J. Biochem. Biophys.

Methods 61 (2004) 47 – 56
www.elsevier.com/locate/jbbm

HPLC–RIA analysis of steroid hormone profile in

a virilizing stromal tumor of the ovary
M. Szécsi a,*, I. Tóth a, J. Gardi a, T. Nyári b, J. Julesz a
a
Endocrine Unit and Research Laboratory, University of Szeged, 8 Korányi fasor, H-6720 Szeged, Hungary
b
Department of Medical Informatics, University of Szeged, 9 Korányi fasor, H-6720 Szeged, Hungary
Received 4 November 2003; received in revised form 7 April 2004; accepted 9 April 2004

Abstract

The pathological steroid biosynthesis of a virilizing ovarian tumor was examined via high
performance liquid chromatography – radioimmunoassay (HPLC – RIA) determination of the intra-
tissular steroid concentrations. Sex cord-stromal tumor of the ovary was obtained surgically from an
18-year-old female patient with extremely high androst-4-ene-3,17-dione (4-en-dione) and
testosterone (Test) blood serum levels. The tissue specimen was extracted with ethyl acetate and
the extract was then purified on a C18 mini-column with methanol – water eluents. Steroids were
isolated by reversed-phase HPLC on a C18-silica gel column with 51%, 55% and 64% v/v
methanol – water eluents. Steroids in the collected eluent fractions were detected by the radioactivity
of tritiated internal standards and then quantified by specific RIAs. In the tumor specimen, very high
17a-hydroxyprogesterone (17-OH-Prog; 6300 fmol/g), dehydro-epiandrosterone (2870 fmol/g),
androst-4-ene-3,17-dione (3000 fmol/g), testosterone (5700 fmol/g) concentrations, and less
progesterone (PROG; 320 fmol/g) and androst-5-ene-3h,17h-diol (5-en-diol; 320 fmol/g), were
determined. Tissue levels of 5a-dihydrotestosterone (DHT), 5a-androstane-3a,17h-diol (3a-diol),
5a-androstane-3h,17h-diol (3h-diol), and 17h-estradiol were found to be 71, 20, 28, and 12 fmol/g,
respectively. Steroid profile analysis verified a pathological steroid biosynthesis in the ovarian tumor
and suggested that the 17a-hydroxylase (17a-H), 17,20-lyase (17,20-L), and 3h-hydroxysteroid
dehydrogenase/D5 – 4-isomerase (D5-3h-HSD) activities were particularly elevated in this tumorous
tissue. Present data demonstrate that the analysis of intratissular steroid profile by a HPLC – RIA
method may valuably contribute to the steroidal pathophysiology of endocrine tumors.
D 2004 Elsevier B.V. All rights reserved.

Keywords: Steroid hormone biosynthesis; Steroid isolation and purification; High-performance liquid
chromatography; Radioimmunoassay; Ovarian neoplasms; Sex cord-stromal tumor; Virilism

* Corresponding author. Tel.: +36-62-545215; fax: +36-62-545211.

E-mail address: szecsim@endoc.szote.u-szeged.hu (M. Szécsi).

0165-022X/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jbbm.2004.04.008
48 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56

1. Introduction

Determination of intratissular steroid concentrations is a convenient way of investigat-

