Abstract
The incidence of endometrial cancer in the UK has risen steeply since the 1990s. Survival has improved significantly over several decades but outcomes for 25% of women remain poor. Surgery is the initial treatment approach in most cases. Sentinel node surgery has largely replaced systematic lymphadenectomy for surgical staging in apparent early-stage disease. Radiotherapy and chemotherapy are used to reduce recurrence, and for upfront primary treatment in selected cases. Different endometrial cancer sub-types have different aetiologies, histological and molecular characteristics and prognoses. Molecular classification of tumours is now indicated in all patients treated for endometrial cancer. Molecular characterization has important implications for treatment options, both at initial presentation and any subsequent recurrence, including access to new immunotherapy treatments. In addition, “mainstreaming” of genetic testing in those at risk of Lynch syndrome, will improve identification of at-risk families and help prevent future endometrial, ovarian and bowel cancers.
Introduction
Endometrial cancer is the sixth most common female cancer worldwide. In the United Kingdom, the incidence has risen by 59% since the 1990s due to an increase in endometroid-type endometrial cancer. These account for approximately 80% of all endometrial cancers. Incidence rises sharply above the age of 45–50 years and peaks between 75 and 79 years. Seventy-five percent of endometrial cancers are diagnosed in post-menopause. The 10-year survival following a diagnosis of endometrial cancer is 72% reflecting most diagnoses at early stage. Despite significant improvements in overall long-term survival in the past 40 years, due to limited treatment options for advanced or recurrent disease, their outcomes have remained poor.
Pathology and aetiology
Endometrial cancers develop in the epithelial cells of endometrial glands. Endometrioid endometrial adenocarcinomas (EEC) arise from atypical hyperplasia. Risk factors for atypical hyperplasia and EEC are shown in Box 1 . Endometrial hyperplasia without atypia, has minimal risk of progression to endometrial cancer (<5%) but endometrial hyperplasia with nuclear atypia (atypical hyperplasia) has a considerable risk of progression to invasive cancer and hysterectomy is advised. Patients should be informed that 43% of hysterectomy specimens removed for atypical hyperplasia contain invasive cancer. Endometrioid tumours are further sub-classified according to grade. Grade 3 tumours are referred to as “high grade” and have been associated with higher risk of metastasis and worse prognosis. Grade 1 and 2 tumours are “low grade”, frequently express oestrogen and progesterone receptors and are associated with less aggressive biological behaviour Box 1 .
-
Obesity
-
Prolonged anovulation, e.g. polycystic ovarian syndrome
-
Diabetes mellitus
-
Hypertension
-
Infertility
-
Nulliparity
-
Early menarche and late menopause
-
Current or past tamoxifen use
-
Genetic predisposition (Lynch syndrome)
-
Oestrogen-secreting ovarian tumour
-
Unopposed oestrogen replacement therapy
Serous endometrial carcinomas, clear cell carcinomas and endometrial carcinosarcomas account for <15% of endometrial cancers but are responsible for a disproportionate number of deaths and are biologically more aggressive, with increased risk of metastasis. Serous carcinomas are the most common of this group and are high-grade tumours. Endometrial intra-epithelial carcinoma (EIC) is the pre-cursor lesion for these.
Traditional histopathological interpretation is subject to inter-observer error, particularly for high-grade tumours, where interpretation of histological sub-type can differ in up to 36% of cases. The recent identification of four distinct molecular subtypes of endometrial cancer has not only improved our understanding of these tumours but is also of practical importance, and now included in the updated FIGO tumour staging for endometrial cancer and clinical care algorithms. These subtypes are p53-abnormal (p53abn), POLE-ultramutated (POLEmut), mismatch repair–deficient (MMRd), and no specific molecular profile (NSMP). POLE-ultramutated (POLEmut) tumours confer an excellent prognosis even when assessed as high grade by conventional histopathology. In contrast, individuals with p53abn cancer have poor prognosis. MMRd or NSMP tumours have intermediate clinical outcomes. This was confirmed in a large cohort of fully staged patients with endometrial cancer (full systematic lymphadenectomy for all patients), where p53abn tumours were associated with poor prognosis even when confirmed to be stage 1.
Endometrial cancers grow directly into myometrium and through the uterine serosa, inferiorly into the cervical stroma and laterally to the para-uterine tissues. There may be trans-tubal infiltration with involvement of ovaries and peritoneal cavity. Trans-peritoneal spread is a feature of serous tumours. Grade and depth of myometrial invasion both increase the risk of lymph node metastasis. Affected nodes are pelvic (iliac and obturator nodes) and para-aortic nodes. Isolated para-aortic node involvement is uncommon. Haematogenous spread leads to lung metastases. Vaginal metastases arise via haematogenous or lymphatic spread. Presence or absence of lymphovascular space invasion (LVSI) is routinely included on histopathology reports. LVSI is an independent adverse prognostic factor and the presence of substantial LVSI is a new addition to the latest FIGO staging system for endometrial cancer.
