08 - Radiol Clin N Am 2007 - Imaging of Uterine Cancer.pdf - Bio Med 21 - Aivree

167

RADIOLOGIC

CLINICS

OF NORTH AMERICA

Radiol Clin N Am 45 (2007) 167–182

Imaging of Uterine Cancer

Oguz Akin, MD a,b, *, Svetlana Mironov, MD a,b ,

Neeta Pandit-Taskar, MD a,b ,LucyE.Hann, MD a,b

- Endometrial cancer

- Cervical cancer

- Endometrial cancer

Screening and diagnosis

- Tumor detection and staging

Ultrasonography

MR imaging

- The role of imaging in treatment planning

- Posttreatment follow-up

- Cervical cancer

Screening and diagnosis

Tumor detection and staging

- The role of imaging in treatment planning

- Posttreatment follow-up

- Summary

- Acknowledgments

- References

Endometrial cancer

The American Cancer Society estimates that in

2006, 41,200 new cases of cancer of the uterine cor-

pus, mostly endometrial, will be diagnosed and

7350 women will die from this disease in the

United States [1] .

Endometrial cancer may develop from endome-

trial hyperplasia caused by unopposed estrogen

stimulation; it also may develop spontaneously.

Risk factors for developing endometrial cancer

include conditions leading to increased estrogen

exposure, such as estrogen replacement therapy

(without progestin), obesity, tamoxifen use, early

menarche, late menopause, nulliparity, and history

of polycystic ovary disease. Pregnancy and use of

oral contraceptives reduce the risk of endometrial

cancer.

Up to 90% of endometrial cancers are adenocar-

cinomas. Depending on the glandular pattern, they

are classiﬁed as well-differentiated (grade 1) to an-

aplastic (grade 3) tumors. Prognostic factors

include tumor grade and stage, depth of myome-

trial invasion, and lymph node status.

Most endometrial cancers are detected at an early

stage because of clinical assessment for postmeno-

pausal bleeding. Treatment options include surgery,

radiation, hormones, and chemotherapy, depend-

ing on the stage of the disease.

The 1-year relative survival rate for uterine corpus

cancer is 94%. The 5-year survival rate is 96% for

local disease, but it decreases to 66% for disease

with regional spread and 25% for disease with

distant spread.

Cervical cancer

The American Cancer Society estimates that in

2006, 9716 new cases of invasive cervical cancer

will be diagnosed and 3700 women will die from

this disease in the United States [1] . As Papanico-

laou (Pap) smearing has become more common,

incidence rates of cervical cancer have decreased

and preinvasive lesions of the cervix are far more

a Weill Medical College of Cornell University, New York, NY, USA

b Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY

10021, USA

* Corresponding author. Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York

Avenue, New York, NY 10021.

E-mail address: akino@mskcc.org (O. Akin).

doi:10.1016/j.rcl.2006.10.009

radiologic.theclinics.com

168

Akin et al

commonly diagnosed than invasive cervical cancer.

Mortality rates also have declined as a result of pre-

vention and early detection.

Risk factors for developing cervical cancer

include infection with certain types of human pap-

illomavirus, early age at ﬁrst sexual intercourse,

multiple sexual partners, multiparity, history of sex-

ually transmitted diseases, and low socioeconomic

status.

Cervical intraepithelial neoplasia (CIN) is con-

sidered a precursor lesion of cervical cancer. CIN

is characterized in three groups depending on cellu-

lar dysplasia: CIN 1, minor dysplasia; CIN 2, mod-

erate dysplasia; and CIN 3, severe dysplasia or

carcinoma in situ. Up to 40% of CIN 3 lesions

could develop into invasive cervical cancer if left

untreated. Squamous cell carcinoma accounts for

80% to 90% of cases of cervical cancer. Adenocarci-

nomas are rare but have a worse prognosis.

Preinvasive lesions (ie, lesions that have not yet

transgressed the basement membrane) can be

treated with electrocoagulation, cryotherapy, laser

ablation, or local surgery. Invasive cervical cancers

are treated with surgery, radiation, or chemotherapy

or a combination of these three methods.

Relative 1-year and 5-year survival rates for cervi-

cal cancer patients are 88% and 73%, respectively.

The 5-year survival rate is approximately 92% for

localized cervical cancer [1] .

