Imaging of endometrial adenocarcinoma

Clinical Radiology (2006) 61, 545–555

REVIEW

Imaging of endometrial adenocarcinoma T.D. Barwicka, A.G. Rockalla, D.P. Bartonb, S.A. Sohaibc,* Departments of aRadiology, St Bartholomew’s Hospital, West Smithfield, bGynaecological Oncology, and c Radiology, Royal Marsden Hospital, London, UK Received 4 November 2005; received in revised form 6 March 2006; accepted 15 March 2006

Endometrial cancer is the most common gynaecological malignancy and the incidence rising. Prognosis depends on age of patient, histological grade, depth of myometrial invasion and cervical invasion and lymph node metastases. Myometrial invasion and accurate cervical involvement cannot be predicted clinically. Pre-treatment knowledge of these criteria is advantageous in order to plan treatment. The clinical challenge is to effectively select patients at risk of relapse for more radical treatment whilst avoiding over treating low risk cases. This is important as endometrial cancer predominately occurs in postmenopausal women with co-morbidities. Modern imaging provides important tools in the accurate pre-treatment assessment of endometrial cancer and may optimize treatment planning. However, there is little consensus to date on imaging in the routine preoperative assessment of endometrial carcinoma and practice varies amongst many gynaecologists. Transvaginal ultrasound is often the initial imaging examination for women with uterine bleeding. However, once the diagnosis of endometrial cancer has been made, contrast-enhanced magnetic resonance imaging (MRI) provides the best assessment of the disease. The results of contrast-enhanced MRI may identify patients who need more aggressive therapy and referral to a cancer centre. In this article we review the role of imaging in the diagnosis and staging/preoperative assessment of endometrial carcinoma. Q 2006 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Introduction Endometrial adenocarcinoma is the most common gynaecological malignancy and the fourth most frequent site of malignancy in females in North America and Europe. Incidence is rising due to increased life expectancy and rises in obesity.1,2 Despite being a common malignancy it is not a leading cause of cancer deaths. This is due to early presentation with 75–80% presenting with stage I disease.3 The prognosis in stage I disease is good with 5-year overall and cancer specific survival rates of 80–85% and 90–95%, respectively.4 Prognosis also depends on age of patient, histological tumour grade, depth of myometrial invasion, cervical invasion, and lymph node metastases.3,5,6 Staging in endometrial cancer is surgical–pathological based on the FIGO (International Federation of Gynaecology and Obstetrics) system (Table 1).7 This is partly because most patients are treated * Guarantor and correspondent: S.A. Sohaib, Department of Radiology, Royal Marsden Hospital, 203 Fulham Road, London SW3 6JJ, UK. Tel.: C44 207 80822046; fax: C44 207 3520726.

surgically and also as clinical staging is inaccurate and often underestimates the extent of disease.3,8,9 Therefore since 1988 FIGO has required hysterectomy, bilateral salpingo-oopherectomy and pelvic/para aortic lymphadenectomy/lymph node sampling to stage endometrial cancer.7 This surgical–pathological staging system is independent of radiological staging or assessment. However, since the introduction of surgical staging there have been advances in imaging. Modern imaging provides important tools in the preoperative assessment of endometrial cancer and may optimize treatment planning and the surgery to be undertaken.10 Without imaging the surgeon has to rely on preoperative tumour grade, which is often inaccurate, to guide operative management. However, there is no consensus on the role of imaging in the routine preoperative assessment of endometrial carcinoma and practice varies amongst many gynaecologists.11 The use of magnetic resonance imaging (MRI) in the assessment of myometrial invasion and cervical invasion is increasingly being recognized.10,12,13 More recently lymph nodespecific contrast agents are emerging as useful tools

0009-9260/$ - see front matter Q 2006 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.crad.2006.03.011

546

Table 1

T.D. Barwick et al.

Endometrial cancer staging: summary.86

FIGO stage

TMN stage

Extent of disease

I

T1 T1a T1b T1c T2 T2a T2b T3GN1 T3a

Confined to corpus Limited to endometrium Less than half of myometrium More than half of myometrium Extension to cervix Endocervical glandular only Cervical stroma Local G regional disease Serosa/adnexa/positive peritoneal cytology Vaginal involvement Regional lymph node metastasis Mucosa of bladder/bowel Distant metastasis

IA IB IC II IIA IIB III IIIA IIIB IIIC IVA IVB

T3b N1 T4 M1

in determining metastatic disease in lymph nodes.14 In this article we review the role of imaging in the diagnosis and staging/preoperative assessment of endometrial adenocarcinoma.

Primary diagnosis Endometrial carcinoma most commonly presents with abnormal uterine bleeding, usually postmenopausal. Most patients presenting with postmenopausal bleeding have a benign cause such as endometrial atrophy, endometrial polyps, and non-uterine causes. Endometrial carcinoma accounts for 7–14% of cases of postmenopausal bleeding.15–18 Consensus regarding evaluation of women with postmenopausal bleeding has not been reached and practice varies depending on resources and the physician’s assessment of patient risk.19 Transvaginal ultrasound (TVS) is the imaging procedure of choice to evaluate patients with postmenopausal bleeding.19,20 Ultrasound signs of endometrial carcinoma include heterogeneity and irregular endometrial thickening (Fig. 1). However, these signs are non-specific and ultrasound cannot reliably distinguish between benign proliferation, hyperplasia, polyps and cancer.20,21 Sonohysterography (TVS with instillation of fluid into the uterine cavity to outline intrauterine pathology) may help to differentiate polyps, hyperplasia and carcinoma.20,22,23 This improves specificity of TVS in differentiating endoluminal masses from diffuse endometrial thickening, but it is not widely available.20 An endometrial thickness of less than 4 mm generally excludes cancer.18,19,24 A meta-analysis of TVS in the diagnosis of endometrial carcinoma using a thickness of 5 mm showed a sensitivity of 96% with a specificity of 61%.17 However, using a

Figure 1 TVS transverse image through the uterus shows thickened and heterogeneous endometrium (between callipers). (Courtesy of Dr J.A.W. Webb, with permission.)

cut-off of 4–5 mm for normal endometrial thickness may miss 1–4% cancers.17,25 Some authors find this false-negative rate unacceptable and advocate biopsy in all patients.25 Most authors consider TVS a highly sensitive non-invasive test and advise biopsy only if endometrial thickness greater than 4 mm or focal thickening is seen.16–18,20,24 However, if the bleeding continues the endometrial (and endocervical) tissues need to be sampled. Ultrasound also images the pelvis and adnexae and rarely a cause of post-menopausal bleeding is an ovarian or fallopian tube cancer.23 Histopathological diagnosis and tumour grade can either be obtained by outpatient clinic endometrial biopsy (Pipelle, Prodimed, Neuillyen-Thelle, France) or by hysteroscopic assessment. When sampling is sufficient, detection rates for endometrial cancer are up to 99%.20,26 Falsenegative rates of office-based endometrial biopsy and of dilatation and curettage (D&C) are 5–15% and 2–6% respectively.16,17,27 Sampling errors occur when a focal abnormality is not sampled or the sample is insufficient.20 Up to 28% of endometrial biopsies are non-diagnostic/insufficient.23,27,28 Furthermore (random) endometrial biopsy can be inaccurate with 15–30% of patients with a preoperative biopsy of grade 1 having a final grade 2 or 3 in the histological specimen.26,29–31 Both office endometrial biopsy (pipelle) and hysteroscopic biopsy have equivalent accuracy in determining tumour grade.29,30 In summary TVS is highly sensitive at detecting endometrial abnormalities but due to low specificity biopsy is required if there are abnormal findings. Like all techniques TVS will fail to identify cancer in some women but the false-negative rates

Imaging of endometrial adenocarcinoma

compare favourably with endometrial biopsy techniques. The cost of TVS compares favourably with endometrial biopsy in the evaluation of postmenopausal bleeding.23

Staging/preoperative assessment The clinical challenge in endometrial cancer is to effectively select patients at risk of relapse for more radical surgery (e.g. lymphadenectomy) and adjuvant treatment whilst avoiding over treating low-risk cases.32 This is important as endometrial cancer predominately occurs in postmenopausal women with co-morbidities such as obesity, hypertension and diabetes.33,34 Recent modifications of surgical (such as laparoscopic) techniques are emerging as alternatives for patients with early disease.35,36–38 Surgical staging is intended as the initial treatment and to identify patients who may require adjuvant therapy. When determining extent of adjuvant therapy surgeons consider tumour grade, depth of myometrial invasion and lymph node status.39 The depth of myometrial invasion is an important factor in predicting nodal metastases.3,5 Incidence of lymph node metastases increases from 3% with superficial myometrial invasion (stage IB) to more than 40% with deep myometrial invasion (stage IC).3,8 In addition the prevalence of para-aortic lymph node metastases varies from 3% in patients with tumour confined to the endometrium or superficial myometrium to up to 46% in patients with deep myometrial invasion.40,41 At surgery the depth of myometrial invasion is determined by gross visual inspection or frozen section. However, gross visual inspection has been reported to be only 65% accurate in grade 2 lesions and 31% accurate in grade 3 lesions,42 and only 71% accurate in cases of deep (O50%) myometrial invasion.43 Frozen section analysis is more accurate6,44 but is still inaccurate in a significant number of patients45. There is controversy regarding the role of lymphadenectomy and practice varies considerably.32,46 Lymphadenectomy carries a risk of complications but these are minimized by an experienced surgical team working in a cancer centre. Therefore careful selection of high-risk patients is important for specialist surgical referral to a gynaecological oncology team.33,47 Recognized indications for lymphadenectomy include grade 1 or 2 with O50% myometrial invasion, all grade 3 disease, high-risk histology (serous papillary and clear cell) and cervical extension.29,48 There is also

547

a range of opinion (and clinical practice) as to whether node sampling should be performed or a formal node dissection and whether both pelvic and para-aortic nodes should be removed.32 A pivotal issue is the role of lymphadenectomy in early stage 1 carcinoma (stage IA and B). It is not clear whether lymph node sampling itself conveys a therapeutic benefit with variable outcomes reported.29,48–51 A large randomized trial on endometrial cancer (the ASTEC trial) has been completed in the UK. The questions this trial proposed to address were the role of lymphadenectomy and the role of adjuvant radiotherapy in this disease.52 The results should be published within the next 12 months. Imaging may assist in preoperative assessment and surgical planning by predicting depth of myometrial invasion, cervical involvement, distant spread, and lymph node involvement. Preoperative knowledge of these factors is highly important as patients at high risk of extrauterine spread and lymph node metastases may have pre/postoperative radiotherapy and lymphadenectomy. Importantly preoperative MRI may decrease the number of unnecessary lymph node dissections.11 In addition patients with cervical involvement may have preoperative radiotherapy and arguably more radical surgery.53,54 However, there is considerably more morbidity with the more radical surgery and adjuvant radiotherapy and the optimal management of endometrial carcinoma with cervical involvement (stage II disease) remains to be established.55 Imaging may provide other information that can influence management. For example information on uterine size, tumour volume, ascites and adnexal pathology can help in determining whether the surgical route is transabdominal or transvaginal. In patients at high risk for peri-operative morbidity imaging may assist in the planning of non-surgical treatment such as radiotherapy or in early stage disease (IA) treatments such as intrauterine progesterone.56