ing physiology of steroid-hormone-secreting endocrine glands or steroid-hormone-depen-
dent target organs [1,2]. Adequate studies in steroid metabolism require specific
measurement of major steroid hormones together with their precursors and metabolites.
Tissue sample is a complex matrix containing large amount of lipid substances and cross-
reacting steroid compounds. Quantitation of intratissular steroids, therefore, is a challeng-
ing task and requires highly specific and usually meticulous analytical techniques. High-
performance liquid chromatography coupled to radioimmunoassay (HPLC – RIA) soon
became popular in this field as it combines low detection limit of RIA with enhanced
specificity due to the chromatographic separation [3]. A great variety of methods has been
developed on this ground. Extraction and prepurification procedures have been modified
according to the diverse tissue specimens with various matrix characteristics [4,5].
Isolation and separation may be optimized with regard to the chromatographic behaviour
of the steroid set to be determined [6,7]. Steroid profile studies of human tissues were
primarily focussed on the androgen action in prostate tumors[8 –11]. In some cases,
hormone metabolism in the brain [12,13], adrenal gland, and ovary [14], as well as steroid
levels in blood [15 –17], was investigated with HPLC – RIA.
Ovary is the major source of gonadal steroids in the female organism. Approximately
50% of androst-4-ene-3,17-dione (4-en-dione), 40% of 5a-dihydrotestosterone (DHT),
25% of testosterone (Test) and 10% of dehydroepiandrosterone (DHEA) total production
originated from the ovary. Depending on the menstrual cycle, an estimated 25– 95% of the
17a-hydroxyprogesterone (17-OH-Prog), 70 – 95% of the progesterone (Prog), and 85–
90% of the etradiol (E2) are secreted by this organ in a healthy woman in reproductive age
[18,19]. Functioning ovarian neoplasms in humans are rare. These tumors can produce
estrogens, progesterones, androgens, or a combination thereof [2,20]. Functioning ovarian
tumors are, therefore, usually associated with elevated steroid hormone levels in blood and
consequently with severe endocrine manifestations.
Blood serum steroids or urinary metabolites do not reflect ovarian steroid biosynthesis
properly. Steroid release from other organs and biotransformation in peripheral or target
tissues also contribute to the overall steroid production. Investigations of primary steroid
productions of normal and pathological ovaries have been attempted with measurement of
steroid concentrations in the ovarian venous effluent [18,21 – 28]. These studies provided
significant insight into the in vivo steroidogenic activity, but application of the method
remained limited. Major disadvantages of this technique are that blood sample has to be
taken at a difficult catheterization and the steroid concentrations refer only to the
fluctuating momentary secretory activity of the ovary.
Ovarian steroidogenesis can be examined via determination of intratissular steroid
profile. This paper describes a HPLC – RIA method with a feasible tissue sample
processing for the determination of the key gonadal steroids in the metabolic pathway
(Fig. 1). With this method, the steroid profile, i.e., the tissue concentrations of Prog, 17-
OH-Prog, DHEA, androst-5-ene-3h,17h-diol (5-en-diol), 4-en-dione, Test, DHT, 5a-
androstane-3a,17h-diol (3a-diol), 5a-androstane-3h,17h-diol (3h-diol), and E2 were
determined in a virilizing sex cord-stromal tumor of the ovary.
M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56 49

Fig. 1. Major pathways of gonadal steroid biosynthesis.

50 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56

2. Method

2.1. Tissue sample

Ovarian tumor specimen was obtained from a unilateral oophorectomy. The 18-year-old
virilized female patient had secondary amenorrhea and diagnostic hormone assays
measured highly elevated Prog, 17-OH-Prog, DHEA, 4-en-dione, Test levels, and normal
E2 concentration in her peripheral blood serum (Table 1). Following surgery, serum steroid
levels returned to the normal range and the patient resumed normal ovarian function.
Metastasis or recurrence was not observed. The resected tumor was 13117 cm in size
and designated a sex cord-stromal tumor of the ovary. Histology of the tumor resembled a
sclerosing stromal tumor, however, the tumor tissue was atypic and contained poorly
differentiated heterolog elements.

2.2. Steroid standards

Radioactive steroids with specific activities of 1.5– 2.5 TBq/mmol were purchased
from New England Nuclear (Great Britain) and the Radiochemical Centre, Amersham
(Great Britain). Non-radoactive steroids were obtained from Sigma (USA) and Makor
Chemical (Israel).

2.3. Extraction and purification

Tissue specimen was minced and four samples of 1.0– 1.2 g weight were prepared. For
monitoring the procedural loss, 50 000 dpm tritiated internal standards of each steroid were
added to the extraction tubes. Samples were extracted with 43 ml ethyl acetate via
homogenization with Ultra-Turrax and 3 – 3 min ultrasonic agitation. Solvents were
separated from the tissue by centrifugation and evaporated under nitrogen. Dry residue
was then resuspended twice in 30% v/v methanol – water and loaded on a 500-mg Amprep
C18 microcolumn. Impurities were washed out with 10 ml water and the steroids were
eluted in 2.5 ml 90% v/v methanol – water mixture.

