Provincial Health Services Authority (PHSA) improves the health of British Columbians by seeking province-wide solutions to specialized health care needs in collaboration with BC health authorities and other partners.
Updated 2023 March
NB: Omentum or peritoneal disease = Stage IVB
Endometrial cancer (EC) is the most common gynecologic malignancy in Canada, and the incidence and mortality continue to rise globally. Historically, Bokhman described two types of EC based on distinct epidemiological factors and clinical behaviour. Type I EC corresponded to the estrogen dependent low-grade endometrioid adenocarcinomas that typically had a favourable prognosis. Type II EC corresponded to the estrogen independent non-endometrioid carcinomas which typically showed more aggressive disease with a poor prognosis. The main problem with this historic classification system was the difficulty in applying it to diagnostic practice, as grade and histotype assignment in EC have both been shown to be poorly reproducible, even amongst expert pathologists. This inconsistency can mean that depending on who interprets a pathology specimen, an individual patient could be assigned to a different risk group and receive different treatment recommendations. Inconsistency in diagnosis has also meant that biologically diverse tumours are grouped together in clinical trials making interpretation of treatment efficacy challenging or impossible.
To address these problems, in 2013 The Cancer Genome Atlas (TCGA) Project applied in depth genomic, proteomic, and transcriptomic methodologies to a cohort of ECs identifying four molecular subtypes with distinct prognoses; POLE ultramutated, microsatellite instability (MSI) hypermutated, copy-number low and copy-number high (1). Following this discovery, two research teams independently worked to develop and validate a pragmatic EC classification system that recapitulated the survival curves of the TCGA and identified molecular subtypes through clinically applicable, comparably low cost and easy to interpret assays (2-6). This uses a combination of focused next generation sequencing (NGS) for the detection of pathogenic POLE mutations, and immunohistochemistry (IHC) to assess mismatch repair proteins and p53 status, to assign EC to one of four groups: POLEmut, mismatch repair deficient (MMRd), p53abn and NSMP (no specific molecular profile) (Table below and See MOLECULAR TESTING section for [A1] further details on these molecular tests).
Unlike histopathological evaluation, the molecular subtyping of EC offers an objective and reproducible classification system that has been shown to have strong prognostic value and therapeutic implications. In Dec 2020, the World Health Organization (WHO) recommended the integration of molecular parameters in to standard pathology reporting where available and molecular classification is also now incorporated in many Societal guidelines (NCCN, ESGO/ESTRO/ESP).
NB MMR and p53 IHC are available in any pathology laboratory and is now considered standard of care for all ECs in BC. POLE testing has been approved in BC since May 2022. Testing in the current funding scheme is being paid for, for patients in whom information on POLE status would impact adjuvant treatment decisions (see POLE testing section below).
An excess of unopposed estrogen is a major risk factor for developing EC. Excess estrogen can be attributed to obesity (conversion of androstenedione to estrone in adipose tissue), anovulatory conditions (polycystic ovarian syndrome, which results in a deficiency of progesterone), and exogenous estrogen (estrogen hormone replacement therapy, or tamoxifen, a selective estrogen receptor modulator, which acts as an estrogen agonist in the uterus). Although associated with obesity, type 2 diabetes is also an independent risk factor for ECs, related to hyperinsulinemia. EC is also seen as part of Lynch syndrome (Hereditary non-polyposis colorectal cancer, HNPCC). This hereditary cancer syndrome is characterized by a very high lifetime risk of endometrial and colorectal cancers (both approximately 60%), as well as ovarian (lifetime risk of 10%), gastric, small bowel, ureter, and renal pelvis. This syndrome is caused by an inherited mutation in one of the DNA mismatch repair genes (mutations in MLH1, MSH2, MSH6, PMS2 account for 99% of all Lynch Syndrome), with the consequent tumours showing high microsatellite instability (MSI-H). These MSI-H tumours can be reliably identified by mismatch repair (MMR) protein immunohistochemistry, as described above.
