1 Department of Obstetrics and Gynaecology, Division of Gynaecological Endocrinology and Reproductive Medicine, 2 Department of Internal Medicine I, Division of Oncology, Medical University of Vienna, Waehringer Gürtel 1820, 1090 Vienna, 3 Department of Medical Statistics, Vienna University, Schwarzspanierstrasse 6, 1090 Vienna, Austria and 4 Department of Obstetrics and Gynaecology, Medical Faculty of the Technical University of Aachen, Pauwelsstrasse 30, 52057 Aachen, Germany
5 To whom correspondence should be addressed. Email: walter.tschugguel{at}meduniwien.ac.at
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: aromatase/endometriosis/endometrium/menstrual characteristics/predictive model
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Endometriosis is regarded as a polygenically inherited disease of complex multifactorial aetiology (Olive and Schwartz, 1993; Bulun et al., 1997
; Arvanitis et al., 2003
). Among all theories, transplantation of endometrial tissue via retrograde menstruation is the most widely accepted hypothesis (Sampson, 1927
). Although retrograde menstruation can be observed in up to 90% of all cycling women, the prevalence of endometriosis is far lower (Halme et al., 1984
; Lebovic et al., 2001
). Hence, besides a massive exposure to menstrual efflux, a defect of clearance from pelvic peritoneum suggests that other factors, such as aberrant expression of certain cytokines and tissue matrix metalloproteinases, as well as aberrant expression of aromatase and deficiency of 17
-hydroxysteroid dehydrogenase, may contribute to the development of the disease (Osteen et al., 1996
; Zeitoun et al., 1998
; Zeitoun and Bulun, 1999
; Bergqvist et al., 2001
). Eutopic endometrium of patients seems to share alterations with ectopic endometrial implants. There is evidently a difference between eutopic endometrium of women suffering from endometriosis and disease-free women (Bulun et al., 2002
; Leyendecker et al., 2002
). Hence, a predisposing factor for endometriosis might be found within eutopic endometrium, probably leading to the establishment of ectopic disease.
The key enzyme converting C19 steroids into estrogen is cytochrome P450 (CYP 19) aromatase, converting androstendione into estrone and testosterone into estradiol. Aberrant aromatase expression has been clearly documented in ectopically located endometriotic lesions (Noble et al., 1997; Fang et al., 2002
) as well as in eutopic endometrium of women suffering from endometriosis or other estrogen-dependent diseases (Noble et al., 1996
; Kitawaki et al., 1997
). First reports on the usefulness of eutopic aromatase as a target for a screening test are contradictory (Kitawaki et al., 1999
; Dheenadayalu et al., 2002
). However, apart from other screening markers suggested (Bedaiwy et al., 2002
; Harada et al., 2002
; Brosens et al., 2003
), patients' histories provide valuable information about clinical symptoms typically associated with endometriosis (Cramer et al., 1986
; Darrow et al., 1993
; Kuohung et al., 2002
).
The aim of this study was to prospectively evaluate whether screening for the presence of CYP 19 aromatase in eutopic endometrium in combination with select patients' characteristics might assist in developing a predictive model for endometriosis.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Sixty-four women aged 2148 years (mean: 32.9 years) attending this tertiary care centre for diagnosis and/or treatment of unexplained infertility, dysmenorrhoea, dyspareunia or chronic pelvic pain were selected (Table I). Prior to enrolment, estrogen-dependent diseases (EDD) other than endometriosis were clinically excluded by performing gynaecological exploration as well as transvaginal ultrasound examination. None of the patients was diagnosed with endometriosis or another EDD prior to inclusion or received any endocrine therapy such as GnRH analogues or danazol at that time. Five women using oral contraceptives were excluded from this trial due to possibly altered endocrine regulation and aromatase activity in the endometrium. Menstrual characteristics as well as other demographic data were collected in a standardized interview. Furthermore, all subjects completed a questionnaire including visual analogue scales (VAS) to assess the level of dysmenorrhoea, dyspareunia and pelvic pain irrespective of menstrual symptoms and sexual intercourse during a period of 30 days.
|
RNA extraction and real-time RTPCR analysis
For RNA extraction, frozen tissue samples were triturated and total RNA was extracted using the TRI Reagent method by MRC (Molecular Research Centre, Inc., USA). RNA concentration was determined by measuring the optical density at 260 nm. RNA was reverse-transcribed into first strand cDNA using Superscript (Invitrogen Ltd, UK).
