Clinical Expert Series

Endometrial Cancer
Joel I. Sorosky, MD

The epidemiology, prevention, diagnosis, treatment, prognosis, and new International Federa-
tion of Gynecology and Obstetrics staging system of endometrial carcinoma are reviewed.
Endometrial cancer has increased 21% in incidence since 2008, and the death rate has increased
more than 100% over the past two decades. Precursor lesions of complex hyperplasia with atypia
are associated with an endometrial carcinoma in more than 40% of cases. Endometrial cancer in
white women occurs at twice the incidence as in black women, but, stage for stage, black women
have a less favorable prognosis. Preoperative imaging cannot accurately assess lymph node
involvement. Gross examination of depth of myometrial invasion does not have the sensitivity,
specificity, positive predictive value, or negative predictive value to select women who can have
lymphadenectomy safely omitted from the surgical procedure. Although surgical staging remains
the most accurate method of determining the extent of disease, the therapeutic value of pelvic
lymphadenectomy has not been established. The anatomical extent of lymphadenectomy and
the number of lymph nodes removed to establish prognostic and therapeutic benefit are
controversial. Research efforts are directed at identifying women with early stage endometrial
cancer who only require total hysterectomy and bilateral salpingo-oophorectomy. Minimally
invasive surgical techniques have become established as standard therapy for treating women
with endometrial cancer. Women with a family history of hereditary nonpolyposis colorectal
cancer are at increased risk for endometrial cancer. Conservative treatment to allow for
childbearing is possible in select situations. Women with endometrial cancer should be managed
by physicians experienced in the complex multimodality treatment of this disease.
(Obstet Gynecol 2012;120:383–97)
DOI: 10.1097/AOG.0b013e3182605bf1

E ndometrial cancer is the most common gyneco-

logic malignancy in the United States and the fourth
most common cancer in women after breast, lung, and
rics & Gynecology in 2008.2 The death rate per 100,000
population has increased more than 100% during the
past 20 years and 8% since 2008. This is distressing
colorectal cancers. The American Cancer Society esti- given that overall death rates from cancer have de-
mates there will be 47,130 new cases and 8,010 deaths in creased 1.6% per year in women, and these declines
2012.1 The incidence of endometrial cancer has in- have been consistent since 2001.1 Ninety percent of
creased 21% since this topic was last reviewed in Obstet- women present with abnormal uterine bleeding and
nearly 75% of women present with early stage disease. It
is postulated that the increased death rate may be related
From the Department of Obstetrics and Gynecology, Hartford Hospital and the to an increasing life span and coexisting medical comor-
Hospital of Central Connecticut, The University of Connecticut, Hartford,
Connecticut.
bidities in these women. Women with an advanced
Continuing medical education for this article is available at [Link]
stage diagnosed may have aggressive disease, denial of
com/AOG/A309. symptoms, genetic risk, confounding symptoms, or mis-
Corresponding author: Joel I. Sorosky, Department of Obstetrics and diagnosis. This review is directed toward the general
Gynecology, Hartford Hospital and the Hospital of Central Connecticut, The obstetrician gynecologist. Emphasis is directed to scien-
University of Connecticut, 80 Seymour Street, Hartford, CT 062102; tific advances in this disease because this topic was last
e-mail: jsorosk@[Link].
reviewed in Obstetrics & Gynecology in 2008. A new
Financial Disclosure
The author did not report any potential conflicts of interest.
International Federation of Gynecology and Obstetrics
(FIGO) staging system has been implemented and min-
© 2012 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins. imally invasive surgery has become accepted surgical
ISSN: 0029-7844/12 therapy, but the roles of lymph node staging and

VOL. 120, NO. 2, PART 1, AUGUST 2012 OBSTETRICS & GYNECOLOGY 383
postoperative adjuvant radiation therapy have become
more controversial. The Ovid database was queried
from 2007 through the first week of February 2012. The
key words “endometrial neoplasms including diagno-
sis,” “epidemiology,” “genetics,” “pathology,” “preven-
tion and control,” “radiotherapy,” “surgery,” and “ther-
apy” were used to search for manuscripts in English.
Sarcomas of the uterus are not discussed here.
The mean age for endometrial adenocarcinoma is
61 years, with most cases diagnosed in women be-
tween the ages of 50 and 60 years.2 Ninety percent of
cases occur in women older than 50 years. Endo-
metroid is the most common histologic type. Approx-
imately 20% of women have diagnoses before meno-
pause and approximately 5% of women will have
development of disease before age 40 years. Approx-
imately 72% of endometrial cancers are stage I, 12%
are stage II, 13% are stage III, and 3% are stage IV.
Endometrial cancer commonly has been classi-
fied into three types (Table 1). Type I commonly is Fig. 1. Endometroid adenocarcinoma (hematoxylin and
estrogen-related and occurs in younger, obese, or eosin stain, 40⫻ magnification).
perimenopausal women. These tumors are usually Sorosky. Endometrial Cancer. Obstet Gynecol 2012.
low-grade and arise in a background of hyperplasia
(Figs. 1 and 2). Type I disease represents the majority
Hereditary or genetic disease can have a familial
of endometrial cancers. Endometroid is the most
association or can be part of the Lynch syndrome,
common histology. These tumors may show micro-
hereditary nonpolyposis colorectal cancer. Genetic
satellite instability and mutations in PTEN, PIK3CA,
disease can represent up to 10% of cases, of which 5%
K-ras, and CTNNBI.3 It has not been established with
are Lynch syndrome. A history should be taken from
certainty whether obesity is behavioral, social, ge-
all women with endometrial cancer to determine if
netic, or an admixture. A genetic predisposition to
obesity can increase the risk of endometrial cancer.
Type II disease generally has high-grade tumors, is of
serous or clear cell histology, occurs in an older
cohort of women than type I, and is more common in
black women (Figs. 3 and 4). These tumors may
exhibit p-53 mutations in approximately 10 –30% of
cases. Type II disease represents up to 10% of cases.
The epidemiologic profile of women with type II
disease is not certain.

Table 1. Classification of Endometrial Cancer

Type I Type II Familial

Low-grade High-grade Lynch

Minimal myometrial Deep myometrial invasion
invasion
Arising in a background Serous or clear-cell
of hyperplasia
Perimenopausal
Estrogen-related
Younger age
Obesity Fig. 2. Endometroid adenocarcinoma (hematoxylin and
Reprinted from Sorosky JI. Endometrial cancer. Obstet Gynecol eosin stain, 400⫻ magnification).
2008;111:436 – 47. Sorosky. Endometrial Cancer. Obstet Gynecol 2012.

384 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

 RISK FACTORS
The epidemiology of endometrial cancer includes
women with genotypic and phenotypic risk. A recent
prospective study reported that nearly 70% of women
with early stage endometrial cancer were obese.4 This
is greater than twice the percentage of previous
reports.5 The relative risk (RR) for death increased
with the body mass index (BMI, calculated as weight
(kg)/[height (m)]2). For women with endometrial can-
cer and BMIs of 24 –30, the RR of death was 2.53,
with BMIs of 35– 40 the RR was 2.77, and with BMIs
more than 40 the RR of death increased to 6.25.6,7
Women who survived endometrial cancer had the
highest rate of death when compared with other can-
cers. Women with endometrial cancer have other med-
ical comorbidities that contribute to them dying from
causes other than the cancer. Endometrial cancer survi-
vors have unhealthy lifestyles that put them at risk for
morbidity.8 After diagnosis and treatment, survivors
Fig. 3. Serous endometrial carcinoma (hematoxylin and
eosin stain, 40⫻ magnification).
should be offered muiltibehavorial lifestyle interven-
Sorosky. Endometrial Cancer. Obstet Gynecol 2012. tions. Another study also demonstrated that obese en-
dometrial cancer patients had a higher mortality from
medical comorbidities. Women with BMIs more than
there is a hereditary component. If the woman is at 40 had significantly shorter survival and experienced
risk for hereditary disease, then she may be offered more endometrial cancer– unrelated deaths when com-
genetic counseling and consideration should be given pared with nonobese women.7 A single institution study
for genetic testing. Other family members may be at of 442 women found BMI was also correlated to tumor
risk for hereditary cancers and also may be candidates grade, race, and stage at diagnosis.9
for genetic counseling. Reproductive, menstrual, and medical risk co-
morbidities can increase or decrease the risk of a
woman having development of endometrial cancer.10
Continuous estrogen stimulation, albeit it exogenous
or endogenous, can alter the normal endometrial
cycle. Anovulation results in continuous unopposed
estrogen stimulation because there is no corpus lu-
teum to produce progesterone. Anovulation is com-
mon during perimenopause. Obesity, generally more
than 50 pounds over ideal body weight, can result in
endogenous estrogen because of peripheral conver-
sion of androstenedione into estrone. Women with
BMIs more than 30 have two times to three times the
risk of development of endometrial cancer. Data from
observational studies report that both symptomatic
vaginal bleeding and postmenopausal status in
women with endometrial polyps are associated with
an increased risk of endometrial cancer.11
Estrogen-producing tumors, unopposed estrogen
therapy, cirrhosis, and tamoxifen also may result in
excess estrogen stimulation to the endometrium. Al-
Fig. 4. Serous endometrial carcinoma (hematoxylin and though tamoxifen is an antiestrogen in breast tissue, it
eosin stain, 400⫻ magnification). can have estrogenic activity in the endometrium.12
Sorosky. Endometrial Cancer. Obstet Gynecol 2012. Unopposed estrogen replacement in menopause is

