The American Journal of Surgical Pathology 25(5): 667–672, 2001
, Philadelphia
Absence of Estrogen Receptor-␣ Expression in
Human Ovarian Clear Cell Adenocarcinoma
Compared With Ovarian Serous, Endometrioid, and
Mucinous Adenocarcinoma
Masaki Fujimura, M.D., Takao Hidaka, M.D., Ken Kataoka, M.D.,
Yoshihiro Yamakawa, M.D., Shinobu Akada, M.D., Akiko Teranishi, M.D.,
and Shigeru Saito, M.D.
The mechanism that regulates growth in ovarian clear cell ad-
enocarcinoma (CCA) is not well understood. A high incidence
of concurrent endometriosis with CCA may indicate that estro-
gen is a growth promotor in CCA. To determine estrogen as a
growth promotor, the authors investigated the presence or ab-
sence of estrogen receptor-␣ (ER-␣), ER-␤, progesterone re-
ceptor, and dioxin receptor (i.e., aromatic hydrocarbon recep-
tor) in clinically resected ovarian CCA, serous adenocarcinoma
(SAC), endometrioid adenocarcinoma (EAC), and mucinous
adenocarcinoma (MAC) specimens using an immunohisto-
chemical method. Expression of ER-␣ and ER-␤ messenger
ribonucleic acid was examined by reverse transcription–
polymerase chain reaction in three established CCA cell lines:
KK, RMG-1, and HAC-II. None of the surgically resected CCA
and CCA cell lines showed positive staining for ER-␣. Con-
versely, 97.2% of SACs, 100% of EACs, and 70% of MACs
showed positive nuclear staining for ER-␣ (p <0.001). Con-
versely, positive ER-␤ staining for CCA (39.3%) was similar to
that of SAC (41.7%) and MAC (30.0%). EAC (75%) showed a
higher expression of ER-␤ (p <0.02). Progesterone receptor
was detected in only 10.7% of CCA, compared with SAC and
EAC (SAC, 86.1%; EAC, 91.7%; p <0.01). Aromatic hydro-
carbon receptor was detected in all histologic types at an inci-
dence of approximately 50% to 60%. Messenger ribonucleic
acid of ER-␣ and ER-␤ was not detected in the three CCA cell
lines. These findings indicate biologic characteristics that dis-
tinguish CCA from other types of ovarian epithelial cancer.
From the Department of Obstetrics and Gynecology (M.F., T.H.,
K.K., Y.Y., S.S.), Toyama Medical and Pharmaceutical University; the
Department of Obstetrics and Gynecology (S.A.), Habikino Prefectural
Hospital; and the Department of Obstetrics and Gynecology (A.T.),
Nara Medical University, Japan.
Supported in part by grant no. 09671667 from the Ministry of
Education (M.F.).
Address correspondence and reprint requests to Masaki Fujimura,
MD, Department of Obstetrics and Gynecology, Toyama Medical and
Pharmaceutical University, 2630 Sugitani Toyama-City, 930–0194,
Toyama, Japan.
667
© 2001 Lippincott Williams & Wilkins, Inc.
Key Words: Ovarian clear cell adenocarcinoma—Estrogen
receptor—Progesterone receptor—Aromatic hydrocarbon
receptor.
Am J Surg Pathol 25(5): 667–672, 2001.
Clear cell adenocarcinoma (CCA) of the ovary is a
strongly chemoresistant tumor among the many histo-
pathologic types of ovarian surface epithelial carci-
noma.4,34 Because of this characteristic feature, CCA of
the ovary remains one of the most difficult tumors to
treat in the field of gynecologic oncology. The 5-year
survival rate of patients with stage I ovarian CCA ranged
from 90% of patients4,34 to approximately 50% to 60%
of patients,2,7,12,14,15,24,28,30 which is comparable with
the other histologic types of surface epithelial ovarian
carcinoma. However, the 5-year survival rate for stage II
disease declines to only 17% to 29%,15,24,30 which is
worse, probably because of the chemoresistance of CCA.
Optimal treatment of CCA is believed to be complete
resection of the tumor; however, a complete resection is
difficult to achieve in patients with advanced disease. An
effective treatment modality for this tumor is urgently
needed.
CCA of the ovary is frequently associated with pelvic
endometriosis, with a reported incidence ranging from
20% to 70%.4,6,15,24,27,31 Endometriosis is promoted by
intrinsic estrogen activity. If estrogen promotes the
growth of CCA, the estrogen receptor (ER) would likely
be present in CCA; however, this has never been studied.
Furthermore, the incidence of endometriosis is said to
have increased in recent years because of an increase in
environmental hazards, such as dioxin and related com-
668 M. FUJIMURA ET AL.
pounds.22 Many biologic responses to dioxin are known
to be mediated through ligand binding to the aromatic
hydrocarbon receptor (AhR).21 No information, how-
ever, is yet available about the relationship between
ovarian CCA and AhR status.
A higher expression of ER-␣ relative to the expression
of ER-␤ has been observed in surface epithelial ovarian
cancer.29 The frequent occurrence of ovarian surface ep-
ithelial carcinoma, especially of serous adenocarcinoma
(SAC), in patients with mutated BRCA-1 was observed.1