The resulting cDNA was amplified and quantified using the TaqMan® Gene Expression Assay for the detection of exon 9 of the human CYP 19 aromatase gene in real-time RTPCR according to the standardized protocol (Applied Biosystems, USA). The human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified in parallel reactions as a housekeeping reference gene serving as an internal control for the quantity and quality of the cDNA.
The quantification of the target gene expression (CYP 19 aromatase) was performed according to the delta-delta CT method for relative quantification (Livak and Schmittgen, 2001): expression levels of aromatase were related to the expression of a reference housekeeping gene (GAPDH) in the same sample. The difference in expression (=ratio) was calculated to explore the relative quantity of aromatase expression in each endometrial biopsy specimen.
Immunohistochemical analysis
Frozen tissue samples were cut into 6 µm sections and fixed using acetone, air-dried and washed in phosphate-buffered saline (pH 7.4). To block endogenous peroxidase activity, the sections were treated with 3% H2O2 in methanol for 15 min at room temperature. To block non-specific protein binding the sections were incubated with Dako Protein Block Serum-Free (Dako, USA) for 10 min at room temperature. For the detection of human aromatase cytochrome P450 we used the monoclonal mouse anti-human antibody MAb 2-C-2 (HauptmannWoodward Medical Research Inc., USA) in a 1:500 dilution. All sections were incubated overnight at 4 °C.
Visualization of the immunoreaction was carried out with the ChemMate Antibody Detection Kit (Dako) according to the manufacturer's instructions. Gill's haematoxylin (Merck, Germany) was applied as a nuclear counterstain. As a positive control for aromatase expression, human placental tissue was obtained from a patient at 39 weeks of gestation with an uneventful obstetric history. Sections serving as negative control for aromatase cytochrome P450 were incubated with a 1:500 dilution of non-immunized mouse immunoglobulin G serum (Dako).
Six sections of each tissue sample were examined microscopically by three independent trained physicians who were unaware of the patients' history. Each slide was graded into negative (0) or positive (1). Due to small sample sizes, no further analysis of subgroups was performed.
Western blot analysis
Whole cell extracts were prepared from endometrial tissue, homogenized in lysis buffer [10 mmol/l Tris, 30 mmol/l Na4P2O7, 50 mmol/l NaCl, 50 mmol/l NaF, 1% Triton X-100, pH 7.1; 1 mmol/l phenylmethylsulphonyl fluoride; 0.1 mmol/l Na3VO4 and protease inhibitor cocktail (1:25; Boehringer Ingelheim GmbH, Germany)]. Insoluble material was removed by centrifugation (14 000 rpm for 5 min) and the protein concentration was determined by the Bradford assay (Bio-Rad Laboratories Inc., USA). Thirty micrograms of protein extract were separated on 10% SDSPAGE gels in Mini-Protean 3 Gel Tank (Bio-Rad) at 200 V for 1 h, followed by electrophoretic transfer to nitrocellulose membranes (Amersham Pharmacia, Germany). Membranes were blocked for 1 h at room temperature using 5% non-fat milk powder dissolved in Tris-buffered saline containing 0.02% Tween 20 and incubated overnight at 4 °C with the goat polyclonal antiserum directed against human aromatase at a 1:400 dilution (Santa Cruz Biotechnology Inc., USA). Membranes were then incubated for 1 h at room temperature in peroxidase-conjugated donkey anti-goat secondary antibody (Jackson Immuno Research, USA). Proteins were detected by using an enhanced chemiluminescence detection system (Pierce Chemical Co., USA).
Statistical analysis
For the comparison of metric variables between groups we used the Student's t-test, whereas for categorical variables the 2-test or Fisher's exact test was applied. To evaluate associations between continuous variables, Pearson's correlation coefficient was computed. The prevalence of disease according to the study sample was 52.1%.