Imaging assessment of endometrial cancer Histological diagnosis before staging is essential as imaging cannot accurately differentiate endometrial carcinoma from blood clot or hyperplasia.57 MRI is considered the most accurate imaging technique for preoperative assessment of endometrial cancer due its excellent soft-tissue contrast resolution. Overall staging accuracies have been reported at 83–92% (See Table 2).12,58,59 In

548

T.D. Barwick et al.

comparative studies between computed tomography (CT), MRI and TVS; TVS is considered more accurate than CT for assessment of myometrial invasion but less accurate than MRI.12,60,61 TVS has been shown in some studies to be up to 80–100% accurate in assessing deep myometrial invasion.60,62–64 However, if strict assignment of FIGO stage is used accuracies drop to 60– 77%.12,60,63,65 The accuracy of CT staging of endometrial carcinoma has been reported at 61–76%.12,66– 68 Furthermore a study of the impact of preoperative CT showed routine preoperative CT altered treatment in only 8%.69 To date no studies have assessed multidetector CT (MDCT). However, it is unlikely MDCT will improve much the T staging in endometrial cancer as there is relatively little contrast difference between tumour and myometrium.

MRI Our MRI technique protocol is shown in Table 3. The normal zonal uterine anatomy is clearly delineated on T2-weighted imaging. Endometrial carcinoma typically is isointense to myometrium on T1-weighted sequences and lower signal intensity than endometrial lining on T2-weighted sequences. Tumours are usually of lower signal intensity than the brightly enhancing normal myometrial tissue after contrast medium administration, and on dynamic contrast enhancement enhance more slowly than the myometrium. Dynamic MRI depicts inner muscle layer with higher signal intensity than outer muscle layer in the early phase.70–72 Maximum contrast between inner and outer layer occurs at 50 s.72 In addition sub-endometrial enhancement is frequently observed in postmenopausal women when the junctional zone (JZ) is absent.70,71

Myometrial invasion (stage I disease) Stage 0 tumour (carcinoma in situ) is not visualized on MRI. Tumours are diagnosed as confined to the endometrium (stage IA) when the JZ appears intact (Fig. 2). Myometrial invasion is detected by disruption or discontinuity of the JZ. In patients in whom the JZ in not visible or indistinct, a smooth interface between the endometrium and the myometrium is considered to represent an intact myometrium. Conversely an irregular interface suggests myometrial invasion (stage IB or C; Fig. 3). In stage IC tumour disrupts the JZ and extends O50% into myometrium, however, an outer

Figure 2 Stage IA: oblique axial T2-weighted image shows an intermediate signal intensity tumour within the endometrial cavity but the junctional zone is intact (arrow). Note the signal void from the intrauterine contraceptive device (arrowhead).

stripe of normal myometrial tissue remains intact.41,57,59,70,73 The literature is confusing regarding accuracy of MRI in assessing depth of invasion and staging as each study quotes different parameters. From a management point of view differentiating stage IC from lesser degrees of invasion is important as in some centres this determines who will undergo lymphadenectomy. Therefore in Table 2 we have quoted the sensitivity, specificity and accuracy of various sequences in differentiating deep myometrial invasion from lesser degrees of invasion. In addition overall staging accuracy is also quoted. Accuracies in assessing depth of myometrial invasion with dynamic contrast enhanced MRI are between 83.9–92.6% (Table 2).35,70–72,74,75 Some studies have shown a significant improvement in accuracy of depth of myometrial tumour invasion with dynamic contrast-enhanced MRI over T2weighted imaging or non-dynamic contrastenhanced T1-weighted images.71,72,75 Whereas other authors have not found any statistically significant difference. However, it is felt to improve specificity and PPV in differentiating deep myometrial invasion from superficial and also level of diagnostic confidence.74,76 Our experience is that the overall accuracy in determining depth of

Imaging of endometrial adenocarcinoma

549

hyperintense. In stage IIA disease, invasion of the endocervix, there is widening of the endocervical canal and internal os with preservation of the cervical stroma. In stage IIB there is invasion of cervical stroma (Fig. 5). Dynamic contrast-enhanced MRI in combination with T2-weighted imaging is useful in assessing cervical involvement.13 The accuracy of MRI in cervical involvement is reported as 90–92% with sensitivities of 75–80% and specificities of 94– 96%.13,35,79 MRI is superior to TVS and CT in predicting cervical stromal invasion but cannot reliably identify superficial mucosal involvement and in these cases hysteroscopy is useful.79 Polypoidal tumour protruding from the endometrial cavity into the endocervix is a recognized pitfall79.

Locally advanced and nodal disease (stage III & IVA)

Figure 3 Stage IB: sagittal T2-weighted image (a) and T1-weighted post-contrast image (b) show intermediate signal intensity tumour invading the inner half of the myometrium (arrows).

myometrial invasion and cervical invasion is not different between T2 and dynamic contrastenhanced images.76 This is probably because some of the sequences are complimentary and doing both sequences adds to diagnostic confidence in assessing myometrial invasion. In determining depth of myometrial invasion one needs to be aware of a number of pitfalls. These are due to polypoid tumours significantly compressing the myometrium;41,77 distension of the endometrial cavity by pyometria, presence of leiomyomata57,77 and poor tumour/myometrial contrast (Fig. 4).41,59,60,75,78

Cervical invasion (stage II disease) On T2-weighted imaging normal cervical stroma is hypointense and the endometrial carcinoma is

Stage III disease spreads outside the uterus but not beyond the true pelvis. In stage IIIA disease (extension to serosa, adnexa or positive peritoneal cytology) the integrity of the outer myometrium may be disrupted with direct extension to ovaries or discrete ovarian metastases. In stage IIIB disease there is tumour extension into upper vagina. In stage IIIC there is lymph node involvement (Fig. 6). The location of nodal metastases corresponds to the drainage sites of the involved portions of the uterus. The middle and lower aspects of the uterus drain to the parametrium, paracervical and obturator lymph nodes. The upper corpus and fundus drain to common iliac and para-aortic lymph nodes. The inguinal nodes may be involved by spread along the round ligaments and rarely there is a direct para-aortic pathway.4 There is no definite correlation between endometrial tumour size and risk of nodal disease, unlike cervical cancer where there is a correlation between tumour size and pelvic lymph node metastases.80,81 In one study macroscopic measurement of tumour diameter was shown to correlate with pelvic lymph node metastases but to date MRI studies of tumour size have not shown a correlation.22,76 This may be due to the fact it is difficult to get accurate and reproducible measurement of size of the primary lesions. The diagnosis of lymph node metastases on CT and MRI use size criterion with a cut off of 1 cm minimum axial diameter.82 The use of size criteria results in a low sensitivity for detection of lymph node metastases.14,35,83 A recent MRI study in

Sensitivity, specificity and accuracy of magnetic resonance imaging (MRI) in differentiating deep myometrial invasion (stage 1C) from lower stages (stages 1A & 1B).

Author

No. of patients

Tesla

Sensitivity (%)

Specificity (%)

A. T2 weighted imaging (T2WI) 39 Hricak et al. 57 Hricak et al. 87 18

0.5 1.5

100a 100a

Lien et al. 59 Sironi et al. 41

33 56

1.5 0.5

90.9 83

63.6 87

81.8 86a

Yamashita et al. 60 Yamashita et al. 71

1.5 1.5

77 40.0a

89 95.8a

85 a 92.4a

Ito et al. 70 Joja et al. 72

40 53 (27 normals) 22 46

Seki et al. 75 Savci et al. 74 Lee et al. 78

Pre 19 Post 27 40 20 46

97a 75a

Accuracy (%) deep versus lower depths of myometrial invasion

Accuracy (%) of actual depth of invasion

97a 83

77 72 – 72a 67 67.9

1.5 1.5

100a 89a

58a 79a

77 83a

77a 73.9

1.5 1.0 1.5

100a 85a 69 88 50a

93a 64a 89 91 84a

95a 74a 82a 90 78a

84.2 66.7 58

100a 43a

100a 57a

100a 52a

80 33

B. Non-dynamic contrast-enhanced T1-weighted imaging (CET1) Hricak et al. 87 18 1.5 100a 92a

94

94

93a

88a

–

Pre 25 Post 21

Sironi et al.

41

– 59

0.5

94

92

Yamashita et al. 60 Kim et al. 12 Joja et al. 72

40 26 46

1.5 0.5 1.5

85 90 94a

96 87.5 86a

93a 88.5 89a

Seki et al. 75 Lee et al. 78

Pre 19 Post 27 40 46

1.5 1.5

100a 92a 77 88a

93a 79a 100 89a

95a 85a 92a 89a

84.2 77.8 68 61

100a 86a

92a 86a

92a 86a

44 81

Pre 25 Post 21

85 80.4

Note

No statistically significant difference in accuracy between T2 and CET1 CET1 significantly more sensitive than T2 for depth of invasion Accuracy of CET1 significantly better than T2 Accuracy of DET1 significantly higher than T2

DET1 significantly more accurate in determining depth of invasion in premenopausal women but not postmenopausal women

Accuracy of DET1 significantly higher than T2 No statistically significant difference between T2 & DET1 T2 significantly more accurate in premenopausal, CET1 more accurate in postmenopausal

No statistically significant difference in accuracy between T2 and CET1 CET1 significantly more sensitive than T2 for depth of invasion Accuracy of CET1 significantly better than T2 T2 and CET1 reviewed together DET1 significantly more accurate in determining depth of invasion in premenopausal women but not postmenopausal women

Accuracy of DET1 significantly higher than T2 T2 significantly more accurate in premenopausal, CET1 more accurate in postmenopausal

T.D. Barwick et al.

56

550

Table 2

1.0 1.5 1.5 Savci et al. 74 Seki et al. 75 Manfredi et al. 35

Pre, premenopausal; post, postmenopausal; CETI, contrast-enhanced T1-weighted imaging; DET1, dynamic contrast-enhanced T1-weighted imaging. a Indicates when a figure was calculated from the study data.

No statistically significant difference between T2 and DET1 Accuracy of DET1 significantly higher than T2 T2 and DET1 reviewed together 100a 96a 90 92a 91.9 100a 100a 100 100 95.2 100a 85a 78 77 86.6 Pre 19 Post 27 20 40 37

100a 1.5 46 Joja et al. 72

89a

96a

–

85 83.9

89.5 92.6

91.3

DET1 performed at single level where invasion most suspected on T2 DET1 significantly more accurate in determining depth of invasion in premenopausal women but not postmenopausal women 86a 91 83a

Accuracy of DET1 significantly higher than T2 96.2

C. Dynamic contrast enhanced T1 weighted imaging (DET1) Yamashita et al. 71 53 1.5 80.0 (27 normals) Ito et al. 70 22 1.5 100a

97.9

84.9

Imaging of endometrial adenocarcinoma

551

Figure 4 Axial T2-weighted image shows low signal intensity fibroid (star) distorting the uterine cavity and making assessment of myometrial invasion by the intermediate signal intensity tumour (arrows) difficult. Radiologically thought to be stage IA, but was histologically stage IB.

assessment of lymph node metastases in endometrial carcinoma using a 1 cm cut off showed a sensitivity of only 50% with a specificity of 95%.35 Therefore presently the radiologist may indicate the presence and site of enlarged lymph nodes to guide lymphadenectomy and treatment planning.35 To date no imaging technique can identify microscopic metastases, which may account for up to 39% of nodal metastases.84 However, the sensitivity for detection of lymph node metastases may be increased by the use of lymph node-specific MRI contrast agents- ultra small particles of iron oxide (USPIO).14 A recent study of USPIO in diagnosis of lymph node metastases showed improvement in sensitivity of MRI in diagnosis of lymph node metastases with no loss of specificity (Fig. 7).14 The implications of USPIO on the management of endometrial cancer could be that: (1) the high negative predictive values of USPIO for nodal metastases would obviate the need for lymph node dissection: and (2) a positive USPIO diagnosis for nodal metastases would direct either surgical de-bulking or targeting of adjuvant radiotherapy. However, this requires further investigation. Stage IV disease extends beyond the true pelvis or invades bladder or rectum. In stage IVA disease invasion of bladder or rectum may be identified by focal loss of the low signal intensity wall of bladder or rectum.85

552

Figure 5 Stage IIB endometrial adenocarcinoma. Axial (a) and sagittal (b) T2-weighted images show an endometrial cancer (arrows) extending into and invading the cervical stroma. Note multiple uterine fibroids are present.

Metastatic spread (stage IVB)

Distant spread to lung, liver and bones is rare at presentation and usually occurs haematogenously. The lung is the most common site of distant metastases and is involved in 2–3% of patients.5 Patients with distant spread are more likely to have poorly differentiated/anaplastic cancers. Table 3

Protocol for imaging of endometrial cancer. Plane

Section thickness/gap (mm)

Field of view

Reason

T1-weighted T2-weighted T2-weighted T2-weighted

Axial Axial Sag (SE) Oblique axial (SE) (perpendicular to long axis of uterus G cervix) Sagittal

5–7/1 5–7/1 3 / 0.3 3/0.3

Standard (35–40 cm) Standard (35–40 cm) Standard (35–40 cm) Small (16–18 cm)

Mid-renal hilum to symphysis lymph node staging Mid-renal hilum to symphysis lymph node staging

3D acquisition 2/2 mm

Standard (35–40 cm)

T1-weighted C fat saturated Pre and post i.v. contrast medium at 60 and 180 s

To view the relationship between the primary tumour and the myometrium in a second plane To optimize the assessment of the possibility of myometrial invasion

Pelvic phased array coil is used. Intravenous bowel relaxants are helpful for improving the quality of images by reducing motion artefact.

T.D. Barwick et al.

Sequence

Imaging of endometrial adenocarcinoma

553

myometrial invasion and cervical involvement correlates with risk of lymph node metastases, MRI may be of use in selecting patients for lymph node sampling or lymphadenectomy and therefore identify patients who require specialist gynaecological oncology referral. MRI/CT are not sensitive for diagnosing lymph node metastases by size criteria. The use of lymph node specific MRI contrast agent appears promising. In patients at high risk of peri-operative morbidity imaging may indicate suitability for non-surgical management of stage I disease. MRI/CT may identify retroperitoneal nodes indicating extended irradiation fields.

Figure 6 Stage IIIC endometrial adenocarcinoma. Axial T2-weighted images show an enlarged left external iliac node (white arrow). Note the intermediate signal intensity endometrial carcinoma with serosal involvement (black arrows).

Figure 7 T2*-weighted axial image 24 h after administration of USPIO. Left external iliac node is of high signal intensity (arrow). This was correctly diagnosed as malignant on USPIO criteria.

Summary TVS in combination with pipelle biopsy or hysteroscopy is best for diagnosis of endometrial carcinoma. Radiological staging should only be carried out after histopathological confirmation of endometrial carcinoma as imaging cannot differentiate adequately between carcinoma and other causes of endometrial thickening. MRI is reliable in predicting depth of myometrial invasion and cervical extension with greater sensitivity and accuracy than TVS and CT. As depth of

References 1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55:10—30. 2. Amant F, Moerman P, Neven P, et al. Endometrial cancer. Lancet 2005;366:491—505. 3. Boronow RC, Morrow CP, Creasman WT, et al. Surgical staging in endometrial cancer: clinical-pathologic findings of a prospective study. Obstet Gynecol 1984;63:825—32. 4. Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355: 1404—11. 5. Berman ML, Ballon SC, Lagasse LD, Watring WG. Prognosis and treatment of endometrial cancer. Am J Obstet Gynecol 1980;136:679—88. 6. Larson DM, Connor GP, Broste SK, Krawisz BR, Johnson KK. Prognostic significance of gross myometrial invasion with endometrial cancer. Obstet Gynecol 1996;88:394—8. 7. Shepherd JH. FIGO Revised staging for gynaecological cancer. Br J Obstet Gynaecol 1989;96:889—92. 8. Creasman WT, Morrow CP, Bundy BN, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987; 60(Suppl.):2035—41. 9. Creasman WT, DeGeest K, Disaia PJ, Zaino RJ. Significance of true surgical pathologic staging: a Gynecologic Oncology Group Study. Am J Obstet Gynecol 1999;181:31—4. 10. Kinkel K, Kaji Y, Yu KK, et al. Radiologic staging in patients with endometrial cancer: a meta-analysis. Radiology 1999; 212:711—8. 11. Hardesty LA, Sumkin JH, Nath ME, et al. Use of preoperative MR imaging in the management of endometrial carcinoma: cost analysis. Radiology 2000;215:45—9. 12. Kim SH, Kim HD, Song YS, Kang SB, Lee HP. Detection of deep myometrial invasion in endometrial carcinoma: comparison of transvaginal ultrasound, CT, and MRI. J Comput Assist Tomogr 1995;19:766—72. 13. Seki H, Takano T, Sakai K. Value of dynamic MR imaging in assessing endometrial carcinoma involvement of the cervix. AJR Am J Roentgenol 2000;175:171—6. 14. Rockall AG, Sohaib SA, Harisinghani MG, et al. Diagnostic performance of nanoparticle-enhanced magnetic resonance

554

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

29.

30.

31.

32.

imaging in the diagnosis of lymph node metastases in patients with endometrial and cervical cancer. J Clin Oncol 2005;23:2813—21. Lidor A, Ismajovich B, Confino E, David MP. Histopathological findings in 226 women with post-menopausal uterine bleeding. Acta Obstet Gynecol Scand 1986;65:41—3. Karlsson B, Granberg S, Wikland M, et al. Transvaginal ultrasonography of the endometrium in women with postmenopausal bleeding—a Nordic multicenter study. Am J Obstet Gynecol 1995;172:1488—94. Smith-Bindman R, Kerlikowske K, Feldstein VA, et al. Endovaginal ultrasound to exclude endometrial cancer and other endometrial abnormalities. JAMA 1998;280: 1510—7. Gull B, Karlsson B, Milsom I, Granberg S. Can ultrasound replace dilation and curettage? A longitudinal evaluation of postmenopausal bleeding and transvaginal sonographic measurement of the endometrium as predictors of endometrial cancer Am J Obstet Gynecol 2003;188:401—8. Goldstein RB, Bree RL, Benson CB, et al. Evaluation of the woman with postmenopausal bleeding: Society of Radiologists in Ultrasound-Sponsored Consensus Conference statement. J Ultrasound Med 2001;20:1025—36. Dubinsky TJ, Parvey HR, Maklad N. The role of transvaginal sonography and endometrial biopsy in the evaluation of periand postmenopausal bleeding. AJR Am J Roentgenol 1997; 169:145—9. Grasel RP, Outwater EK, Siegelman ES, et al. Endometrial polyps: MR imaging features and distinction from endometrial carcinoma. Radiology 2000;214:47—52. O’Connell LP, Fries MH, Zeringue E, Brehm W. Triage of abnormal postmenopausal bleeding: a comparison of endometrial biopsy and transvaginal sonohysterography versus fractional curettage with hysteroscopy. Am J Obstet Gynecol 1998;178:956—61. Weber AM, Belinson JL, Bradley LD, Piedmonte MR. Vaginal ultrasonography versus endometrial biopsy in women with postmenopausal bleeding. Am J Obstet Gynecol 1997;177: 924—9. Ferrazzi E, Torri V, Trio D, et al. Sonographic endometrial thickness: a useful test to predict atrophy in patients with postmenopausal bleeding. An Italian multicenter study. Ultrasound Obstet Gynecol 1996;7:315—21. Tabor A, Watt HC, Wald NJ. Endometrial thickness as a test for endometrial cancer in women with postmenopausal vaginal bleeding. Obstet Gynecol 2002;99:663—70. Stovall TG, Photopulos GJ, Poston WM, Ling FW, Sandles LG. Pipelle endometrial sampling in patients with known endometrial carcinoma. Obstet Gynecol 1991;77:954—6. Stovall TG, Solomon SK, Ling FW. Endometrial sampling prior to hysterectomy. Obstet Gynecol 1989;73:405—9. Batool T, Reginald PW, Hughes JH. Outpatient pipelle endometrial biopsy in the investigation of postmenopausal bleeding. Br J Obstet Gynaecol 1994;101:545—6. Ben Shachar I, Pavelka J, Cohn DE, et al. Surgical staging for patients presenting with grade 1 endometrial carcinoma. Obstet Gynecol 2005;105:487—93. Daniel AG, Peters III WA. Accuracy of office and operating room curettage in the grading of endometrial carcinoma. Obstet Gynecol 1988;71:612—4. Eltabbakh GH, Shamonki J, Mount SL. Surgical stage, final grade, and survival of women with endometrial carcinoma whose preoperative endometrial biopsy shows well-differentiated tumors. Gynecol Oncol 2005;99:309—12. Creutzberg CL. Lymphadenectomy in apparent early-stage endometrial carcinoma: do numbers count? J Clin Oncol 2005;23:3653—5.

T.D. Barwick et al.

33. Orr Jr JW, Holloway RW, Orr PF, Holimon JL. Surgical staging of uterine cancer: an analysis of perioperative morbidity. Gynecol Oncol 1991;42:209—16. 34. Tozzi R, Malur S, Koehler C, Schneider A. Analysis of morbidity in patients with endometrial cancer: is there a commitment to offer laparoscopy? Gynecol Oncol 2005;97: 4—9. 35. Manfredi R, Mirk P, Maresca G, et al. Local-regional staging of endometrial carcinoma: role of MR imaging in surgical planning. Radiology 2004;231:372—8. 36. Fagotti A, Ferrandina G, Longo R, Mancuso S, Scambia G. Minilaparotomy in early stage endometrial cancer: an alternative to standard and laparoscopic treatment. Gynecol Oncol 2002;86:177—83. 37. Eltabbakh GH, Shamonki MI, Moody JM, Garafano LL. Laparoscopy as the primary modality for the treatment of women with endometrial carcinoma. Cancer 2001;91: 378—87. 38. Wong CK, Wong YH, Lo LS, Tai CM, Ng TK. Laparoscopy compared with laparotomy for the surgical staging of endometrial carcinoma. J Obstet Gynaecol Res 2005;31: 286—90. 39. Gretz HF, Economos K, Husain A, et al. The practice of surgical staging and its impact on adjuvant treatment recommendations in patients with stage I endometrial carcinoma. Gynecol Oncol 1996;61:409—15. 40. Piver MS, Lele SB, Barlow JJ, Blumenson L. Paraaortic lymph node evaluation in stage I endometrial carcinoma. Obstet Gynecol 1982;59:97—100. 41. Sironi S, Colombo E, Villa G, et al. Myometrial invasion by endometrial carcinoma: assessment with plain and gadolinium-enhanced MR imaging. Radiology 1992;185:207—12. 42. Goff BA, Rice LW. Assessment of depth of myometrial invasion in endometrial adenocarcinoma. Gynecol Oncol 1990;38:46—8. 43. Doering DL, Barnhill DR, Weiser EB, et al. Intraoperative evaluation of depth of myometrial invasion in stage I endometrial adenocarcinoma. Obstet Gynecol 1989;74: 930—3. 44. Quinlivan JA, Petersen RW, Nicklin JL. Accuracy of frozen section for the operative management of endometrial cancer. BJOG 2001;108:798—803. 45. Frumovitz M, Slomovitz BM, Singh DK, et al. Frozen section analyses as predictors of lymphatic spread in patients with early-stage uterine cancer. J Am Coll Surg 2004;199: 388—93. 46. Maggino T, Romagnolo C, Landoni F, et al. An analysis of approaches to the management of endometrial cancer in North America: a CTF study. Gynecol Oncol 1998;68:274—9. 47. Frei KA, Kinkel K, Bonel HM, et al. Prediction of deep myometrial invasion in patients with endometrial cancer: clinical utility of contrast-enhanced MR imaging-a metaanalysis and Bayesian analysis. Radiology 2000;216:444—9. 48. COSA-NZ-UK ENDOMETRIAL CANCER STUDY GROUPS. Pelvic lymphadenectomy in high risk endometrial cancer. Int J Gynecol Cancer 1996; 6:102–107. 49. Kilgore LC, Partridge EE, Alvarez RD, et al. Adenocarcinoma of the endometrium: survival comparisons of patients with and without pelvic node sampling. Gynecol Oncol 1995;56: 29—33. 50. Trimble EL, Kosary C, Park RC. Lymph node sampling and survival in endometrial cancer. Gynecol Oncol 1998;71: 340—3. 51. Cragun JM, Havrilesky LJ, Calingaert B, et al. Retrospective analysis of selective lymphadenectomy in apparent earlystage endometrial cancer. J Clin Oncol 2005;23:3668—75.

Imaging of endometrial adenocarcinoma

52. ASTEC: A Study in the Treatment of Endometrial Cancer. Protocol available at: http://www.ctu.mrc.ac.uk/studies/ documents/ASTECv3.pdf. Accessed Mar 2006. 53. Rubin SC, Hoskins WJ, Saigo PE, et al. Management of endometrial adenocarcinoma with cervical involvement. Gynecol Oncol 1992;45:294—8. 54. Elia G, Garfinkel DA, Goldberg GL, Davidson S, Runowicz CD. Surgical management of patients with endometrial cancer and cervical involvement. Eur J Gynaecol Oncol 1995;16: 169—73. 55. Menczer J. Management of endometrial carcinoma with cervical involvement. An unsettled issue. Eur J Gynaecol Oncol 2005;26:245—55. 56. Montz FJ, Bristow RE, Bovicelli A, Tomacruz R, Kurman RJ. Intrauterine progesterone treatment of early endometrial cancer. Am J Obstet Gynecol 2002;186:651—7. 57. Hricak H, Stern JL, Fisher MR, et al. Endometrial carcinoma staging by MR imaging. Radiology 1987;162:297—305. 58. Hricak H, Rubinstein LV, Gherman GM, Karstaedt N. MR imaging evaluation of endometrial carcinoma: results of an NCI cooperative study. Radiology 1991;179:829—32. 59. Lien HH, Blomlie V, Trope C, Kaern J, Abeler VM. Cancer of the endometrium: value of MR imaging in determining depth of invasion into the myometrium. AJR Am J Roentgenol 1991;157:1221—3. 60. Yamashita Y, Mizutani H, Torashima M, et al. Assessment of myometrial invasion by endometrial carcinoma: transvaginal sonography vs contrast-enhanced MR imaging. AJR Am J Roentgenol 1993;161:595—9. 61. Hardesty LA, Sumkin JH, Hakim C, Johns C, Nath M. The ability of helical CT to preoperatively stage endometrial carcinoma. AJR Am J Roentgenol 2001;176:603—6. 62. Fleischer AC, Dudley BS, Entman SS, et al. Myometrial invasion by endometrial carcinoma: sonographic assessment. Radiology 1987;162:307—10. 63. Olaya FJ, Dualde D, Garcia E, et al. Transvaginal sonography in endometrial carcinoma: preoperative assessment of the depth of myometrial invasion in 50 cases. Eur J Radiol 1998; 26:274—9. 64. Artner A, Bosze P, Gonda G. The value of ultrasound in preoperative assessment of the myometrial and cervical invasion in endometrial carcinoma. Gynecol Oncol 1994;54: 147—51. 65. Teefey SA, Stahl JA, Middleton WD, et al. Local staging of endometrial carcinoma: comparison of transvaginal and intraoperative sonography and gross visual inspection. AJR Am J Roentgenol 1996;166:547—52. 66. Dore R, Moro G, D’Andrea F, et al. CT evaluation of myometrium invasion in endometrial carcinoma. J Comput Assist Tomogr 1987;11:282—9. 67. Walsh JW, Goplerud DR. Computed tomography of primary, persistent, and recurrent endometrial malignancy. AJR Am J Roentgenol 1982;139:1149—54. 68. Balfe DM, Van Dyke J, Lee JK, Weyman PJ, McClennan BL. Computed tomography in malignant endometrial neoplasms. J Comput Assist Tomogr 1983;7:677—81. 69. Connor JP, Andrews JI, Anderson B, Buller RE. Computed tomography in endometrial carcinoma. Obstet Gynecol 2000;95:692—6. 70. Ito K, Matsumoto T, Nakada T, et al. Assessing myometrial invasion by endometrial carcinoma with dynamic MRI. J Comput Assist Tomogr 1994;18:77—86.

555

71. Yamashita Y, Harada M, Sawada T, et al. Normal uterus and FIGO stage I endometrial carcinoma: dynamic gadoliniumenhanced MR imaging. Radiology 1993;186:495—501. 72. Joja I, Asakawa M, Asakawa T, et al. Endometrial carcinoma: dynamic MRI with turbo-FLASH technique. J Comput Assist Tomogr 1996;20:878—87. 73. Sironi S, Taccagni G, Garancini P, Belloni C, DelMaschio A. Myometrial invasion by endometrial carcinoma: assessment by MR imaging. AJR Am J Roentgenol 1992;158: 565—9. 74. Savci G, Ozyaman T, Tutar M, et al. Assessment of depth of myometrial invasion by endometrial carcinoma: comparison of T2-weighted SE and contrast-enhanced dynamic GRE MR imaging. Eur Radiol 1998;8:218—23. 75. Seki H, Kimura M, Sakai K. Myometrial invasion of endometrial carcinoma: assessment with dynamic MR and contrast-enhanced T1-weighted images. Clin Radiol 1997; 52:18—23. 76. Meroni R, Rockall AG, Sohaib SA, Shepherd JH, Reznek RH. T2 weighted MRI in the pre-operative assessment of myometrial and cervical invasions in endometrial carcinoma. British Journal of Radiology Congress Series 2004;30: 6—6-. 77. Scoutt LM, McCarthy SM, Flynn SD, et al. Clinical stage I endometrial carcinoma: pitfalls in preoperative assessment with MR imaging. Work in progress. Radiology 1995;194: 567—72. 78. Lee EJ, Byun JY, Kim BS, Koong SE, Shinn KS. Staging of early endometrial carcinoma: assessment with T2-weighted and gadolinium-enhanced T1-weighted MR imaging. RadioGraphics 1999;19:937—45. 79. Toki T, Oka K, Nakayama K, Oguchi O, Fujii S. A comparative study of pre-operative procedures to assess cervical invasion by endometrial carcinoma. Br J Obstet Gynaecol 1998;105: 512—6. 80. Burghardt E, Pickel H. Local spread and lymph node involvement in cervical cancer. Obstet Gynecol 1978;52: 138—45. 81. Trattner M, Graf AH, Lax S, et al. Prognostic factors in surgically treated stage ib–iib cervical carcinomas with special emphasis on the importance of tumor volume. Gynecol Oncol 2001;82:11—16. 82. Kim SH, Kim SC, Choi BI, Han MC. Uterine cervical carcinoma: evaluation of pelvic lymph node metastasis with MR imaging. Radiology 1994;190:807—11. 83. Bipat S, Glas AS, van d V, et al. Computed tomography and magnetic resonance imaging in staging of uterine cervical carcinoma: a systematic review. Gynecol Oncol 2003;91: 59—66. 84. Chuang L, Burke TW, Tornos C, et al. Staging laparotomy for endometrial carcinoma: assessment of retroperitoneal lymph nodes. Gynecol Oncol 1995;58:189—93. 85. Ascher SM, Reinhold C. Imaging of cancer of the endometrium. Radiol Clin North Am 2002;40:563—76. 86. Sobin LH, Wittekind CH (editors). In: TNM classification of malignant tumours. Gynaecological tumours. 6th ed. New York: Wiley-Liss; 2002. p. 160—4. 87. Hricak H, Hamm B, Semelka RC, et al. Carcinoma of the uterus: use of gadopentetate dimeglumine in MR imaging. Radiology 1991;181:95—106.