Content uploaded by Po Yee Loung
Author content
All content in this area was uploaded by Po Yee Loung
Content may be subject to copyright.
Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org 463
Introduction
The widespread application of sophisticated imaging techniques means adrenal nodules,
which have a prevalence of around 6% in the general population,1 are being detected
more often. Most are benign adrenocortical adenomas (ACA) but early recognition of the
much rarer adrenocortical carcinoma (ACC) is important because its mortality rate is high
when the diagnosis is delayed.2
The adrenal cortex produces cortisol, corticosterone, and C19 steroids. Adrenal
steroids are metabolised in the liver and the kidney, then excreted in the urine either in the
form of free steroids or conjugated with sulphuric or glucuronic acid. Adrenal carcinoma
tissues have been reported to express enzymes in the steroidogenic pathways aberrantly,
leading to the increased production of normal adrenal steroids as well as unusual steroids
such as metabolites of 11-deoxycortisol (compound S), steroid hormone precursors, and
neonatal steroids.2-7 Most of these adrenal steroids and metabolites can be unselectively
identified and quantified by urinary steroid profiling (USP) using gas chromatography–
mass spectrometry (GC-MS). In this article, we describe the clinical and laboratory findings
of five cases of ACC identified in our centre between 2003 and 2005, and compare the
USP findings of these patients with those of 76 patients with benign ACA and 172 healthy
controls.
Subjects
Patients with adrenocortical carcinoma
Patient 1
This was a 51-year-old female who was referred to us in 2003 with Cushing’s syndrome
(CS). She had hypertension, oligomenorrhoea, a moon face, a buffalo hump, striae over
her abdomen and thighs, truncal obesity, and mild hirsutism. Investigations revealed
hypokalaemia of 3.0 mmol/L. Her urinary-free cortisol (UFC) levels were 1733 and 2019
nmol/d on two separate collections (reference interval [RI]: 100-379). Her cortisol levels
after being given 1 mg and 8 mg overnight dexamethasone suppression were 546 nmol/L
and 545 nmol/L, respectively. Her adrenocorticotropic hormone (ACTH) level was less
than 2.2 pmol/L (RI: <10.1) and her dehydroepiandrosterone sulphate (DHEAS), erect
plasma renin activity (PRA), and aldosterone levels were all normal. Urinary steroid
profiling detected increased excretion of total cortisol metabolites (FM), and an excess
of tetrahydro-11-deoxycortisol (THS) [665 µg/d; RI: 9-59], 5-pregnene-3α,16α,20α-triol (128
µg/d; RI: 5-44), and 5-pregnene-3β,16α,20α-triol (34 µg/d; RI: <10) [Table 1, Figs 1a, 2a-c].
The FM/THS ratio was 39 (RI: 96-390). Computed tomography (CT) showed a 7.9×5.6 cm
left adrenal tumour, with calcification and possible compression on the left renal vein.
Use of urinary steroid proling for diagnosing and
monitoring adrenocortical tumours
M E D I C A L
PRACTICE
Key words
Adrenocortical adenoma; Adrenocortical
carcinoma; Steroids/urine
Hong Kong Med J 2009;15:463-70
Queen Elizabeth Hospital, 30 Gascoigne
Road, Kowloon, Hong Kong:
Department of Medicine
SC Tiu, MD, FRCP
CH Choi, FRCP, FHKAM (Medicine)
Chemical Pathology Laboratory,
Department of Pathology
AOK Chan, MB, ChB, FHKCPath
CC Shek, MB, BS, FRCPath
Department of Clinical Biochemistry,
King’s College Hospital, London, United
Kingdom
NF Taylor, PhD, FRCPath
Department of Paediatrics, Caritas
Medical Centre, Hong Kong
CY Lee, FRCP (Edin), FHKAM (Paediatrics)
Department of Paediatrics and
Adolescent Medicine, Princess Margaret
Hospital, Laichikok, Hong Kong
PY Loung, MB, ChB, MRCPCH
Correspondence to: Dr SC Tiu
E-mail: tscz01@ha.org.hk
SC Tiu
Angel OK Chan
Norman F Taylor
CY Lee
PY Loung
CH Choi
CC Shek
張秀祥
陳安琪
李靜賢
梁寶兒
蔡祥熙
石志忠
It has been suggested that urinary steroid profiling may be used to provide information
aiding the diagnosis and monitoring of adrenocortical carcinoma. Nonetheless, the abnormal
patterns suggestive of adrenal malignancy are not well defined. We retrospectively studied
the urinary steroid profiles of five patients with adrenocortical carcinoma at presentation
and at follow-up, and compared these results with those from 76 patients with benign
adrenocortical adenoma and 172 healthy controls. Three abnormal patterns of urinary
steroid excretion were identified in patients with adrenocortical carcinoma at presentation
and/or follow-up of residual disease: (1) hypersecretion in multiple steroid axes; (2) excretion
of unusual metabolites, notably 5-pregnene-3α,16α,20α-triol, 5-pregnene-3β,16α,20α-triol,
and neonatal steroid metabolites in the post-neonatal period; (3) increase of tetrahydro-11-
deoxycortisol relative to total cortisol metabolites. These preliminary findings offer ways in
which urinary steroid profiling performed using gas chromatography–mass spectrometry can
be helpful in the diagnosis and monitoring of adrenocortical carcinoma.
