Steroid profile in urine: a useful tool in the diagnosis and follow up of adrenocortical carcinoma
Staffan Gröndal1, Barbro Eriksson4, Lars Hagenäs2, Sigbritt Werner3 and Tore Curstedt1
Departments of Clinical Chemistry’, Pediatric Medicine and Endocrinology’, Karolinska Hospital, Stockholm, and Department of Internal Medicine1, University Hospital, Uppsala, Sweden
Abstract. The urinary steroid profile was determined in 24 patients with adrenocortical carcinoma. Seventeen of the patients had Cushing’s syndrome, virilization or feminization, and 7 had no signs of endocrine disease. Seven of the 11 patients still alive are free of disease, after a follow-up period of 5-75 months. The steroid profile varied widely between the patients with adrenocortical carcinoma. Patients with Cushing’s syndrome had in- creased levels of cortisol metabolites and those with viril- ism had raised excretion of androgen metabolites. Six of the patients with adrenocortical carcinoma showed normal values of these metabolites. In 23 of the 24 pa- tients the excretion of 36-hydroxy-5-ene steroids and/or metabolites of cortisol precursors, such as tetrahydro-11- deoxycortisol, were significantly increased, compared with healthy controls or patients with adrenal adenomas. These findings suggest a relative deficit or low activity of 36-hydroxysteroid dehydrogenase/ 43-4 isomerase and/or 11ß-hydroxylase in tumour tissue. In the single patient where the steroid profile failed to indicate malignancy, hypercortisolism was seen and the tumour mass was small. The steroid excretion normalized after radical sur- gery and decreased in patients responding to chemother- apy. During recurred disease the metabolites of 36-hy- droxy-5-ene steroids and/or cortisol precursors increased, but in some patients the excretory pattern then was dif- ferent from that seen before treatment.
Adrenocortical carcinoma is a rare disease with an incidence of approximately 2 per million inhabit- ants. The symptoms are often diffuse and 50% of the patients have no signs of excessive hormone production (1). Many patients have metastases when subjected to surgery, and treatment of recur-
rent disease is often delayed. Benign and malig- nant adrenocortical tumours are often difficult to differentiate by histopathology, and the predictive value of findings such as nuclear polymorphism, mitotic rate, necrosis, vascular and capsular inva- sion are debated (2). Various chemical methods have been used to analyse steroids in serum and urine, but owing to the low incidence of adreno- cortical carcinoma these studies often represent only a few cases (3-7).
The aim of the present investigation was to study the excretory pattern of urinary steroid metabolites in patients with adrenocortical carcinoma and to evaluate whether increased levels of some specific steroids could be of value to discriminate malignant from benign tumours.
Patients and Methods
Patients
We analysed urinary samples from 24 patients, 18 females and 6 males, with adrenocortical carcinoma (Table 1). Samples were taken before and after surgery in 19 pa- tients and after surgery in 5 patients. Diagnosis was based mainly on tumour size (median 12 cm in diameter, range 3-20) and 11 of the 24 patients had metastases at diag- nosis. In the remaining 13 ones, 4 had capsular invasion, 2 numerous mitoses, one angioinvasion; all had nuclear pleomorphism in varying degrees at histopathologic ex- amination. Five patients had Cushing’s syndrome, 5 pa- tients virilism, 5 patients Cushing’s syndrome and viril- ism. One man had gynecomastia, one woman hyperal-
Table 1. Clinical data in 24 patients with adrenocortical carcinoma. Tumour size was determined at surgery except in patients Nos. 12 and 23, where it was estimated by computed tomography. Asterisks indicate patients lacking preoperative determination of a urinary steroid profile.