The BRCA-1 gene was found to inhibit the signaling by
the ligand-activated ER-␣ through the estrogen-
responsive enhancer element and to block the transcrip-
tional activation function AF-2 of ER-␣.11 When the
BRCA-1 gene mutates, it loses this inhibitory effect, re-
sulting in tumorigenesis of the breast and ovary. The
inhibitory and maturing effect of the BRCA-1 gene in the
proliferation of human mammalian tissue was also de-
scribed recently in one clinical study.16 These observa-
tions strongly suggest that estrogen has some role in the
carcinogenesis of common epithelial surface ovarian
cancer.
Taking these previous studies into account, we used
the immunohistochemical method to examine the pres-
ence or absence of ER-␣ and ER-␤ in surgically resected
ovarian CCA specimens, in SAC, in endometrioid ad-
enocarcinoma (EAC), and in mucinous adenocarcinoma
(MAC) specimens, and in three established human ovar-
ian CCA cell lines. The results demonstrated that ER-␣
was not detected in surgically resected ovarian CCA
cases nor was it detected in CCA cell lines, suggesting
that the biologic response to estrogen may be different
from that of other ovarian surface epithelial cancer.
MATERIALS AND METHODS
Culture Conditions For Three Human Ovarian
CCA Cell Lines
KK,33 RMG-1,17 and HAC-II26 were cultured in
Dulbecco’s Modified Eagle’s Medium (Nissui, Tokyo,
Japan) supplemented with 10% fetal calf serum
(Mitsubishi Chemicals Co., Tokyo, Japan) at 37°C in 5%
CO2 under stable growth conditions.
ER-␣ and ER-␤ Progesterone Receptor (PR) and
AhR Status: Immunohistochemical Staining
Twenty-eight patients with ovarian CCA who had
been treated initially in the Department of Obstetrics and
Gynecology, Toyama Medical and Pharmaceutical
University Hospital and Nara Medical University
Hospital from 1982 to 1999 were registered in this study.
The age of these patients ranged from 32 to 65 years
(mean age, 51.9 years; median age, 51.0 years). Eight
Am J Surg Pathol, Vol. 25, No. 5, 2001
patients (28.6%) were associated with concomitant en-
dometriosis. Thirty-six patients with ovarian SAC, 12
with ovarian EAC, and 10 with ovarian MAC who were
also treated in our institute were the control subjects.
After pathologic review by one of the authors (M.F.), an ap-
propriate block was selected for each of the 28 patients
and 58 control subjects, and then continuous, 5-␮m-thick
paraffin sections were obtained from each block. Immu-
nohistochemical staining was performed using the usual
streptavidin–biotin method using an immunohistochem-
ical staining kit (Nichirei Co., Tokyo, Japan) after a high-
temperature antigen unmasking procedure (121°C, 15
minutes) and blocking of intrinsic peroxidase were per-
formed. The first antibody used was rabbit polyclonal
anti-ER-␣ antibody HC-20 (Santa Cruz Biotechnology,
CA, USA) at a concentration of 1:100 of the primary,
as applied by the manufacturer, at 4°C overnight; goat
polyclonal anti-ER-␤ antibody N-19 (Santa Cruz
Biotechnology) at a concentration of 1:100 at 4°C over-
night; mouse monoclonal anti-PR antibody (NCL-PGR;
Nobocastra Laboratories, Tyne, UK) at a concentration
of 1:50 at 4°C overnight; and goat polyclonal antibody
raised against a peptide mapping at the carboxy terminus
of AhR of human origin C-18 (Santa Cruz Biotechnology)
at a concentration of 1:100 at 4°C overnight. Diamino-
benzidine was used as the staining substrate. The result
was interpreted as positive when more than 10% of the
tumor cell nuclei were stained positively under the high-
power field of the microscope.
Immunohistochemical study for ER-␣ and ER-␤ was
also performed on the three ovarian CCA cell lines cul-
tured on glass slides using the same procedure.
Expression of ER-␣ and ER-␤ Messenger
Ribonucleic Acid (mRNA) in the Three Established
Ovarian CCA Cell Lines
Total RNAs were prepared from the three ovarian
CCA cell lines using RNAzol according to the manufac-
turer’s instructions (TEL-TEST Inc., Friendswood, TX,
USA). Two micrograms of total RNA were reverse tran-
scribed using an mRNA selective PCR Kit (Takara Co.,
Tokyo, Japan). Subsequently, one-quarter volume of the
resulting complementary deoxyribonucleic acid (cDNA)
samples were used for each polymerase chain reaction
(PCR) examination.
The primer sequences used were 5⬘-TGCCAAG-
GAGACTCGCTA-3⬘ and 5⬘-TCAACATTCTCCCTCC-
TC-3⬘ (263 base pairs [bp]) for amplifying ER-␣,36 and
5⬘-GTCCATCGCCAGTTATCACATC-3⬘ and 5⬘-GCC-
TTACATCCTTCACACGA-3⬘ (242 bp) for ER-␤.8 PCR
was performed using an mRNA selective PCR Kit
(Takara Co.) to avoid genomic DNA amplification. The
PCR buffer contained 1× mRNA selective PCR buffer, 5
mM MgCl2, 1 mM deoxyribonucleoside triphosphate
 ABSENCE OF ESTROGEN RECEPTOR-␣ EXPRESSION 669