# Tiu et al #
464 Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org
尿中類固醇激素水平測試有助確診和監測腎上腺皮質癌；不過，反映
腎上腺癌的異常模式仍未明確。本文回顧五名腎上腺皮質癌患者於入
院和隨訪期間的尿中類固醇激素水平，並將其結果與76名良性腎上
腺皮質腺瘤患者和屬對照組的172人作比較。在因腫瘤入院和／或隨
訪期間發現腫瘤殘留的患者中，發現三種異常尿類固醇排泄模式，
包括：（1）多個類固醇軸出現分泌過多情況；（2）異常代謝物的
分泌，尤以5-pregnene-3α,16α,20α-triol、5-pregnene-3β,16α,20α-
triol，和後新生兒的新生兒類固醇代謝物為甚；（3）tetrahydro-11-
deoxycortisol於總氫化皮質醇代謝物的比例上升。這些初步研究結
果，有助以氣相色譜質譜法進行尿中類固醇激素水平測試，從而確診
和監測腎上腺皮質癌。
以尿中類固醇激素水平測試確診和監測腎上
腺皮質癌
after surgery showed markedly decreased levels
of FM compared with the preoperative sample. 5-
Pregnene-3β,16α,20α-triol was no longer detectable,
and 5-pregnene-3α,16α,20α-triol had also decreased
to normal (16 µg/d). Nevertheless, her THS remained
elevated (168 µg/d), with the FM/THS ratio remaining
low, at 17. Computed tomography of thorax, abdomen
and pelvis revealed suspicious lymph nodes in the
perisplenic region. Positron emission tomography
(PET) was negative. Three months later, multiple
peritoneal and retroperitoneal nodules consistent
with metastases were detected on CT. Chemotherapy
in the form of cisplatinum and etoposide, gemcitabine
and carboplatin, and thalidomide failed to arrest
the disease progression. A follow-up USP revealed
elevated steroid marker levels, with a high THS (1061
µg/d), 5-pregnene-3α,16α,20α-triol (281 µg/d), and 5-
pregnene-3β,16α,20α-triol (48 µg/d). Her FM remained
low (863 µg/d), with a FM/THS ratio of less than 1.
She developed multiple intracranial haemorrhages 3
months later, possibly due to brain metastases, and
died 28 months after surgery and 31 months after
presentation.
Patient 2
This was a 58-year-old female patient who presented
in May 2005 with hypertension of 195/115 mm Hg,
hypokalaemia of 2.9 mmol/L, and increased urinary
potassium excretion. She had no Cushingoid features,
and no features of virilisation. Her UFC levels were
At adrenalectomy, the tumour was partially fixed to
the left renal vein. A radical left nephrectomy and
left adrenalectomy were performed. A histological
examination revealed cords and trabeculae with a
rich vascular network and a diffuse growth pattern.
