Clin Transl Oncol (2010) 12:783-787
DOI 10.1007/s12094-010-0599-0
E D U C AT I O N A L S E R I E S Blue Series
MOLECULAR AND CELLULAR BIOLOGY OF CANCER
Classification of ovarian carcinomas based on pathology
and molecular genetics
Emanuela D’Angelo · Jaime Prat
Received: 7 September 2010 / Accepted: 13 October 2010
Abstract Malignant epithelial tumours (carcinomas) are
the most common ovarian cancers and the most lethal
gynaecological malignancies. Based on light microscopy
and molecular genetics, ovarian carcinomas are subdivided
into at least five main subtypes that account for over 95%
of cases and are inherently different diseases, as indicated
by differences in epidemiological and genetic risk factors,
precursor lesions, patterns of spread, molecular events dur-
ing oncogenesis, response to chemotherapy and outcome.
For successful subtype-specific treatment, reproducible
pathological diagnosis of tumour cell type is critical. Re-
cent investigations have also demonstrated that a signifi-
cant number of cancers traditionally thought to be primary
ovarian tumours (particularly serous, endometrioid and
clear cell carcinomas) originate in the fallopian tube and
the endometrium and involve the ovary secondarily. In this
review we summarise recent advances in the molecular
pathology, which have greatly improved our understanding
of the biology of ovarian carcinoma and are also relevant to
patient management.
Keywords Ovary · Carcinoma · Pathology ·
Molecular genetics
E. D’Angelo · J. Prat (쾷)
Department of Pathology
Hospital de la Santa Creu i Sant Pau
Autonomous University of Barcelona
C/ Sant Antoni M. Claret, 167
ES-08025 Barcelona, Spain
e-mail: jprat@santpau.cat
Ovarian epithelial tumours are heterogeneous and primar-
ily classified according to cell type into serous, mucinous,
endometrioid, clear cell, transitional and squamous cell
tumours [1]. According to the World Health Organiza-
tion (WHO) classification [2], neoplasms in each of these
categories are further subdivided into benign, borderline
(intermediate) and malignant (carcinoma) forms, which are
associated with different prognoses. This is done micro-
scopically according to the amount of epithelial cell pro-
liferation, the degree of nuclear atypia (mild, moderate and
severe) and the presence or absence of stromal invasion [1,
2]. Borderline tumours (also designated as tumours of low
malignant potential) show epithelial proliferation greater
than that seen in their benign counterparts, variable nuclear
atypia, but an absence of stromal invasion, and are associ-
ated with much better prognosis than carcinomas. Despite
this lack of invasiveness within the ovary, serous borderline
tumours can either implant on peritoneal surfaces or be as-
sociated with independent foci of primary serous peritoneal
neoplasia and, rarely (about 10% of peritoneal implants),
invasion of the underlying tissues occurs. The biologic be-
haviour of invasive peritoneal implants is similar to that of
well differentiated (low-grade) serous carcinomas.
Malignant epithelial tumours (carcinomas) are the most
common ovarian cancers, accounting for 90% of cases,
and are the most lethal gynaecological malignancies [1].
In spite of significant improvements in cytoreduction and
chemotherapy, the overall outcome of patients continues
to be poor. Unlike colorectal carcinoma, a progression
model for ovarian carcinoma has not been described. Cur-
rently, however, based on light microscopy and molecular
genetics, ovarian carcinoma is subdivided into at least five
main subtypes which, in descending order of frequency, are
high-grade serous carcinomas (HGSC), clear cell carcino-
mas (CCC), endometrioid carcinomas (EC), mucinous car-
784
RHO-GTP RAC-GTP CDC-GTP
ROCK PAK
CRNPH
LIMK
SSP
1/2
PP1/2A
TESK1/2
NESK
COF
Actin reorganization
cinomas (MC) and low-grade serous carcinomas (LGSC)
(Fig. 1, Table 1) [3]. These tumours account for 98% of
ovarian carcinomas, can be reproducibly diagnosed and are
inherently different diseases, as indicated by differences in
epidemiological and genetic risk factors, precursor lesions,
patterns of spread, molecular events during oncogenesis,
response to chemotherapy and outcome. These differences
are the subject of this review. With progress towards sub-
type-specific management of ovarian carcinoma, reproduc-
ible pathological diagnosis of tumour cell type is critical.