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 385
associated with a fourfold to eightfold increased risk The majority of cases of endometrial adenocarci-
of disease, whereas estrogen and progesterone re- noma results from estrogenic stimulation with com-
placement therapy in the menopause decreases the plex hyperplasia with atypia or endometrial intraepi-
risk of disease. The extent of reduction of cancer risk thelial neoplasia as a precursor lesion. Approximately
with progestin is based on the type of progestin 10% of endometrial adenocarcinoma is not related to
administered, dosage, and duration of use. Progestin- estrogen excess, that is, type II. These tumors tend to
containing or combination oral contraceptives de- occur at an older age than estrogen-associated lesions
crease the risk of disease by approximately 50%.13 and are poorly differentiated with serous or clear-cell
Nulliparity and diabetes are associated with a two- histology. Uterine papillary serous carcinoma is an
fold to threefold increased incidence of disease, whereas aggressive tumor that histologically resembles papil-
hypertension appears to be related to obesity and dia- lary serous ovarian carcinoma. There are no risk
betes and is not an independent risk factor. Nulliparity is factors, as opposed to type 1 disease. Serous carci-
believed to be related to infertility rather than inten- noma of the endometrium is more commonly seen in
tional prevention of pregnancy. Infertility related to older women and in black women. Clear-cell uterine
anovulation and progesterone deficiency increase the adenocarcinoma also is an aggressive tumor that is
risk as opposed to infertility related to tubal factors. histologically similar to clear-cell adenocarcinoma
The incidence of endometrial cancer in white that is seen in the ovary or cervix.
women is twice the incidence of that of black
women.14,15 Stage for stage, black women have a less ENDOMETRIAL HYPERPLASIA
favorable prognosis. It is unclear if this survival Women with endometrial hyperplasia are usually
disadvantage for black women is related to tumor identified during evaluation of menometrorrhagia or
aggressiveness, effectiveness of treatment, differences postmenopausal bleeding. Diagnosis is made by his-
in the extent of disease within similar stages, or tologic evaluation of the endometrium. From 1985
biologic variation by race. More research is needed to until 2006, it was believed that complex hyperplasia
clarify these racial disparities. with atypia had a 29% likelihood of progressing to
Tamoxifen, when used for chemoprevention of endometrial cancer, whereas simple hyperplasia with or
breast cancer, increased the risk of endometrial can- without atypia and complex hyperplasia without atypia
cer nearly threefold. Whereas the majority of tamox- had a small incidence of progression to cancer.19 This
ifen-induced carcinomas were endometroid in histol- retrospective study had a small number of patients.
ogy, low-grade, and staged, there are reports of Medical therapy with progestins was recommended for
tamoxifen-associated high-grade tumors and sarco- women with simple hyperplasia with or without atypia
mas. Women with a history of tamoxifen therapy and complex hyperplasia without atypia diagnosed.
should have continued surveillance after completion Hysterectomy was recommended for those women with
of tamoxifen to allow for early diagnosis of uterine complex hyperplasia with atypia. Medical therapy with
cancer. In one study of 106 women, nearly 40% of progestins was recommended in instances when future
women who had development of uterine cancer after childbearing was desired.
tamoxifen therapy did so more than 12 months after The Gynecologic Oncology Group (GOG) sought
discontinuation with a median time of 33 months.12,16,17 to validate the reproducibility of the referring institu-
Screening asymptomatic women using tamoxifen with tion’s pathologic diagnosis of complex hyperplasia with
ultrasonography or endometrial biopsy is not recom- atypia (atypical endometrial hyperplasia) in a multi-
mended. These women should be educated about institutional prospective manner.20 In the studies cited,
reported symptoms of increased vaginal discharge or complex hyperplasia with atypia and atypical endome-
bleeding to their physician. Symptomatic women trial hyperplasia are used interchangeably. A panel of
should undergo evaluation for endometrial cancer or three pathologists independently reviewed the speci-
its precursors. mens from 306 women with the diagnosis based on
Oral estrogen replacement is associated with a either endometrial biopsy or curettage. These women
2-fold to 12-fold increased risk of endometrial cancer. underwent hysterectomy within 12 weeks of biopsy
Increasing dosage and duration of use are associated diagnosis of atypical endometrial hyperplasia. The re-
with increased risk. This risk continues until 2 to 3 ferring institution’s pathologist’s diagnosis of atypical
years after cessation of estrogen therapy. Tumors endometrial hyperplasia was supported by the majority
associated with estrogen use are usually early stage at of the panel in only 38% of cases. The majority diagnosis
diagnosis and of lower grade.18 These tumors gener- was adenocarcinoma in 29% of cases, cycling endome-
ally arise in a background of hyperplasia. trium in 7%, and hyperplasia without atypia in 18% of

386 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

cases. Unanimous agreement for any diagnosis was In a population-based retrospective cohort study
reached among all three pathologists in only 40% of of 824 women within a large health maintenance
cases. Reproducibility of the referring institution’s pa- organization, dilation and curettage (D&C) performed
thologist’s diagnosis of atypical endometrial hyperplasia after a biopsy-proven diagnosis of complex atypical
by a panel of gynecologic pathologists was poor. Both hyperplasia lowered the rate of unexpected cancer
underestimation and overestimation of the severity of compared with biopsy alone.22 Overall, 48% of
the lesion were very common. The level of reproduc- women with complex atypical hyperplasia were
ibility among the three reference pathologists also was found to have cancer. Preoperative D&C was associ-
poor. Better criteria and better sampling are needed to ated with a statistically significant decrease in the risk
improve reproducibility of atypical endometrial hyper- of having cancer found at surgery when compared
plasia if this diagnosis is to be used for clinical decisions. with at biopsy alone. Although 18% of women still
Given that the reproducibility of the diagnosis of atypi- had an invasive endometrial carcinoma at hysterec-
cal endometrial hyperplasia on permanent section is tomy, D&C is recommended to reduce the incidence
poor, clinical decisions based on a frozen-section diag- of unanticipated cancer in the final hysterectomy
nosis of atypical endometrial hyperplasia would be specimen. This could allow for triage and evaluation
similar if not less reproducible. The use of intraoperative of cancer before definitive surgery. In this study, the
frozen-section diagnosis is not recommended. risk of unexpected cancer was strongly related to age.
In a multi-institutional prospective study, GOG Given the incidence of more than 50% myometrial
also estimated the prevalence of concurrent carci- invasion in women with a preoperative diagnosis of
noma in women with a referring institution’s patho- atypical endometrial hyperplasia and an overall 42.6%
logic diagnosis of atypical endometrial hyperplasia.21 incidence of malignancy, a preoperative gynecologic
A panel of three pathologists independently reviewed oncologic consultation may be prudent in women with
the specimens from 306 women with the diagnosis atypical endometrial hyperplasia until the criteria for
based on either endometrial biopsy or curettage. reproducibility of atypical endometrial hyperplasia and
These women underwent hysterectomy within 12 distinction from invasive carcinoma are improved.
weeks of biopsy diagnosis of atypical endometrial Women with simple and complex hyperplasia
hyperplasia. Of 289 specimens that met inclusion without atypia and those with simple hyperplasia with
criteria, the study panel review found 25.6% were atypia may be treated with progestins because the risk
diagnosed as less than atypical endometrial hyperpla- of endometrial carcinoma is low. Common regimens
sia, 39.8% were diagnosed as atypical endometrial include cyclic medoxyprogesterone 10 mg daily for
hyperplasia, and 29.1% (independent of the review of 14 days per month or continuous megesterol acetate
the previous biopsy specimen) were diagnosed as 20 – 40 mg daily. Alternatively, a progestin-releasing
endometrial cancer. In 5.5% there was no consensus intrauterine device has been reported to be effective
on the diagnosis. The overall rate of endometrial in treating hyperplasia.23
carcinoma was 42.6% (including biopsy specimen and Women with complex hyperplasia with atypia
hysterectomy specimen, 123 of 289 specimens). Of have a more than 40% incidence of a coexistent
these, 30.9% (38 of 123 specimens) had myometrial adenocarcinoma. Hysterectomy is the treatment of
invasion and 10.6% (13 of 123 specimens) had inva- choice for complex hyperplasia with atypia. For those
sion to the outer half of the myometrium. Among the individuals with atypical endometrial hyperplasia de-
women who had hysterectomy specimens with carci- siring childbearing or those with medical comorbidi-
noma, 14 of 74 women (18.9%) had a study panel ties that preclude surgery, treatment with progestins
consensus diagnosis of less than atypical endometrial to allow for subsequent pregnancy has been re-
hyperplasia and 54 of 84 women (64.3%) had a study ported.24 The endometrium should be evaluated
consensus diagnosis of carcinoma. Among women within 3– 4 month. After childbearing is completed in
who had no consensus in their biopsy diagnosis, 10 of women with complex hyperplasia with atypia, hyster-
16 women (62.5%) had carcinoma in their hysterec- ectomy is recommended.
tomy specimens. These authors concluded that the A retrospective cohort study of 1,443 women in
prevalence of endometrial carcinoma in women with an integrated health plan with endometrial complex
a community hospital biopsy diagnosis of atypical atypical hyperplasia compared those prescribed pro-
endometrial hyperplasia was high (42.6%). Clinicians gestin with those not prescribed progestin.25 The risk
and patients should acknowledge the high rate of of endometrial carcinoma was decreased threefold to
concurrent carcinoma of more than 40% when plan- fivefold in women with complex or atypical endome-
ning therapy for atypical endometrial hyperplasia. trial hyperplasia diagnosed and who were adminis-