Mitosis was frequent and vascular permeation
and capsular invasion were observed. The overall
histological picture was consistent with ACC, with
tumour involvement of the raw surface, tumour
thrombus in the left renal vein, and tumour invasion
into adjacent adipose tissue. After operation, the
clinical features of CS, hypokalaemia, and UFC level
returned to normal. A follow-up USP done 6 months
Findings ACC patient No. Reference interval
1 2 3 4 Male Female
Serum studies
Spot serum cortisol level (nmol/L) 546 645 280 32.5 7-10 am: 171-536
4-8 pm: 64-340
Erect plasma renin activity (ng/mL/h) 1.42 2.91 <0.10 3.48 0.97-4.18
Erect aldosterone level (pmol/L) 317 1469 >3300 256 111-860
DHEAS level (µmol/L) 1.4 25.1 3.8 Not done 2.2-15.2 0.9-11.7
ACTH level (pmol/L) <2.2 <2.2 <2.2 <2.2 <10.1
Urine free cortisol level (nmol/d) 2019 3538 193 28 100-379
USP
AM (µg/d) 778 1001 1371 1452 1047-5509 377-3205
DHAM (µg/d) 223 44 403 939 66 273-5255 98-3020
FM (µg/d) 26186 8727 9397 3431 3504-14 867 1906-7839
THS (µg/d) 665 404 335 83 10-90 9-59
5-Pregnene-3α,16α,20α-triol (µg/d) 128 1872 209 11 15-89 5-44
5-Pregnene-3β,16α,20α-triol (µg/d) 34 1881 43 <1 <51 <10
FM/THS ratio 39 22 28 41 116-542 96-390
TABLE 1. Laboratory findings of adult patients with adrenocortical carcinoma (ACC)*†
* Serum and urine results of patient 4 were available only after removal of the primary tumour; data in bold are values that reflect increased steroid secretions
† ACTH denotes adrenocorticotropic hormone; AM total androstenedione metabolites; DHAM total dehydroepiandrosterone metabolites; DHEAS
dehydroepiandrosterone sulphate; FM total cortisol metabolites; THS tetrahydro-11-deoxycortisol; and USP urinary steroid profiling
# Urinary steroid profiling for adrenocortical tumours #
Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org 465
(a) (b) (c)
(d) (e)
FIG 1. Urinary steroid profiling chromatograms of (a) patient 1, (b) patient 2, (c) patient 5, (d) a female patient with a cortisol-secreting adrenal
adenoma, and (e) a healthy 56-year-old female volunteer
A: 5α-androstane-3α,17α-diol; B: stigmasterol; C: cholesteryl butyrate; 1: androsterone; 2: aetiocholanolone; 3: dehydroepiandrosterone; 4: 11-
hydroxyandrosterone; 5: 16α-hydroxydehydroepiandrosterone; 6: pregnanediol; 7: pregnanetriol; 8: pregnanediol; 9: androstenetriol; 10: pregnanetriol; 11:
tetrahydrocortisone; 12: tetrahydro-11-dehydrocorticosterone; 13: tetrahydrocorticosterone; 14: allo-tetrahydrocorticosterone; 15: tetrahydrocortisol; 16:
allo-tetrahydrocortisol; 17: α-cortolone; 18: β-cortolone and β-cor tol; 19: α-cortol; 20: cortisol; 21: tetrahydro-11-deoxycortisol; 22: 5-pregnene-3α,16α,20α-
triol; 23: 5-pregnene-3β,16α,20α-triol; 24: 16α-hydroxypregnenolone; N: non-steroidal contaminants
FIG 2. Levels of steroid metabolite excretion in different groups of adult female patients: (a) tetrahydro-11-deoxycortisol (THS); (b) 5-pregnene-
3α,16α,20α-triol after log10; (c) 5-pregnene-3β,16α,20α-triol after log10; and (d) total cortisol metabolites (FM)/THS ratio after log10
Group 1: female patients with adrenocortical carcinoma; group 2: female patients with adrenal incidentalomas; group 3: female patients with Conn’s
syndrome; group 4: female patients with Cushing’s syndrome; and group 5: female healthy volunteers. No 5-pregnene-3β,16α,20α-triol was detected in the
two patients with Cushing’s syndrome
1
0
Patient/subject groups
THS (µg/d)
200
400
600
2 3 4 5
(a)
1
0.00
Patient/subject groups
Log (5-pregnene-3α,16α,20α-triol) [µg/d]
1.00
2.00
3.00
2345
(b)
1
0.00
Patient/subject groups
Log (5-pregnene-3β,16α,20α-triol) [µg/d]
1.00
2.00
3.00
2 3 4 5
(c) (d)
1
1.60
Patient/subject groups
Log (FM/THS ratio) [µg/d]
2.00
2.40
2.80
2 3 4 5
# Tiu et al #
466 Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org
3408 and 3538 nmol/d on two separate occasions.
She had no diurnal cortisol rhythm, with morning
and evening serum cortisol levels of 645 nmol/L
and 615 nmol/L, respectively. After being given 1 mg
overnight dexamethasone, her cortisol level was
559 nmol/L. The ACTH level was less than 2.2 pmol/L,
and her DHEAS level was elevated at 25.1 µmol/L (RI:
0.9-11.7). Her erect PRA was 2.91 ng/mL/h (RI: 0.97-
4.18), and aldosterone level 1469 pmol/L (RI: 111-860).
Urinary steroid profiling detected marked increases
in total DHEA metabolites (DHAM) and FM, as well as
an excess of THS and 5-pregnene-3,16,20-triols (Table
1, Figs 1b, 2a-c). The FM/THS ratio was 22. She was also
excreting high levels of 16α-hydroxypregnenolone, a
neonatal steroid metabolite that is virtually absent
in the post-neonatal period. Computed tomography
showed an 8.0-cm right-sided tumour mass, with
invasion into the inferior vena cava and possibly the
right kidney. She also had multiple lung metastases
and enlarged abdominal and hilar lymph nodes.
She was given mitotane but failed to respond. Her
general condition deteriorated rapidly, and she
died 10 months after presentation. A post-mortem
examination was not performed.