Serous carcinomas
Recently, it has been recognised that LGSC and HGSC are
fundamentally different tumour types [4, 5]. The former are
associated with a serous borderline component in most cas-
es and are not related to BRCA abnormalities. In contrast,
HGSCs are not associated with serous borderline tumours
Clin Transl Oncol (2010) 12:783-787
Fig. 1 Representative examples of
the 5 main subtypes of ovarian car-
cinoma, which together account for
PLC-Gamma* GFS 98% of cases: (a) high-grade serous
carcinoma; (b) clear cell carcinoma;
(c) endometrioid carcinoma; (d) mu-
cinous carcinoma; (e) low-grade se-
rous carcinoma
and may arise from the fallopian tube in a significant num-
ber of cases [6, 7]. Because of the fundamental differences
between HGSC and LGSC, they are considered distinct
subtypes and will be discussed separately.
High-grade serous carcinomas (HGSC)
These are the most common ovarian carcinomas and most
patients present with high-stage disease; tumour confined
to the ovary at diagnosis is distinctly uncommon [8]. The
most distinctive growth pattern consists of highly stratified
epithelium with slit-like spaces. The tumour cells are typi-
cally of intermediate size, with scattered bizarre mononu-
clear giant cells; prominent nucleoli are common. The mi-
totic rate is very high (Fig. 1a). In contrast to LGSCs, these
tumours show more than 3-fold variation in nuclear size.
Most HGSCs have abnormalities of BRCA1 or BRCA2
(germline or somatic mutation or, in the case of BRCA1,
promoter methylation and loss of expression) and TP53
Clin Transl Oncol (2010) 12:783-787
Table 1 Ovarian carcinoma: clinical and molecular features of the 5 most common subtypes
HGSC LGSC
Risk factors BRCA1/2 ? ?
Precursor lesions Tubal intraepithelial Serous borderline
carcinoma tumour
Pattern of spread Very early Transcoelomic
transcoelomic spread spread
Molecular abnormalities BRCA, p53 BRAF, KRAS
Chemosensitivity High Intermediate
Prognosis Poor Intermediate
a
Hereditary non-polyposis colorectal carcinoma
(mutation or deletion) [5, 9]. These changes occur early
during oncogenesis and result in loss of ability to repair
DNA double strand breaks, which in turn leads to chro-
mosomal instability. There is also a very high proliferative
rate in the earliest in situ lesions. Tubal intraepithelial car-
cinoma cells also react for gamma-H2AX, which localises
at sites of DNA double-strand breaks. From this point for-
ward during oncogenesis, there is rapid accrual of addi-
tional genetic abnormalities, such that ovarian carcinomas
typically show aneuploidy with considerable intratumoral
genetic heterogeneity [10]. The fundamental underlying
abnormalities in these cells are the inability to repair DNA
and increased proliferation. Carcinomas arising in patients
with germline BRCA1 or BRCA2 mutations are almost
invariably of high-grade serous type. Among diagnostic
immunomarkers, WT1 immunoreaction occurs in approxi-
mately 80% of cases of HGSC and LGSC, but in less than
5% of ovarian carcinomas of other subtypes. Oestrogen
receptor is expressed in approximately two thirds of cases
of HGSC, and is also expressed in LGSC and endometrioid
carcinoma, but is negative in almost all clear cell carcino-
mas and all mucinous carcinomas [11].
Primary surgical debulking is the initial approach in
most patients. Most (70–80%) HGSCs show a response to
platinum/taxane chemotherapy, but most patients will sub-
sequently experience a recurrence, at which point cure is
not possible [12]. Because of the poor outcome for patients
with HGSC, there is a desperate need for new treatments.
One possibility, currently in clinical trials, is Parp inhibi-
tors that target the underlying molecular abnormality in
HGSC, i.e., loss of BRCA function and inability to repair
double-strand breaks in DNA [13].
Low-grade serous carcinomas (LGSC)
LGSCs are uncommon and account for less than 5% of all
cases of ovarian carcinoma. Small foci of LGSC in a bor-
derline tumour is associated with an excellent prognosis.