FIG. 1. (A–D) Immunohistochemical staining for estrogen receptor (ER)-␣ in clear cell adenocarcinoma (CCA) (A), serous
adenocarcinoma (SAC) (B), endometrioid adenocarcinoma (EAC) (C), and mucinous adenocarcinoma (MAC) (D). Im-
munohistochemical staining for ER-␣ showed no positive finding in CCA cells; however, staining for ER-␣ showed strong
positive reactions in the nuclei of SAC, EAC, and MAC cells.

(dNTP)/analog mixture, 2.5 U/␮L Avian Myeloblastosis Statistical Analysis

Virus (AMV)-optimized Taq polymerase, and 0.4 ␮M
sense- and antisense-specific primers. A total of 5 ␮L
cDNA from each cell line and sterile distilled water was
added to adjust the final concentration to 50 ␮L. As a
positive control for ER-␣, cDNA from the Ishikawa hu-
man endometrial cancer cell line was used, and human
granulosa cells collected from a woman of reproductive
age after informed consent was obtained were used as a
positive control for ER-␤. All transcripts were analyzed
in a DNA thermal cycler (Perkin–Elmer, Norwalk, CT,
USA) with the following cycle: a denaturation step of
85°C for 1 minute to avoid genomic DNA amplification,
an annealing step of 50°C for 1 minute, and an extension
step of 72°C for 1 minute, repeated for 30 cycles. A final
primer extension step of 72°C for 5 minutes followed the
30 cycles of PCR. PCR products were analyzed by elec-
trophoresis through a 6% polyacrylamide gel and visu-
alized by ethidium bromide staining under ultraviolet
illumination.
To evaluate the results of ER-␣, ER-␤, PR, and AhR
status in surgical specimens, we compared ovarian SAC,
EAC, and MAC cases and ovarian CCA cases using the
␹2 test at a significance of p <0.05. All statistical analy-
ses were performed using StatView for the Macintosh
(version 4.5; Abacus Concepts, Berkeley, CA, USA).
RESULTS
ER-␣, ER-␤, PR, and AhR Status:
Immunohistochemical Staining
ER-␣ was not detected in any of the 28 surgically
resected ovarian CCA specimens (Fig 1A). However,
ER-␣ was detected in 35 of 36 cases of ovarian SAC
(97.2%; p <0.001), in 12 of 12 cases of ovarian EAC
(100.0%; p <0.001), and in seven of 10 cases of ovarian
MAC (70.0%; p <0.001; Fig. 1B–D and Table 1). ER-␤,