Patient 3
This was a 42-year-old female patient who presented
with lower limb weakness due to hypokalaemia of
2.1 mmol/L in June 2003. Her blood pressure (BP) was
138/78 mm Hg. She had no Cushingoid features and
no features of virilisation. Investigations confirmed
urinary potassium loss and metabolic alkalosis. Her
erect PRA was <0.10 ng/mL/h and her aldosterone
level was >3300 pmol/L. A normal saline suppression
test confirmed primary aldosteronism. Her serum
and urinary cortisol levels and DHEAS were all
normal. Quantification of her steroid metabolites,
however, revealed an elevated excretion of FM.
Her total androstenedione metabolites (AM) and
DHAM levels were normal. She was also excreting
excessive levels of THS, 5-pregnene-3α,16α,20α-
triol, and 5-pregnene-3β,16α,20α-triol (335, 209, and
43 µg/d, respectively) [Table 1, Fig 2a-c]. Her FM/
THS ratio was 28. Computed tomography showed a
5.7×5.3 cm left adrenal tumour with no evidence of
internal fat. No local or distant invasion was evident.
A left adrenalectomy was performed. A histological
examination revealed a highly cellular tumour,
which consisted of broad nests and diffuse sheets of
moderately pleomorphic polygonal cells with round
stippled nuclei, distinct nucleoli, multiple foci of
necrosis, and mitoses of up to 11 per 10 high-power
fields. Lymphovascular permeation was evident in
the adjacent adrenal tissue. No capsular invasion was
seen. The overall histological picture was consistent
with ACC, stage T2.8 After surgery her potassium,
renin, and aldosterone levels returned to normal.
Computed tomography, a PET scan, and her USP
were also normal. She remained well at 52 months
of follow-up.
Patient 4
This was a 24-year-old male patient who presented
with recurrent lower abdominal pain in 2003. At
operation for presumed appendicitis, an incidental
abdominal mass was found and resected. On
pathological examination it was found an 8.2×6.2 cm
adrenal tumour with moderate nuclear pleomorphism
and distinct nucleoli, localised foci of coagulative
necrosis and vascular invasion, compatible with ACC.
When referred to us, his BP was 110/62 mm Hg, and
a physical examination revealed left gynaecomastia
only. His potassium level was normal. His UFC level
was 28 nmol/d and his DHEAS, renin and aldosterone
levels were normal. His total testosterone level was
14.6 nmol/L, and oestradiol level 94 pmol/L. His USP
was normal, though the THS was on the high side (83
µg/d, RI for males: 10-90) and his FM/THS ratio was
low (41; RI: 116-542) [Table 1, Fig 2a-c]. His 5-pregnene-
3,16,20-triol levels were not elevated. Computed
tomography showed absence of a right adrenal and a
normal left adrenal gland. Two suspicious hypodense
lesions were detected in the dome and segment 6
of his liver. Positron emission tomography and fine-
needle aspiration of the liver lesions were negative.
An exploratory hepatectomy revealed metastatic
ACC. Computed tomography performed 4 months
later revealed another lesion in segment 4 of the
liver and PET became positive 3 months afterwards.
The patient underwent further debulking surgery,
followed by adjuvant chemotherapy. Use of mitotane,
cisplatinum and etoposide or gemcitabine, and
thalidomide failed to control his disease. He died 28
months after presentation.
Patient 5
This was a Pakistani girl who presented in
September 2004 with recurrent convulsions and
impaired consciousness caused by hypertensive
encephalopathy at the age of 25 months. Her BP was
persistently around 160/110 mm Hg, and she had
increased body hair, pubic hair growth, progressive
abdominal distension, and a marked increase in
appetite and body weight from the age of 18 months.
Physical examination revealed marked hirsutism,
acne, a moon face, truncal obesity, and striae over
her abdomen. A firm abdominal mass was palpable
3 cm below the left costal margin. Her BP was
controlled with a labetolol infusion. She had severe
hypokalaemia of 1.9 mmol/L but her serum sodium
and creatinine levels were normal. Her cortisol levels
were 856 nmol/L in the morning and 696 nmol/L at
night. Her testosterone level was 19.1 nmol/L (RI: 0.1-
0.6), oestradiol level 183 pmol/L (RI: 22.0-99.1), and
# Urinary steroid profiling for adrenocortical tumours #
Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org 467
progesterone level 61.9 nmol/L (RI: 0.22-1.65). Her
urine catecholamines were normal. Urinary steroid
profiling revealed a significant increase in AM (378 µg/d;
RI: <101), FM (7458 µg/d; RI: 389-2730), corticosterone
(3260 µg/d; RI: 33-409), pregnanediol (1541 µg/d;
RI: <29), THS (1067 µg/d; RI: <49), 5-pregnene-
3α,16α,20α-triol (827 µg/d; RI: <17), and 5-pregnene-
3β,16α,20α-triol (930 µg/d; RI: <2). The concentration
of 16α-hydroxypregnenolone was highest among all
steroid metabolites (Fig 1c). Computed tomography
of her abdomen showed a large left supra-renal mass
suggestive of an adrenal tumour. Complete resection
of the mass was performed. Histology confirmed
ACC. She was put on hydrocortisone replacement
and maintained satisfactory growth. Unusual steroid
metabolites were no longer detectable 4 months and
2 years after the operation.