| Pat. No. | Sex | Age | Endocrine symptoms | Tumour size (cm) | Treatment | Outcome | Survival after diagnosis (months) | |||
|---|---|---|---|---|---|---|---|---|---|---|
| Cushing's syndrome | Virilism | Other | Surgery | Chemo- therapy | ||||||
| 1 | F | 5 | + | 4×4 | + | Alive | 12 | |||
| 2 | F | 5 | + | + | 15×11 | + | + | + | 14 | |
| 3 | F | 8 | + | + | 12×12 | + | + | + | 5 | |
| 4 | F | 9 | + | 7×5 | + | Alive | 75 | |||
| 5 | F | 32 | 13×10 | + | + | Alive | 43 | |||
| 6 | F | 33 | + | 14×12 | + | + | + | 40 | ||
| 7 | M | 33 | 7×5 | + | Alive | 27 | ||||
| 8 | F | 40 | + | 12×12 | + | + | Alive+metastases | 13* | ||
| 9 | M | 42 | 15×10 | + | + | Alive | 5 | |||
| 10 | F | 47 | + | 10×7 | + | + | Alive+metastases | 7 | ||
| 11 | F | 49 | + | 17×7 | + | + | + | 8* | ||
| 12 | M | 53 | >10×10 | + | ៛ | 10* | ||||
| 13 | M | 54 | +1) | 10×9 | + | + | Alive+metastases | 6 | ||
| 14 | F | 57 | + | + | 11×9 | + | + | ៛ | 15* | |
| 15 | F | 58 | 16×14 | + | Alive | 40 | ||||
| 16 | F | 59 | + | +2) | 7×6 | + | + | + | 13 | |
| 17 | F | 60 | + | + | 3×3 | + | + | + | 24 | |
| 18 | M | 60 | + | 14×8 | + | + | + | 1* | ||
| 19 | F | 60 | 9×8 | + | + | Alive+ metastases | 5 | |||
| 20 | M | 61 | 15×12 | + | + | 2 | ||||
| 21 | F | 63 | + | 20×20 | + | + | + | 6 | ||
| 22 | F | 65 | + | + | 10×8 | + | + | 1 | ||
| 23 | F | 68 | + | >10×10 | + | + | 2 | |||
| 24 | F | 69 | + | 10×6 | + | Alive | 27 | |||
1) Gynecomastia; 2) Hyperaldosteronism
dosteronism and virilism. Seven patients had no signs of endocrine dysfunction. Age at diagnosis ranged between 5 and 69 years.
Controls
Ten patients with Cushing’s syndrome, and 8 patients with primary hyperaldosteronism, all owing to surgically verified adrenal adenomas were also studied with regard to urinary steroid profile. As healthy controls 16 fertile women volunteered, all employees at the department of Clinical Chemistry (8).
Analytical procedure
Free and conjugated steroids were isolated from 2.5 ml of a 24-h urinary sample by chromatography on a Sep-pak C18 cartridge (Waters Ass, USA) (9). The steroid glucu- ronides were hydrolyzed with Helix Pomatia digestive juice (Pharmindustrie, France) and the steroid sulphates
with tetrahydrofurane acidified with 4 mol/l aqueous sul- phuric acid (10). The free steroids obtained were passed through a TEAP-Sephadex LH-20 column in carbonate form (10,11). The solvents were removed and stigmaste- rol was added as an internal standard. The steroids were analysed as O-methyloximetrimethylsilyl derivatives by gas-liquid chromatography, using a 30 m × 0.32 mm fused silica capillary column coated with methylsilicone (DB-1, J&W Scientific, USA). The identity of the gas- liquid chromatography peaks was confirmed by gas chro- matography-mass spectrometry, using a Finnigan 1050 instrument (Finnigan MAT, San Jose, Ca.) with an elec- tron energy of 40 eV and equipped with a similar column as described above. A well trained laboratory technician can perform this analytical procedure of altogether 14 samples during a period of five days. The intra-assay co- efficient of variation was below 10% and the inter-assay coefficient of variation was between 6 and 21% for the different steroids.
Free, glucuronized, mono- and disulphated steroids were separated in 5 patients with adrenocortical carci- noma and analysed as previously described (11).
