E-Poster Presentation 63rd Endocrine Society of Australia Annual Scientific Meeting 2020

Introduction

Adrenocortical carcinoma (ACC) is rare (incidence: one to two cases per million adults) and is more common in females, with a sex ratio of 4.2¹. ACC diagnosed during pregnancy is exceedingly rare and studies are limited to case reports². Cushing’s syndrome (CS) is uncommon during pregnancy due to reduced fertility associated with hypercortisolism and elevated androgens². Confirming the diagnosis of pathological hypercortisolism is also problematic due to changes in the hypothalamic-pituitary-adrenal (HPA) axis during pregnancy, which results in physiological hypercortisolism. Investigations commonly used in the non-pregnant state are not well validated during pregnancy³. Confirming the diagnosis of CS is imperative due to the adverse maternal and fetal outcomes experienced during pregnancy⁴. We present the case of a 31 year old pregnant woman who presented for management of recently diagnosed gestational diabetes mellitus (GDM) and was found to have features of CS on clinical examination. ACTH-independent CS and a 4cm adrenal tumour were subsequently diagnosed. The pregnancy was complicated by premature rupture of membranes and the patient delivered a female infant at 33 weeks gestation. Four weeks post-partum she underwent a left adrenalectomy, which subsequently identified an ACC on histopathology. 

Case Report

A 31 year old caucasian woman (gravida 1, para 0) of 28 weeks gestation presented to the Endocrine Outpatient Department for the management of recently diagnosed GDM  confirmed on 75g OGTT according to Australian criteria⁵ (fasting glucose 5.3 mmol/L, 60 min glucose 14.0 mmol/L 120 min glucose 8.8 mmol/L). She had a past history of Hashimoto’s hypothyroidism and was taking adequate thyroxine replacement therapy (TSH 0.3 mIU/L, reference range [RR] 0.31-2.75 mIU/L for 24-30 weeks gestation) and obesity (pre-pregnancy BMI 33.8 kg/m²). In the initial consultation, she described developing coarse hair growth, facial and chest acne during the pregnancy. There was no history of polycystic ovarian syndrome, prior glucocorticoid use or adrenal disease. There was no family history of diabetes or adrenal disease. She had conceived naturally and her pregnancy was otherwise uncomplicated. An obstetric ultrasound at 22 weeks gestation demonstrated a singleton pregnancy with normal fetal morphology. On clinical examination, hirsuitism with male facial and chest hair distribution, acne and violaceous abdominal striae were confirmed (Figures 1a-c). The was no evidence of a myopathy. She was hypertensive (blood pressure 134/94 mmHg) and clinically euthyroid. 

A 24-hour urinary free cortisol (UFC) was elevated (968 nmol/24h, RR <166 nmol/24h), as was testosterone (9.3 nmol/L, RR <2.0 nmol/L), and androstenedione (20.2 nmol/L, RR 0.9-7.5 nmol/L). Adrenocorticotropic hormone (ACTH) was suppressed (3.6ng/L, RR 7.2-63.3 ng/L) (table 1). Serum biochemistry, liver function, metanephrines, normetanephrines, aldosterone:renin ratio and DHEAS were within normal limits. Serially elevated 24 hour UFC and suppressed ACTH levels were used to confirm ACTH-independent hypercortisolism. A dexamethasone suppression test (DST) and midnight salivary cortisol levels were not performed due to lack of validation in pregnancy². An abdominal ultrasound demonstrated a left adrenal lesion measuring 28x22x23mm. Her gestational diabetes was treated with metformin and subcutaneous basal glargine and bolus aspart insulin, which was subsequently well controlled. Endocrine surgery and obstetric medicine input was sought regarding further assessment and management of CS. As there was no evidence of foetal distress and the patient was otherwise stable, close and regular observation of the mother and foetus was decided upon until induction of labour after 34 weeks gestation. Postpartum, the patient would undergo a dedicated adrenal computed tomography (CT) for further characterisation of the lesion, before proceeding to surgery. 

