Radiology-pathology correlation of endometrial carcinoma assessment on magnetic resonance imaging.

MRI plays an important role in staging endometrial carcinoma particularly in the evaluation of depth of invasion into the myometrium. Factors such as large intracavitary tumour, cornual tumour location, variable tumour appearance, adenomyosis and fibroid can affect the MRI accuracy in assessing the depth of myometrial invasion. Measurement of depth of invasion is best performed on oblique axial post-contrast sequence—paralleling the pathologists’ histological assessment. MRI plays an important role in staging endometrial carcinoma particularly in the evaluation of depth of invasion into the myometrium. Factors such as large intracavitary tumour, cornual tumour location, variable tumour appearance, adenomyosis and fibroid can affect the MRI accuracy in assessing the depth of myometrial invasion. Measurement of depth of invasion is best performed on oblique axial post-contrast sequence—paralleling the pathologists’ histological assessment.

The MRI protocol for the imaging of endometrial carcinoma as well as the value of the individual sequences is detailed in Table 2 . High-resolution T2 imaging helps to delineate the anatomy of the uterus and its relationship to adjacent organs. DWI is sensitive for tumour which shows diffusion restriction but has low spatial resolution. Dynamic contrast-enhanced (DCE) sequences with arterial phase allow visualisation of the normal early sub-endometrial myometrial enhancement. Contrast-enhanced sequence is the most reliable in differentiating between the tumour and the myometrium and optimal for the measurement of the depth of tumour invasion. Table 2 MRI protocol and assessment Sequence Plane (s) Slice thickness Tumour characteristics and what to look for T1W Axial T1W 5 mm - Endometrial cancer is isointense to myometrium on T1W - Haematometros is often seen post-biopsy and hyperintense T2W Sagittal T2W 4 mm - Endometrial carcinoma has an intermediate signal on T2W - Best sequences for anatomical delineation - Assess depth of myometrial invasion and local staging Axial T2W 4 mm Oblique axial T2W 3 mm DWI Oblique axial DWI (b50, 500, 800) 4 mm - Endometrial carcinoma shows diffusion restriction - Assess depth of myometrial invasion - Detection of small tumour deposits in the cervix, adnexa, or vagina - Detection of lymph nodes—however, does not distinguish between malignant and benign nodes DCE and post-contrast T1W Sagittal DCE T1W 3 mm - Tumour enhancement is homogeneous, slower and less avid compared to the myometrium - Early enhanced imaging helps in detecting junctional zone (subendometrial myometrium) to exclude myometrial invasion - Loss of the normal rim of enhancement of the outer myometrium indicates serosal involvement - Assess depth of myometrial invasion - Assess presence of cervical or vaginal invasion 3D Axial T1W FS with oblique axial reformat 1 mm T1W T1-weighted, T2W T2-weighted, DWI diffusion-weighted imaging, DCE dynamic contrast-enhanced sequence, FS fat-saturated MRI protocol and assessment - Endometrial cancer is isointense to myometrium on T1W - Haematometros is often seen post-biopsy and hyperintense - Endometrial carcinoma has an intermediate signal on T2W - Best sequences for anatomical delineation - Assess depth of myometrial invasion and local staging Oblique axial DWI (b50, 500, 800) - Endometrial carcinoma shows diffusion restriction - Assess depth of myometrial invasion - Detection of small tumour deposits in the cervix, adnexa, or vagina - Detection of lymph nodes—however, does not distinguish between malignant and benign nodes - Tumour enhancement is homogeneous, slower and less avid compared to the myometrium - Early enhanced imaging helps in detecting junctional zone (subendometrial myometrium) to exclude myometrial invasion - Loss of the normal rim of enhancement of the outer myometrium indicates serosal involvement - Assess depth of myometrial invasion - Assess presence of cervical or vaginal invasion T1W T1-weighted, T2W T2-weighted, DWI diffusion-weighted imaging, DCE dynamic contrast-enhanced sequence, FS fat-saturated