Results
Steroid profile in controls Urinary steroids from healthy subjects are domi- nated by metabolites of cortisol and androgens
(Fig. 1A). In the 10 patients with cortisol-producing adrenocortical adenomas the same steroids were identified, but the levels of cortisol metabolites were raised and the tetrahydrocortisol-tetrahydro- cortisone ratio was increased (median 1.3, range 1.0-1.7; reference value <1.0). A representative steroid profile from one of these 10 patients with Cushing’s syndrome is shown in Fig. 1B. The 8 patients with aldosterone-secreting adrenal ade- nomas had normal levels of cortisol and androgen
A
B
14
13
Std
1
13
Std
5
1
14
15
17
15
2
16
16
17
2
5
18
18
3
8
15
20
25
30
35
40
15
20
25
30
35
40
Minutes
Minutes
D
C
1
14
12
Std
2
11
13
9
3
1
Std
7
2
17
5
5
11
8
15
6
13
7
16
18
10
8
10
18
14
15
17
5
10
15
20
25
30
15
20
25
30
35
40
Minutes
Minutes
45
| Patient No. | 36-hydroxy-5- ene steroids | Pregnane- triol | THS | |||||
|---|---|---|---|---|---|---|---|---|
| DHA | Androst-5- ene-38,178- diol | 16a- hydroxy- DHA | Androst-5- ene-36,16a, 17B-triol | Pregn-5- ene-38,20a- diol | Pregn-5- ene-38,17a, 20a-triol | |||
| 1 | 1.4 | <0.1 | 3.0 | 2.2 | 7.5 | 9.2 | 6.7 | 1.3 |
| 2 | 23 | 13 | 12 | 4.2 | 1.3 | 9.2 | 14 | 16 |
| 3 | 770 | 100 | 83 | 41 | 34 | 64 | 3.0 | <0.1 |
| 4 | 15 | 2.9 | 5.4 | 3.5 | 12 | 22 | 6.6 | 30 |
| 5 | 100 | 17 | 18 | 5.9 | <0.1 | 85 | 9.2 | 15 |
| 6 | 2.1 | 1.9 | 2.4 | 6.6 | <0.1 | 2.5 | 6.0 | 9.0 |
| 7 | 13 | <0.1 | 5.7 | 2.4 | <0.1 | 4.4 | 7.2 | 26 |
| 8 | 88 | 13 | 38 | 11 | 43 | 32 | 16 | 140 |
| 9 | 0.8 | <0.1 | <0.1 | <0.1 | <0.1 | 4.2 | 7.5 | 24 |
| 10 | <0.1 | <0.1 | <0.1 | 4.0 | <0.1 | 5.8 | 13 | 25 |
| 11 | <0.1 | <0.1 | <0.1 | 61 | 25 | 18 | 5.5 | 39 |
| 12 | 3.7 | <0.1 | <0.1 | 1.6 | <0.1 | 0.7 | 5.7 | 25 |
| 13 | 34 | 3.9 | 12 | 3.4 | 7.2 | 29 | 17 | 36 |
| 14 | 88 | 24 | 13 | 7.8 | 3.4 | 13 | 4.7 | 6.5 |
| 15 | 1.3 | 1.7 | 11 | 16 | 12 | 15 | 3.0 | <0.1 |
| 16 | 3.8 | <0.1 | <0.1 | 3.9 | <0.1 | <0.1 | 4.4 | 21 |
| 17 | 0.5 | <0.1 | <0.1 | <0.1 | <0.1 | <0.1 | 2.2 | 1.3 |
| 18 | 98 | 71 | 26 | 4.7 | <0.1 | 140 | 36 | 38 |
| 19 | 8.0 | <0.1 | <0.1 | <0.1 | <0.1 | 2.4 | 7.3 | 8.4 |
| 20 | 25 | 11 | <0.1 | 12 | 14 | 38 | 28 | 14 |
| 21 | 13 | 9.6 | 24 | 18 | 9.0 | 28 | 160 | 280 |
| 22 | 11 | 25 | 18 | 32 | 63 | 48 | 15 | 250 |
| 23 | 41 | 32 | 25 | 65 | <0.1 | 33 | 15 | <0.1 |
| 24 | 0.3 | <0.1 | <0.1 | <0.1 | <0.1 | <0.1 | 10 | 39 |
| Reference value | <4.3 | <1 | <2 | <1 | <1 | <1 | <4.2 | <0.1 |
metabolites (chromatogram not shown). Metabo- lites of aldosterone precursors were found, but the levels were low and not significantly higher than those found in healthy controls. Estrogens were not detected in any patient with the present method owing to their relatively low excretion.