The prognosis for patients with advanced-stage disease is
less favourable, although the disease may follow a relative-
ly indolent course, with long periods between recurrences
[14]. LGSCs have uniform nuclei and differentiated archi-
785
MC EC CCC
HNPCCa ?
Cystadenoma/borderline Endometriosis Endometriosis
tumour?
Usually confined Usually confined Usually confined
to ovary to pelvis to pelvis
KRAS, HER2 PTEN HNF1
Low High Low
Favourable Favourable Intermediate
tecture with papillary growth; numerous psammoma bod-
ies are a common feature (Fig. 1e). The uniformity of the
nuclei is the principal criterion for distinguishing between
LGSC and HGSC, with less than 3-fold variability [15].
BRAF or KRAS mutations are present in most LGSCs
[4]. LGSCs do not show chromosomal instability and lack
the complex genetic abnormalities seen in high-grade se-
rous carcinomas. LGSCs are not associated with BRCA
germline mutations. The response rate to conventional ther-
apy for this subset is difficult to determine as this group has
only recently been recognised. There are no studies on well
characterised groups of cases that are known to lack muta-
tions in BRCA/TP53, and existing data may reflect case se-
ries that include some cases of HGSC. In most cases, these
tumours do not respond to conventional ovarian carcinoma
chemotherapy [16].
Mucinous carcinomas (MC)
The tumour cells of MCs may resemble those of the endo-
cervix, gastric pylorus or intestine [1] (Fig. 1d). The large
majority of these tumours show gastrointestinal differen-
tiation and are the focus of this discussion. Although mu-
cinous tumours account for 10–15% of all primary ovarian
tumours, approximately 80% are benign and most of the
remainder are borderline tumours. Only 3–4% of ovarian
carcinomas are of mucinous type, if metastases to the ovary
are carefully excluded. Large size (>13 cm) and unilateral-
ity are features suggestive of a primary MC, while metas-
tases are typically smaller and bilateral. Primary MCs of
the ovary are usually confined to the ovary, without ovarian
surface involvement or pseudomyxoma peritonei. Muci-
nous ovarian tumours are often heterogeneous. Benign-
appearing, borderline, non-invasive carcinoma and invasive
patterns may coexist within an individual neoplasm, and
adequate sampling for histological examination is critical
[17]. The category of mucinous borderline tumour with in-
traepithelial carcinoma is used for those tumours that lack
the architectural features to support a diagnosis of invasive
carcinoma but show areas where the cytological features
of cells lining the glandular spaces are unequivocally ma-
786
lignant. Mucinous borderline tumours with intraepithelial
carcinoma have a very low likelihood of recurrence, of less
than (probably much less than) 5% [1].
KRAS mutations are common in mucinous ovarian car-
cinomas and are an early event in mucinous tumorigenesis
[18]. HER2 is amplified in 15–20% of ovarian carcinomas
of mucinous type. Primary ovarian mucinous tumours are
almost always (up to 80%) immunoreactive for cytokeratin
7 (CK7) whereas colorectal adenocarcinomas are usually
CK7 negative [19]. Ovarian MCs are immunoreactive for
CK20 in 65% of cases, but the reaction is typically weak
and focal; staining for cdx-2 is similar. In contrast, colorec-
tal adenocarcinomas are diffusely and strongly reactive for
CK20 and cdx-2. Loss of Dpc4 immunoreactivity occurs
in almost 50% of metastatic carcinomas of the pancreas,
whereas most primary ovarian MCs are focally or diffusely
positive. Cervical adenocarcinomas metastatic to ovary are
positive for HPV and show strong diffuse positivity for
p16. MCs are uniformly negative for oestrogen receptor
(ER) and WT1, in contrast to endometrioid (ER+) and se-
rous (ER+ and WT1+) carcinomas.
As older case series invariably include an admixture
of metastatic and primary MCs, it is difficult to ascertain
the response rate of MC to adjuvant chemotherapy. Recent
studies have indicated lower response rates (15–35%) to
platinum-based chemotherapy compared to HGSC [20],
prompting a search for alternative therapies. Trastuzumab
therapy may prove effective in those patients whose tu-
mours show high-level amplification of the HER2 gene, but
clinical studies to establish the effectiveness of trastuzumab
in this setting have not been done.
Endometrioid carcinomas (EC)
Endometrioid tumours of the ovary closely resemble those
encountered more frequently in the endometrium. ECs ac-
count for 10% of all ovarian carcinomas and most of them
are diagnosed at stage I or II [1]. ECs occur most frequent-
ly in women in the perimenopausal or postmenopausal age
groups. Up to 42% of cases are associated with ipsilateral
ovarian or pelvic endometriosis. ECs of ovary are bilateral
in 28% of cases and are associated in 15–20% of cases with
carcinoma of the endometrium [21]. Most are grade 1 or 2
and show round or tubular glands lined by stratified non-
mucin-containing epithelium. Cribriform or villoglandular
patterns may be present (Fig. 1c). Squamous differentiation
occurs in 50% or more of cases [1].
Somatic mutations of the beta-catenin (CTNNB1) and
PTEN genes are the most common genetic abnormali-
ties identified in ovarian ECs [22–24]. Compared with
uterine EC, ECs of the ovary have a similar frequency of
beta-catenin abnormalities but lower rate of microsatellite
instability (MI) and PTEN alterations [24]. The incidence
of CTNNB1 mutations ranges from 38% to 50%. Mutations
have been described in exon 3 (codons 32, 33, 37 and 41)
Clin Transl Oncol (2010) 12:783-787
and involve the phosphorylation sequence for glycogen
synthase kinase 3-beta. These mutations probably render a
fraction of cellular beta-catenin insensitive to APC-medi-
ated down-regulation and are responsible for its accumula-
tion in the nuclei of the tumour cells. Beta-catenin is im-
munohistochemically detectable in carcinoma cells in more
than 80% of the cases. PTEN is mutated in approximately
20% of tumours and in 46% of those with 10q23 LOH.
PTEN mutations occur between exons 3 and 8. The finding
of 10q23 LOH and PTEN mutations in endometriotic cysts
adjacent to ECs with similar genetic alterations provides
additional evidence for the precursor role of endometriosis
in ovarian carcinogenesis [25]. ECs are the subtype most
commonly encountered in patients with hereditary non-
polyposis colon cancer syndrome. ECs are immunoreac-
tive for vimentin, cytokeratins (CK7, 97%; CK20, 13%),
epithelial membrane antigen (EMA), and oestrogen and
progesterone receptors. Immunoreaction for alpha-inhibin,
WT-1 and calretinin are negative in most ECs.
Because the diagnosis of EC has been used with less
stringent criteria in the past, the response to therapy of
these tumours is difficult to evaluate as different case series
are not directly comparable; nonetheless, when diagnosed
accurately, this is the subtype of ovarian carcinoma with
the most favourable prognosis. ECs are lower grade and
lower stage than other ovarian carcinoma subtypes, which
accounts for much of the favourable prognosis, but chemo-
responsiveness may also be a contributory factor.
Clear cell carcinomas (CCC)
This is the most enigmatic subtype of ovarian carcinoma.
The presence of “clear cells” alone is not sufficient for a
diagnosis of CCC, as cells with clear cytoplasm can be
seen in HGSC and EC. The accurate diagnosis of CCC
depends on consideration of the constellation of architec-
tural and cytological findings characteristic of this tumour
type. CCCs account for approximately 10% of ovarian
carcinomas and patients typically present with stage 1 or 2
disease. CCCs are associated with a relatively unfavourable
prognosis, compared to other low-stage ovarian carcino-
mas. As with EC, there is an association with endometrio-
sis, and CCCs associated with endometriosis have a favour-
able prognosis [26]. Characteristic microscopic features of
CCC include: (a) multiple complex papillae; (b) densely
hyaline basement membrane material expanding the cores
of the papillae; and (c) hyaline bodies, which are present
in approximately 25% of cases (Fig. 1b). Mitoses are less
frequent than in other types of ovarian carcinomas (usually
less than 5/10 HPFs).
Very little is known about the genetic alterations of
CCCs. They lack the BRCA abnormalities, chromosomal
instability or complex karyotypes of HGSC [27]. CCCs are
usually positive for HNF1-beta and are negative for ER and
WT1 in more than 95% of cases [3].
Clin Transl Oncol (2010) 12:783-787 787

CCCs are less likely to respond to chemotherapy than Conclusions

HGSCs [12]. The reported differences in response rates
(15–45%) may reflect inclusion of HGSC with clear cell
change in some case series. The lowest response rates are
reported from Japan, where HGSC is relatively less com-
mon and the CCC case series may therefore be less likely
to have admixed cases of HGSC. Whereas highly prolif-
erative cells that lack the ability to repair double-stranded
DNA (i.e., high-grade serous carcinoma cells) can be
anticipated to show sensitivity to platinum-based chemo-
therapy, the less proliferative, genomically stable cells of
CCC can similarly be anticipated to be less sensitive to
platinum compounds [28]. At this point there are no clearly
superior alternatives to platinum-based chemotherapy.
Recently, however, postoperative whole abdominal radio-
therapy was shown to be effective in patients with stage Ic–
III CC, compared to a historical control group treated with
platinum-based chemotherapy [29].
Significant progress has been made in the histopathological
subclassification of ovarian surface epithelial carcinomas.
The five main subtypes of ovarian carcinoma (in descend-
ing order of frequency: HGSC, CCC, EC, MC and LGSC)
account for 98% of ovarian carcinomas, can be reproduc-
ibly diagnosed, and are inherently different diseases, as in-
dicated by differences in risk factors, molecular abnormali-
ties, natural history and response to chemotherapy. For a
successful subtype-specific treatment of ovarian carcinoma,
accurate subtype assignment by pathologists is becoming
increasingly important.
Acknowledgements This study is supported by Grants RTICC
RD06/0020/0015 and FIS 080410, Department of Health, Spain.
Conflict of interest The authors declare that they have no conflict of
interest relating to the publication of this manuscript.

References
1. Prat J (2004) Pathology of the ovary. Saunders,
Philadelphia
2. Lee KR, Tavassoli FA, Prat J et al (2003) Surface
epithelial-stromal tumours (in Ch. 2: Tumours
of the ovary and peritoneum). In: Tavassoli FA,
Devilee P (eds) World Health Organization clas-
sification of tumours: pathology and genetics of
tumours of the breast and female genital organs.
IARC Press, Lyon, pp 117–145
3. Gilks CB, Prat J (2009) Ovarian carcinoma pa-
thology and genetics: recent advances. Hum
Pathol 40:1213–1223
4. Singer G, Oldt R 3rd, Cohen Y et al (2003) Muta-
tions in BRAF and KRAS characterize the devel-
opment of low-grade ovarian serous carcinoma. J
Natl Cancer Inst 95:484–486
5. Singer G, Stohr R, Cope L et al (2005) Patterns of
p53 mutations separate ovarian serous borderline
tumors and low- and high-grade carcinomas and
provide support for a new model of ovarian car-
cinogenesis. Am J Surg Pathol 29:218–224
6. Medeiros F, Muto MG, Lee Y et al (2006) The
tubal fimbria is a preferred site for early adeno-
carcinoma in women with familial ovarian cancer
syndrome. Am J Surg Pathol 30:230–236
7. Kindelberger DW, Lee Y, Miron A et al (2007)
Intraepithelial carcinoma of the fimbria and pelvic
serous carcinoma: evidence for a causal relation-
ship. Am J Surg Pathol 31:161–169
8. Leitao MM, Boyd J, Hummer A et al (2004) Clin-
icopathological analysis of early-stage sporadic
ovarian carcinoma. Am J Surg Pathol 28:147–159
9. Esteller M, Silva JM, Dominguez G et al (2000)
Promoter hypermethylation is a cause of BRCA1
inactivation in sporadic breast and ovarian tumors.
J Natl Cancer Inst 92:564–569
10. Khalique L, Ayhan A, Weale ME et al (2007)
Genetic intra-tumour heterogeneity in epithelial
ovarian cancer and its implications for molecular
diagnosis of tumours. J Pathol 211:286–295
11. Al-Hussaini M, Stockman A, Foster H, McClug-
gage WG (2005) WT-1 assists in distinguishing
ovarian from uterine serous carcinoma and in
distinguishing between serous and endometrioid
ovarian carcinoma. Histopathology 46:468
12. duBois A, Luck HJ, Meier W et al (2003) A ran-
domized clinical trial of cisplatin/paclitaxel versus
carboplatin/paclitaxel as first-line treatment of
ovarian cancer. J Natl Cancer Inst 95:1320–1329
13. Farmer H, McCabe N, Lord CJ et al (2005) Target-
ing the DNA repair defect in BRCA mutant cells
as a therapeutic strategy. Nature 434:917–921
14. Gershenson DM, Sun CC, Lu KH et al (2006)
Clinical behavior of stage II-IV low-grade serous
carcinoma of the ovary. Obstet Gynecol 108:361–
368
15. Malpica A, Deavers MT, Lu K et al (2004) Grad-
ing ovarian serous carcinoma using a two-tier
system. Am J Surg Pathol 28:496–504
16. Crispens MA, Bodurka D, Deavers M et al (2002)
Response and survival in patients with progres-
sive or recurrent serous ovarian tumors of low
malignant potential. Obstet Gynecol 99:3–10
17. Rodriguez IM, Prat J (2002) Mucinous tumors of
the ovary. A clinicopathologic analysis of 75 bor-
derline tumors (of intestinal type) and carcinomas.
Am J Surg Pathol 26:139–152
18. Cuatrecasas M, Villanueva A, Matias-Guiu X, Prat
J (1997) K-ras mutations in mucinous ovarian tu-
mors. A clinicopathologic and molecular study of
95 cases. Cancer 79:1581–1586
19. Park SY, Kim HS, Hong EK, Kim WH (2002) Ex-
pression of cytokeratins 7 and 20 in primary car-
cinomas of the stomach and colorectum and their
value in the differential diagnosis of metastatic
carcinomas to the ovary. Hum Pathol 33:1078–
1085
20. Hess V, A’Hern R, Nasiri N et al (2004) Mucinous
epithelial ovarian cancer: a separate entity requir-
ing specific treatment. J Clin Oncol 22:1040–1044
21. Irving JA, Catasus L, Gallardo A et al (2005) Syn-
chronous endometrioid carcinomas of the uterine
corpus and ovary: alterations in the beta-catenin
(CTNNB1) pathway are associated with inde-
pendent primary tumors and favorable prognosis.
Hum Pathol 36:605–619
22. Obata K, Morland SJ, Watson RH et al (1998)
Frequent PTEN/MMAC mutations in endometri-
oid but not serous or mucinous epithelial ovarian
tumors. Cancer Res 58:2095–2097
23. Palacios J, Gamallo C (1998) Mutations in the
beta-catenin gene (CTNNB1) in endometrioid
ovarian carcinomas. Cancer Res 58:1344–1347
24. Catasús L, Bussaglia E, Rodríguez IM et al (2004)
Molecular genetic alterations in endometrioid
carcinomas of the ovary: similar frequency of
beta-catenin abnormalities but lower rate of mi-
crosatellite instability and PTEN alterations than
in uterine endometrioid carcinomas. Hum Pathol
35:1360–1368
25. Sato N, Tsunoda H, Nishida M et al (2000) Loss
of heterozygosity on 10q23.3 and mutation of
the tumor suppressor gene PTEN in benign en-
dometrial cyst of the ovary: possible sequence
progression from benign endometrial cyst to en-
dometrioid carcinoma and clear cell carcinoma of
the ovary. Cancer Res 60:7052–7056
26. Komiyama S, Aoki D, Tominaga E et al (1999)
Prognosis of Japanese patients with ovarian clear
cell carcinoma associated with pelvic endometrio-
sis: clinicopathologic evaluation. Gynecol Oncol
72:342–346
27. Press JZ, de Luca A, Boyd N et al (2008) Ovarian
carcinomas with genetic and epigenetic BRCA1
loss have distinct molecular abnormalities. BMC
Cancer 8:17
28. Itamochi H, Kigawa J, Sugiyama T et al (2002)
Low proliferation activity may be associated with
chemoresistance in clear cell carcinoma of the
ovary. Obstet Gynecol 100:281–287
29. Nagai Y, Inamine M, Hirakawa M et al (2007)
Postoperative whole abdominal radiotherapy in
clear cell adenocarcinoma of the ovary. Gynecol
Oncol 107:469–473