Whilst rare, endometrial cancer can invade adjacent pelvic organs such as bladder and bowel (Fig.  18 ). This would occur in late presentation of the disease. Such pattern of advance pelvic invasion is more commonly seen with cervical carcinoma. Fig. 18 Stage IVA endometrial carcinoma. Sagittal T2W MR images ( a ) show direct tumour extension into the bladder mucosa (arrow). The air anteriorly in the bladder is due to a fistula. Chips of a bladder tumour ( b , H&E stain, × 4 magnification) initially thought to be primary urothelial tumour show fragments of necrotic endometrioid adenocarcinoma (arrow) confirmed on immunohistochemistry. The normal bladder transitional epithelium is indicated by the arrowheads Stage IVA endometrial carcinoma. Sagittal T2W MR images ( a ) show direct tumour extension into the bladder mucosa (arrow). The air anteriorly in the bladder is due to a fistula. Chips of a bladder tumour ( b , H&E stain, × 4 magnification) initially thought to be primary urothelial tumour show fragments of necrotic endometrioid adenocarcinoma (arrow) confirmed on immunohistochemistry. The normal bladder transitional epithelium is indicated by the arrowheads The high-resolution T2 sequences are optimal for the assessing the relationship of the tumour to the adjacent organs. The local invasion on MRI shows tumour signal that disrupts the adjacent organs such as bowel or bladder. Bladder involvement requires extension to the bladder urothelium. Isolated detrusor muscle involvement is not considered to be stage IV disease. Often, involvement of the urinary bladder detrusor muscle leads to an oedematous mucosa (bullous oedema) in the bladder which can be seen on MRI or cystoscopy. Protuberant tumour in the bladder lumen with acute angle relative to the bladder inner layer would be highly concerning for urothelial involvement. The finding of a utero-vesical fistula also invariably infers the presence of bladder involvement.

Endometrial cancer is the most common gynaecological cancer in developed countries [ 1 , 2 ]. It predominantly affects post-menopausal women; however, younger cases can be seen in association with diabetes, obesity and Lynch syndrome, for example. Magnetic resonance imaging (MRI) has emerged as an informative imaging modality for local staging of endometrial cancer. This review aims to demonstrate the correlative radiological and pathological findings of endometrial carcinoma in various FIGO (International Federation of Gynecology and Obstetrics) stages as well as give examples of potential pitfalls where MR imaging is limited.

Endometrial carcinoma is the most common gynaecological cancer. Whilst most cases are low grade with low-volume disease, it can also present with higher grade and more advanced disease with higher risk of nodal spread. MRI can assist in the local staging of the endometrial carcinoma preoperatively and guide the surgical and management decisions. MRI can assess the depth of myometrial invasion and distinguish between stage 1A and stage 1B disease. Its accuracy can, however, be affected by factors such as fibroids, adenomyosis, atrophic myometrium, bulky tumour, indistinct or variegated endometrium/myometrium junction and atypical tumour appearance on MRI. Understanding these potential pitfalls would help the reporter to improve the accuracy in detection of deep myometrial invasion. MRI with its higher tissue contrast resolution also has a role in identifying the extra-uterine pelvic disease. The extra-uterine disease can be due to different patterns of spread including contiguous, drop metastases, transcoelomic and lymphatic nodal spread. We present in this paper the radiology–pathology correlation at the different FIGO stages to help reporting radiologists to have a better understanding of the local spread of endometrial carcinoma and its imaging appearance.