Steroid profile in patients with adrenocortical carcinoma
In the patients with adrenocortical carcinoma sev- eral other urinary steroids were also identified. In- creased levels of metabolites of cortisol precursors, such as tetrahydro-11-deoxycortisol (THS) and pregnanetriol, and/or of 3ß-hydroxy-5-ene steroids were observed. The major 3ß-hydroxy-5-ene ste- roids were either metabolites of dehydroepian-
drosterone (DHA) (e.g. 16a-hydroxy-DHA, androst-5-ene-36,170-diol and androst-5-ene- 36,16a,17ß-triol), pregnenolone (pregn-5-ene- 36,20a-diol) or 17a-hydroxypregnenolone (pregn- 5-ene-36,17a,20a-triol). The steroid levels showed great variation, but 23 of the 24 patients had raised levels of THS and/or 30-hydroxy-5-ene steroids (Table 2). According to the steroid excretion, the patients could be divided into three groups. One group had increased levels of mainly THS (Fig. 1C), another group of mainly 30-hydroxy-5-ene steroids, and the third group of both THS and 3B-hydroxy-5-ene steroids (Fig. 1D). THS was re- covered as a glucuronide, and 30-hydroxy-5-ene steroids as mono- or disulphates.
Androgen metabolites were increased in 13 of 24
| Peak No. | Trivial name | Generic name |
|---|---|---|
| 1 | Androsterone | 5a-Androstane-3a-ol-17-one |
| 2 | Etiocholanolone | 5B-Androstane-3a-ol-17-one |
| 3 | Dehydroepiandrosterone (DHA) | Androst-5-ene-3B-ol-17-one |
| 4 | A5diol | Androst-5-ene-36,17ß-diol |
| 5 | f 1 11-hydroxyandrosterone 17a-hydroxypregnanolone | 5a-Androstane-3a, 118-diol-17-one 5a-Pregnane-3a, 17a-diol-20-one |
| 6 | 16a-hydroxy-DHA | Androst-5-ene-36,16a-diol-17-one |
| 7 | Pregnanediol | 5B-Pregnane-3a,20a,diol |
| 8 | Pregnanetriol | 56-Pregnane-3a, 17a,20a-triol |
| 9 | P5diol | Pregn-5-ene-38,20a-diol |
| 10 | A5triol | Androst-5-ene-36,16a,17ß-triol |
| 11 | Tetrahydro-11-deoxycortisol (THS) | 5B-Pregnane-3a, 17a,21- triol-20-one |
| 12 | Potriol | Pregn-5-ene-38,17a,20a-triol |
| 13 | Tetrahydrocortisone | 5B-Pregnane-3a,17a,21-triol-11,20-dione |
| 14 | Tetrahydrocortisol | 5B-Pregnane-3a, 116,17a,21-tetrol-20-one |
| 15 | Allo-Tetrahydrocortisol | 5a-Pregnane-3a, 118,17a,21-tetrol-20-one |
| 16 | a-Cortolone | 5B-Pregnane-3a,17a,20a,21-tetrol-11-one |
| 17 | [ 1 B-Cortolone ß-Cortol | 58-Pregnane-3a,17a,208,21-tetrol-11-one 5B-Pregnane-3a,118,17a,208,21-pentol |
| 18 | a-Cortol | 5B-Pregnane-3a,118,17a,20a,21-pentol |
patients. The levels of cortisol metabolites were raised in 11 of 24 patients, but only 3 of them had a tetrahydrocortisol-tetrahydrocortisone
ratio more than 1 (median 0.8, range 0.5-1.5).
One patient (No. 17) had a rapid development of Cushing’s syndrome and increased levels of cortisol metabolites, but initially only a minimal increase in THS, as seen in patients with cortisol-producing adrenal adenomas. She was therefore considered to have a benign adrenocortical tumour, but when the disease recurred the THS levels were signifi- cantly increased, with increased cortisol excretion.
Steroid profiles in recurrent disease
In 19 of the 24 patients urinary samples were ob- tained for analysis both before and after surgery or chemotherapy. The steroid profile normalized after radical surgery but raised levels of 30-hy- droxy-5-ene steroids and/or THS or other metab- olites of cortisol precursors were seen when the disease recurred.