FIG. 2. (A–D) Immunohistochemical staining for estrogen receptor (ER)-␤ in clear cell adenocarcinoma (CCA) (A), serous
adenocarcinoma (SAC) (B), endometrioid adenocarcinoma (EAC) (C), and mucinous adenocarcinoma (MAC) (D). Im-
munohistochemical staining for ER-␤ showed positive findings in the nuclei of CCA, SAC, EAC, and MAC cells.

Am J Surg Pathol, Vol. 25, No. 5, 2001

670 M. FUJIMURA ET AL.

TABLE 1. Results of immunohistochemical positive staining

ER-␣ (%) ER-␤ (%) PR (%) AhR (%)

Clear cell 0/28 (0.0) 11/28 (39.3) 3/28 (10.7) 15/28 (53.6)
with EMosis 0/8 (0.0) 3/8 (37.5) 0/8 (0.0) 5/8 (62.5)
without EMosis 0/20 (0.0) 8/20 (40.0) 3/20 (14.3) 10/20 (50.0)
Serous 35/36 (97.2)* 15/36 (41.7) 31/36 (86.1)* 19/36 (52.8)
Endometrioid 12/12 (100.0)* 9/12 (75.0)† 11/12 (91.7)* 7/12 (58.3)
Mucinous 7/10 (70.0)* 3/10 (30.0) 1/10 (10.0) 6/10 (60.0)

EMosis, endometriosis.
* vs clear cell adenocarcinoma p <0.01.
† vs clear cell adenocarcinoma p <0.02.

however, was detected in 11 cases of CCA (39.3%), 15 DISCUSSION

cases of SAC (41.7%), nine cases of EAC (75.0%;
p <0.02), and three cases of MAC (30.0%; Fig. 2 and
Table 1). In the three ovarian CCA cell lines, neither
ER-␣ nor ER-␤ was detected by immunohistochemical
staining. PR was detected in only three cases of ovarian
CCA (10.7%). Conversely, 31 of 36 SACs (86.1%; p
<0.01) and 11 of 12 EACs (91.7%; p <0.01) were posi-
tive for PR (Table 1). One of 10 MACs (10%) were
positive for PR. AhR was detected in 53.6%, 52.8%,
58.3%, and 60.0% of CCA, SAC, EAC, and MAC re-
spectively. No significant difference in AhR status was
observed among these four histologic types of surface
epithelial ovarian cancer. In CCA cases, ER status was
compared with the presence or absence of endometriosis.
ER-␤ was found in 37.5% of cases with endometriosis
compared with 40.0% of cases without endometriosis.
Also 0% versus 14.3% was found in PR, and 62.5%
versus 50.0% was found in AhR. No significant differ-
ence was found in ER-␤, PR, and AhR status in cases
with or without concomitant endometriosis.
Expression of ER-␣ and ER-␤ mRNA in the Three
Established Ovarian CCA Cell Lines
In the three ovarian CCA cell lines examined, no
specific bands for ER-␣ and ER-␤ mRNA were found
(Fig. 3).
Because of the dramatic evolution of anticancer che-
motherapy in recent years, especially after the clinical
introduction of platinum-containing chemotherapeutic
regimens and the recent appearance of paclitaxel for
ovarian cancer, therapeutic effect, prognosis, and quality
of life for ovarian cancer patients have improved.23
Ovarian CCA, however, because of its resistance to these
regimens, has eluded an effective treatment modality.15
Although the importance of developing a new approach
to this tumor is apparent, little information about the
biologic characteristics of this tumor that would yield
clues for developing a new regimen has been available so
far. However, one of its characteristics, association with
endometriosis, is well-known. Because endometriosis is
related closely to intrinsic estrogen activity, growth regu-
lation of ovarian CCA could be promoted by estrogen.
Komiyama et al.18 reported the prognostic advantage in
cases with endometriosis compared with those without
endometriosis. They assumed that some negative influ-
ence affected the growth of this tumor when concomitant
endometriotic tissue was present. In the current study,
28.6% of cases were associated with endometriosis.
However, the cases with endometriosis did not show any
significant difference of expression of ER-␣, ER-␤, PR,
and AhR (Table 1) or survival benefit compared with the
cases without endometriosis in our study (data not
shown).