Genetic predisposition to endometrial cancer
Whilst only 3% of endometrial cancers are heritable, identifying people at risk is important as preventative options are available for them and affected relatives. Lynch syndrome is an autosomal dominant condition whereby germline mutations in mismatch repair (MMR) genes lead to inability to accurately repair DNA faults leading to increased risk of endometrial and colorectal cancers. Affected individuals have a 40–60% lifetime risk of developing endometrial cancer compared to 3% in the general population. Additionally, the risk of ovarian cancer is 10–12% compared to the general population risk of 1.6%. Once they have completed their family, those affected by Lynch syndrome are offered hysterectomy and removal of fallopian tubes and ovaries. Many affected individuals have gone undiagnosed due to selective testing criteria based on age at diagnosis and family history. New NICE guidance published in October 2020, states that every individual with endometrial cancer should have the opportunity for MMR deficiency testing by immunohistochemistry on tumour samples. Those with tumour MMR deficiency can then be offered germline mutation testing. People with Lynch syndrome are offered treatment to reduce the risk of developing colorectal cancer and surveillance to diagnose it at an early stage. There has been variable application of the testing guidelines nationally which represents many missed opportunities to offer affected individuals’ surveillance or preventative measures. In response a national transformation project has identified “mainstreaming” as the best way to improve testing coverage, reduce national variation in practice and identify more individuals with Lynch syndrome more quickly. “Mainstreaming” refers to the treating clinicians discussing and arranging the genetic tests themselves rather than making a separate referral to regional genetics services. As most patients receive surgery as their first (or only) treatment, this requires all gynaecologists who undertake surgery for endometrial cancer to be capable of discussing Lynch syndrome and testing for Lynch syndrome with patients. It is expected that the treating gynaecologist will now be responsible for discussing and requesting the genetic tests themselves.
Clinical presentation
Post-menopausal bleeding (PMB) is the most common symptom in those diagnosed with endometrial cancer. Endometrial cancer should also be considered in peri-menopausal or pre-menopausal individuals aged >45 years with new onset chaotic menstruation or menstrual dysfunction which does not improve after 3 months of treatment. Other presentations include persistent post-menopausal vaginal discharge, pyometra, and abnormal endometrial cells reported on routine cervical cytology specimens. Patients with the latter should be referred for endometrial assessment and not for colposcopy.
Nationally there has been a large (43%) increase in referrals to diagnostic cancer services for unscheduled bleeding on HRT over the past few years, which coincides with a year-on-year increase in HRT use nationally. This does not correspond to a significant increase in diagnosed endometrial cancers but places a high burden on diagnostic services. Recently published guidance from the British Menopause Society, in collaboration with other organisations, aids clinicians in prioritizing investigations for those most likely to have endometrial cancer as a cause for their unscheduled bleeding.
Initial assessment and diagnosis
Initial assessment includes enquiry about risk factors and a comprehensive review of general health and fitness, as several chronic health conditions can adversely impact the ability to subsequently provide standard treatment. Body mass index (BMI) and functional ability should be recorded. Pelvic examination, including speculum examination of the cervix, is important to exclude vulval and cervical cancer. Trans-vaginal ultrasound scan (TV-USS) is the initial investigation for measurement of endometrial thickness and to identify any significant ovarian mass (e.g. granulosa cell tumour). Assessment by ultrasound scan alone (without full pelvic examination) is not acceptable.
Different endometrial threshold measurements between 3 and 5 mm have been evaluated for the detection of endometrial disease. The negative predictive value of a thin endometrium (<4 mm) is very high (98%) and 4 mm has been adopted as the cut-off value in many hospitals although 3 and 5 mm also have acceptable performance. Out-patient aspiration endometrial biopsy is indicated for women with PMB and an endometrium measuring ≥4 mm on TV-USS. This successfully diagnoses most cancers. A histopathology report indicating an “inadequate” sample is acceptable provided the sampling device was inserted more than 4 cm through the cervix ( Figure 1 ). It is important to record the depth of insertion of the sampling device in the medical record for this reason.
Hysteroscopy (usually in the out-patient setting) is a common adjunct in many hospitals and specifically indicated if out-patient aspiration biopsy is not feasible or if there is localized endometrial thickening on TV-USS.
TV-USS is not reliable in assessing individuals with PMB on Tamoxifen because Tamoxifen causes benign cystic change in the stroma which is commonly misinterpreted as endometrial thickening ( Figure 2 ). Hysteroscopy is mandatory in this scenario.
Pre-treatment assessment and investigation
Pre-operative investigations determine likely extent of disease, fitness for surgery, appropriate route of surgery, and are needed to exclude unresectable or widely metastatic disease.
Thorough physical assessment to include speculum examination and bimanual pelvic examination is important in every case, to detect vaginal metastases, visible cervical involvement and to determine suitability for minimal access surgery. Full blood count, renal biochemistry and liver function tests are required. Other co-morbidities should be addressed as part of the pre-treatment evaluation.
For low grade tumours (endometrioid type grades 1 and 2) a chest X-ray to exclude lung metastases and a pelvic magnetic resonance imaging (MRI) scan (to identify pelvic lymph node enlargement, cervical extension and presence and extent of myometrial invasion) are indicated. Where examination suggests more advanced disease, or for those with high grade (grade 3) or non-endometrioid tumours, Computed Tomography (CT) scan of the chest, abdomen, and pelvis in indicated, to identify metastases which may lead to change of treatment plan. MRI scan is more useful for evaluating extension to the parametrial and paracervical tissues. Examination under anaesthesia may be necessary where there is uncertainty about operability due to local pelvic extension. Positron Emission Tomography (PET) is not recommended for staging. The tumour marker Ca125, is not recommended as a routine test for the assessment of PMB nor in those with proven endometrial cancer, as it neither informs diagnosis nor treatment.
Once imaging investigations are complete, provisional stage is ascribed. The staging system of the International Federation of Gynaecology and Obstetrics (FIGO) was revised in 2023 to reflect the latest evidence for survival and prognostic data related to surgical and histopathological findings ( Table 1 ). Additional sub-stages for certain molecular sub-types are included ( Table 2 ).
Full access? Get Clinical Tree