A 33-year-old woman (No. 6, Tables 1 and 2) with a large localized tumour and Cushing’s syn- drome had increased levels of cortisol metabolites and THS preoperatively. After a radical tumour resection the steroid profile normalized. Nine
months later, she had no endocrine symptoms and cortisol metabolites were low, but the excretion of THS was increased to 4.5 umol/24 h. A recurrent disease was suggested and further investigations demonstrated a local tumour recurrence, pulmo- nary and hepatic metastases. Chemotherapy with 1,1-dichloro-2-(o-chlorophenyl)-2-(chlorophenyl)- ethane (o,p’DDD) and streptozotocin was initiated. The THS levels decreased initially, but remained increased and the patient died 40 months postope- ratively.
In 7 patients (Nos. 1, 4, 5, 7, 9, 15, 24) the steroid excretion has remained normal postoperatively and all are free of disease after 5-75 months (Table 1). In these patients, all treated for adrenocortical carcinoma, a urinary steroid profile is determined at least twice a year.
Discussion
Steroids secreted from the adrenals, testes and ova- ries are metabolized in the liver and excreted in the urine as free steroids or conjugated to glucuronic or sulphuric acid. Cortisol metabolites, which are the dominating steroids in normal subjects are
almost exclusively excreted as glucuronides. Enzy- matic digestion with Helix Pomatia digestive juice hydrolyzes the steroid glucuronides efficiently, but the capacity for sulphated conjugates is limited (10). In patients with adrenocortical carcinoma, with a high excretion of steroid sulphates, digestion with only glucuronidase or Helix Pomatia digestive juice leads to a low recovery and an underestima- tion of 36-hydroxy-5-ene steroids. However, sol- volysis in tetrahydrofurane-sulphuric acid decon- jugates most of the steroid sulphates and the free steroids obtainol are then detected by gas-liquid chromatography (10).
The increased excretion of THS and/or 36-hy- droxy-5-ene steroids thus made it possible to dis- tinguish nearly all patients with carcinomas from patients with adenomas and healthy subjects before operation. The presence of a malignant tumour was, however, obvious in the 11 patients where me- tastases were found preoperatively, and the histo- pathological examination demonstrated confident signs of malignancy in another 7 patients. Two of the remaining 6 patients have developed me- tastases which indicate that a urinary steroid profile gives additional information of the biological be- haviour in adrenocortical tumours.
Despite a limited follow-up time, none of the pa- tients with Cushing’s syndrome are alive without recurrent disease, whereas 3 patients with virilism
and 4 patients without signs of hormone hyperse- cretion are alive without metastases. This tendency of a better prognosis for patients without Cushing’s syndrome is in agreement with a retrospective study of 54 patients with adrenocortical carcinoma, who were followed for a long period of time (12).
High levels of 11-deoxycortisol in plasma (13) or of THS in urine (14) are described in patients with Cushing’s syndrome owing to adrenocortical carci- noma. However, in our study we found that the excretion of THS was independent of the endo- crine symptoms, indicating that the deficiency of the mitochondrial 11ß-hydroxylase concerns most patients with adrenocortical carcinoma and not ex- clusively those with Cushing’s syndrome ( Fig. 2). Analysis of serum steroids in some of the patients shows only moderately increased levels of 11-de- oxycortisol, also in the patients with a high excre- tion of THS in the urine (S Gröndal and T Cur- stedt, to be published). This discrepancy may be due to a rapid metabolism of 11-deoxycortisol to THS-glucuronide, which is almost instantly ex- creted in the urine. This interpretation may also explain why the increased level of 11-deoxycortisol in plasma could not be used to distinguish benign from malignant adrenocortical tumours in chil- dren (5).
The excessive amounts of pregnanediol and pregnanetriol in some patients could reflect a de-
Cholesterol
P diol
Pregnenolone
5 P triol
1
2
Pregnanediol
Progesterone
17xOH-
pregnenolone
2
1
Pregnanetriol
17xOH-progesterone
DHA
3
THS
11-deoxycortisol
16x OH-DHA
4
Cortisol
5 A diol
5 A triol
creased 17a- or 21-hydroxylase activity (Fig. 2). In- creased levels of metabolites of cortisol precursors were found in patients with both high and low ex- cretion of cortisol metabolites (Tables 1 and 2). This implies that the expression of the enzyme de- ficiency could be of varying degree.