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 387
tered progestin. The hysterectomy risk was also de- is not widely used. If the endometrial intraepithelial
creased. Complex endometrial hyperplasia with neoplasia terminology is widely adopted by the pa-
atypia should be treated with hysterectomy if possible thology community, then this diagnosis may be more
because of the high incidence of progression to cancer useful for clinical decisions than the current classifi-
and the inability of endometrial biopsy and D&C to cation system for hyperplasia.
reliably exclude cancer. This study suggested that
among women who did not have a hysterectomy or a DIAGNOSIS OF ENDOMETRIAL CANCER
diagnosis of endometrial carcinoma treated with proges- Abnormal uterine bleeding including postmeno-
tin, the risk of endometrial carcinoma was lower than in pausal bleeding, menorrhagia, or metrorrhagia are
those women not treated with progestin. The majority of the most common presenting symptoms for women
these women had carcinoma diagnosed in the year after with endometrial hyperplasia or carcinoma. Atypical
their diagnosis of atypical hyperplasia. None of these glandular cells on cytologic screening should be eval-
women died from disease complications. This large uated with colposcopy and endocervical curettage
study provides additional validation that progestin ther- and an endometrial biopsy in women older than age
apy may be used to treat atypical endometrial hyperpla- 35 years or those with risk factors for endometrial
sia in the short-term in selected women. This study was cancer.28,29 Endometrial sampling should be recom-
not able to address the long-term response to ongoing mended in women older than 40 years with abnormal
progestin therapy. bleeding or in younger women with risk factors for
Endometrial intraepithelial neoplasia is an alter- disease. A D&C should be performed if complex
native classification designed to replace the hyperpla- hyperplasia with atypia is detected because of the
sia terminology. Endometrial intraepithelial neoplasia high incidence of a coexistent carcinoma. Hysteros-
is a precursor to endometrial adenocarcinoma that is copy is generally reserved for those women who
characterized by a distinctive histologic appearance continue to have symptoms that cannot be explained
(Fig. 5) and a monoclonal growth of mutated cells.26 by office biopsy. Hysteroscopy is better than curet-
Application of molecular standards and computer- tage at detecting polyps and submucosal leiomyomas.
aided morphometric criteria for the diagnosis of en- Based on data from observational studies, both symp-
dometrial intraepithelial neoplasia may be more re- tomatic vaginal bleeding and postmenopausal status
producible than the visual diagnosis of complex in women with endometrial polyps are associated
hyperplasia with atypia.27 Currently, this terminology with an increased risk of endometrial cancer.11
Whereas sonohysterography also can detect these
lesions, a tissue diagnosis may be warranted depend-
ing on the clinical context (Figs. 1 and 2).
The use of hysteroscopy for evaluation of abnor-
mal bleeding is common. When endometrial cancer is
diagnosed after hysteroscopy sampling of the perito-
neal fluid during standard cancer staging, surgery has
sometimes led to positive cytologic findings. The
clinical significance of malignant cytology and the
potential for tumor dissemination after hysteroscopic
after diagnosis of endometrial cancer is uncertain. A
search of the Cochrane Central Trials Registry, the
Web of Science, and PubMed for publications about
the role of hysteroscopy, laparoscopy, and saline
infusion ultrasonography on dissemination of endo-
metrial cancer cells and the prognostic significance of
positive peritoneal washings found no evidence for
altered prognosis when these procedures were used to
diagnose endometrial cancer.30 Another meta-analy-
sis of 19 studies including 1,099 women with endo-
metrial cancer who underwent preoperative hysteros-
Fig. 5. Endometrial intraepithelial neoplasia (hematoxylin copy demonstrated that hysteroscopy resulted in a
and eosin stain, 400⫻ magnification). statistically higher rate of positive peritoneal cytology
Sorosky. Endometrial Cancer. Obstet Gynecol 2012. but found no evidence to support an association

388 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

between preoperative hysteroscopy and a worse prog- endometrial polyps are associated with an increased
nosis.31 This supports the FIGO decision in 2009 to risk of endometrial cancer.11
omit malignant peritoneal cytology (Table 2) from
staging.32 Most gynecologic oncologists plan treat- PREOPERATIVE EVALUATION OF
ment according to other risk factors, including myo- ENDOMETRIAL CANCER
metrial invasion, grade, histology, serosal involve- In addition to age-associated preoperative blood and
ment, extrauterine disease, and nodal status. electrocardiogram studies, women with a diagnosis of
Several studies have evaluated the effectiveness endometrial cancer and suspected metastatic disease
of transvaginal ultrasonography in diagnosing or may have a preoperative CA 125 test. Sood et al36
screening for endometrial cancer in menopausal reported that elevated CA 125 levels have been
women. In a study of 448 women, an endometrial reported to be the single most significant predictor of
thickness less than 5 mm had a negative predictive extrauterine disease. It is common for women with
value of 99%.33 In 100 menopausal women, an endo- endometrial cancer to have comorbidities such as obe-
metrial stripe of less than 4 mm had a 100% negative sity, hypertension, coronary artery disease, or pulmo-
predictive value.34 Another study found that 96% of nary disease. Preoperative risk assessment is important
women with carcinoma had an endometrial thickness to optimize therapy of concurrent medical comorbidi-
more than 5 mm.35 The use of transvaginal ultra- ties. These women are at increased risk for venous
sonography does not replace a tissue diagnosis but thromboembolic disease and prophylaxis should be
may be a useful adjunct in medically compromised used. Routine computed tomography scan and mag-
women. Persistent vaginal bleeding must be evaluated netic resonance imaging (MRI) rarely alter management
regardless of ultrasonographic findings. Based on data and are poor predictors of nodal disease.37 However,
from observational studies, both symptomatic vaginal MRI is better at determining myometrial invasion than
bleeding and postmenopausal status in women with computed tomography.38 Neither computed tomogra-
phy nor MRI is recommended for routine preoperative
evaluation. Screening for age-appropriate and risk-ap-
Table 2. Revised 2009 International Federation of propriate cancers should be performed. If age-appropri-
Gynecology and Obstetrics Staging of ate, then mammography and colonoscopy should be
Endometrial Cancer performed before surgery.
Stage Description However, MRI for evaluating the depth of myo-
metrial invasion preoperatively does not have the
I Tumor confined to uterine corpus
sensitivity, positive predictive value, or accuracy to
IA No or less than half myometrial invasion
IB Invasion to or in more than half of the myometrium make clinical decisions.39 Although MRI has been
II Tumor invades the cervical stroma but does not demonstrated to have a negative predictive value, the
extend beyond the uterus probability of the absence of myometrial invasion is
III Local or regional spread of the tumor approximately 40%. Also, MRI should not be used to
IIIA Tumor invades the serosa or adnexae or serosa and
determine which women should and should not have
adnexae
IIIB Vaginal or parametrial involvement lymphadenectomy. In endometrial cancer, up to 50%
IIIC Metastases to the pelvic or paraarotic or pelvic and of lymph node metastases are smaller than 1 cm in
paraaortic lymph nodes diameter. Fludeoxyglucose (18F)-positron emission to-
IIIC1 Positive pelvic nodes mography was found to be only moderately sensitive in
IIIC2 Positive paraaortic lymph nodes with or without
predicting lymph node metastasis in women with endo-
positive pelvic nodes
IV Tumor invades bladder or bowel mucosa or distant metrial cancer.40 Preoperative evaluation with this test
metastases cannot replace pelvic lymphadenectomy.
IVA Tumor invasion of bladder or bowel or bladder and According to the 2005 American College of Obste-
bowel mucosa tricians and Gynecologists Practice Bulletin, Manage-
IVB Distant metastasis, including intraabdominal
ment of Endometrial Cancer, when practical and feasi-
metastases, inguinal lymph nodes, or
intraabdominal metastases and inguinal lymph ble, preoperative consultation with a physician with
nodes advanced training and demonstrated competence such
The stage also includes the grade of the tumor. as a gynecologic oncologist may be recommended.
Grade 1, 5% or less of the tumor is solid; grade 2, 6 –50% of the
tumor is solid; grade 3, more than 50% of the tumor is solid. SURGERY FOR ENDOMETRIAL CANCER
Modified from Pecorelli S. Revised FIGO staging for carcinoma
of the vulva, cervix, and endometrium. Int J Gynecol Obstet Total hysterectomy with bilateral salpingo-oophorec-
2009;105:103– 4, with permission from Elsevier. tomy is the primary treatment for women with endo-