Imaging has become an important adjunct to the

clinical assessment of uterine cancer. When inte-

grated with clinical ﬁndings, imaging ﬁndings can

optimize cancer care and aid in the development

of a treatment plan tailored to the individual pa-

tient. Traditionally, the pretreatment evaluation of

uterine cancer consisted of clinical evaluation,

laboratory tests, and conventional radiographic

studies. The conventional imaging studies for clini-

cal staging are being replaced by cross-sectional im-

aging studies, namely ultrasound (US), CT, MR

imaging, and positron emission tomography

(PET). This article focuses on the role of cross-

sectional imaging in the management of endome-

trial cancer and cervical cancer.

age 35 should be offered to women with or at risk

for hereditary nonpolyposis colon cancer [1] .

Deﬁnitive diagnosis of endometrial cancer is

made with endometrial sampling with endometrial

biopsy or dilatation and curettage. The tissue ob-

tained by endometrial sampling is examined under

a microscope and evaluated for cancerous or pre-

cancerous abnormalities.

Transvaginal US may be used in the initial evalu-

ation of women with postmenopausal bleeding

[3–5] . US diagnosis is based on endometrial thick-

ness measurements in the anteroposterior dimen-

sion. The Society of Radiologists in Ultrasound

Consensus Panel recommends a cut-off value of

5mm [3] , but others have reported optimal results

using 4 mm as the upper limit for normal endome-

trial thickness [6,7] . With normal endometrial

thickness on transvaginal US, the risk of cancer is

in the range of 1% to 5.5% [7,8] . Transvaginal US

is reported to be useful for diagnosis of endometrial

abnormalities and carcinoma in women with ab-

normal bleeding even when endometrial biopsy

and hysteroscopy produce negative results [9,10] .

Abnormal uterine bleeding is an early symptom

of endometrial carcinoma, and there is no evidence

that screening asymptomatic women is of any ben-

eﬁt, even in high-risk groups [11,12] . Women who

undergo tamoxifen treatment for breast carcinoma

have a 7.5% relative risk of endometrial cancer,

but routine screening is not recommended [13] .

Women with hereditary nonpolyposis colon cancer

have a 40% to 60% lifetime risk of endometrial can-

cer, but US surveillance in the absence of symptom-

atic bleeding does not offer any prognostic

advantage [14] .

Tumor detection and staging

Staging of endometrial cancer is based on surgico-

pathologic International Federation of Gynecology

and Obstetrics (FIGO) criteria. The TNM staging

system is based on the same criteria as the FIGO sys-

tem ( Table 1 ) [15,16] . The FIGO staging system

uses ﬁndings from exploratory laparotomy, total

abdominal hysterectomy, bilateral salpingo-oopho-

rectomy, peritoneal washings, sampling, and

lymphadenectomy.

Surgical staging, however, is not suitable for

women who are not good surgical candidates

because of older age, obesity, and other medical

problems. Noninvasive cross-sectional imaging is

particularly helpful in such cases to depict the depth

of myometrial invasion, tumor extent, and presence

of lymphadenopathy. Pretreatment imaging im-

proves patient care by assisting in determining the

type and extent of surgery or radiation treatment.

Endometrial cancer

Screening and diagnosis

Endometrial cancer is most commonly seen in el-

derly women with dysfunctional uterine bleeding

[1] . Approximately 12% of endometrial cancers oc-

cur in premenopausal women, however [2] . The

American Cancer Society recommends that all post-

menopausal women be informed about the risks

and symptoms of endometrial cancer and encour-

aged to report any bleeding or spotting. Annual

screening with endometrial biopsy beginning at

Imaging of Uterine Cancer 169

Table 1: TNM and International Federation of Gynecology and Obstetrics staging systems

for endometrial cancer

TNM

FIGO

T - Primary Tumor

Primary tumor cannot be assessed

Tis

Carcinoma in situ

Tumor conﬁned to corpus uteri

T1a

Tumor limited to endometrium

T1b

Tumor invades less than one half of the myometrium

T1c

Tumor invades one half or more of the myometrium

Tumor invades cervix but does not extend beyond uterus

T2a

IIA

Endocervical glandular involvement only

T2b

IIB

Cervical stromal invasion

III

Local and/or regional spread as speciﬁed in T3a, b, and/or N1

and FIGO IIIA, B, and C below

T3a

IIIA

Tumor involves serosa and/or adnexa (direct extension or

metastasis) and/or cancer cells in ascites or peritoneal

washings

T3b

IIIB

Vaginal involvement (direct extension or metastasis)

IIIC

Metastasis to the pelvic and/or para-aortic lymph nodes

IVA

Tumor invades bladder mucosa and/or bowel mucosa (bullous

edema is not sufﬁcient to classify a tumor as T4)