A relative lack or low activity of the microsomal enzyme 3ß-hydroxysteroid dehydrogenase/ 4 5-4 iso- merase decreases the transformation of pregneno- lone to progesterone, which can explain the in- creased levels of 30-hydroxy-5-ene steroids. The levels of these steroids and thereby even this enzyme deficiency showed a wide individual vari- ation. Thirteen of the patients with adrenocortical carcinoma had significantly increased excretion of both 36-hydroxy-5-ene steroids and metabolites of cortisol precursors, indicating that a combination of enzyme deficiencies was common. These multi- ple enzyme deficiencies have also been demon- strated in vitro, where decreased 11ß-hydroxylase, 17a-hydroxylase and 3ß-hydroxysteroid dehy- drogenase/45-4 isomerase activities were found (15,16).
In some patients, the steroid profile was differ- ent in primary and recurrent disease and during chemotherapy. In one patient (No. 6) both cortisol metabolites and THS were increased preoperati- vely, but when the tumour recurred only the THS levels were raised. When chemotherapy was initi- ated in another patient (No. 5) a normalization of DHA and androst-5-ene-30-17ß-diol was accompa- nied by an increase in the 16a-oxygenation of these steroids (8). This indicates that it is not sufficient to analyse a single steroid as a tumour marker for recurrent disease. Determination of a steroid pro- file is then necessary to follow the biochemical ac- tivity of the tumour.
Both patients with and without adrenal tumours may have pathological steroid profiles. Increased excretion of metabolites of cortisol precursors or 3ß-hydroxy-5-ene steroids are seen in patients with inborn deficiencies of steroid hydroxylases or 3- hydroxysteroid dehydrogenase/45-4 isomerase (17). The steroid profile is in these patients ex- plained by a deficiency or low activity of one spe- cific enzyme. The patients have no adrenal tumour and the disorder is usually discovered in early childhood.
Hirsutism and polycystic ovary syndrome are as- sociated with a mild hypersecretion of adrenal an- drogens in women (18). An increase in 3B-hydroxy- 5-ene steroids in some of these patients may sug-
gest a relative 36-hydroxysteroid dehydrogenase/ 4 5-4
isomerase deficiency (data not shown). These disorders can be distinguished from adrenocortical carcinoma by the history of disease, the clinical signs and symptoms, the specific steroid profile, and the absence of a unilateral adrenal mass.
Steroid-producing tumours may also originate from other tissues than the adrenal (19). In one female patient (data not shown) we found an ex- tremely high excretion of androgen metabolites, such as androsterone, etiocholanolone, and DHA. This patient had a malignant ovarian tumour and after surgery the steroid profile normalized.
The incidence of adrenal tumours has increased with the widespread use of computed tomography (20). The tumours in two of the patients in this study (Nos. 7 and 9) were incidentally discovered during work-up for renal disorders. Both patients had large tumours and increased levels of THS and some 3B-hydroxy-5-ene steroids.
In the single patient (No. 17) judged to have a benign tumour from the steroid profile, the histo- pathologic examination revealed an adrenal tumour of 3x3 cm with invasive tumour growth and local metastases. Not until the tumour re- curred and the tumour mass was significantly larger did the steroid profile demonstrate in- creased levels of THS, indicating a malignant tumour.
Tumour size has so far been the best single pre- dictor for malignancy of adrenocortical tumours, despite careful studies of several histopathological parameters (21). The present results suggest that a urinary steroid profile is a reliable tool to distin- guish malignant from benign primary adrenocor- tical tumours. However, determination of a steroid profile is time-consuming and cannot be used for screening of all patients with adrenocortical disor- ders. As a diagnostic tool, these steroid analyses are recommended for patients with large (>3 cm) ad- renal tumours and for patients with progressive clinical symptoms, indicating excessive steroid pro- duction. Thus, development of simplified methods for analysis of the specific cortisol precursors and 36-hydroxy-5-ene steroids in urine and serum should be of clinical importance.
Acknowledgment
Supported by grants from the Karolinska Institute and the Swedish Medical Research Council (No. 2330).
