Rosiglitazone for prevention or adjuvant treatment of Nelson's syndrome after bilateral adrenalectomy

doi:10.1530/eje.1.01994

HOME

HELP

FEEDBACK

SUBSCRIPTIONS

Rosiglitazone for prevention or adjuvant treatment of Nelson’s syndrome after bilateral adrenalectomy

Mikkel Andreassen and Lars Østergaard Kristensen

Department of Endocrinology and Internal Medicine, J 106, Herlev Hospital, University of Copenhagen, Herlev Ringvej, 2730 Herlev, Denmark

(Correspondence should be addressed to M Andreassen; Email: andreassenmikkel{at}hotmail.com)

Abstract

Objective: To investigate the effect of Rosiglitazone in threepatients treated with bilateral adrenalectomy followed by hyperpigmentationand hypersecretion of ACTH.

Patients and methods: One patient had increasing ACTH afterprevious transsphenoidal surgery for Nelson’s syndrome,and two patients without pituitary adenomas had recurrence ofCushing’s disease after primary and repeated transsphenoidalsurgery with need for bilateral adrenalectomy. The patientsdeveloped hyperpigmentation and increasing ACTH at nadir 2–4h after morning hydrocortisone dose. ACTH during Rosiglitazonetherapy (4 mg/day for 4 weeks and then 8 mg/day) was measuredat regular intervals 24 h after the latest dose of hydrocortisone.

Results: In two patients there was a decrease in ACTH by 40%after 5 months. The first of these patients showed an escapewith increasing ACTH to the initial value after 11 months. Inthe third patient no effect was observed. Tumour developmentor progression on magnetic resonance imaging was not observed.

Conclusion: Rosiglitazone might represent an adjuvant therapyin patients with ACTH hypersecretion. Larger long-term studiesare needed.

Introduction

Bilateral adrenalectomy is an option of treatment after unsuccessfultranssphenoidal surgery (TSS) and recurrence of Cushing’sdisease. A considerable fraction of these patients need bilateraladrenalectomy (1). A complication of bilateral adrenalectomyis Nelson’s syndrome with adrenocorticotrophin (ACTH)hypersecretion, hyperpigmentation and possibly development ofpituitary tumour. Radiotherapy is partially prophylactic, butimplicates the risk of pituitary insufficiency and also damageof the optic chiasma. Therefore, the recent suggestion for treatmentis TSS in cases with tumour, but tumour is not always present.An acute effect of bromocriptine in Nelson’s syndromehas been observed (2) and long-term remission with cabergolinehas been reported in a single patient (3).

Peroxisome proliferator-activating receptor- ${gamma}$ (PPAR- ${gamma}$ ) is a memberof the nuclear hormone receptor superfamily (4, 5). PPAR- ${gamma}$ receptorsare predominantly expressed in adipose tissue (6, 7). Importantlyin this context it has recently been shown that PPAR- ${gamma}$ is alsoexpressed in normal ACTH-secreting cells and abundantly in ACTH-secretingtumour cells (8).

Rosiglitazone is a PPAR- ${gamma}$ ligand. It is a member of the thiazolidinedioneclass of insulin-sensitizing drugs. In vitro treatment of humanACTH-secreting pituitary tumour cells with Rosiglitazone leadsto enhanced apoptosis and decreased cell division and in vivostudies reveal that Rosiglitazone is able to inhibit ACTH secretionand tumour growth in mice (8). Furthermore, Rosiglitazone hasbeen used as short-term therapy in Cushing’s disease inhumans (9). Based on these observations we have employed Rosiglitazonein three patients with Nelson’s syndrome.

Patients and methods

Patients

Patient 1 was a 67-year-old female with Cushing’s diseasediagnosed in 1970. Primary treatment was bilateral adrenalectomy.Gradually increasing ACTH and hyperpigmentation was observedfrom about 1990. In 1998 magnetic resonance imaging (MRI) showeda pituitary adenoma with suprasellar extension close to theoptic chiasma. TSS was performed with temporary improvement,but in 2002 there was recurrence of Nelson’s syndromewith increasing ACTH and relapse of a small (7 x 7 x 3 mm) pituitaryadenoma.

Patient 2 was a 74-year-old male with Cushing’s diseasediagnosed in 1993. Initially successful treatment with TSS wasperformed, but in 1999 recurrence was diagnosed. Another TSSwas attempted without success, and in 2000 bilateral adrenalectomywas performed. 2 years later hyperpigmentation and increasingACTH occurred. There was no visible tumour on MRI.