For the assessment of a predictive model for endometriosis, stepwise logistic regression was performed for clinically relevant patients' characteristics and the applied testing methods; the significance level for a variable to enter and stay in this model was 15%. The discriminating power of the resulting multiple logistic model outcome (linear combination of the selected explanatory variables in the logistic model) was visualized by a receiver operating characteristics curve (ROC) and quantified by the area under the curve (AUC).
P<5% was considered to be statistically significant. All computations were performed using SAS 8.2 software (SAS Institute, USA).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The mean age of all patients was 32.9 years (±6.4 years), 25 out of 48 (52.1%) women were diagnosed with endometriosis; 23 out of 48 (47.9%) had no evidence of disease. Diagnosis of endometriosis was performed according to the current gold standard (Brosens, 1997). Consequently, three out of 48 patients, who revealed putative endometriotic lesions at laparoscopy that could not be confirmed during further histopathologic assessment, were considered to be disease-free.
Additional pelvic disease was discovered in two patients; one patient was diagnosed with a mild myomatosis of the uterus (which was not detected by pelvic ultrasound prior to surgery) in coexistence with endometriosis. The other patient undergoing laparoscopy for diagnostic infertility work-up and treatment of an ovarian cyst appeared to have a borderline tumour in one ovary, whereas endometriosis could not be confirmed.
Expression and localization of aromatase in eutopic endometrium
According to real-time RTPCR, aromatase mRNA expression was found in all the 48 samples, ranging from 1- to 100.7-fold expression after relative quantification according to the delta-delta CT method (Figure 1). The quantity of CYP 19 aromatase transcripts was strongly associated with the presence of endometriosis (Pearson's correlation coefficient r=0.49, P<0.001). When calculated with regard to the stage of disease, the correlation coefficient increased to r=0.51 (P<0.001).
|
Fifteen out of 25 (60.0%) cases and three out of 23 (13.1%) controls were aromatase positive (rAFS I and II: 50.0%; rAFS III and IV: 63.9%). There was a clear correlation between aromatase protein in endometrial stromal and/or endometrial epithelial cells and the presence of endometriosis (r=0.48, P<0.001). The sensitivity of IHC testing for aromatase expression in endometrial epithelial and stromal cells was 60.0%, the specificity 86.9%; the positive and negative predictive values were 83.3% and 66.7% respectively. Representative sections are shown in Figure 2.
|
To confirm the specificity of our IHC data, we performed western blot analyses in a series of select samples. The sample size was limited to the residual material available following generation of data that pertain to the model. Hence, in 21 of the 64 samples we had sufficient material to perform western analyses. We found aromatase protein in 12 of the 21 samples (57.1%). Ten of these 12 patients suffered from endometriosis whereas in two patients showing aromatase protein in western blot analysis endometriosis was not confirmed (sensitivity: 83.3%) (Figure 3). Of 10 endometriosis patients, in nine aromatase activity was found concordantly in IHC. In the two controls, one showed no signal for aromatase in IHC in stromal and epithelial cells whereas in the other patient aromatase protein was found in endometrial stromal cells in IHC.
|
Devising a predictive model for endometriosis
In a multivariate analysis using the applied testing methods and clinical data from patients' histories, a series of gynaecological characteristics was different between cases and controls (Table II). The presence of endometriosis clearly correlated with the severity of dysmenorrhoea (r=0.44, P<0.01) whereas for pelvic pain (r=0.23, P>0.05) and for dyspareunia (r=0.11, P>0.05) no significant association was found.