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 389
metrial carcinoma. Approximately 75% of women complications, shorter hospital stay, and improved
will have stage I disease curable by surgery alone. short-term quality of life. Long-term outcomes of
Those with more advanced disease may require che- women undergoing minimally invasive surgery for
motherapy or radiation therapy. Complete surgical endometrial cancer are not available. The cost benefit
staging includes total hysterectomy with bilateral sal- among the minimally invasive procedures is contro-
pingo-oophorectomy and pelvic and paraaortic versial.53 Most gynecologic oncologists recommend
lymphadenectomy. Pelvic washings are no longer minimally invasive surgery for endometrial cancer
part of FIGO staging but may be reported separately. when appropriate and feasible.
Surgical staging allows for the most accurate assess- Before the 2005 American College of Obstetri-
ment of extent of disease. The FIGO staging has been cians and Gynecologists Practice Bulletin, Manage-
surgical since 1988. Surgical staging provides prog- ment of Endometrial Cancer, which was reaffirmed in
nostic and uncertain therapeutic benefits to women 2011, the treatment of endometrial cancer had been
with endometrial cancer. Some authors have reported controversial with respect to whether an individual
pelvic and paraaortic lymphadenectomy to be associ- with subspecialty training in gynecologic oncology
ated with improved survival.41– 44 Two large random- performed the surgery, the indications for lymphade-
ized European trials have reported that lymphadenec- nectomy, and the route of surgery (laparoscopic com-
tomy had no effect on survival for women with early pared with open). The recommendations in 2005
stage endometrial cancer.45,46 Whereas critics of these clarified many of these issues. Pelvic and paraarotic
two studies have suggested methodologic concerns, lymphadenectomy were recommended to accurately
many women with endometrial cancer do not undergo stage endometrial cancer. Preoperative imaging can-
lymph node staging.47 There is absolutely no consensus not accurately assess lymph node involvement. Ap-
regarding which patients require lymph node staging or proximately 50% of lymph node metastases in endo-
what constitutes an adequate lymphadenectomy in metrial cancer are less than 1 cm in diameter.
terms of number of nodes removed and extent of Visualization and palpation of retroperitoneal lymph
lymphadenectomy. Surgical therapy may be via lapa- nodes cannot accurately predict the presence of me-
rotomy, laparoscopic-assisted vaginal hysterectomy, to- tastasis. Gross examination of depth of myometrial
tal laparoscopic hysterectomy, or robotic total hysterec- invasion does not have the sensitivity, specificity,
tomy with pelvic and paraaortic lymphadenectomy.48 positive predictive value, or negative predictive value
Vaginal hysterectomy with bilateral salpingo-oophorec- to select women who can have lymphadenectomy
tomy is occasionally recommended in women with safely omitted from the surgical procedure. In the
medical comorbidities that preclude abdominal or lapa- absence of ideal noninvasive preoperative testing,
roscopic procedures. Primary radiation therapy is an surgical staging remains the most accurate method of
option for the medically inoperable.49 determining the extent of disease.54
The GOG, in the LAP2 study, demonstrated that There has been controversy regarding whether to
laparoscopic surgical staging for endometrial cancer is surgically stage grade 1 endometrial adenocarcinoma
feasible and safe in terms of short-term outcomes and in women. In a study of 181 women with grade 1
results in fewer complications and shorter hospital endometrial carcinoma, 82% underwent surgical stag-
stay.50 Despite a 26% conversion to laparotomy rate, ing. In women who have undergone staging, 3.2%
there was a short-term quality-of-life benefit demon- had complications. Nineteen percent of cases were
strated at 6 weeks in the laparoscopy arm that disap- histologically upgraded. Lymph node involvement
peared by 6 months. A follow-up study demonstrated was present in 4.4% of women. High-risk uterine
the estimated 5-year overall survival was almost iden- features including more than 50% myometrial inva-
tical in both arms at 89.8%.51 When robotic-assisted sion, grade 3 histology (despite a preoperative grade 1
hysterectomy for endometrial cancer was compared diagnosis), high-risk histologic variants, and cervical
with traditional laparoscopic and laparotomy ap- involvement were found in 26% of women. Overall,
proaches, perioperative clinical outcomes for robotic surgical staging in women presenting with grade 1
and laparoscopic hysterectomy appeared similar, with disease significantly affected postoperative treatment
the exception of less blood loss for robotic cases and decisions in 29% of women. Omitting lymphadenec-
longer operative times for robotic and laparoscopy tomy in women presenting with grade 1 disease may
cases.52 These studies have demonstrated that mini- lead to inappropriate postoperative treatment.55
mally invasive surgery is equivalent to traditional Pelvic and paraaortic lymph node sampling are
laparotomy in terms of adequacy of surgical resection useful in surgical staging to provide accurate prognos-
and lymph node counts with decreased postoperative tic evaluation. A therapeutic role for lymphadenec-

390 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

tomy is not universally accepted. A randomized trial undergoing abdominal hysterectomy for endometrial
to evaluate therapeutic lymphadenectomy in endo- cancer between 2003 and 2007, perioperative surgical
metrial cancer was performed by the Medical Re- complications, medical complications, and intensive
search Council in England.56 The study found no care unit requirements were lower for women treated
therapeutic benefit for the procedure. This study has by high-volume (more than 30 procedures per year)
been criticized because paraaortic lymph node dissec- surgeons.62 Two studies demonstrated that primary
tion was not mandated and postoperative adjuvant management by a gynecologic oncologist resulted in
radiation therapy was not directed by the nodal status. an efficient use of health care resources and mini-
Research efforts are ongoing to identify a subset of mized potential morbidity.63,64 Women with complex
women who may not require surgical staging.57 hyperplasia with atypia have a 40% incidence of a
No specific criteria exist for management of coexistent endometrial adenocarcinoma. The role of
women who have undergone total hysterectomy with the gynecologic oncologist in caring for these women
bilateral salpingo-oophorectomy without lymphade- is unclear. Preoperative consultation with a gyneco-
nectomy for initial treatment of endometrial cancer. A logic oncologist is recommended for women with a
recent study of Surveillance, Epidemiology, and End known endometrial cancer, for women with an endo-
Results data revealed that approximately 62% of metrial cancer diagnosed unexpectedly at hysterec-
women (24,436 of 39,396) underwent hysterectomy tomy, and for women with advanced stage disease.
for endometrial cancer without lymphadenectomy.58
Using data retrospectively reviewed from GOG in the ADJUVANT RADIATION THERAPY
LAP2 study, criteria were used to aid in treatment No adjuvant therapy is recommended for women
planning for reoperation in women who did not with surgical staging indicating low risk for recur-
undergo complete surgical staging for endometroid rence. Radiation therapy traditionally had been rec-
adenocarcinoma.59 Women at low risk with tumor ommended as adjuvant therapy for women at risk for
size of 2 cm or less, grade 1 or 2 disease, and recurrence. Several randomized trails have failed to
myometrial invasion of 50% or less were found to demonstrate a survival advantage for adjuvant whole
have a rate of nodal metastasis of only 0.8%. pelvic radiation therapy for stage I disease.65,66 All
Serous and clear-cell endometrial carcinoma can women in the GOG study underwent surgical staging
arise in a background of an atrophic endometrium. and then were randomized to radiation therapy or
The precursor lesion is considered to be endometrial observation.43 The Post Operative Radiation Therapy
intraepithelial carcinoma rather than hyperplasia. Se- in Endometrial Carcinoma study determined the
rous carcinoma can be multifocal and distant disease value of radiation therapy in women after surgical
may be detected in the absence of myometrial inva- therapy without comprehensive surgical staging.44
sion. Comprehensive surgical staging similar to ovar- Although local control was better in the treatment
ian carcinoma should be performed. Omentectomy arm, there was no difference in survival. In the GOG
along with peritoneal and upper abdominal biopsies study, a subgroup of patients at high intermediate risk
are recommended. was defined as those with: 1) moderate to poorly
Women with advanced stage disease may benefit differentiated tumor, presence of lymphovascular in-
from removal of the bulk disease.60 Maximal surgical vasion, and outer one third myometrial invasion; 2)
effort should be directed at these women. A meta- age 50 years or older with any two risk factors listed
analysis of 14 retrospective cohorts including 672 here; or 3) age of at least 70 years with any risk factor
women with advanced or recurrent endometrial cancer listed here. All other eligible participants were con-
undergoing complete surgical cytoreduction or adjuvant sidered to be in a subgroup at low intermediate risk.
radiation or receiving adjuvant chemotherapy demon- The GOG study concluded that adjunctive radiation
strated that complete cytoreduction to no gross disease therapy in early-stage intermediate-risk endometrial
was associated with superior overall survival.61 carcinoma decreases the risk of recurrence but should
The route of primary surgery, the value and be limited to patients whose risk factors fit a high
extent of lymphadenectomy, and the value of postop- intermediate risk definition. Between July 1996 and
erative adjuvant radiation therapy are now very con- March 2005, 905 women from seven countries includ-
troversial. Most cases of endometrial cancer are cured ing the United States with intermediate-risk or high-
after total hysterectomy with bilateral salpingo-oo- risk early stage disease were randomized to observa-
phorectomy. Also controversial is the role of the tion or to external beam radiation therapy.56 There
general obstetrician gynecologist in the management was no evidence that overall survival with external
of endometrial cancer. In a study of 6,015 women beam radiation therapy was better than with observa-