Patient 3 was a 37-year-old female with Cushing’s diseasediagnosed in 2002. TSS was performed as a primary treatmentwithout cure and a second TSS was also ineffective, whereafterlaparascopically bilateral adrenalectomy was done. Due to residualdisease activity a second open operation was performed on theleft adrenal gland. Clinically the patient was cured and hydrocortisone-replacementtherapy was initiated but, apparently, low endogenous cortisolproduction was still present. Since 2003 the patient has shownhyperpigmentation and increasing ACTH, but there is no visibletumour on MRI.

All patients received hydrocortisone-replacement therapy of20–30 mg/day divided into two or three doses.

Methods

During Rosiglitazone therapy (4 mg/day for 4 weeks and then8 mg/day) ACTH was measured at regular intervals 24 h afterthe latest dose of hydrocortisone at nadir cortisol, when zenithACTH could be expected.

Plasma ACTH was measured with a high-detectability immunoradiometricassay (DYNOtest; BRAHMS Diagnostica GmbH, Berlin, Germany).The normal range is 1.8–12 pM. Serum cortisol was measuredwith RIA (Diagnostic System Laboratories, Webster, TX, USA);the lower limit of detection was 10 nM.

No side effects were reported, including any episodes of hypoglycaemia.Liver-function tests were performed at every visit and remainednormal during the study. This retrospective report was approvedby the Danish Data Protection Agency, Copenhagen.

Results

Figure 1 shows ACTH during Rosiglitazone treatment. Patients1 and 2 showed a gradual decrease in ACTH by 40% after 3–6months (patient 1 from 1290 to 765 pM and patient 2 from 478to 289 pM). However, in patient 1 there was an escape after5 months of treatment, with ACTH increasing to the initial value,and the treatment has now been paused. Patient 3 received Rosiglitazonefor 1 year, during which time no effect on ACTH was observed,and the treatment has recently been stopped and so far no increasein ACTH has been observed. Hyperpigmentation did not improvein any of the patients.

View larger version (17K):
[in this window]
[in a new window]

Figure 1 Plasma ACTH (pM) during Rosiglitazone treatment. Measurements of ACTH were taken 24 h after the latest hydrocortisone dose.

MRI of the pituitary showed no evidence of tumour progressionin patient 1 and no development of tumours in patients 2 and3.

Discussion

Hypersecretion of ACTH with hyperpigmentation is the key characteristicof Nelson’s syndrome and an obvious pituitary tumour isnot always present (10). According to this criterion there isno doubt that our three patients had Nelson’s syndrome.

In two of the patients Rosiglitazone treatment lead to a significantdecrease in ACTH, assumed to reflect secretion, but in one ofthese patients there was an escape of the effect. The thirdpatient did not respond. This patient had low residual endogenouscortisol production, which might be of some importance in reducingcorticotrophic ACTH secretion and cell proliferation. Concerningcell proliferation it should be noted that the residual tumourin one patient did not progress and that tumour developmentwas not observed in the other two patients.

In order to separate potential responders from nonrespondersit would have been of interest to evaluate PPAR- ${gamma}$ receptor expressionin tumours already removed, but such an analysis was not performed.However, Ambrosi et al. (9) observed no correlation betweenPPAR- ${gamma}$ expression and response to Rosiglitazone in patients withCushing’s disease.

Rosiglitazone did not lead to normalization of ACTH in any ofour patients. However, it seems worthwhile to consider an attemptof such treatment in patients with Nelson’s syndrome withoutobvious tumours and possibly in patients after unsuccessfulTSS for a pituitary adenoma hypersecreting ACTH. Furthermore,the medical treatment might be used in concert with radiotherapyduring the time period awaiting the effect of radiation.

PPAR- ${gamma}$ is also expressed in a variety of other pituitary tumours(6). This current information and the present results pointto the possibility that thiazolidinedione represents an optionof treatment for Nelson’s syndrome and, potentially, otherpituitary tumors. Larger long-term studies are needed to elucidatethe answers to these questions. A new generation of thiazolidinedionesdeveloped selectively for this specific purpose would be ofconsiderable interest. This may bring about a change in thestrategy towards medical treatment of a variety of pituitarytumors, analogous to the well-established first-line medicaltreatment of prolactinomas with dopamine agonists.