|
|
![]() |
![]() |
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Histological confirmation of endometriosis by laparoscopy revealed a prevalence of 52.1% for the disease among symptomatic patients attending our tertiary care centre for dysmenorrhoea, dyspareunia, chronic pelvic pain and unexplained infertility. This figure is similar to recently published data about epidemiological facts on endometriosis (Mahmood and Templeton, 1991; Spaczynski and Duleba, 2003
). The present study confirms that high levels of CYP 19 aromatase in eutopic endometrium are conclusively associated with the presence of endometriosis. Several earlier studies have shown that endometriosis and other pelvic pathologies are strongly associated with aberrantly high aromatase expression (Noble et al., 1996
; Bulun et al., 1997
; Kitawaki et al., 1997
). For instance, Kitawaki et al. (1999)
found aromatase expression and activity in endometrial specimens from patients with EDD such as endometriosis, adenomyosis and leiomyoma of the uterus, but not in disease-free women. In their retrospective analysis of 105 patients using non-quantitative RTPCR and IHC analysis, they found a sensitivity and specificity of 91 and 100% respectively. A prospective trial with 56 patients showed a far lower sensitivity and specificity using non-quantitative PCR techniques (82 and 59% respectively) (Dheenadayalu et al., 2002
). These differences in sensitivity and specificity of aromatase testing might be due to a series of limitations, such as small sample sizes in both trials and a lack of comparability due to polymorbid patients in the retrospective trial (Kitawaki et al., 1999
) as well as selection bias, or the restriction to non-quantitative PCR techniques in the prospective trial (Dheenadayalu et al., 2002
). However, recently, aromatase was found at particularly low levels in any endometrial tissue, rising to aberrantly high levels in association with EDD (Brosens et al., 2004
). Along with these data, our results showed aromatase expressionalthough at very low levelsin each single endometrial sample ranging from 1- to 100.7-fold expression. Therefore, efforts to predict endometriosis on the basis of non-quantitative aromatase detection might not be sustainable.
Despite advanced attempts to generate laboratory tests or appropriate imaging techniques for the diagnosis of endometriosis, to date no simple non-invasive tests have been available to reduce the number of uninformative laparoscopies. The use of clinical symptoms and signs together with laboratory markers appears to be a potentially effective method to predict endometriosis (Eskenazi et al., 2001; Falcone and Mascha, 2003
; Gagne et al., 2003
). Early diagnosis and treatment of endometriosis is crucial for the management of this chronic disease (Donnez et al., 2003
). While some women suffering from pelvic pain possibly associated with endometriosis might benefit from an initial medical rather than from a surgical therapy (Olive and Pritts, 2001
), others might benefit from immediate excision of lesions by laparoscopic surgery (Abbott et al., 2003
). The statistical formula proposed here may help the clinician to identify patients suffering from endometriosis to reconcile appropriate treatment. However, our preliminary data need to be confirmed by a larger series before routine clinical application can be proposed.
The limitations of our screening method comprise a possible dependency on the phase of the menstrual cycle for endometrial biopsy and the influence of concomitant medication [i.e. oral contraceptives (Meresman et al., 2002) and GnRH analogues (Ishihara et al., 2003
)]. Additionally, a further confounding influence might occur by using a subjective pain score (VAS) to describe the severity of dysmenorrhoea. In order to minimize this limitation, we used mean pain scores of the complete phase of menstrual efflux as described (Chapron et al., 2003
; Walsh et al., 2003
) and found them indicative for disease prediction.
In conclusion, this model, based on cytochrome P450 aromatase screening of eutopic endometrial biopsy specimens combined with select patients' characteristics, represents a novel, non-surgical diagnostic method which can easily be performed in every outpatient infertility or chronic pelvic pain clinic.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
American Society for Reproductive Medicine (1997) Revised American Society for Reproductive Medicine classification of endometriosis. Fertil Steril, 817821.
Arvanitis DA, Koumantakis GE, Goumenou AG, Matalliotakis IM, Koumantakis EE and Spandidos DA (2003) CYP1A1, CYP19, and GSTM1 polymorphisms increase the risk of endometriosis. Fertil Steril 1, 702709.
Bedaiwy MA, Falcone T, Sharma RK, Goldberg JM, Attaran M, Nelson DR and Agarwal A (2002) Prediction of endometriosis with serum and peritoneal fluid markers: a prospective controlled trial. Hum Reprod 17, 426431.