N—Regional Lymph Nodes

Regional nodes cannot be assessed

No regional nodal metastasis

Regional nodal metastasis

M—Distant Metastasis

Distant metastasis cannot be assessed

No distant metastasis

IVB

Distant metastasis (includes metastasis to intra-abdominal

lymph nodes other than para-aortic, and/or inguinal lymph

nodes; excludes metastasis to vagina pelvic serosa, or adnexa)

Ultrasonography

Ultrasonography, especially with a transvaginal ap-

proach, is the initial imaging modality in patients

with suspected endometrial cancer. Endometrial

cancer most often appears as thickened endome-

trium that is more than 5 mm in a postmenopausal

woman or 15 mm in a premenopausal woman

( Fig. 1 ). Echogenicity varies, but alteration of endo-

metrial texture or focal increased echogenicity may

be seen [17] . These appearances are not speciﬁc and

can be observed in endometrial hyperplasia and

polyps [18] . Saline infusion sonohysterography

improves diagnosis for endometrial cancer with

reported 89% sensitivity, 46% speciﬁcity, 16% pos-

itive predictive value, and 97% negative predictive

value [19,20] . Risk of disseminating malignant cells

by saline infusion sonohysterography is small,

approximately 7% [21] .

Color Doppler US often reveals increased vascu-

larity with a multivessel pattern, in contrast to the

pedicle artery sign seen in endometrial polyps

[22–24] . Spectral Doppler indices may have low-

impedance ﬂow, but there is signiﬁcant overlap in

Doppler indices of benign and malignant condi-

tions of the endometrium [25] .

Myometrial invasion is depicted as irregularity of

the endometrium-myometrium border and disrup-

tion of the subendometrial halo. The accuracy of US

for diagnosing the depth of invasion is approxi-

mately 73% to 93%, but US is better for grade 2-3

tumors and should not be used as the sole criterion

for the decision to perform extensive surgery

[26–28] . Although US can be used to estimate

depth of invasion, a recent meta-analysis has shown

that contrast-enhanced MR imaging has better over-

all performance [29] .

On CT, endometrial cancer remains relatively low

attenuation compared with myometrium after con-

trast administration ( Fig. 1 ).

Early studies with conventional CT reported 84%

to 88% staging accuracy for endometrial cancer

[30,31] . A more recent study with helical CT

reported a sensitivity of 83% and a speciﬁcity of

42% for the assessment of depth of myometrial

No evidence of primary tumor

170

Akin et al

Fig. 1. A 57-year-old woman with endometrial cancer. Transvaginal US (A), contrast-enhanced CT (B), and sagittal

T2-weighted (C) and postcontrast T1-weighted MR imaging (D) show a large endometrial mass (M). Note that

the low signal intensity junctional zone is intact (arrow)(C) and there is a smooth interface between the

mass and the myometrium (arrow)(D). These ﬁndings rule out myometrial invasion.

invasion and a sensitivity of 25% and a speciﬁcity of

70% for the depiction of cervical invasion [32] .CT

is limited for the evaluation of cervical extension

and depth of myometrial invasion. CT is most com-

monly used in the assessment of advanced disease.

CT can demonstrate invasion to the adjacent or-

gans, such as bladder and rectum. Distant metasta-

ses from endometrial cancer are most often seen in

the extrapelvic lymph nodes and peritoneum. CT is

a reliable method in the assessment of enlarged

lymph nodes. Peritoneal metastases on CT appear

as peritoneal thickening, soft-tissue masses, and as-

cites. Detection of small lymph node metastases

and peritoneal implants is difﬁcult not only with

CT but also with other imaging methods, however.

presence of myometrial invasion is suspected if

there is an irregular endometrium-myometrium in-

terface. Dynamic postcontrast images are especially

valuable in demonstrating myometrial invasion

because endometrial cancer enhances less than

myometrium ( Figs. 1 and 2 ).

Determining the presence of myometrial inva-

sion is a critical factor because in patients with

deep myometrial invasion (invasion >50% thick-

ness of myometrium), there is a six- to sevenfold in-

creased prevalence of pelvic and lumboaortic

lymph node metastases compared with patients

with myometrial invasion that is absent or less

than 50% [33] . The preoperative determination of

myometrial invasion helps in planning the extent

of lymphadenectomy.

Conditions that may render MR imaging evalua-

tion of endometrial cancer difﬁcult include the

presence of an indistinct junctional zone in a post-

menopausal woman, an irregular and thickened

junctional zone in adenomyosis, myometrial thin-

ning by a large tumor, and myometrial distortion

by a large leiomyoma.

