ONCOLOGY
Prognostic analysis of ovarian cancer
associated with endometriosis
Sanjeev Kumar, MD; Adnan Munkarah, MD; Haitham Arabi, MD; Sudeshna Bandyopadhyay, MD;
Assaad Semaan, MD; Kinda Hayek, MD; Gunjal Garg, MD; Robert Morris, MD; Rouba Ali-Fehmi, MD
OBJECTIVE: The objective of the study was to evaluate the prognosis of
ovarian cancer arising in endometriosis.
STUDY DESIGN: We retrospectively compared 42 cases of endometri-
osis-associated ovarian cancer (EAOC) with 184 cases of ovarian carci-
noma without endometriosis (OC).
RESULTS: The median age in the EAOC group was 52 vs 59 years in OC
(P .05). In comparison with OC, the EAOC patients were more likely to
have low-grade (21% vs 8%; P = .04) and early-stage tumors (Interna-
tional Federation of Gynecology and Obstetrics I and II combined) (49%
vs 24%; P = .002). Clear cell (21% vs 2%) and endometrioid (14% vs
3%) tumors were more frequent in EAOC, whereas mucinous tumors
were more prevalent in OC (P = .001). The median survival (199 vs 62
months) and the 5 year survival (62% vs 51%) were better for EAOC
when compared with OC (P = .038). After controlling for age, stage,
grade, and treatment, association with endometriosis was not an inde-
pendent predictor of better survival in ovarian cancer.
CONCLUSION: As such, EAOC has a much better survival rate than OC.
This could be explained by the higher prevalence of early-stage and
low-grade tumors in EAOC when compared with OC.
Key words: endometriosis, ovarian cancer, prognosis, survival
Cite this article as: Kumar S, Munkarah A, Arabi H, et al. Prognostic analysis of ovarian cancer associated with endometriosis. Am J Obstet Gynecol
2011;204:63.e1-7.
O
varian cancer is the gynecologic
malignancy with the highest case
fatality ratio. The high mortality of ovar-
ian cancer is due to the fact that the vast
majority of cases would have already dis-
seminated to the extraovarian sites at the
time of diagnosis. The late detection has
been partly attributed to the nonspecific
signs and symptoms of this disease in
early phases of cancer growth.
1
Endometriosis is characterized by pro-
liferation of endometrial glands and
stroma at an ectopic site, other than the
endometrium. It affects approximately
3-10% women in the reproductive age
group, 2-5% of postmenopausal women,
and 25-80% of infertile women.
2
Endo-
metrioid and clear cell subtypes of ovar-
ian cancer are both known to be closely
associated with endometriosis.
3
Al-
though endometriosis has not been clas-
sified as a premalignant condition, the
association of endometriosis and ovar-
ian cancer has been well described.
2
Endometriosis exhibits phenotypic fea-
tures of both a benign and malignant dis-
ease. The features of endometriosis, which
resemble that of cancer, are local and dis-
tant spread, loss of control of cell prolifer-
ation, and in some circumstances, inva-
sion. Brinton et al
4
in 2004 reported that
patients suffering from endometriosis
have a standard incidence ratio (SIR) of
4.19 (95% confidence interval [CI], 2.0 –
7.7) for ovarian cancer, when compared
with patients with no endometriosis.
Other investigators have reported a much
higher risk associated with endometriosis,
as depicted by the study by Kobayashi et al
5
in 2007 in which SIR for ovarian cancer
was 8.95 (95% CI, 4.12–15.3).
It is also evident from these studies
that the longer the duration of endome-
triosis, the greater is the risk for develop-
ing ovarian cancer. Numerous studies
document the progression of endometri-
osis from a benign histological lesion to a
frankly invasive malignant tissue. How-
ever, the precise mechanism for transfor-
mation from endometriosis to ovarian
cancer remains to be elucidated.
2
One keenly debated issue in the litera-
ture is whether endometriosis-associated
ovarian cancer (EAOC) is a distinct disease
entity or whether it is similar to the ovarian
cancer not associated with endometriosis
(OC). The existing literature on this issue
has several limitations. Some investigators
have reported a better overall survival
when ovarian cancer is associated with en-
dometriosis,
6,7
whereas others have failed
to find such a difference.
8,9
The histology reporting for EAOC is
extremely heterogeneous, making com-
parisons between the various reports
suboptimal. For example, some authors
specifically focused on clear cell tumors
6
or
From the Division of Gynecologic Oncology,
Department of Obstetrics and Gynecology
(Drs Kumar, Semaan, Morris, and Ali-
Fehmi), and the Department of Pathology
(Drs Bandyopadhyay, Hayek, and Ali-
Fehmi), Wayne State University School of
Medicine, and the Division of Gynecologic
Oncology, Department of Obstetrics and
Gynecology, Henry Ford Health System (Dr
Munkarah), Detroit, and the Division of
Gynecologic Oncology, Department of
Obstetrics and Gynecology, University of
Michigan School of Medicine, Ann Arbor
(Dr Garg), MI, and the Department of
Pathology, Emory University School of
Medicine, Atlanta, GA (Dr Arabi).
Received June 20, 2010; revised July 27,
2010; accepted Aug. 16, 2010.
Reprints: Rouba Ali-Fehmi, MD, Departments
of Pathology & Obstetrics-Gynecology, Wayne
State University and Karmanos Cancer
Institute, 3990 John R., Detroit, MI 48201.
rali@med.wayne.edu.
0002-9378/$36.00
© 2011 Mosby, Inc. All rights reserved.
doi: 10.1016/j.ajog.2010.08.017
Research www. AJOG.org
JANUARY 2011 American Journal of Obstetrics & Gynecology 63.e1