Patients with adrenal adenoma
The clinical and laboratory data of 83 patients with
adrenal nodules who had USP performed between
2003 and 2007 were retrieved for comparison. Of these,
seven subjects were excluded from analysis: four
had congenital adrenal hyperplasia, one had adrenal
lymphoma, one had pituitary Cushing’s disease,
and one presented with clinical features of primary
hyperaldosteronism but defaulted follow-up after 1
month. All those adenomas found to be functional
after biochemical investigations were resected, and
the diagnosis of benign functional ACA confirmed by
pathology and clinical improvement after operation.
Patients in whom the diagnosis of ‘incidentaloma’ was
made because of negative biochemical findings had
at least one repeated CT after 1 year that showed no
increase in the size of the adrenal nodule(s). Of the
76 patients with benign ACA, 57 had incidentaloma, 16
had primary aldosteronism, two had adrenal CS, and
one had subclinical CS; 35 were male and 41 female
patients. Their age ranged from 34 to 87 years, with a
mean of 57 years. In only three was the adrenal nodule
≥4 cm in diameter (4.0, 4.5, and 4.6 cm). Their clinical
and laboratory findings are listed in Tables 2 and 3.
Figure 1d shows the urine steroid gas chromatogram
of a female patient with CS. There was a gross
elevation in the level of excretion of FM; AM were
virtually absent. A gas chromatogram of a healthy
56-year-old female subject is shown in Figure 1e for
comparison. Figure 2 shows the levels of excretion
of THS, 5-pregnene-3,16,20-triols, and the FM/THS
ratio in different groups of adult female patients and
the healthy volunteers. Patients with ACC had the
highest excretion levels of these metabolites (cases
1 to 3) among all the groups of patients with adrenal
tumours and sex-matched healthy subjects (Fig 2a-c).
The log (FM/THS) ratio in all the female ACC patients
was lower than all other groups of subjects studied
(Fig 2d). The male data are listed in Table 3.
None of the subjects were taking steroids or
hormonal pills during the urine and serum studies.
Verbal consent for the investigations was given by all
subjects.
Urinary steroid profiling
The methodology used for USP and the adult RI
of steroid metabolites have been described by
our group previously.9 Reference intervals for girls
of less than 6 years were derived from USP data
from nine healthy age-matched control subjects.
The AM excretion was defined as the sum of
androsterone and aetiocholanolone. The DHAM
excretion was defined as the sum of DHEA, 16α-
hydroxydehydroepiandrosterone, and androstenetriol.
The FM excretion was defined as the sum of
tetrahydrocortisone, tetrahydrocortisol (THF), 5α-THF,
α-cortolone, β-cortolone, β-cortol, and α-cortol.
Discussion
Endocrinology services are receiving an increasing
number of patients referred for investigations after
the incidental discovery of an adrenal nodule.1 Along
with investigating hormonal hyperfunction, it is also
important to consider ACC so it can be identified
early. The only means of achieving long-term survival
of this highly aggressive condition is early detection
and radical surgical extirpation.10 Unfortunately,
like most endocrine malignancies, differentiating
between benign and malignant adrenal tumours can
be difficult before metastases develop, even with
histological examination.11 Clues that should alert
clinicians to the possibility of ACC include a tumour
size of greater than 4 to 5 cm in diameter, imaging
features of calcification, blurred margins, an irregular
shape, heterogeneous contrast enhancement and
distant metastases on CT, a high signal intensity on
MRI T2-weighted images, negative uptake on an
iodocholesterol scan, a positive uptake on an 18F-
FDG PET scan, and hormonal overproduction in more
than one adrenocortical axis.12-15 Hormone secretion
was noted in up to 79% of ACC.5,16 In this article, we
focus our discussion on the value of USP as a tool for
differentiating ACC from benign adenomas.