In the early 1990s, the overall staging accuracy of

MR imaging was reported to be 83% to 92%

[34–36] . A more recent study conﬁrmed these early

reports and showed that MR imaging had 87%

MR imaging

MR imaging is the most accurate modality for the

pretreatment evaluation of endometrial cancer. En-

dometrial carcinoma is usually seen as a mass that

is hypo- to isointense on T1-weighted images and

hyperintense or heterogeneous on T2-weighted im-

ages compared with myometrium. On T2-weighted

images if the normal low signal intensity junctional

zone is intact, myometrial invasion can be

excluded. If the junctional zone is not well seen be-

cause of atrophy or distention caused by a mass, the

Imaging of Uterine Cancer 171

Fig. 2. A 76-year-old woman with endometrial cancer. Transverse (A) and sagittal (B) T2-weighted MR imaging

show a large endometrial mass (M) that extends to the cervix (arrows). Note that the junctional zone is disrupted

and the mass extends to the uterine serosa (short arrow)(A).

sensitivity and 91% speciﬁcity in assessing myome-

trial inﬁltration, 80% sensitivity and 96% speciﬁcity

for cervical invasion, and 50% sensitivity and 95%

speciﬁcity for lymph node assessment. There was

signiﬁcant agreement between MR imaging and

surgicopathologic ﬁndings in assessment of myo-

metrial invasion (P < 0.001) [37] . Like all other

cross-sectional imaging methods, MR imaging is

limited in the assessment of lymph node status

because it does not allow clear differentiation be-

tween metastatic and nonmetastatic lymph nodes

of similar size.

A study using meta-analysis and bayesian analy-

sis showed that ﬁndings from contrast-enhanced

MR imaging signiﬁcantly affected the posttest prob-

ability of deep myometrial invasion in patients with

endometrial cancer. In this study, the mean

weighted pretest probabilities of deep myometrial

invasion in patients with tumor grades 1, 2, and 3

were 13%, 35%, and 54%, respectively. Posttest

probabilities of deep myometrial invasion for

grades 1, 2, and 3 increased to 60%, 84%, and

92%, respectively, with positive MR imaging ﬁnd-

ings and decreased to 1%, 5%, and 10%, respec-

tively, with negative MR imaging ﬁndings [38] .

of myometrial invasion is important. Tumor exten-

sion into the cervix affects the type of surgery, and

parametrial invasion requires radiation as the ini-

tial treatment or a more radical surgical approach.

The value of US, CT, and MR imaging for diagno-

sis of myometrial invasion and cervical extension

has been assessed. Several reports have indicated

that MR imaging, being more accurate than CT

and US, is the most advantageous technique for

the evaluation of endometrial cancer [40–42] .A

meta-analysis showed no signiﬁcant differences in

the overall performance of CT, US, and MR imag-

ing. For the assessment of myometrial invasion,

however, contrast-enhanced MR imaging per-

formed signiﬁcantly better than did non-enhanced

MR imaging or US (P < 0.002) and demonstrated

a trend toward better results, as compared with

CT. The lack of data on the assessment of cervical

invasion at CT or US prevented meta-analytic com-

parison with data obtained at MR imaging [43] .

The following guidelines can be used for staging

endometrial cancer [43] :

1. No imaging is required for a patient with grade 1

tumor and a non-enlarged uterus at physical

examination because the pretest probability of

myometrial, cervical, or nodal involvement is

low. If results from the physical examination

are inconclusive or if there is concomitant pelvic

disease, US, CT, or MR imaging can be used for

the initial radiologic investigation.

2. Patients with high-grade papillary or clear cell

tumors should undergo CT or MR imaging be-

cause there is a high pretest probability of nodal

involvement.

3. Patients with possible cervical involvement at

physical examination or with positive or incon-

clusive results from endocervical curettage

should undergo MR imaging, because this is

The role of imaging in treatment planning

Morphologic prognostic factors, including depth of

myometrial invasion, cervical extension, and lymph

node metastasis, inﬂuence the prognosis and treat-

ment options in endometrial cancer [39] . Lympha-

denectomy and pre- or postoperative radiation

therapy are indicated in patients at high risk of ex-

trauterine disease or lymph node metastasis. Be-

cause the probability of extrauterine disease and

lymph node metastasis correlates with the depth

of myometrial invasion, preoperative knowledge

08 - Radiol Clin N Am 2007 - Imaging of Uterine Cancer.pdf

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: