Article Text

Carcinoid-associated ectopic ACTH syndrome with variable response to octreotide


The case is presented of a 31-year-old woman who developed florid clinical and biochemical Cushing's syndrome due to metastatic hepatic carcinoid tumour from a probable pancreatic primary. Hypercortisolaemia was controlled with metyrapone and ketoconazole, but high doses of octreotide failed to affect plasma cortisol and urinary 5-hyroxyindole acetic acid (5HIAA) levels, or prevent rapid tumour growth. Hepatic polystyrene embolisation failed, and she was treated by liver transplantation with initial excellent results, and normalisation of cortisol and 5HIAA levels. Ten months later, however, she relapsed with bony and pelvic tumour recurrence, and high and symptomatic levels of cortisol and 5HIAA. At this time, octreotide in similar doses to those used previously appeared to normalise her biochemically, although she died soon after. This variable responsiveness to octreotide could be related to somatostatin receptor changes, or cyclical tumour secretion patterns.

Statistics from

Carcinoid tumour is a well-recognised though unusual cause of Cushing's syndrome due to ectopic adrenocorticotropin (ACTH) production. The primary tumour may be bronchial and solitary, in which case surgical cure is possible. Mid- or hind-gut tumours, however, usually present with hepatic metastases.1 Here, drugs which block steroid production, such as metyrapone or ketoconazole, may be of symptomatic and biochemical benefit,2 but there is particular interest in therapy with the somatostatin analogue drug octreotide, as, in addition to inhibiting ACTH production from carcinoid cells, octreotide may retard cell growth.3

We present here a case of fulminant Cushing's syndrome due to hepatic carcinoid tumour associated with ectopic ACTH. The case is of interest in that octreotide was initially ineffective and a liver transplantation was performed with good initial result. Unfortunately, tumour later recurred and, at this stage, was biochemically responsive to octreotide.

Case history

A 31-year-old woman presented with a 6-week history of weight gain, acne, hirsutes and weakness. Examination confirmed florid cushingoid features, with truncal obesity, striae, buffalo hump, acne, hypertension and proximal myopathy. She was not pigmented.

Biochemical investigations are summarised in the table. Gross hypercortisolaemia (unsuppressed by dexamethasone), hyperglycaemia and hypokalaemia suggested an ectopic ACTH syndrome. Computed tomographic (CT) imaging showed normal pituitary and hypothalamus, mild bilateral adrenal hyperplasia, and multiple hepatic tumour deposits. No primary malignant focus in chest or abdomen was seen. CT-guided liver biopsy revealed tumour cells of neuroendocrine origin consistent with carcinoid tumour, and urinary 5-hydroxyindole-acetic acid (5HIAA) was high (see table). A diagnosis of ectopic ACTH syndrome due to metastatic carcinoid tumour was made.

Table Biochemical investigations in a case of ectopic ACTH syndrome due to hepatic carcinoid

Further investigation showed elevated ACTH and grossly high ACTH-precursor levels. Plasma corticotropin-releasing factor was normal, as was β-melanocyte-stimulating hormone and gut hormone profile, except for an elevated gastrin level (this was consistently found, but there was no dyspepsia and gastroscopy was negative) (see table). Indium-labelled octreotide scan confirmed liver metastatic uptake, but none elsewhere. MIBG scan was entirely negative. Exhaustive investigations (including selective venous catheterisation studies) failed to show a primary tumour site.

Treatment with metyrapone 1 g tid normalised plasma cortisol, glucose and potassium levels with much symptomatic improvement (figure). Marked virilisation occurred, however, and plasma testosterone rose to 13.9 nmol/l. Ketoconazole 200 mg tid was added and plasma testosterone fell to 2.3 nmol/l. A trial of ketoconazole alone, however, led to relapse of hypercortisolaemia. Octreotide in doses of up to 500 μg tid subcutaneously had no clear effect on plasma cortisol or urinary HIAA levels. She was, however, maintained on octreotide, as well as metyrapone and ketoconazole, in the hope of tumour growth-suppressing activity.

Figure  Sequential changes in plasma cortisol and urinary 5HIAA in a patient with ACTH-secreting carcinoid tumour

Polystyrene embolisation of the liver was attempted 4 months after diagnosis, but failed due to distortion of the hepatic artery, presumably due to external compression by tumour deposits at the porta hepatis. Following this there was symptomatic and biochemical deterioration, despite continued metyrapone, octreotide, and ketoconazole; and insulin was needed for glycaemic control. Her general condition deteriorated and liver pain became difficult to control. Liver transplantation was performed 9 months after presentation. At operation a presumed primary tumour was also removed from the body of the pancreas. Postoperative immunosuppression was with cyclosporin (dose varying with blood levels), and prednisolone 20 mg daily, the dose of which did not alter.

Initial response was excellent, both clinically and biochemically, and apart from cyclosporin all treatment was stopped. Unfortunately, 10 months post-transplant she relapsed with intractable nausea, diarrhoea and ascites. A large pelvic tumour was palpable and bone scan showed multiple bony metastases, though CT scan of the liver was normal. Plasma cortisol was 2490 nmol/l and urinary HIAA 1885 μmol/24 h. Octreotide 500 μg tid was restarted with dramatic biochemical response (2 weeks later plasma cortisol was 401 nmol/l and urinary HIAA 581 μmol/24 h). Unfortunately there was no associated clinical benefit, and she deteriorated and died, almost 2 years after presentation.


This patient had typical features of the ectopic ACTH syndrome. Octreotide scanning4 and measurement of ACTH precursors5 were especially useful. Some carcinoid tumours with associated Cushing's syndrome produce corticotropin-releasing factor instead of, or as well as, ACTH; but this was not the case here. Our patient responded biochemically to metyrapone, though this resulted in troublesome raised testosterone levels as hypercortisolaemia became controlled. A combination of ketoconazole and metyrapone was eventually successful, though ketoconazole alone did not control plasma cortisol. Ketoconazole blocks cortisol production directly, and possibly also directly reduces ACTH production by tumour cells.2 7 It is normally at least partially effective in the control of ectopic ACTH syndrome.

Somatostatin analogue was first shown directly to reduce ACTH production from a bronchial carcinoid tumour in 1988,8 and this effect has been confirmed elsewhere.9 Initially, our patient had no response hormonally to octreotide, and others have reported similar lack of response.10 Interestingly, late in her illness she did show biochemical responsiveness to octreotide. This was, however, at a stage of widespread tumour spread, and clinical deterioration continued. There are five subtypes of somatostatin receptors (sst1–5) on carcinoid tumour cells, and not all (in particular sst1 and sst4) bind to octreotide.11 It may be that sst receptor status changed as the tumour spread and became more dysplastic, favouring octreotide binding. On the other hand, the positive octreoscan early in the illness does indicate the presence of octreotide-binding receptors. An alternative explanation may be that tumour secretion of ACTH and serotonin was cyclical, and response or lack of response to octreotide was coincidental.

Liver transplantation as an option for managing hepatic carcinoid tumour is now accepted as a rare, but sometimes reasonable option.12 It is usually palliative, but occasional long-term survival can occur. In our patient, useful life was extended by about 12 months. Interestingly, our patient developed widespread bony metastases pre-terminally, though the transplanted liver was free of disease. This unusual pattern has been observed in other carcinoid patients after liver transplant and may represent an effect of immunosuppression.


We are grateful to Professor Roger Williams and Dr Martin Lombard for their assistance.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.