At 32 weeks gestation, the patient experienced premature prelabour rupture of membranes and was admitted to hospital. At 33 weeks gestation, she began to labour spontaneously and subsequently delivered a live female infant weighing 1879g via forceps-assisted vaginal delivery. The newborn was admitted to the special care nursery for observation and received a short course of dexamethasone to cover for possible adrenal insufficiency. There was no neonatal hypoglycaemia and the infant demonstrated normal appearing external genitalia, in contrast to previous case reports⁵. 

An adrenal CT scan performed post-partum confirmed the presence of a lesion measuring 3.2cm in maximal diameter. The lesion was homogenous (32 Hounsfield units). The absolute and relative washout ratios were 66% and 42% respectively, suggestive of an adenoma (Figure 2). Repeat 24h UFC performed post-partum remained elevated (693 nmol/24hr) and ACTH remained suppressed (1.9 ng/L). Four weeks post-partum, the patient underwent a laparoscopic left adrenalectomy, which was uncomplicated. Histopathology revealed a cortical lesion measuring 40x25x25mm with high-grade nuclear features with pleomorphism, presence of giant cells, bizarre nuclei, myxoid deposits and zones of tumour necrosis. Immunohistochemical staining was positive for inhibin, melanA. A Ki67 of 12% was identified in a few areas, but otherwise was very low (1-2%) for most of the tumour (Figure 3). There was no lymphovascular or capsular invasion. Overall the features were consistent with an adrenocortical carcinoma. The patient underwent fluorodeoxyglucose (FDG)-positron emission tomography (PET), which did not demonstrate evidence of residual or metastatic disease. The case was discussed at a regional Endocrine-Oncology Multi-disciplinary meeting and in consultation with the patient, a period of close surveillance was weighed against further surgery and chemo-radiotherapy. Due to the malignant nature of the lesion and strong preference to ensure microscopic clearance to protect against local or regional recurrence, the patient underwent further surgery to excise the residual left adrenal tissue.

Discussion

CS due to ACC during pregnancy is rare and associated with significant morbidity and mortality in both foetus and mother². Comorbid hypertension, pre-eclampsia, weight gain, androgenisation and catabolic effects of hypercortisolism can occur. Diabetes mellitus is reported in 25% of pregnant women with CS and premature labour has been reported in 50% of cases^2-4. Adrenal lesions are responsible for approximately 50% of CS during pregnancy, which is higher than CS in non-pregnant women (around 15%)². Confirming pathological hypercortisolism during pregnancy is problematic due to physiological changes that occur within the hypothalamic-pituitary axes³. Total cortisol levels are elevated twofold-threefold during pregnancy due to an increase in corticosteroid-binding globulin (CBG) and an increase in corticotropin secretion from both hypothalamic and placental sources². The diurnal pattern of cortisol secretion is maintained in pregnancy, however suppression of cortisol by dexamethasone (DST) is reduced when measured against standard reference intervals derived from the non-pregnant population^2,4. Whilst 24h UFC levels are elevated in second and third trimesters, measurement is considered the most reliable method of diagnosing pathological hypercortisolism during pregnancy, with concentrations greater than three times the upper limit of normal considered diagnostic^7-8. An abdominal ultrasound was performed as the initial imaging modality due to its safety in pregnancy. The safety of magnetic resonance imaging (MRI) requires further evaluation, however it has been suggested as the preferred imaging modality in pregnancy⁹. There is no consensus on management of ACC during pregnancy and a paucity of data exists due to its rarity⁹. Compared with the non-pregnant population, patients diagnosed with ACC during pregnancy are reported to have a less favourable prognosis. One retrospective cohort study identified pregnant patients were more likely to be diagnosed at a later stage compared to non-pregnant women of childbearing age and had poorer overall survival¹⁰. Fortunately, our patients tumour was detected at an early stage and the patient remains disease free after surgical treatment alone. Despite delivery of a premature infant at 33 weeks gestation, foetal outcomes were also favourable.