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 391
tion. The study concluded that adjuvant external FAMILIAL ENDOMETRIAL CANCER
beam radiation therapy could not be recommended There are numerous reports describing family history
as routine treatment for women with intermediate-risk as a risk factor for endometrial cancer. Women with a
or high-risk early stage endometrial cancer with the family history of colon cancer are at increased risk.
aim of improving survival. The absolute benefit of Hereditary nonpolyposis colorectal cancer, Lynch
external beam radiation therapy in preventing iso- syndrome, is an autosomal-dominant condition, with
lated local recurrence was small and was not without endometrial cancer being the most common cancer
toxicity. seen in association with hereditary nonpolyposis colo-
A survey of members of the Society of Gyneco- rectal cancer.70 Other malignancies associated with
logic Oncologists in 2005 queried the practice of hereditary nonpolyposis colorectal cancer can occur
surgical staging and adjuvant radiation therapy.67 in the brain and gastric, biliary, intestinal, ovarian,
Society of Gynecologic Oncologists members were and urinary tracts. Hereditary nonpolyposis colorec-
more likely to perform complete surgical staging tal cancer is associated with germline mutations in
during all surgeries for endometrial cancer in 2005
any one of six genes, PMS2 on chromosome 7, MLH1
than in 1999 (71% compared with 48%; P⬍.001). A
on chromosome 3, and MSH6, MSH2, or PMS1 on
higher percentage of respondents described surgery
chromosome 2. Mismatch repair failure leads to
as a complete lymphadenectomy (76% compared
microsatellite instability. Hereditary nonpolyposis
with 44%; P⬍.001) and believed this was therapeutic
colorectal cancer increases the population risk of
(71% compared with 66%; P⫽.04). Approximately
endometrial cancer from 0.2% to 20% by age 50 and
half of Society of Gynecologic Oncologists members
from 1.5% to 60% by age 70. The lifetime risk for
used laparoscopic-assisted staging in the primary treat-
hereditary nonpolyposis colorectal cancer gene carriers
ment of endometrial cancer. Since 1999, there has a
significant decrease in the recommendation for postop- can be as high as 30%. Endometrial cancers in Lynch
erative radiation therapy. In almost all cases in which syndrome can be of any grade or histology. Up to 35%
radiation therapy was recommended, the use of vaginal of endometrial cancers in this syndrome may be of high
brachytherapy was more common than pelvic radiation stage or adverse histology. Many women with endome-
therapy. In all situations, consultation recommendations trial cancer diagnosed will have a mismatch repair
for additional intervention were more likely if complete deficiency discovered by immunohistochemistry and
surgical staging had not been performed, suggesting that microsatellite instability testing. Of 61 consecutive pa-
all patients with endometrial cancer would benefit from thology specimens in women with endometrial cancer
surgery by a gynecologic oncologist. younger than age 50 years, 34% of the tumors had
Twelve years of experience, from 1993 to 2004, at absence of at least one of the four mismatch repair
Memorial Sloan-Kettering Cancer Center showed an proteins.71 Women with endometrial cancer diagnosed
increase in surgical staging and a decrease in the use of before age 50 years should be considered for immuno-
postoperative radiation therapy.68 When surgical ther- histochemical testing. There are some authors who
apy of endometrial cancer has included lymphadenec- advocate this testing for all women with endometrial
tomy, there has been a decrease in the use of adjuvant cancer. Obese women were less likely than nonobese to
radiation therapy without any decrease in survival. have a mismatch repair deficiency.
Because these women can have development of
Postoperative Therapy colon cancer before age 50 years and because disease
Vaginal brachytherapy can reduce the incidence of may be commonly found in the proximal colon, colono-
vaginal vault recurrence. Using high-dose therapy, scopy should be started every other year beginning at
treatment can be on an outpatient basis with low age 20 years and should be performed annually after age
morbidity. 35 years. Endometrial screening including transvaginal
There has been a single institution study report- ultrasonography, CA 125 examinations, and pelvic ex-
ing a postoperative survival advantage of brachyther- amination should commence at age 30. Endometrial
apy.69 Recommendations for brachytherapy are cur- biopsy should be performed if symptoms of irregular
rently undergoing investigation. The GOG is bleeding or menorrhagia develop. Prophylactic hyster-
currently studying treatment between external beam ectomy with bilateral salpingo-oophorectomy has been
pelvic radiation therapy and vaginal brachytherapy proposed as an effective strategy for preventing endo-
with three cycles of paclitaxel and carboplatin for metrial and ovarian cancer in women with Lynch
women with stage I disease and intermediate-risk or syndrome as a risk-reducing procedure after childbear-
high-risk features (GOG study 249). ing is completed.72

392 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

 Most studies do not suggest that the BRCA1 or endometrial cancer. Endometrial sampling should be
BRCA2 gene is associated with endometroid endome- performed only in symptomatic women.
trial cancer. Several reports suggest that the papillary
serous histologic subtype of endometrial cancer may Follow-Up
be linked to BRCA mutations. There are several Most endometrial cancers recur within the first 2 years
studies suggesting removing the uterus for prophy- after treatment. Women should undergo examination
laxis in BRCA1-positove and BRCA2-positive women every 3– 4 months for the first 2 years, every 6 months
undergoing prophylactic salpingo-oophorectomy for for the next 3 years, and yearly after 5 years. It is unclear
ovarian cancer prevention to prevent serous papillary if annual chest radiographs contribute to early detection
uterine cancer.73–75 Although controversy exists, this of recurrence. It is uncertain whether vaginal cytology
information should be discussed with women contem- contributes to early detection of recurrence because
plating prophylactic risk-reducing surgery for BRCA most lesions detected by cytology have been visually
mutations. Hysterectomy at the time of prophylactic apparent. Most recurrences are detected when diagnos-
bilateral salpingo-oophorectomy in BRCA carriers is tic testing is dictated by symptoms.79,80
not universally recommended.
TREATMENT OF ADVANCED AND
RECURRENT DISEASE
Screening Treatment of recurrent disease depends on the origi-
Screening has not been found to be effective in nal stage, the location of the recurrence, and previous
diagnosing endometrial cancer in asymptomatic treatments. The most common site for recurrence for
women, and no screening criteria have been estab- stage I disease is the vagina.65,81 Isolated vaginal
lished except those for women with Lynch II syn- recurrences in women who did not undergo radiation
drome. Guidelines from the Bethesda system recom- therapy are generally treated with whole pelvic radi-
mend that normal endometrial cells be reported in ation therapy and vaginal brachytherapy. The Post
women age 40 and older. A study of 29,114 asymp- Operative Radiation Therapy in Endometrial Carci-
tomatic postmenopausal women with normal endo- noma trial reported survival after relapse was signifi-
metrial cells on routine screening cytology reported a cantly better in the patient group without previous
prevalence rate of premalignant disease of 6.5% com- radiation therapy.81 Treatment for vaginal relapse was
pared with a rate of 0.2% in those without disease.76 In effective, with 89% complete response and 65%
a study of asymptomatic women with normal endo- 5-year survival in the control group. The excellent
metrial cells on liquid-based cytology, 2.1% had en- survival after treatment for vaginal relapse is among
dometrial pathology.77 Because approximately 50% of the reasons that adjuvant radiation therapy has not
women with endometrial cancer have malignant cells yielded a survival advantage. There was no difference
on screening cytology, cervical cytology is not a useful in survival between patients with pelvic relapse and
screening test. Several studies have demonstrated that those with distant metastases. If the recurrence is
women with malignant cells in routine cytology are limited to the vagina, then survival is better than it
more likely to have more advanced disease than those would be if the disease involved the pelvic sidewall. A
without malignant cytology. Malignant cytology is not multi-institutional retrospective review from the
an established independent prognostic variable. United States found similar results.82 Eighty-one per-
Although evidence-based data are lacking, screen- cent of isolated vaginal recurrences were salvaged
ing may be considered in women at increased risk, such with radiation therapy. The mean time to recurrence
as postmenopausal women using estrogen therapy with- was 24 months, and the mean follow-up was 63
out progestin and premenopausal women with anovu- months. Among women, 18% died from subsequent
latory cycles. In women with a genetic predisposition or recurrent disease. The 5-year overall survival was
using tamoxifen, increased awareness of the signs and 75%. The majority of isolated vaginal recurrences in
symptoms of endometrial cancer may allow for early women with surgical stage I endometrial cancer can
detection. In woman using tamoxifen for breast can- be successfully salvaged with radiation therapy, fur-
cer prevention, transvaginal ultrasonography is asso- ther questioning the role of adjuvant therapy for
ciated with a high false-positive rate because tamox- patients with uterine-confined endometrial cancer at
ifen also induces subendometrial cyst formation. the time of initial diagnosis.
Endometrial biopsy samples from asymptomatic For systemic disease, hormonal therapy with pro-
woman had low yields.78 Woman using tamoxifen gestins has been a favored therapeutic option because
should be counseled about the signs and symptoms of of its minimal toxicity. The response rates have been

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 393
reported in 10 –25% of women. Recent studies have apeutic recommendations. Because a new FIGO stag-
tried to improve this response rate with combined ing system was implemented in 2009, prospective
therapy with tamoxifen to increase the progesterone data for survival are not available.32 Cases of stage IA
receptor expression and medroxyprogesterone ace- and IB were combined into a single stage, stage IA.
tate.83 The response rate was 33%. The median pro- Cervical glandular involvement was omitted from the
gression-free survival was 3 months and median criteria for staging and only those women with cervi-
overall survival was 13 months. The combination of cal stromal invasion are now classified as having stage
daily tamoxifen and intermittent weekly medroxypro- II. Peritoneal cytology has been removed as criteria
gesterone acetate is an active treatment for advanced for staging. Women with nodal metastasis are now
or recurrent endometrial carcinoma. Outcomes are stratified into the same category as those with pelvic
better with chemotherapy, and chemotherapy is now nodal disease IIIC1 and those with paraaortic nodal
used as the primary therapy for advanced and meta- disease IIIC2. A population-based analysis was per-
static disease. formed comparing the performance of the 1988 and
Chemotherapy can be initial therapy for ad- 2009 FIGO staging systems in women with endo-
vanced disease or used for recurrence after hormonal metroid adenocarcinoma treated between 1988 and
therapy. The combination of cisplatin, doxorubicin, 2006 and was recorded in the Surveillance, Epidemi-
and paclitaxel is the most active chemotherapeutic ology, and End Results database;85 81,902 women
regimen reported in advanced or recurrent endome- were classified based on the 1988 FIGO staging
trial cancer. The GOG performed a randomized system compared with the 2009 system. Five-year
prospective study of cisplatin, doxorubicin, and pac- survival was calculated based on disease grade and
litaxel compared with cisplatin and doxorubicin.84 whether lymphadenectomy was performed. The au-
Objective response (57% compared with 34%; thors found survival for stage IA based on the 1988
P⬍.01), progression-free survival (median, 8.3 com- staging system was 90.7% compared with 88.9% for
pared with 5.3 months; P⬍.01), and overall survival the 2009 system. The 2009 FIGO staging system is
(median, 15.3 compared with12.3 months; P⫽.037) highly prognostic based on this study. Prospective
were improved with cisplatin, doxorubicin, and pac- data are currently being collected.
litaxel. Treatment was hematologically well-tolerated,
with only 2% of patients receiving cisplatin and Fertility-Preserving Procedures in the
doxorubicin and 3% of patients receiving cisplatin, Management of Endometrial Cancer
doxorubicin, and paclitaxel experiencing neutropenic The conservative management of women with atypi-
fever. Neurologic toxicity was worse for those receiv- cal endometrial hyperplasia or endometrial cancer
ing cisplatin, doxorubicin, and paclitaxel, with 12% who desire future fertility has been controversial.
with grade 3 and 27% with grade 2 peripheral neu- Younger age has been associated with a more favor-
ropathy compared with 1% and 4%, respectively, in able prognosis. Multiple reports describe therapeutic
those receiving cisplatin and doxorubicin. Patient- success with medroxyprogesterone acetate or meges-
reported neurotoxicity was significantly higher in the terol acetate.24,86 A multicenter prospective study of
cisplatin, doxorubicin, and paclitaxel arm after two 28 women with endometrial carcinoma and 17
cycles of therapy. Cisplatin, doxorubicin, and pacli- women with atypical endometrial hyperplasia were
taxel significantly improved RR, progression-free sur- treated with 600 mg of medroxyprogesterone acetate
vival, and overall survival compared with cisplatin daily with low-dose aspirin for 26 weeks.87 Response
and doxorubicin. Although cisplatin, doxorubicin, was assessed histologically at 8 and 16 weeks of
and paclitaxel is the most active regimen, the toxicity treatment. Complete response was found in 55% of
has led many physicians to treat women with carbo- endometrial carcinomas and 82% of atypical endome-
platin and paclitaxel. At the 2012 Society of Gynecol- trial hyperplasia. During the 3-year follow-up there
ogy Oncology Annual Meeting on Women’s Cancer, were 12 pregnancies. Seven normal deliveries were
an oral abstract reviewed these data and established achieved. There was a 47% recurrence rate between 7
noninferiority of cisplatin and doxorubicin compared and 36 months. One woman died of a synchronous
with cisplatin, doxorubicin, and paclitaxel. ovarian carcinoma. Before conservative management,
MRI to exclude myometrial invasion and pathology
Prognosis review to verify grade I disease should be performed.
Stage is the most important prognostic factor, and Although there are no prospective randomized stud-
surgical staging offers the most accurate prognostic ies and no standard accepted therapy, conservative
information. Surgical staging facilitates adjuvant ther- therapy is an option for a well-informed woman.

394 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

Fertility therapy must follow resolution of the endo- 10. Brinton LA, Berman ML, Mortel R, Twiggs LB, Barrett RJ,
Wilbanks GD, et al. Reproductive, menstrual, and medical risk
metrial pathology. The role of hysterectomy and
factors for endometrial cancer: results from a case-control
bilateral salpingo-oophorectomy after childbearing is study. Am J Obstet Gynecol 1992;167:1317–25.
uncertain. 11. Lee SC, Kaunitz AM, Sanchez-Ramos L, Rhatigan RM. The
oncogenic potential of endometrial polyps: A systemic review
Management of Menopausal Symptoms and meta-analysis. Obstet Gynecol 2010;116:1197–205.

Estrogen replacement therapy probably has no effect 12. Fisher B, Costantino JP, Wickerham DL, Redmond CK,
Kavanah M, Cronin M, et al. Tamoxifen for prevention of
on survival in women with early stage disease. Bara- breast cancer: report of the National Adjuvant Breast and
kat et al reported a GOG study of 1,236 assessable Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371– 88.
women entered into a randomized double-blind study 13. Grimes DA, Economy KE. Primary prevention of gynecologic
of estrogen replacement therapy compared with pla- cancers. Am J Obstet Gynecol 1995;172:227–35.
cebo in women with stage I or II disease. The median 14. Partirdge EE, Shingleton HM, Menck HR. The National
follow-up was 35.7 months. The mean age at diagno- Cancer Data Base report on endometrial cancer. J Surg Oncol
1996;61:111–23.
sis for the 618 women assigned to estrogen therapy
15. Hill HA, Coates RJ, Austin H, Correa P, Robboy SJ, Chen V,
was 57 years. The study was halted before completion et al. Racial differences in tumor grade among women with
after the results of the Women’s Health Initiative were endometrial cancer. Gynecol Oncol 1995;56:154 – 63.
made public. Although this study cannot conclusively 16. Magriples U, Naftolin F, Schwartz PE, Carcangiu ML. High-
refute or support the safety of estrogen with regard to grade endometrial carcinoma in tamoxifen-treated breast can-
risk of recurrence of endometrial cancer, the absolute cer patients. J Clin Oncol 1993;11:485–90.
recurrence rate (2.1%) and the incidence of new 17. Ferguson SE, Soslow RA, Amsterdam A, Barakat RR. Com-
parison of uterine malignancies that develop during and fol-
malignancy were low.88 Hot flushes can be treated lowing tamoxifen therapy. Gynecol Oncol 2006;101:322– 6.
with venlafaxine, clonidine, or progestin therapy. 18. Brinton LA, Hoover RN. Estrogen replacement therapy and
Osteoporosis can be prevented with biphosphospho- endometrial cancer risk: Unresolved issues. The Endometrial
nates or raloxifene. Weight reduction and lifestyle Cancer Collaborative Group. Obstet Gynecol 1993;81:
265–71.
modification should be encouraged in obese women.
Nutritional and exercise counseling should be offered. 19. Kurman RJ, Kaminski PF, Norris HJ. The behavior of endo-
metrial hyperplasia. A long-term study of “untreated” hyper-
plasia in 170 patients. Cancer 1985;56:403–12.
REFERENCES 20. Zaino RJ, Kauderer J, Trimble CL, Silverberg SG, Curtin JP,
1. Siegel R, Naishadham J, Jemal A. Cancer statistics 2012. CA Lim PC, et al. Reproducibility of the diagnosis of atypical
Cancer J Clin 2012;62:10 –29. endometrial hyperplasia: a Gynecologic Oncology Group
2. Sorosky JI. Endometrial cancer. Obstet Gynecol 2008;111: study. Cancer 2006;106:804 –11.
436 – 47. 21. Trimble CL, Kauderer J, Zaino R, Silverberg S, Lim PC, Burke
3. Prat J, Gallardo A, Cuatrecasas M, Catasus L. Endometrial JJ II, et al. Concurrent endometrial carcinoma in women with
carcinoma: pathology and genetics. Pathology 2007;39:1–7. a biopsy diagnosis of atypical endometrial hyperplasia: a
4. Courneya KS, Karvinen KH, Campbell KL, Pearcey RG, Gynecologic Oncology Group study. Cancer 2006;106:812–9.
Dundas G, Capstick V, et al. Associations among exercise, 22. Suh-Burgmann ES, Hung YY, Armstrong MA. Complex atyp-
body weight, and quality of life in a population-based sample ical endometrial hyperplasia: The risk of unrecognized adeno-
of endometrial cancer survivors. Gynecol Oncol 2005;97: carcinoma and value of preoperative dilatation and curettage.
422–30. Obstet Gynecol 2009;114:523–9.
5. Anderson B, Connor JP, Andrews JI, Davis CS, Buller RE, 23. Wildemeersch D, Dhont M. Treatment of nonatypical and
Sorosky JI, et al. Obesity and prognosis in endometrial cancer. atypical endometrial hyperplasia with a levonorgestrel-releas-
Am J Obstet Gynecol 1996;174:1171–78. ing intrauterine system. Am J Obstet Gynecol 2003;188:
6. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. 1297– 8.
Overweight, obesity, and mortality from cancer in a prospec- 24. Lowe MP, Cooper BC, Sood AK, Davis WA, Syrop CH,
tively studied cohort of U.S. adults. N Engl J Med 2003;348: Sorosky JI. Implementation of assisted reproductive technolo-
1625–38. gies following conservative management of FIGO grade I
7. von Gruenigen VE, Tian C, Frasure H, Waggoner S, Keys H, endometrial adenocarcinoma and/or complex hyperplasia
Barakat RR. Treatment effects, disease recurrence, and sur- with atypia. Gynecol Oncol 2003;91:569 –72.
vival in obese women with early endometrial carcinoma: a 25. Reed SD, Newton KM, Garcia RL, Allison KH, Voigt LF,
Gynecologic Oncology Group study. Cancer 2006;107: Jordan CD, et al. Complex hyperplasia with and without
2786 –91. atypia: Clinical outcomes and implications of progestin ther-
8. von Gruenigen VE, Waggoner SE, Frasure HE, Kavanagh MB, apy. Obstet Gynecol 2010;116:365–73.
Janata JW, Rose PG, et al. Lifestyle challenges in endometrial 26. Baak JP, Mutter GL, Robboy S, van Diest PJ, Uyterlinde AM,
cancer survivorship. Obstet Gynecol 2011;117:93–100. Orbo A, et al. The molecular genetics and morphometry-based
9. Temkin SM, Pezzullo JC, Hellmann M, Lee YC, Abulafia O. Is endometrial intraepithelial neoplasia classification system pre-
body mass index an independent risk factor of survival among dicts disease progression in endometrial hyperplasia more
patients with endometrial cancer? Am J Clin Oncol 2007;30: accurately than the 1994 World Health Organization classifi-
8 –14. cation system. Cancer 2005;103:2304 –12.

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 395
27. Mutter GL, Zaino RJ, Baak JP, Bentley RC, Robboy SJ. Benign 44. Chan JK, Cheung MK, Huh WK, Osann K, Husain A, Teng
endometrial hyperplasia sequence and endometrial intraepi- NN, et al. Therapeutic role of lymph node resection in
thelial neoplasia. Int J Gynecol Pathol 2007;26:103–14. endometrioid corpus cancer: a study of 12,333 patients. Cancer
2006;107:1823–30.
28. Sharpless KE, Schnatz PF, Mandavilli S, Greene JF, Sorosky JI.
Dysplasia associated with atypical glandular cells on cervical 45. Benedetti Panici P, Basile S, Maneschi F, Alberto Lissoni A,
cytology. Obstet Gynecol 2005;105:494 –500. Signorelli M, Scambia G, et al. Systemic pelvic lymphadenec-
tomy vs. no lymphadenectomy in early-stage endometrial
29. Schnatz PF, Guile M, O’Sullivan DM, Sorosky JI. Clinical cancer: randomized clinical trial. J Natl Cancer Inst 2008;100:
significance of atypical glandular cells on cervical cytology. 1707–16.
Obstet Gynecol 2006;107:701– 8.
46. ASTEC study group, Kitchener H, Swart AM Qian Q, Amos
30. Guralp O, Kushner DM. Iatrogenic transtubal spill of endo- C, Parmar MK. Efficacy of systemic pelvic lymphadenectomy
metrial cancer: risk of myth. Ach Gynecol Obstet Gynecol in endometreial cancer (MRS ASTEC trial): a randomised
2011;284:1209 –21. study. Lancet 2009;373:125–36.
31. Chang YN, Zhang Y, Wang YJ, Wang LP, Duan H. The effect 47. Seamon LG, Fowler JM, Cohn DE. Lymphadenectomy for
of hysteroscopy on the peritoneal dissemination of endometrial endometrial cancer: the controversy. Gynecol 2012;11:6 – 8.
cancer: a meta-analysis. Fertil Steril 2011;96:957– 61.
48. Homesley HD, Boike G, Spiegel GW. Feasibility of laparo-
32. Creasman W. Revised FIGO staging carcinoma of the endo- scopic management of presumed stage I endometrial carci-
metrium. Int J Gynaecol Obstet 2009;105:109. noma and assessment of accuracy of myoinvasion estimates by
33. Langer RD, Pierce JJ, O’Hanlan KA, Johnson SR, Espeland frozen section: a gynecologic oncology group study. Int J
MA, Trabal JF, et al. Transvaginal ultrasonography compared Gynecol Cancer 2004;14:341–7.
with endometrial biopsy for the detection of endometrial 49. Podzielinski I, Randall ME, Breheny PJ, Escobar PJ, Cohn DE,
disease. Postmenopausal Estrogen/Progestin Interventions Quick AM, et al. Primary radiation therapy for medically
Trial. N Engl J Med 1997;337:1792– 8. inoperable patients with clinical stage I and II endometrial
carcinoma. Gynecol Oncol 2012;124:36 – 41.
34. Gull B, Carlsson SA, Karlsson B, Ylostalo P, Milsom I,
Granberg S. Transvaginal ultrasonography of the endome- 50. Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Sch-
trium in women with postmenopausal bleeding: Is it always laerth JB, Mannel RS, et al. Laparoscopy compared with
necessary to perform an endometrial biopsy? Am J Obstet laparotomy for comprehensive surgical staging for uterine
Gynecol 2000;182:509 –15. cancer: Gynecologic Oncology Group Study LAP2. J Clin
Oncol 2009;27:5331– 6.
35. Smith-Bindman R, Kerlikowske K, Feldstein VA, Subak L,
Scheidler J, Segal M, et al. Endovaginal Ultrasound to exclude 51. Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Sch-
endometrial cancer and other endometrial abnormalities. laerth JB, Mannel RS, et al. Recurrence and survival after
JAMA 1998;280:1510 –7. random assignment to laparoscopy versus laparotomy for
comprehensive surgical staging of uterine cancer: Gynecologic
36. Sood AK, Buller RE, Burger RA, Dawson JD, Sorosky JI, Oncology Group LAP2 study. J Clin Oncol 2012;30:695–700.
Berman M. Value of preoperative CA 125 level in the man-
52. Gaia G, Holloway RW, Santoro L, Ahmad S, Di Silverio E,
agement of uterine cancer and prediction of clinical outcome.
Spinillo A. Robotic-assisted hysterectomy for endometrial can-
Obstet Gynecol 1997;90;441–7.
cer compared with traditional laparoscopic and laparotomy
37. Connor JP, Andrews JI, Anderson B, Buller RE. Computed approaches: A systemic review. Obstet Gynecol 2010;116:
tomography in endometrial carcinoma. Obstet Gynecol 2000; 1422–31.
95:692– 6. 53. Barnett JC, Judd JP, Wu JM, Scales CD Jr, Myers ER,
38. Chung HH, Kang SB, Cho JY, Kim JW, Park NH, Song YS, et Havrilesky LJ. Cost comparison among robotic, laparoscopic,
al. Accuracy of MR imaging for the prediction of myometrial and open hysterectomy for endometrial cancer. Obstet Gyne-
invasion of endometrial carcinoma. J Gynecol Oncol 2007; col 2010;116:685–93.
104:654 –9. 54. Aalders JG, Thomas G. Endometrial cancer–revisiting the
39. Nakao Y, Yokoyama M, Hara K, Koyamatsu Y, Yasunaga M, importance of pelvic and para aortic lymph nodes. Gynec
Araki Y, et al. MR imaging in endometrial carcinoma as a Oncol 2007;104:222–31.
diagnostic tool for the absence of myometrial invasion. Gyne- 55. Ben-Shacher I, Pavelka J, Cohn DE, Copeland LJ, Ramirez N,
col Oncol 2006;102:343–7. Manolitsas T, et al. Surgical staging for patients presenting with
40. Horowitz NS, Dehdashti F, Herzog TJ, Rader JS, Powell MA, grade 1 endometrial carcinoma. Obstet Gynecol 2005;105:
Gibb RK, et al. Prospective evaluation of FDG-PET for 487–93.
detecting pelvic and para-aortic lymph node metastasis in 56. ASTEC/EN.5 Study Group, Blake P, Swart AM, Orton J,
uterine corpus cancer. Gynecol Oncol 2004;95:546 –51. Kitchener H, Whelan T, Lukka H, et al. Adjuvant external
41. Kilgore LC, Partridge EE, Alvarez RR, Austin JM, Shingleton, beam radiotherapy in the treatment of endometrial cancer
HM, Noojin F III, et al. Adenocarcinoma of the endometrium: (MRC ASTEC and NCIC CTG EN.5 randomised trials):
survival comparisons of patients with and without pelvic pooled trial results, systematic review, and meta-analysis Lan-
lymph node sampling. Gynecol Oncol 1995;56:29 –33. cet 2009;373:137– 46.
42. Cragun JM, Havrilesky LJ, Calingaert B, Synan I, Secord AA, 57. Mariani A, Sebo TJ, Katzmann JA, Roche PC, Keeney GL,
Soper JT, et al. Retrospective analysis of selective lymphade- Lesnick TG, et al. Endometrial cancer: can nodal status be
nectomy in apparent early-stage endometrial cancer. J Clin predicted with curettage? Gynecol Oncol 2005;96:594 – 600.
Oncol 2005;23:3668 –75. 58. Chan JK, Wu H, Cheung MK, Shin JY, Osann K, Kapp DS.
43. Lutman CV, Havrilesky LJ, Cragun JM, Secord AA, Calin- The outcomes of 27,063 women with unstaged endometroid
gaert B, Berchuck A, et al. Pelvic lymph node count is an uterine cancer. Gynecol Oncol 2007;106:282– 8.
important prognostic variable for FIGO stage I and II endo- 59. Milam MR, Java J, Walker JL, Metzinger DS, Parker LP,
metrial carcinoma with high-risk histology. Gynecol Oncol Coleman RL, et al. Nodal metastasis risk in endometroid
2006;102:92–7. endometrial cancer. Obstet Gynecol 2012;119:286 –92.

396 Sorosky Endometrial Cancer OBSTETRICS & GYNECOLOGY

60. Havrilesky LJ, Cragun JM, Calingaert B, Synan I, Secord AA, BRCA1-related disease? Case report and review of the litera-
Soper JT, et al. Resection of lymph node metastases influences ture. Gynecol Oncol 1999;75:300 – 4.
survival in stage IIIC endometrial cancer. Gynecol Oncol 76. Siebers AG, Verbeck AM, Massuger LF, Grefte JM, Bulten J.
2005;99:689 –95. Normal appearing endometrial cells in cervical smears of
61. Barlin JN, Puri I, Bristow RE. Cytoreductive surgery for asymptomatic postmenopausal women have predictive value
advanced or recurrenmt endometrial cancer: A meta-analysis. of endometrial pathology. Int J Gyn Gynecol Cancer 2006;16:
Gynecol Oncol 2010;118:14 – 8. 1069 –74.
62. Wright JD, Lewin SN, Deutsch I, Burke WM, Sun X, Herzog 77. Moroney JW, Zahn CM, Heaton RB, Crothers B, Kendall BS,
TJ. Effect of surgical volume on morbidity and mortality of Elkas JC. Normal endometrial cells in liquid-based cytology
abdominal hysterectomy for endometrial cancer. Obstet Gyne- specimens in women aged 40 or older. Gynecol Oncol 2007;
col 2011;117:1051–9. 105:672– 6.
63. Roland PY, Kelly FJ, Kulwicki CY, Blitzer P, Curcio M, Orr 78. Barakat RR, Gilewski TA, Almadrones L, Saigo PE, Venka-
JW Jr. The benefits of a gynecologic oncologist: a pattern of traman E, Hudis C, et al. Effect of adjuvant tamoxifen on the
care study for endometrial cancer treatment. Gynecol Oncol endometrium in women with breast cancer: a prospective
2004;93:125–30. study using office endometrial biopsy. J Clin Oncol 2000;18:
64. MacDonald OK, Sause WT, Lee RJ, Dodson MK, Zempolich 3459 – 63.
K, Gaffney DK. Does oncologic specialization influence out- 79. Reddoch JM, Burke TW, Morris M, Tornos C, Levenback C,
comes following surgery in early stage adenocarcinoma of the Gershenson DM. Surveillance for recurrent endometrial carci-
endometrium? Gynecol Oncol 2005;99:730 –5. noma: development of a follow-up scheme. Gynecol Oncol
65. Keys HM, Roberts JA, Brunetto VL, Zaino RJ, Spirtos NM, 1995;59:221–5.
Bloss JD, et al. A phase III trial of surgery with or without 80. Podczaski E, Kaminski P, Gurski K, MacNeill C, Stryker JA,
adjunctive external pelvic radiation therapy in intermediate Singapuri K, et al. Detection and patterns of treatment failure
risk endometrial adenocarcinoma: a Gynecologic Oncology in 300 consecutive cases of “early” endometrial cancer after
Group study. Gynecol Oncol 2004;92:744 –51. primary surgery. Gynecol Oncol 1992;47:323–7.
66. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, 81. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML,
Jobsen JJ, Warlam-Rodenhuis CC, et al. Surgery and postop- Jobsen JJ, Warlam-Rodenhuis CC, et al. Survival after relapse
erative radiotherapy versus surgery alone for patients with in patients with endometrial cancer: results from a randomized
stage I endometrial carcinoma: multicentre randomised trial. trial. Gynecol Oncol 2003;89:201–9.
PORTEC Study Group Post Operative Radiation Therapy in
Endometrial Carcinoma. Lancet 2000;355:1404 –11. 82. Huh WK, Straughn JM Jr, Mariani A, Podratz KC, Havrilesky
LJ, Alvarez-Secord A, et al. Salvage of isolated vaginal recur-
67. Naumann RW, Coleman RL. The use of adjuvant radiation rences in women with surgical stage I endometrial cancer: a
therapy in early endometrial cancer by members of the Society multiinstitutional experience. Int J Gynecol Cancer 2007;17:
of Gynecologic Oncologists in 2005. Gynecol Oncol 2007;105: 886 –9.
7–12.
83. Whitney CW, Brunetto VL, Zaino RJ, Lentz SS, Sorosky J,
68. Barakat RR, Lev G, Hummer AJ, Sonoda Y, Chi DS, Alektiar Armstrong DK, et al. Phase II study of medroxyprogesterone
KM, et al. Twelve-year experience in the management of acetate plus tamoxifen in advanced endometrial carcinoma: a
endometrial cancer: a change in surgical and postoperative Gynecologic Oncology Group study. Gynecol Oncol 2004;92:
radiation techniques. Gyncol Oncol 2007;105:150 – 6. 4 –9.
69. Lachance JA, Stukenborg GJ, Schneider BF, Rice LW, Jazaeri
84. Fleming GF, Brunetto VL, Cella D, Look KY, Reid GC,
JJ. A cost-effective analysis of adjuvant therapies for the
Munkarah AR, et al. Phase III trial of doxorubicin plus
treatment of Stage I endometrial adenocarcinoma. Gynecol
cisplatin with or without paclitaxel plus filgrastim in advanced
Oncol 2008;108:77– 83.
endometrial carcinoma: a Gynecologic Oncology Group
70. Hemminki K, Li X, Dong C. Second primary cancers after Study. J Clin Oncol 2004;22:2159 – 66.
sporadic and familial colorectal cancer. Cancer Epidemiol
85. Lewin SN, Herzog TJ, Barrena Medel NI, Deutsch I, Burke
Biomarkers Prev 2001;10:793– 8.
WM, Sun X, et al. Comparative performance of the 2009
71. Matthews KS, Estes JM, Connor MG, Manne U, Whitworth International Federation of Gynecology and Obstetrics staging
JM, Huh, WK, et al. Lynch syndrome in women less than 50 system for uterine corpus cancer. Obstet Gynecol 2010;116:
years of age with endometrial cancer. Obstet Gynecol 2008; 1141–9.
111:1161– 6.
86. Lowe MP, Bender D, Sood AK, Davis W, Syrop CH, Sorosky
72. Schmeler KM, Lynch HT, Chen LM, Munsell MF, Soliman JI. Two successful pregnancies after conservative treatment of
PT, Clark MB, et al. Prophylactic surgery to reduce the risk of endometrial cancer and assisted reproduction. Fertil Steril
gynecologic cancers in the Lynch syndrome. N Engl J Med 2002;77:188 –9.
2006;354:261–9.
87. Ushijima K, Yahata H, Yoshikawa H, Konishi I, Yasugi T,
73. Lavie O, Ben-Arie A, Pilip A, Rennert G, Cohen Y, Feiner B, Saito T, et al. Multicenter phase II study of fertility-sparing
et al. BRCA2 germline mutation in a woman with uterine treatment with medroxyprogesterone acetate for endometrial
serous papillary carcinoma. Gynecol Oncol 2005;99:486 – 8. carcinoma and atypical hyperplasia in young women. J Clin
74. Lavie O, Hornreich G, Ben-Arie A, Rennert G, Cohen Y, Oncol 2007;25:2798 – 803.
Keidar R, et al. BRCA germline mutations in Jewish women 88. Barakat RR, Bundy BN, Spiritos N, Bell J, Mannel RS.
with uterine serous papillary carcinoma. Gynecol Oncol 2004; Randomized double-blind trial of estrogen replacement ther-
92:521– 4. apy versus placebo in stage I or II endometrial cancer: a
75. Hornreich G, Beller U, Lavie O, Renbaum P, Cohen Y, Gynecologic Oncology Group Study. J Clin Oncol
Levy-Lahad E. Is uterine serous papillary carcinoma a 2006;24:587–92.

VOL. 120, NO. 2, PART 1, AUGUST 2012 Sorosky Endometrial Cancer 397