FIG. 3. Expression of estrogen re-

ceptor (ER)-␣ and ER-␤ messenger
ribonucleic acid (mRNA) in human
ovarian clear cell adenocarcinoma
(CCA) cell lines: KK, RMG-1, and
HAC-II. Neither band corresponding
to 263-base pair (bp) ER-␣ mRNA
and 242-bp ER-␤ mRNA was ex-
pressed in these ovarian CCA cell
lines, whereas a distinct band corre-
sponding to ER-␣ mRNA was found
in an Ishikawa cell line. A band cor-
responding to ER-␤ mRNA was also
found in human granulosa cells.

Am J Surg Pathol, Vol. 25, No. 5, 2001

 ABSENCE OF ESTROGEN RECEPTOR-␣ EXPRESSION
Our current data indicated no existence of ER-␣ in
CCA specimens and established cell lines. However, the
question still exists whether these observations were ac-
tually the result of histopathologic type rather than
nuclear differentiation. In our previous study, we exam-
ined the nuclear grade of CCA cases according to the
grading of renal cell carcinoma by Fuhrman et al.13 As a
result, all CCA cases were divided into grades 2 and 3
evenly. However, none of these cases were stained posi-
tively for ER-␣, as mentioned earlier. If ER-␣ immuno-
reactivity depended on nuclear grading, some cases in
grade 2 would fall into the positive-staining category.
Also in the cases of SAC, EAC, and MAC, we could
observe evenly distributed nuclear grade according to a
previously proposed nuclear grading system for ovarian
cancer,3 despite the high positive rate of ER-␣ immuno-
reactivity observed in these histopathologic types. From
these observations, we could conclude that ER-␣ immu-
noreactivity was not affected by nuclear grading.
Recently, ER was found to contain two subtypes of
receptor structures: so-called classic ER-␣ and the newly
identified ER-␤.19,25 The function of classic ER was
known to be carried out mainly by ER-␣, and ER-␤ was
supposed to work as a kind of housekeeping gene,9,20
although the function of ER-␤ is still under investigation.
ER-␣ was found to be expressed in many ovarian cancer5
and ovarian cancer cell lines, such as the SKOV-3,
OVCAR3, and BG-1 cell lines.27 ER-␤ was known to be
expressed in the normal gonad27 and granulosa cells.10
Rutherford et al.32 reported recently the absence of ER-␤
expression in metastatic ovarian cancer. They hypoth-
esized the presence of some factors that regulate ER gene
expression between primary and metastatic cells. The
recent work of Lau et al.20 disclosed the disruption of
ER-␣ mRNA expression as well as dramatic downregu-
lation of PR transcript expression in ovarian cancer cells.
Established CCA cell lines did not show the presence of
ER-␣ and ER-␤ in mRNA and protein level in our
investigation.
These findings suggested that the growth of ovarian
CCA is promoted by stimulation other than that of es-
trogen through ER-␣, which may be a growth promotor
of serous, endometrioid, or mucinous types of ovarian
carcinoma. When ER-␣ is depleted, the phenotype of
surface epithelial ovarian carcinoma may become a clear
cell type, resulting in chemoresistant tumor. Further in-
vestigation is needed to clarify this point. Another bio-
logic characteristic that distinguishes CCA from other
types of epithelial ovarian cancer was described by
Shimizu et al.,35 who emphasized lower expression of
p53 and cyclin A protein and higher expression of p21
and cyclin E proteins in resected ovarian CCA speci-
mens. The difference between estrogen-independent and