Determination of the USP has been proposed
as a useful tool for the detection and follow-up of
ACC5,17 but because of the low prevalence of ACC,
the abnormal patterns characterising this condition
and the best way to utilise this tool have not yet
been determined. High levels of a range of steroid
metabolites have been reported in ACC: pregnanediol
(a metabolite of pregnenolone), pregnanetriol
(a metabolite of 17-hydroxypregnenolone), THS
(metabolite of 11-deoxycortisol), androstenetriol (a
metabolite of DHEA), and 5-pregnanetriol.4-6,17,18
Our data show that USP may assist with the
# Tiu et al #
468 Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org
Clinical findings ACC patient No. ACA patients (n=76) Normal (n=172)
1 2 3 4
Age at presentation (years)
Mean±standard deviation (range)
51
-
58
-
42
-
24
-
-
57±14 (34-87)
-
43.1±13.1 (20-85)
Sex Female Female Female Male 41 Females, 35 males 89 Females, 83 males
Lateralisation of adrenal nodule Left Right Left Right 34 Left, 38 right, 4
bilateral
NA
Size of adrenal nodule (cm)†7.9 8.0 5.7 8.2 73 were <4 cm,
3 were ≥4 cm
NA
Endocrine syndrome at presentation Cushing’s Conn’s Conn’s Nil 3 Cushing’s,
16 Conn’s,
and
57 incidentaloma
NA
Duration of follow-up (months) 31
(Died)
10
(Died)
52 28
(Died)
Median 34,
IQR 24.5-47
TABLE 2. Clinical findings of adult patients with adrenocortical carcinoma (ACC) and adrenocortical adenoma (ACA)*
* NA denotes not applicable, and IQR interquartile range
† Maximum diameter on imaging studies
Laboratory findings ACA patients Reference interval
Cushing’s syndrome
(male=1†, female=2)
Conn’s syndrome
(male=8, female=8)
[mean±SD]
Incidentaloma
(male=26, female=31)
[mean±SD]
Serum/plasma/urine studies
Spot serum cortisol level
(nmol/L)
Male
Female
307
766, 787
272±142
257±74
343±273
341±130
7-10 am: 171-536
4-8 pm: 64-340
Erect plasma renin activity
(ng/mL/h)
Male
Female
ND
ND
0.81±0.80
0.21±0.22
2.12±1.85 (n=16)
2.23±2.30 (n=17)
0.97-4.18
Erect aldosterone level
(pmol/L)
Male
Female
ND
ND
751±464
1554±1181
243±123 (n=19)
445±390 (n=18)
111-860
DHEAS level (µmol/L) Male
Female
3.2
<0.5, ND
2.0, 7.8 (n=2)
0.7 (n=1)
3.0±2.2 (n=13)
1.6±0.9 (n=11)
2.2-15.2
0.9-11.7
ACTH level (pmol/L) Male
Female
<2.2
<2.2, <2.2
ND 4.4±2.3 (n=13)
3.9±2.2 (n=11)
<10.1
Urine free cortisol level
(nmol/d)
Male
Female
237
789, 1331
235±75
185±100
201±92
167±68
100-379
USP
AM (µg/d) Male
Female
1232
104, 106
1661±564
813±391
1499±955
868±903
1047-5509
377-3205
DHAM ((µg/d) Male
Female
1360
52, 133
892±795
496±325
601±451
482±998
273-5255
98-3020
FM (µg/d) Male
Female
17 302
26 685, 9881
8791±3817
4935±1974
9567±6167
6901±3178
3504-14 867
1906-7839
THS (µg/d) Male
Female
70
55, 189
48±20
40±24
52±35
38±29
10-90
9-59
5-Pregnene-3α,16α,20α-triol
(µg/d)
Male
Female
54
11, 40
31±16
21±11
35±23
18±13
15-89
5-44
5-Pregnene-3β,16α,20α-triol
(µg/d)
Male
Female
4
0, 0
4±3
2±3
4±5
1±2
<51
<10
FM/THS ratio Male
Female
246
487, 52
206±102
163±112
214±114
248±158
116-542
96-390
TABLE 3. Laboratory findings of patients with adrenocortical adenoma (ACA)*
* ACTH denotes adrenocorticotropic hormone; AM total androstenedione metabolites; DHAM total dehydroepiandrosterone metabolites; DHEAS
dehydroepiandrosterone sulphate; FM total cortisol metabolites; ND not done; SD standard deviation; THS tetrahydro-11-deoxycortisol; and USP urinary steroid
profiling
† The male patient had subclinical Cushing’s syndrome
# Urinary steroid profiling for adrenocortical tumours #
Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org 469
diagnosis of ACC in three ways. Firstly, it reveals the
presence of hormonal overproduction in more than
one adrenocortical axis, as exemplified by patients 1,
3, and 5. Urinary steroid profiling includes metabolites
not measured in conventional serum tests. In
addition, 24-hour urine collections are less affected
by diurnal variations and episodic secretions, and are
thus a more sensitive method of detecting abnormal
levels. In this aspect, USP offered less additional
information in patients 2 and 4 because serum studies
already showed hormonal overproduction in more
than one axis in patient 2, while neither serum nor
USP measurements showed elevated levels in any
axis in the latter. Patient 4 was referred after surgery
so the investigations were done in the absence of the
primary adrenal tumour. Although he still had liver
metastases, dedifferentiation of the metastasised
tissue may have led to normalisation of a previously
abnormal pattern of hormone production or
metabolism.
