Journal of Clinical Pathology, 26, 628-634.
COPE, C.L. & LOIZOU, S. (1975) Deoxycorticosterone excretion in normal, hypertensive and hypokalaemic subjects. Clinical Science and Molecular Medicine, 48, 97-105.
DIDOLKAR, M.S., BESCHER, R.A., ELIAS, E.G. & MOORE, R.H. (1981) Natural history of adrenal cortical carcinoma: a clinicopathological study of 42 patients. Cancer, 47, 2153-2161.
FORMAN, B.H., MARBAN, E., KAYNE, D., PASSARELLI, N.M., BOBROW, S.N., LIVOLSI, V.A., MERINO, M., MINOR, M. & FARBER, L.R. (1979) Ectopic ACTH syndrome due to pheochromocytoma: case report and review of the literature. Yale Journal of Biology and Medicine, 52, 181-189.
HAMWI, G.J., SERBIN, R.A. & KRUGER, F.A. (1957) Does adrenocortical hyperplasia result in adrenocortical carcinoma? New England Journal of Medicine, 257, 1153-1157.
HONOUR, J.W., PRICE, D.A., TAYLOR, N.F., MARSDEN, H.B. & GRANT, D.B. (1984) Steroid biochemistry of virilising adrenal tumours in childhood. European Journal of Pediatrics, 142, 165-169.
HOWLETT, T.A., REES, L.H. & BESSER, G.M. (1985) Cushing’s Syndrome. Clinics in Endocrinology and Metabolism, 14, 911-945.
JEFFCOATE, W.J., REES, L.H., LOWRY, P.J., & BESSER, G.M. (1978) A specific radio-immunoassay for B-lipotropin. Journal of Clinical Endocrinology and Metabolism, 47, 160-167.
KELLY, W.F., BARNES, A.J., CASSAR, J., WHITE, M., MASHITER, K., LOIZOU, S., WELBOURN, R.B. & JOPLIN, G.F. (1979) Cushing’s syndrome due to adrenocortical carcinoma - a comprehensive clinical and biochemical study of patients treated by surgery and chemotherapy Acta Endocrinologica (Copenhagen), 91, 303-318. KING, D.R. & LACK, E.E. (1979) Adrenal cortical carcinoma. A clinical and pathologic study of 49 cases. Cancer, 44, 239-244.
MACKAY, A. (1969) Atlas of human adrenal cortex ultrastructure. In Functional Pathology of the Human Adrenal Gland (Ed T. Symington), pp. 345-489, E. & S. livingstone, Edinburgh.
NABARRO, J.D.N. (1986) Transsphenoidal surgery for Cushing’s syndrome. Journal of The Royal Society of Medicine, 79, 253-254.
NADER, S., HICKEY, R.C., SELLIN, R.V. & SAMAAN, N.A. (1983) Adrenal cortical carcinoma: a study of 77 cases. Cancer, 52, 707-711.
NEVILLE, A.M. & O’HARE, M.J. (1985) Histopathology of the human adrenal cortex. Clinics in Endocrinology and Metabolism, 14, 791-820.
PHILLIPOU, L. (1982) Investigation of urinary steroid profiles as a dynamic method in Cushing’s syndrome. Clinical Endocrinology, 16, 433-435.
REES, L.H., COOK, D.M., KENDALL, J.W., ALLEN, C.F., KRAMER, R.M., RATCLIFFE, J.G. & KNIGHT, R.A. (1971) A radioimmunoassay for rat plasma ACTH. Endocrinology, 89, 254-261.
SILVA, E.G., MACKAY, B., SAMAAN, N.A., & HICKEY, R.C. (1982) Adrenocortical carcinomas: an ultra- structural study of 22 cases. Ultrastructural Pathology, 3, 1-7.
SPARK, R.F., CONNOLLY, P.B., GLUCKIN, D.S., WHITE, R., SACKS, B. & LANDSBERG, L. (1979) ACTH secretion from a functioning pheochromocytoma. New England Journal of Medicine, 301, 181-189.
SUDAT, T., TOMORI, N., YAJIMA, S., ODAGIRI, E., DEMURA, H. & SHIZUME, K. (1986) Characterization of immunoreactive corticotropin and corticotropin releasing factor in human adrenal and ovarian tumours. Acta Endocrinologica, 111, 546-522.
TAYLOR, N.F., BEVAN, B.R., FISHER, R.A. & ISMAIL, A.A.A. (1986) Evidence that 21-deoxycortisol competes for hypothalamic cortisol receptors. Journal of Steroid Biochemistry, 25S, 229.