Endometrial cancer can be broadly divided into two types: Type I and Type II (Fig.  1 ). The most common Type 1 endometrial cancer is endometrioid adenocarcinoma, which accounts for 75–80% of endometrial cancers [ 3 ] and often arises on a background of atypical hyperplasia. Type II is less oestrogen dependent and shows a more aggressive behaviour with tendency to deeper myometrial invasion. Serous, clear-cell and undifferentiated carcinoma makes up the common histological subtypes of the Type II cancer. Fig. 1 Histology images of Type 1 ( a ) and Type 2 ( b ) endometrial carcinomas. a shows a grade 1 endometrioid adenocarcinoma. There are well-formed glands with uniform appearing pencillate nuclei. b shows a serous adenocarcinoma—a Type 2 tumour with increased propensity for early extra-uterine spread. There is a more haphazard and disorderly glandular growth with pale enlarged nuclei and prominent nucleoli Histology images of Type 1 ( a ) and Type 2 ( b ) endometrial carcinomas. a shows a grade 1 endometrioid adenocarcinoma. There are well-formed glands with uniform appearing pencillate nuclei. b shows a serous adenocarcinoma—a Type 2 tumour with increased propensity for early extra-uterine spread. There is a more haphazard and disorderly glandular growth with pale enlarged nuclei and prominent nucleoli The specific histology subtype of the endometrial carcinoma can influence the prognosis as well as the pattern of extra-uterine spread. Endometrioid subtype tends to spread by direct extension and nodal disease initially. In contrast, serous and clear-cell endometrial carcinoma is more aggressive with increased lymphovascular invasion, peritoneal and extra-abdominal spread [ 3 ]. The prognosis of endometrial cancer is dependent on histological subtype, histological grade, lymphovascular invasion and FIGO staging [ 4 , 5 ]. Tumours are at higher risk of lymph node metastases with higher grades or deeper myometrial invasion. In these patients, additional surgical staging with lymphadenectomy or sentinel node biopsy may be performed [ 5 – 7 ]. Lymphadenectomy does not confer survival advantage but allows nodal staging that identify patients that require further adjuvant therapy following surgery. FIGO staging was last updated in 2009 [ 8 ] (Table 1 ). The majority of patients with endometrial carcinoma present at early stage. Over two-thirds (69%) of the women present with stage I disease at diagnosis, whilst other stages are less common (stage II 7%; stage III 10%; stage IV 7%; stage unknown 7%) (10). There has been a recent proposal for updating the staging. The proposal incorporates the histopathologic grade into the definition of stage 1 and eliminates the cervical involvement from the staging [ 9 ]. Table 1 FIGO (2009) classification of endometrial cancer FIGO Description Stage I Tumour confined to the corpus uteri  IA No or less than half myometrial invasion  IB Invasion equal to or more than half of the myometrium Stage II Tumour invades cervical stroma, but does not extend beyond the uterus* Stage III Local and/or regional spread of the tumour  IIIA Tumour invades the serosa of the corpus uteri and/or adnexae (direct extension or metastasis) #  IIIB Vaginal and/or parametrial involvement (direct extension or metastasis) #  IIIC Metastases to pelvic and/or para-aortic lymph nodes #   IIIC1 Positive pelvic nodes   IIIC2 Positive para-aortic lymph nodes with or without positive pelvic lymph nodes Stage IV Tumour invades bladder and/or bowel mucosa, and/or distant metastases  IVA Tumour invasion of bladder and/or bowel mucosa §  IVB Distant metastases, including intra-abdominal metastases and/or inguinal lymph nodes FIGO International Federation of Gynecology and Obstetrics ⁎Endocervical glandular involvement only should be considered as stage I and no longer as stage II # Positive cytology has to be reported separately without changing the stage § Bullous oedema in itself does not indicate mucosal invasion FIGO (2009) classification of endometrial cancer FIGO International Federation of Gynecology and Obstetrics ⁎Endocervical glandular involvement only should be considered as stage I and no longer as stage II # Positive cytology has to be reported separately without changing the stage § Bullous oedema in itself does not indicate mucosal invasion

Endometrial carcinoma is intermediate in signal intensity on T2-weighted imaging and hyperintense compared to the myometrium. The tumour is hypoenhancing relative to the background myometrium on the contrast-enhanced sequences—a useful feature in evaluating the depth of myometrium invasion. Normal endometrium, in contrast, shows delayed enhancement similar to the myometrium. The tumour has signal restriction on diffusion weighted imaging (DWI) due to the increased cellularity. DWI findings help to identify the tumour when small. The tumour can be quite variable in its growth pattern. It may have a large polypoidal intracavitary component, diffuse thickening of the endometrium or minimal intracavitary component with bulk of the tumour growing into the myometrium.