Secondly, USP is useful for revealing production
of unusual metabolites such as 5-pregnene-
3α,16α,20α-triol, 5-pregnene-3β,16α,20α-triol or
16α-hydroxypregnenolone in ACC, as seen in our
patients. These metabolites are unusual in the sense
that, though low levels were detectable in our healthy
subjects, they were not elevated in any of our patients
with benign adrenal nodules, be these functional
or incidental tumours. Indeed, both 5-pregnene-
3α,16α,20α-triol and 5-pregnene-3β,16α,20α-triol were
first isolated from ACC patients.19,20 The structure of
these two compounds is based on chromatographic
retention time as well as the infrared spectrum. A
USP chromatogram as shown in Figure 1 allows an
overall appreciation of grossly abnormal excretion of
particular steroids and steroid metabolites, enabling
unusual metabolites to be identified. This property
is particularly useful in conditions such as ACC, in
which relative enzyme deficiencies in the steroid
metabolic pathway may lead to the production of
unusual steroid metabolites.21
Thirdly, USP may assist with the diagnosis of
ACC by enabling detection of subtle abnormalities
in the quantities of different metabolites. Since
THS is the metabolite of one of the intermediates
(11-deoxycortisol) in the cortisol pathway, it was
not surprising to find elevated THS in ACC patients
and in patients with CS in whom FM was elevated
(Fig 2a). Previous reports on ACC have also noted
that THS is a useful tumour marker.4-6 We observed
that in ACC, THS appeared to be disproportionately
high compared with other cortisol metabolites, as
illustrated by the marked decrease in the FM/THS
ratio (Fig 2d). The decrease in this ratio was less
dramatic in patients with CS. Indeed, in patient 4 in
whom we could only perform USP after removal of
the primary adrenal tumour, suppression of the FM/
THS was the only abnormality identified.
Urinary steroid profiling can also be used
to indicate incomplete removal or relapse of
ACC, as illustrated by patient 1. In this patient, her
cushingoid features and UFC levels returned to
normal after surgery, but her urine THS and the
FM/THS ratio remained abnormal. Her subsequent
course confirmed residual disease. At the terminal
stage of her disease, a rise in THS and 5-pregnene-
3,16,20-triols, and a fall in her FM/THS ratio informed
us of her disease progression. In this patient, the USP
pattern of residual disease was similar to that before
treatment, but change in the secretion pattern has
been described as the tumour or its metastases change
in size, growth rate and differentiation.5 The value of
USP for monitoring ACC was supported by Khorram-
Manesh et al.17 Among five of their patients who had
both pre- and post-operative urine samples, residual
or recurrent ACC was identified in two because of
abnormalities in their USP despite normal imaging
studies. Because age- or sex-related normal values
for USP using GC-MS had not been established at
the time of their study, these investigators compared
postoperative with preoperative samples from the
same patient.
All five of our patients with ACC had some CT
features of adrenal carcinoma, most notably large
tumours, raising the question whether USP really played
an additional role in the diagnosis of this condition.
Nevertheless, there is no clear CT feature—apart from
the demonstration of metastases or adjacent tissue
invasion—that is diagnostic of adrenal carcinoma, so
we still have to rely on a conglomeration of data for
earlier diagnosis. This includes clinical, radiological
as well as biochemical data, and USP contributes by
providing a comprehensive biochemical assessment
of adrenal secretions. In addition, in public health
care, the waiting time for a CT scan can range
from 3 to 6 months, which is very long if one takes
into consideration the poor prognosis of adrenal
carcinoma. An abnormal USP can alert clinicians to
arrange earlier CT scanning and earlier surgery for
the patient. In fact, this was the case with one of our
patients. Urinary steroid profiling may also warn
clinicians of residual disease after surgery, before it
can be definitively identified on CT.
In conclusion, our experience supports the
notion that USP can be useful in differentiating
ACC from ACA. Our data suggest that we can
focus our attention on three aspects: (1) hormonal
hypersecretion in multiple axes, (2) excretion of
unusual metabolites, and (3) subtle alterations in the
metabolic pathways. A low FM/THS ratio and elevated
levels of 5-pregnene-3,16,20-triols are potential
tumour markers for ACC. Because of the small number
of ACC subjects in our study, we can only describe
our findings as preliminary. We cannot define the
role of USP in the diagnosis and management of
ACC at this stage, but we hope to arouse clinicians’
# Tiu et al #
470 Hong Kong Med J Vol 15 No 6 # December 2009 # www.hkmj.org
1. Thompson GB, Young WF Jr. Adrenal incidentaloma. Curr
Opin Oncol 2003;15:84-90.