Table 1
Hormone diagnostic findings in blood serum of patient with ovarian tumor before and 7 days after the unilateral
oophorectomy
Steroid Concentration in peripheral blood serum (fmol/ml) Normal range in blood
Before surgery After surgery serum (fmol/ml)

Prog 4.5 1.1 0.6 – 3.4*

17-OH-Prog 12 2.7 1.2 – 4.2*
DHEA 66 9.7 7.9 – 25
4-en-dione 36 3.5 1.8 – 9.0
Test 13 0.62 0.9 – 2.8
E2 0.18 0.34 0.1 – 0.5
*
Follicular phase.
 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56 51

2.4. HPLC separation

Prepurified eluate was dried up and then resolved and filtered through a 0.2-ı̀m
membrane. The sample in 400 ı̀l volume was injected on a 2504.6 mm column packed
with C18-silica gel (Bioszeparációs Rt., Hungary). The adsorbent had 5 Am particle size
and 100 Å pore diameter. Chromatogram was developed using a three-step gradient
elution. The eluents were between 0 and 65 min 51% v/v, between 65 and 100 min 55% v/
v, and between 100 and 120 min 65% v/v methanol in 0.01 M sodium dihydrogenphos-
phate buffer. Chromatography was performed with a Knauer (Germany) HPLC system, at
ambient temperature with 1.0 ml/min flow rate. Eluted fractions of 1.0 ml were collected
with a HeliFrac automated collector (Pharmacia-LKB, Sweden).
Detection and identification of eluted steroids were based on measurement of
radioactivity of the tritiated internal standards. Radioactivities of 100-ı̀l aliquots of the
fractions were measured by liquid scintillation counting and the residual parts of four
fractions containing the highest amount of steroids were combined for RIA quantification.

2.5. Radioimmunoassay

Isolated steroids in the eluent were dried up and then resolved in RIA buffer. Aliquots
were taken for quantification in specific RIA methods and also for determination of
recoveries by the help of the radioactive internal standards. RIA methods were
developed in our laboratory previously [29,30] and performed on the basis of WHO
principles [31]. Due to the high cross-reactivities of 5-en-diol, 3a-diol and 3h-diol, the
same antiserum (an anti-5-en-diol antiserum) was used for the measurement of these
steroid diols. In these cases, specificity was guaranteed by the chromatographic
separation. Each RIA incubation lasted 18 h at 4 jC in 0.1 M sodium dihydrogen-
phosphate RIA buffer. The free fractions of steroids were eliminated by adsorption on
dextran-coated charcoal, and the radioactivities of the bound fractions were measured by
liquid scintillation.

2.6. Calculations

RIA data were analyzed with a computer program based on calculation of the bound
fraction percentage and a simple spline fitting method for the calibration curve. Radio-
activities and doses of the internal standards were corrected and methodological back-
ground of blank samples were also subtracted [32]. Steroid content of the tissue sample
was calculated with consideration of the recoveries of internal standards. Results were
given in fmol/g wet weight tissue, as the mean value and standard error of two to four
parallel determinations.

3. Results

HPLC method with C18-silica gel adsorbent and a three-step gradient methanol –water
eluent system was found suitable for the separation of 10 main steroid hormones extracted
52 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56

from the ovarian tissue. Strongly apolar C18-silica differentiated well the adsorption of
lipophylic steroids and methanol proved to be highly selective for these molecules
differing in numbers and positions of hydroxy groups. Satisfactory resolution was
achieved with a delicate optimization of the methanol content of the eluent and steroids
were collected in pure fractions prior to radioimmunoassay (Fig. 2). The peak pair of 17-
OH-Prog and Test was an exception with close retention volumes varied between 56 and
58 and between 60 and 62 ml. Coelution of the internal standards was prevented in the
way that 17-OH-Prog and Test were isolated from separate parallel samples in different
HPLC runs. Specific quantifications of these steroids were secured by the very small
(<0.01%) cross-reactivities in the radioimmunoassays.
Tissue extract prepurified on an Amprep C18 mini-column was free from impurities
disturbing chromatographic isolation and the HPLC was reproducible. (Data on the
HPLC –RIA method and the steroid profile results are listed in Table 2.) The simple
prepurification procedure was also efficient enough to allow the on-line photometric
detection and peak annotation of the ultraviolet-active 4-ene-3-oxo steroids present in the
tissue in higher amount.
Steroids were isolated efficiently by this method: recoveries were mainly between 40%
and 60% for the entire procedure. HPLC recoveries of E2 were considerably low
compared to the other steroids, presumably due to the irreversible adsorption caused by
the phenolic hydroxy group of E2. Differences in the recoveries of certain steroids and in
individual samples were monitored by the radioactive internal standards. Recovery
correction improved precision of the method which was usually below 15% variation
constant. Further advantage of the internal standards is that the HPLC –RIA method can be