-dependent cell growth may cause these biologic
changes.
671
AhR, which is known as the receptor of dioxin, was
detected in 53.6% cases of CCA, 52.8% cases of SAC,
58.3% cases of EAC, and 60.0% cases of MAC. This
finding demonstrates that environmental factors may
have some impact on the proliferation or carcinogenesis
of surface epithelial ovarian carcinoma; however, no dif-
ference in expression rate was observed between histo-
logic types.
In conclusion, ER-␣ was not present in ovarian CCA,
but it was present in SAC, EAC, and MAC. This differ-
ence could be a fundamental biologic characteristic that
segregates CCA from other types of ovarian surface ep-
ithelial carcinoma. This could be a clue for establishing
a novel treatment modality for CCA. 䊐
REFERENCES
1. Aida H, Takakuwa K, Nagata H, et al. Clinical features of ovarian
cancer in Japanese women with germ-line mutations of BRCA1.
Clin Cancer Res 1998;4:235–40.
2. Aure JC, Hoeg K, Kolstad P. Mesonephroid tumors of the ovary.
Clinical and histologic studies. Obstet Gynecol 1971;37:860–7.
3. Barber HRK, Sommers SC, Synder R, et al. Histologic and nuclear
grading and stromal reactions as indices for prognosis in ovarian
cancer. Am J Obstet Gynecol 1975;121:795–807.
4. Brescia RJ, Dubin N, Demopoulos RI. Endometrioid and clear cell
carcinoma of the ovary. Factors affecting survival. Int J Gynecol
Pathol 1987;7:287–93.
5. Chu S, Mamers P, Burger HG, et al. Estrogen receptor isoform
gene expression in ovarian stromal and epithelial tumors. J Clin
Endocrinol Metab 2000;85:1200–5.
6. Crozier MA, Copeland LJ, Silva EG, et al. Clear cell carcinoma of
the ovary: a study of 59 cases. Gynecol Oncol 1989;35:199–203.
7. Doshi N, Tobon H. Primary clear cell carcinoma of the ovary. An
analysis of 15 cases with review of the literature. Cancer 1977;
39:2658–64.
8. Dotzlaw H, Leygue E, Watson PH, et al. Estrogen receptor-␤
messenger RNA expression in human breast tumor biopsies: rela-
tionship to steroid receptor status and regulation by progestins.
Cancer Res 1999;59:529–32.
9. Drummond AE, Baillie AJ, Findlay JK. Ovarian estrogen receptor
alpha and beta mRNA expression: impact of development and
estrogen. Mol Cell Endocrinol 1999;149:153–61.
10. Enmark E, Pelto–Huikko M, Grandien K, et al. Human estrogen
receptor beta-gene structure, chromosomal localization, and ex-
pression pattern. J Clin Endocrinol Metab 1997;82:4258–65.
11. Fan S, Wang J-A, Yuan R, et al. BRCA1 inhibition of estrogen
receptor signaling in transfected cells. Science 1999;284:1354–6.
12. Fine G, Clarke H, Horn RC. Mesonephroma of the ovary. A clini-
cal morphological and histogenetic appraisal. Cancer 1973;31:
398–410.
13. Fuhrman SA, Lasky LC, Limas C. Prognostic significance of mor-
phologic parameters in renal cell carcinoma. Am J Surg Pathol
1982;6:655–63.
14. Hayes D. Mesonephroid tumours of the ovary. J Obstet Gynecol Br
Commonw 1972;79:728–36.
15. Jenison EL, Montag AG, Griffiths CT, et al. Clear cell adenocar-
cinoma of the ovary: a clinical analysis and comparison with se-
rous carcinoma. Gynecol Oncol 1989;32:65–71.
16. Jernstrom H, Lerman C, Ghadirian P, et al. Pregnancy and risk of
early breast cancer in carriers of BRCA1 and BRCA2. Lancet
1999;354:1846–50.
17. Kiguchi K, Iwamori M, Mochizuki Y, et al. Selection of human
carcinoma cells with high dissemination potential by repeated pas-
sage of the cells in vivo into nude mice, and involvement of Le(x)-