Bergqvist A, Bruse C, Carlberg M and Carlstrom K (2001) Interleukin 1beta, interleukin-6, and tumor necrosis factor-alpha in endometriotic tissue and in endometrium. Fertil Steril 75, 489495.[CrossRef][ISI][Medline]
Brosens I (1997) Diagnosis of endometriosis. Semin Reprod Endocrinol 15, 229233.[Medline]
Brosens I, Puttemans P, Campo R, Gordts S and Brosens J (2003) Non-invasive methods of diagnosis of endometriosis. Curr Opin Obstet Gynecol 15, 519522.[ISI][Medline]
Brosens J, Verhoeven H, Campo R, Gianaroli L, Gordts S, Hazekamp J, Hagglund L, Mardesic T, Varila E, Zech J and Brosens I (2004) High endometrial aromatase P450 mRNA expression is associated with poor IVF outcome. Hum Reprod 19, 352356.
Bulun SE, Noble LS, Takayama K, Michael MD, Agarwal V, Fisher C, Zhao Y, Hinshelwood MM, Ito Y and Simpson ER (1997) Endocrine disorders associated with inappropriately high aromatase expression. J Steroid Biochem Mol Biol 61, 133139.[CrossRef][ISI][Medline]
Bulun SE, Gurates B, Fang Z, Tamura M, Sebastian S, Zhou J, Amin S and Yang S (2002) Mechanisms of excessive estrogen formation in endometriosis. J Reprod Immunol 55, 2133.[CrossRef][ISI][Medline]
Chapron C, Fauconnier A, Dubuisson JB, Barakat H, Vieira M and Breart G (2003) Deep infiltrating endometriosis: relation between severity of dysmenorrhoea and extent of disease. Hum Reprod 18, 760766.
Cramer DW, Wilson E, Stillman RJ, Berger MJ, Belisle S, Schiff I, Albrecht B, Gibson M, Stadel BV and Schoenbaum SC (1986) The relation of endometriosis to menstrual characteristics, smoking, and exercise. J Am Med Assoc 255, 19041908.[Abstract]
Darrow SL, Vena JE, Batt RE, Zielezny MA, Michalek AM and Selman S (1993) Menstrual cycle characteristics and the risk of endometriosis. Epidemiology 4, 135142.[ISI][Medline]
Dheenadayalu K, Mak I, Gordts S, Campo R, Higham J, Puttemans P, White J, Christian M, Fusi L and Brosens J (2002) Aromatase P450 messenger RNA expression in eutopic endometrium is not a specific marker for pelvic endometriosis. Fertil Steril 78, 825829.[CrossRef][ISI][Medline]
Donnez J, Pirard C, Smets M, Jadoul P and Squifflet J (2003) Pre- and post-surgical management of endometriosis. Semin Reprod Med 21, 235242.[CrossRef][ISI][Medline]
Eskenazi B, Warner M, Bonsignore L, Olive D, Samuels S and Vercellini P (2001) Validation study of nonsurgical diagnosis of endometriosis. Fertil Steril 76, 929935.[CrossRef][ISI][Medline]
Falcone T and Mascha E (2003) The elusive diagnostic test for endometriosis. Fertil Steril 80, 886888.[CrossRef][ISI][Medline]
Fang Z, Yang S, Gurates B, Tamura M, Simpson E, Evans D and Bulun SE (2002) Genetic or enzymatic disruption of aromatase inhibits the growth of ectopic uterine tissue. J Clin Endocrinol Metab 87, 34603466.