Fig. 2. HPLC isolation of steroid hormones. Steroids extracted from ovarian tumor tissue were chromatographed
on a 2504.6 mm C18-silica gel column with methanol – 0.01 M NaH2PO4 buffer eluent (51%, 55% and 64% v/v
methanol for 0 – 65 – 100 – 120 min). Eluted steroids were identified by the radioactivity of tritiated internal
standards and the fractions shaded were combined for RIA. (Coeluting 17-OH-Prog and Test were isolated from
separate samples in different HPLC runs). Steroids bearing UV-absorbing 3-oxo-4-ene structure were also
detected by spectrometry at 254 nm wavelength.
 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56 53

Table 2
Steroid profiling results of the ovarian tumor tissue measured with HPLC – RIA method
Steroid Retention volume Total recovery Limit of Constant of ConcentrationFS.E.
(ml of the main (%) detection variation in tumorous ovarian
fraction) (fmol) (c.v.%) tissue, n=2 – 4 (fmol/g)
Prog 112 – 113 36 – 51 0.7 7.7 320F12
17-OH-Prog 56 – 58 19 – 42 0.7 8.0 6300F360
DHEA 67 – 70 38 – 44 0.5 9.9 2870F140
5-en-diol 77 – 78 49 – 58 0.7 16 320F25
4-en-dione 46 – 48 64 – 66 0.4 1.6 3000F28
Test 60 – 62 38 – 59 0.4 13 5700F380
DHT 89 – 91 26 – 35 0.4 13 71F5
3a-diol 107 – 108 41 – 56 0.7 22 20F2
3h-diol 94 – 96 25 – 43 0.9 6.6 28F1
E2 53 – 54 27 – 29 0.5 0.9 12F0.1

used for determination of steroids in other type of tissue samples without performing
laborious validation processes [33,34].
Steroid profile of the ovarian tumor obtained with the HPLC –RIA method (Fig. 2)
revealed very high concentrations of 17-OH-Prog (6300 fmol/g), DHEA (2870 fmol/g), 4-
en-dione (3000 fmol/g), and testosterone (5700 fmol/g) in the tissue specimen. Concen-
trations of Prog and 5-en-diol were determined 320 fmol/g equally. Tissue levels of DHT,
3a-diol, 3h-diol, and E2 were considerably lower and found to be 71, 20, 28, and 12 fmol/
g, respectively.

4. Discussion

Ovarian steroidogenesis was thoroughly investigated in the last decades. Theca and
granulosa cells are believed the primary sites of steroidogenesis and the ‘‘two-cell
concept’’ is widely acknowledged to explain their activities [35,36]. Theca cells possess
all the enzyme machinery necessary for the biosynthesis of sexual steroids from
cholesterol and androgenic C19 steroids—mainly 4-en-dione and also DHEA, Test, and
DHT—are predominantly produced by the theca cells. Granulosa cells, on the other hand,
are deficient in 17a-hydroxylase (17a-H) and 17,20-lyase (17,20-L) activities, and they
are not capable of transforming C21 to C19 steroids. Granulosa cells are the principal
source of progestagens. Estrogens are formed by cooperation of the two cell types: Thecal
4-en-dione is converted to estrone by the highly active aromatase of the granulosa cells
and estrone is then readily transformed to E2 by the 17h-hydroxysteroid oxidoreductase
(17-HSOR). Much less attention has been paid to the steroid synthesis of ovarian stroma.
Steroidogenic potential of stromal cells is acknowledged, but the extent is in doubt and the
mechanisms are poorly understood [37 – 42].
Virilizing sex cord-stromal tumors are uncommon neoplasms of the ovary. Five to eight
percent of all ovarian neoplasms are sex cord-stromal tumors and most of them are
estrogen producing granulosa cell tumors. Only two or three percent of sex cord-stromal
tumors secrete androgens and cause virilization with various degrees [2,20,43]. Sex cord-
54 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56

stromal tumors often contain heterolog elements as a result of differentiation stimulated by