References
1. Nader S, Hickey RC, Sellin RV, Samaan NA. Adrenal cortical carcinoma-A study of 77 cases. Cancer 1980; 52:707-11.
2. Weiss LM. Comparative study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984;8:163-9.
3. Huhtaniemi I, Kahri AI, Pelkonen R, Salmenpera M, Sivula A, Vihko R. Ultrastructural and steroidogenic characteristics of an androgen-producing adrenocor- tical tumor. Clin Endocrinol (Oxf) 1978;8:305-14.
4. Honour JW, Price DA, Taylor NF, Marsden HB, Grant DB. Steroid biochemistry of virilizing adrenal tumours in childhood. Eur J Pediatr 1984; 142:165-9.
5. Doerr HG, Sippell WG, Drop SLS, Bidlingmaier F, Knorr D. Evidence of 11 beta-hydroxylase deficiency in childhood adrenocortical tumors. Cancer 1987; 60:1625-9.
6. Minowada S, Kinoshita K, Hara M, Isurugi K, Uchi- kawa, T, Niijima T. Measurement of urinary steroid profile in patients with adrenal tumor as screening method for carcinoma. Endocrinol Jpn 1985;32:29- 37.
7. Homoki J, Hol R, Teller WM. Harnsteroidprofile bei Cushing Syndrom und Tumoren der Nebennieren- rinde. Klin Wochenschr 1987;65:719-26.
8. Eriksson B, Öberg K, Curstedt T, et al. Treatment of hormone producing adrenocortical cancer with o,p’DDD and streptozocin. Cancer 1987;59:1398- 403.
9. Shackleton CHL, Whitney JO. Use of Sep-pak® car- tridges for urinary steroid extraction: evaluation of the method for use prior to gas chromatographic analysis. Clin Chim Acta 1980;107:231-43.
10. Axelson M, Sahlberg B-L, Sjövall J. Analysis of pro- files of conjugated steroids in urine by ion-exchange separation and gas chromatography-mass spectro- metry. J Chromatogr Biomed Appl 1981;224:355- 70.
11. Cronholm T, Curstedt T, Schmidt DN, Sjövall J. Ste- roid profiles in urine and plasma of alcoholics during withdrawal. Alcohol 1985;2:677-82.
12. Gröndal S, Cedermark B, Eriksson B, et al. Adreno- cortical carcinoma. A retrospective study of a rare tumor with a poor prognosis. Eur J Surg Oncol (in press).
13. Lipsett MB, Hertz R, Ross GT. Clinical and pa- thophysiologic aspects of adrenocortical carcinoma. Am J Med 1963;35:374-83.
14. Levine AC, Mitty HA, Gabrilove JL. Steroid content of the peripheral and adrenal vein in Cushing’s syn- drome due to adrenocortical adenoma and carci- noma. J Urol 1988; 140:11-5.
15. Degenhart HJ, Drop SLS, Hoogerbrugge J, van Seters AP, de Vries HR. In vitro studies of enzymatic activities in human adrenal tumors. Horm Res 1982;16:96-9.
16. d’Agata RD, Malozowski S, Cassola F, Loriaux D. Ste- roid biosynthesis in human adrenal tumors. Horm Metab Res 1987;19:386-8.
17. Knorr D, Bidlingmaier F, Kuhnle U. Diagnosis and monitoring of therapy of the various enzymatic de- fects causing congenital adrenal hyperplasia by semi- automatic capillary gas-liquid chromatography. Horm Res 1982; 16:201-8.
18. McKenna TJ. Pathogenesis and treatment of polycys- tic ovary syndrome. N Engl J Med 1988;9:558-62.
19. Mattox JH, Phelan S. The evaluation of adult females with testosterone producing neoplasms of the adre- nal cortex. Surg Gynecol Obstet 1987;164:98-101.
20. Copeland PM. The incidentally discovered adrenal mass. Ann Intern Med 1983;98:940-5.
21. Cagle PT, Hough AJ, Pysher J, et al. Comparison of adrenal cortical tumors in children and adults. Cancer 1986;57:2235-7.
Received December 1st, 1989. Accepted March 2nd, 1990.
Dr Staffan Gröndal, Department of Surgery, Karolinska Hospital, S-104 01 Stockholm, Sweden.