Gagne D, Rivard M, Page M, Lepine M, Platon C, Shazand K, Hugo P and Gosselin D (2003) Development of a nonsurgical diagnostic tool for endometriosis based on the detection of endometrial leukocyte subsets and serum CA-125 levels. Fertil Steril 80, 876885.[CrossRef][ISI][Medline]
Halme J, Hammond MG, Hulka JF, Raj SG and Talbert LM (1984) Retrograde menstruation in healthy women and in patients with endometriosis. Obstet Gynecol 64, 151154.[Abstract]
Harada T, Kubota T and Aso T (2002) Usefulness of CA19-9 versus CA125 for the diagnosis of endometriosis. Fertil Steril 78, 733739.[CrossRef][ISI][Medline]
Ishihara H, Kitawaki J, Kado N, Koshiba H, Fushiki S and Honjo H (2003) Gonadotropin-releasing hormone agonist and danazol normalize aromatase cytochrome P450 expression in eutopic endometrium from women with endometriosis, adenomyosis, or leiomyomas. Fertil Steril 1, 735742.[CrossRef]
Kitawaki J, Noguchi T, Amatsu T, Maeda K, Tsukamoto K, Yamamoto T, Fushiki S, Osawa Y and Honjo H (1997) Expression of aromatase cytochrome P450 protein and messenger ribonucleic acid in human endometriotic and adenomyotic tissues but not in normal endometrium. Biol Reprod 57, 514519.[Abstract]
Kitawaki J, Kusuki I, Koshiba H, Tsukamoto K, Fushiki S and Honjo H (1999) Detection of aromatase cytochrome P-450 in endometrial biopsy specimens as a diagnostic test for endometriosis. Fertil Steril 72, 11001106.[CrossRef][ISI][Medline]
Kuohung W, Jones GL, Vitonis AF, Cramer DW, Kennedy SH, Thomas D and Hornstein MD (2002) Characteristics of patients with endometriosis in the United States and the United Kingdom. Fertil Steril 78, 767772.[CrossRef][ISI][Medline]
Lebovic DI, Mueller MD and Taylor RN (2001) Immunobiology of endometriosis. Fertil Steril 75, 110.[CrossRef][ISI][Medline]
Leyendecker G, Herbertz M, Kunz G and Mall G (2002) Endometriosis results from the dislocation of basal endometrium. Hum Reprod 17, 27252736.
Livak KJ and Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)). Methods 25, 402408.[CrossRef][ISI][Medline]
Mahmood TA and Templeton A (1991) Prevalence and genesis of endometriosis. Hum Reprod 6, 544549.[Abstract]
Meresman GF, Auge L, Baranao RI, Lombardi E, Tesone M and Sueldo C (2002) Oral contraceptives suppress cell proliferation and enhance apoptosis of eutopic endometrial tissue from patients with endometriosis. Fertil Steril 77, 11411147.[CrossRef][ISI][Medline]
Noble LS, Simpson ER, Johns A and Bulun SE (1996) Aromatase expression in endometriosis. J Clin Endocrinol Metab 81, 174179.[Abstract]
Noble LS, Takayama K, Zeitoun KM, Putman JM, Johns DA, Hinshelwood MM, Agarwal VR, Zhao Y, Carr BR and Bulun SE (1997) Prostaglandin E2 stimulates aromatase expression in endometriosis-derived stromal cells. J Clin Endocrinol Metab 82, 600606.
Olive DL and Pritts EA (2001) Treatment of endometriosis. N Engl J Med 345, 266275.
Olive DL and Schwartz LB (1993) Endometriosis. N Engl J Med 328, 17591769.
Osteen KG, Bruner KL and Sharpe-Timms KL (1996) Steroid and growth factor regulation of matrix metalloproteinase expression and endometriosis. Semin Reprod Endocrinol 14, 247255.[ISI][Medline]
Sampson J (1927) Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol, 422429.
Spaczynski RZ and Duleba AJ (2003) Diagnosis of endometriosis. Semin Reprod Med 21, 193208.[CrossRef][ISI][Medline]
Vessey MP, Villard-Mackintosh L and Painter R (1993) Epidemiology of endometriosis in women attending family planning clinics. Br Med J 306, 182184.[ISI][Medline]
Walsh TM, LeBlanc L and McGrath PJ (2003) Menstrual pain intensity, coping, and disability: the role of pain catastrophizing. Pain Med 4, 352361.[CrossRef][ISI][Medline]
Zeitoun KM and Bulun SE (1999) Aromatase: a key molecule in the pathophysiology of endometriosis and a therapeutic target. Fertil Steril 72, 961969.[CrossRef][ISI][Medline]
Zeitoun K, Takayama K, Sasano H, Suzuki T, Moghrabi N, Andersson S, Johns A, Meng L, Putman M, Carr B and Bulun SE (1998) Deficient 17beta-hydroxysteroid dehydrogenase type 2 expression in endometriosis: failure to metabolize 17beta-estradiol. J Clin Endocrinol Metab 83, 44744480.
Submitted on June 10, 2004; resubmitted on December 17, 2004; accepted on January 16, 2005.
|