the stromal cell proliferation. Systematic investigations and classification are difficult
because of wide variety of the morphologic features and because some of the tumors have
only been identified in a few patients. Sex cord-stromal tumors are usually benign.
Immunhistochemistries revealed 17,20-L and 3h-hydroxysteroid dehydrogenase/D5 – 4-
isomerase (D5-3h-HSD) enzymes in majority of these tumorous stromal tissues [2,44,45].
Primary steroid biosynthesis in the ovary can be examined via determination of steroid
concentrations in the tissue itself. Surgically removed tumorous ovaries provide adequate
experimental specimens for studies of the steroidal pathophysiology. Evaluation of the
intratissular steroid results are not easy in the absence of normal values, but the steroid
profile of the tissue well represents the overall processes of production, accumulation and
secretion.
Steroid profile of the sex cord-stromal tumor of the ovary measured with the HPLC –
RIA method revealed very high testosterone concentration in the tissue. Test concen-
trations in the tissue verified the ovarian overproduction of this androgenic steroid which
led to the virilization of the patient. In addition to Test, other androgenic steroids (DHEA
and 4-en-dione) as well as 17-OH-Prog were found in highest concentrations in the tissue.
The significant accumulations of these steroid hormones can be results of high activities of
17a-H as well as 17,20-L and D5-3h-HSD enzymes which were reported abundant in
ovarian sex cord-stromal tumors [2,44,45]. 5a-Reduced C19 steroids and E2 were found in
considerably lower levels demonstrating low or missing activities of the 5a-reductase and
aromatase enzymes in this ovarian tumor. Tumor concentrations of 17-OH-Prog, 4-en-
dione, and Test are found very similar with those measured by Kessler [14] with HPLC –
RIA and HPLC –ultraviolet spectrometry in a histologically unidentified ovarian tumor.
(17-OH-Prog, 4-en-dione, and Test concentrations published by Kessler were 6300, 3000,
and 5700 fmol/g, respectively).
Peripheral blood serum levels of Test, DHEA, 4-en-dione, and 17-OH-Prog measured
with routine diagnostic assays were also elevated before the tumor resection (Table 1).
Relative concentrations of steroids in the tumor and in the serum differed markedly and
this discrepancy highlights again the importance of steroid determination in the tumor
itself.
Steroid profile results of the tumorous ovarian tissue demonstrate that the HPLC –RIA
method is suitable for accurate and reproducible determination of intratissular steroids.
Ethyl acetate extraction and the prepurification on a mini-column isolated steroids with
high yields and removed lipid impurities efficiently. Reversed-phase HPLC with metha-
nol – water eluent system was reliable for the separation and identification of steroid
hormones feature very similar chromatographic behaviour. RIA methods quantified the
isolated steroids specifically in a wide dose range. Recovery correction using the tritiated
internal standards improved precision. This internal standard technique makes the HPLC –
RIA method applicable for the determination of other steroids in different tissue samples
with no further validations.
In conclusion, intratissular steroid concentrations verified a pathological steroid
biosynthesis in the virilizing sex cord-stromal tumor of the ovary. Very high concentrations
of 17-OH-Prog and C19 androgenic steroids suggest high 17a-H, 17,20-L and D5-3h-HSD
activity in the tissue. Present data demonstrate that analysis of intratissular steroid profile
 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56 55

by a HPLC – RIA method offers an adequate tool for studying steroidal pathophysiology of
endocrine organs and tumors with presumed steroidogenic activities.

Acknowledgements

The authors wish to thank Mrs. Katalin Dobó-Bárkányi and Mrs. Katalin Varga-Bodó
for their excellent technical assistance.

References

[1] Hanukoglu I. Steroidogenic enzymes: structure, function, and role in regulation of steroid hormone bio-
synthesis. J Steroid Biochem Mol Biol 1992;43:779 – 804.
[2] Sasano H. Steroid-producing tissues. In: Kovacs K, Asa S, editors. Functional Endocrine Pathology.
Geneva: Blackwell; 1998. p. 707 – 714.
[3] Schöneshöfer M, Dulce HJ. Comparison of different high-performance liquid chromatographic systems for
the purification of adrenal and gonadal steroids prior to immunoassay. J Chromatogr 1979;164:17 – 28.
[4] Sjövall J, Axelson M. Newer approaches to the isolation, identification, and quantification of steroids in
biological materials. Vitam Horm 1982;39:31 – 144.
[5] al-Alousi LM, Anderson RA. A relatively simple and rapid multi-component method for analysis of steroid
profiles in blood, fecal and liver samples. Steroids 2002;67:269 – 275.
[6] Heftmann E, Hunter IR. High-pressure liquid chromatography of steroids. J Chromatogr 1979;165:283 – 299.
[7] Capp MW, Simonian MH. Separation of the major adrenal steroids by reversed-phase high-performance
liquid chromatography. Anal Biochem 1985;147:374 – 381.
[8] Albert J, Geller J, Geller S, Lopez D. Prostate concentrations of endogenous androgens by radioimmuno-
assay. J Steroid Biochem 1976;7:301 – 307.
[9] Hammond GL. Endogenous steroid levels in the human prostate from birth to old age: a comparison of
normal and diseased tissues. J Endocrinol 1978;78:7 – 19.
[10] Bartsch W, Kozak I, Gorenflos P, Becker H, Voigt KD. Concentrations of 3beta-hydroxy androgens in
epithelium and stroma of benign hyperplastic and normal human prostate. Prostate 1986;8:3 – 10.
[11] Belanger A, Couture J, Caron S, Roy R. Determination of nonconjugated and conjugated steroid levels in
plasma and prostate after separation on C-18 columns. Ann N Y Acad Sci 1990;595:251 – 259.
[12] Hammond GL, Hirvonen J, Vihko R. Progesterone, androsterone, testosterone, 5a-dihydrotestosterone
and androsterone concentrations in specific regions of the human brain. J Steroid Biochem
1983;18:185 – 189.
[13] Appelblad P, Irgum K. Separation and detection of neuroactive steroids from biological matrices. J Chro-
matogr A 2002;955:151 – 182.
[14] Kessler MJ. Analysis of steroids from normal and tumor tissue by HPLC. Clin Chim Acta 1982;125:21 – 30.
[15] Schöneshöfer M, Fenner A, Dulce HJ. Assessment of eleven adrenal steroids from a single serum sample by
combination of automatic high-performance liquid chromatography and radioimmunoassay (HPLC – RIA).
J Steroid Biochem 1981;14:377 – 386.
[16] Boschi S, De Iasio R, Mesini P, Bolelli GF, Sciajno R, Pasquali R, et al. Measurement of steroid hormones
in plasma by isocratic high performance liquid chromatography coupled to radioimmunoassay. Clin Chim
Acta 1994;231:107 – 113.
[17] Fernandes VT, Ribeiro-Neto LM, Lima SB, Vieira JG, Verreschi IT, Kater CE. Reversed-phase high-
performance liquid chromatography separation of adrenal steroids prior to radioimmunoassay: applica-
tion in congenital adrenal hyperplasia. J Chromatogr Sci 2003;41:251 – 254.
[18] Tagatz GE, Gurpide E. Hormone secretion by the normal human ovary. In: Greep RO, editor. Endocrino-
logy. London: American Physiological Society; 1973. p. 603 – 613.
[19] Azziz R. The role of the ovary in the genesis of hyperandrogenism. In: Adashi EY, Leung PCK,
editors. The Ovary. Washington, DC: Raven Press; 1993. p. 582 – 595.
56 M. Szécsi et al. / J. Biochem. Biophys. Methods 61 (2004) 47–56

[20] Aiman J. Virilizing ovarian tumors. Clin Obstet gynecol 1991;34:835 – 847.
[21] Meldrum DR, Abraham GE. Peripheral and ovarian venous concentrations of various steroid hormones in
virilizing ovarian tumors. Obstet Gynecol 1979;53:36 – 43.
[22] Laatikainen TJ, Apter DL, Paavonen JA, Wahlstrom TR. Steroids in ovarian and peripheral venous blood in
polycystic ovarian disease. Clin Endocrinol (Oxf.) 1980;13:125 – 134.
[23] Imperato-McGinley J, Peterson RE, Dawood MY, Zullo M, Kramer E, Saxena BB, et al. Steroid hormone
secretion from a virilizing lipoid cell tumor of the ovary. Obstet Gynecol 1981;57:525 – 531.
[24] Munemura M, Inoue S, Himeno R, Koyama N, Mizumoto J, Maeyama M, et al. Endocrine studies on
ovarian androblastomas (Sertoli – Leydig tumors). J Endocrinol Invest 1984;7:615 – 622.
[25] Moltz L, Sörensen R, Schwartz U, Hammerstein J. Ovarian and adrenal vein steroids in healthy women with
ovulatory cycles—selective catheterization findings. J Steroid Biochem 1984;20:901 – 905.
[26] van den Brule FA, Bourque J, Gaspard UJ, Hustin JF. Massive ovarian edema with androgen secretion.
Horm Res 1994;41:209 – 214.
[27] Sainz de la Cuesta R, Rock JA, Homsi R, Symonds DA, Horowitz IR. Analysis for steroid production of a
Sertoli – Leydig cell tumor. Gynecol Oncol 1994;52:276 – 280.
[28] Kaltsas GA, Mukherjee JJ, Kola B, Isidori AM, Hanson JA, Dacie JE, et al. Is ovarian and adrenal venous
catheterization and sampling helpful in the investigation of hyperandrogenic women? Clin Endocrinol
2003;59:34 – 43.
[29] Faredin I, Tóth I, Janáky T, Tekeres A, Faluhelyi S, Szabó I, et al. A direct radioimmunoassay of proges-
terone in bovine serum and milk. Acta Vet Acad Sci Hung 1987;35:373 – 382.
[30] Faredin I, Tóth I, Szécsi M, Somlai Cs. Androstenedione radioimmunoassay from human serum and its
application in clinical practice. Exp Lab Med 1998;25:10 – 20 (Hungarian).
[31] Sufi SB, Donaldson A., Jeffcoate S.L. WHO Matched Reagent Programme Method Manual. Geneva:
WHO; 1991.
[32] Szécsi M, Tóth I, Faredin I. The levels of androgens in rat prostate and serum after castration. Endocr Regul
1995;29:107 – 113.
[33] Szécsi M, Tóth I, Rengei B, Scultéty S, Faredin I. Androgenic steroids in normal and hyperplastic tissue
samples from human prostates. In: Görög S, Heftman E, editors. Advances in Steroid Analysis. Geneva:
Akadémiai Kiadó; 1993. p. 469 – 474.
[34] László FA, Tóth S, Kocsis J, Pávó I, Szécsi M. Testosterone-secreting gonadotropin-responsive adrenal
adenoma and its treatment with the antiandrogen flutamide. J Endocrinol Invest 2001;24:622 – 627.
[35] Moon YS, Duleba AJ. Comparative studies of androgen metabolism in theca and granulosa cells of human
follicles in vitro. Steroids 1982;39:419 – 430.
[36] Adashi EY. Endocrinology of the ovary. Hum Reprod 1994;9:36 – 51.
[37] Barbieri RL. Human ovarian 17-ketosteroid oxidoreductase: unique characteristics of the granulosa – luteal
cell and stromal enzyme. Am J Obstet Gynecol 1992;166:1117 – 1123.
[38] Nagamani M, Stuart CA, Doherty MG. Increased steroid production by the ovarian stromal tissue of
postmenopausal women with endometrial cancer. J Clin Endocrinol Metab 1992;74:172 – 176.
[39] Sluijmer AV, Heineman MJ, Koudstaal J, Theunissen PH, de Jong FH, Evers JL. Relation between ovarian
production of estrone, estradiol, testosterone, and androstenedione and the ovarian degree of stromal
hyperplasia in postmenopausal women. Menopause 1998;5:207 – 210.
[40] Marelli G, Carinelli S, Mariani A, Frigerio L, Ferrari A. Sclerosing stromal tumor of the ovary. Report of
eight cases and review of the literature. Eur J Obstet Gynecol Reprod Biol 1998;76:85 – 89.
[41] Leong S, Trivedi AN. A case of ovarian stromal hyperplasia causing hirsutism in a post-menopausal
woman. Aust N Z J Obstet Gynaecol 2001;41:102 – 103.
[42] Jongen VH, Hollema H, van der Zee AG, Santema JG, Heineman MJ. Ovarian stromal hyperplasia and
ovarian vein steroid levels in relation to endometrioid endometrial cancer. BJOG 2003;110:690 – 695.
[43] Serov SF, Scully RE, Sobin LH. International Histological Classification of Tumours No. 9. Histological
typing of Ovarian Tumours. Budapest: World Health Organizationm; 1973.
[44] Costa MJ, Morris R, Sasano H. Sex steroid biosynthesis enzymes in ovarian sex-cord stromal tumors. Int J
Gynecol Pathol 1994;13:109 – 119.
[45] Ishikura H, Sasano H. Histopathologic and immunhistochemical study of steroidogenic cells in the stroma
of ovarian tumors. Int J Gynecol Pathol 1998;17:261 – 265.