2. Abiven G, Coste J, Groussin L, et al. Clinical and biological
features in the prognosis of adrenocortical cancer: poor
outcome of cortisol-secreting tumors in a series of 202
consecutive patients. J Clin Endocrinol Metab 2006;91:2650-
5.
3. Boyar RM, Nogeire C, Fukushima D, Hellman L, Fishman J.
Studies of the diurnal pattern of plasma corticosteroids and
gonadotropins in 2 cases of feminizing adrenal carcinoma:
measurements of estrogen and corticosteroid production. J
Clin Endocrinol 1977;44:39-45.
4. Minowada S, Kinoshita K, Hara M, Isurugi K, Uchikawa T,
Niijima T. Measurement of urinary steroid profile in patients
with adrenal tumor as a screening method for carcinoma.
Endocrinol Jpn 1985;32:29-37.
5. Gröndal S, Eriksson B, Hagenäs L, Werner S, Curstedt T.
Steroid profile in urine: a useful tool in the diagnosis and
follow up of adrenocortical carcinoma. Acta Endocrinol
(Copenh) 1990;122:656-63.
6. Kikuchi E, Yanaihara H, Nakashima J, et al. Urinary steroid
profile in adrenocortical tumors. Biomed Pharmacother
2000;54 Suppl 1:194s-197s.
7. Yoshimoto T, Naruse M, Ito Y, et al. Adrenocortical
carcinoma manifesting pure primary aldosteronism: a case
report and analysis of steroidogenic enzymes. J Endocrinol
Invest 2000;23:112-7.
8. Norton JA, Doppinan JI, Jensen RT. Cancer of the endocrine
system. In: Cancer: principles and practice of oncology.
Philadelphia: JB Lippincott Co; 1989: 1269-344.
9. Chan AO, Taylor NF, Tiu SC, Shek CC. Reference intervals of
urinary steroid metabolites using gas chromatography-mass
spectrometry in Chinese adults. Steroids 2008;73:828-37.
10. Ng L, Libertino JM. Adrenocortical carcinoma: diagnosis,
evaluation and treatment. J Urol 2003;169:5-11.
11. Weiss LM. Comparative histologic study of 43 metastasizing
and nonmetastasizing adrenocortical tumors. Am J Surg
Pathol 1984;8:163-9.
12. Kloos RT, Gross MD, Francis IR, Korobkin M, Shapiro
B. Incidentally discovered adrenal masses. Endocr Rev
1995;16:460-84.
13. Allolio B, Fassnacht M. Clinical review: adrenocortical
carcinoma: clinical update. J Clin Endocrinol Metab
2006;91:2027-37.
14. Schteingart DE. Current perspective in the diagnosis and
treatment of adrenocortical carcinoma. Rev Endocr Metab
Disord 2001;2:323-33.
15. Roman S. Adrenocortical carcinoma. Curr Opin Oncol
2006;18:36-42.
16. Tauchmanovà L, Colao A, Marzano LA, et al. Adrenocortical
carcinomas: twelve-year prospective experience. World J
Surg 2004;28:896-903.
17. Khorram-Manesh A, Ahlman H, Jansson S, et al.
Adrenocortical carcinoma: surgery and mitotane for
treatment and steroid profiles for follow-up. World J Surg
1998;22:605-11; discussion 611-2.
18. Honour JW, Price DA, Taylor NF, Marsden HB, Grant
DB. Steroid biochemistry of virilising adrenal tumours in
childhood. Eur J Pediatr 1984;142:165-9.
19. Fukushima DK, Smulowitz M, Williams KI. ∆5-Pregnene-
3∝,16∝,20∝-triol and related steroids in adrenocortical
carcinoma. J Biol Chem 1961;236:3147-50.
20. Hirschmann H, Hirschmann FB. Steroid excretion in a case
of adrenocortical carcinoma. IV. delta 5 pregnanetriol-3
beta, 16 alpha, 20 alpha. J Biol Chem 1950;184:259-82.
21. Lipsett MB, Hertz R, Ross GT. Clinical and pathophysiologic
aspects of adrenocortical carcinoma. Am J Med 1963;35:374-
83.
References
awareness of this tool, so that more experience can be
accumulated in the future. Adrenocortical carcinoma
is a rare, highly malignant yet curable disease whose
preoperative identification by clinical and imaging
features is difficult. Thus, it is important to make a
concerted effort to document the role of different
investigation modalities in differentiating between
benign and malignant adrenal nodules.