Am J Surg Pathol, Vol. 25, No. 5, 2001
672 M. FUJIMURA ET AL.
determinant in the dissemination potential. Jpn J Cancer Res 1989;
89:923–32.
18. Komiyama S, Aoki D, Tominaga E, et al. Prognosis of Japanese
patients with ovarian clear cell adenocarcinoma associated with
pelvic endometriosis: clinicopathologic evaluation. Gynecol Oncol
1999;72:342–6.
19. Kuiper GG, Gustafsson JA. The novel estrogen receptor-beta sub-
type: potential role in the cell- and promotor-specific actions of
estrogens and anti-estrogens. FEBS Lett 1997;410:87–90.
20. Lau KM, Mok SC, Ho SM. Expression of human estrogen recep-
tor-alpha and -beta, progesterone receptor, and androgen receptor
mRNA in normal and malignant ovarian epithelial cells. Proc Natl
Acad Sci USA 1999;96:5722–7.
21. Lucier GW, Portier CJ, Gallo MA. Receptor mechanisms and
dose–response models for the effects of dioxins. Environ Health
Perspect 1993;101:36–44.
22. Mayani A, Barel S, Soback S, et al. Dioxin concentration in
women with endometriosis. Hum Reprod 1997;12:373–5.
23. Mcguire WP, Hoskins WJ, Brady MF, et al. Cyclophosphamide
and cisplatin compared with paclitaxel and cisplatin in patients
with stage III and IV ovarian cancer. N Engl J Med 1996;334:1–6.
24. Montag AG, Jenison EL, Griffiths CT, et al. Ovarian clear cell
carcinoma: a clinicopathologic analysis of 44 cases. Int J Gynecol
Pathol 1989;8:85–96.
25. Mosselman S, Polman J, Dijkema R. ER beta: identification and
characterization of a novel human estrogen receptor. FEBS Lett
1996;392:49–53.
26. Nishida M, Iwasaki H. Establishment of a human ovarian clear-cell
carcinoma cell line (HAC-2 cell) in vitro. Hum Cell 1988;1:345–6.
27. Omura GA, Brady MF, Homesley HD, et al. Long-term follow-up
Am J Surg Pathol, Vol. 25, No. 5, 2001
and prognostic factor analysis in advanced ovarian carcinoma: the
Gynecological Oncological Group experience. J Clin Oncol 1991;
9:1138–50.
28. Parker TM, Dockerty MB, Randall LM. Mesonephric clear cell
carcinoma of the ovary. Am J Obstet Gynecol 1960;80:417.
29. Pujol P, Rey J-M, Nirde P, et al. Differential expression of estrogen
receptor-␣ and -␤ messenger RNAs as a potential marker of ovar-
ian carcinogenesis. Cancer Res 1998;58:5363–73.
30. Rogers LW, Julian CG, Woodruff JD. Mesonephroid carcinoma of
the ovary: a study of 95 cases from the Emil Novak ovarian tumor
registry. Gynecol Oncol 1972;1:76–89.
31. Russel P. Surface epithelial–stromal tumors of the ovary. In:
Kurman RJ, ed. Blaustein’s Pathology of the Female Genital
Tract, ed 4. New York: Springer–Verlag, 1995:705–82.
32. Rutherford T, Brown WD, Sapi E, et al. Absence of estrogen
receptor-␤ expression in metastatic ovarian cancer. Obstet Gynecol
2000;96:417–21.
33. Sasa H, Ishii K, Hirata J, et al. Establishment and characterization
of a CA-125-producing human ovarian clear cell carcinoma cell
line. Hum Cell 1993;6:279–86.
34. Shevchuk MM, Winkler–Monsanto B, Fenoglio CM, et al. Clear
cell carcinoma of the ovary: a clinicopathologic study with review
of the literature. Cancer 1981;47:1344–51.
35. Shimizu M, Nikaido T, Toki T, et al. Clear cell carcinoma has an
expression pattern of cell cycle regulatory molecules that is unique
among ovarian adenocarcinomas. Cancer 1999;85:669–77.
36. Speirs V, Parkes AT, Kerin MJ, et al. Coexpression of estrogen
receptor ␣ and ␤: poor prognostic factors in human breast cancer?
Cancer Res 1999;59:525–8.