With respect to prevention, weight loss in obese women and improving glycemic control in diabetic women may have the most potential for reducing risk in these specific populations. The use of combination oral contraceptives has been shown to decrease risk by 50% if used for 5 or more years (1). The addition of a progestin to estrogen replacement therapy counteracts the adverse effects of unopposed estrogen on the endometrium (2). Women who are identified as having Lynch Syndrome are advised to consider risk-reducing surgery (hysterectomy and bilateral salpingo-oophorectomy by the age of 40) to reduce their risk of endometrial and ovarian cancers (3). Screening for endometrial and ovarian cancer using endometrial biopsy, transvaginal ultrasound, and CA125 has not been proven to decrease the incidence or mortality from cancer in these high-risk women.
Screening for endometrial cancer (endometrial biopsy, transvaginal ultrasound, CA125) has never proven to be effective in asymptomatic women. There is a role for endometrial biopsy to gauge response to therapy for women who are using progestins as fertility-sparing or conservative management for endometrial hyperplasia/cancer. Women on tamoxifen have an increased risk of endometrial cancer. However, screening asymptomatic women on tamoxifen is not recommended. Only those who develop abnormal uterine bleeding need endometrial sampling.
Patients with endometrioid ovarian cancers and adult granulosa cell tumours of ovary have an increased risk for endometrial cancer, and hysterectomy is recommended routinely as part of their management. If fertility preservation is important, this should be discussed with the gynecologic oncologists at BC Cancer.
It must be emphasized that Pap smear is not a satisfactory screening procedure for endometrial cancer (it is only a screening test for cervical cancer). All cases of post-menopausal, intermenstrual bleeding or menorrhagia require investigation, which includes a pelvic examination and endometrial tissue sampling.
NB: A normal endometrial thickness on ultrasound does not rule out endometrial cancer.
All patients with potential clinical involvement of the cervix should be referred for assessment regarding possible surgical staging. Patients who are deemed to have inoperable disease by the gynecologic oncology team will be referred for neoadjuvant chemotherapy (see
section 6.B below).
We recommend all patients referred to BC Cancer have a pathology review.
The rational for this referral approach in endometrial carcinoma (EC) includes:
NB All components of this pathology criteria for referral (grade, histotype, MMR IHC, p53 IHC, and ER IHC) can and should be done in your local centre pathology laboratory and are considered standard of care in BC.
Clinical Stage II (bulky cervical involvement) that is not amendable to surgical resection
Clinical State IIIC (with bulky and unresectable nodes)
Clinical Stage IV (widespread metastatic disease on preoperative imaging, or bladder/bowel involvement)
We recommend all patients with advanced disease be presented at a multidisciplinary tumour conference to review optimal management strategies. Many patients will receive neoadjuvant chemotherapy, followed by repeat imaging and assessment by a gynecologic oncologist after the 3rd cycle of chemotherapy to determine response to treatment and suitability for surgery. If there has been a good radiological response to treatment, surgery will usually be considered after the 3rd cycle of chemotherapy. This will be followed by post-operative chemotherapy usually to a total of 6 cycles, with or without radiotherapy, to be determined by the multidisciplinary team (see sections on adjuvant therapies).
The prognostic value of molecular classification has been demonstrated repeatedly; across unselected population-based series, clinical trials, and even narrowly defined age- or histotype-stratified subsets. Specifically, patients with
POLEmut EC have highly favorable outcomes with almost no recurrence events, in contrast to p53abn EC where >50% of patients recur and die from their disease. MMRd and NSMP have intermediate outcomes and need further prognostic refinement. Recent data also supports the predictive implications of molecular subtype assignment. Molecular sub-group analysis of clinical trial data (PORTEC-3 trial (1)) and a large retrospective series (cohort of 2500 ECs (2)) have both demonstrated improved survival outcomes in patients with p53abn ECs with adjuvant platinum-based chemotherapy when used with radiation compared to radiation alone in patients with European Society of Medical Oncology (ESMO) defined high risk EC. In contrast, there was no apparent benefit of chemotherapy observed in these series for patients with high risk MMRd ECs, compared to radiation alone. Individual patient data meta-analysis of all published and available international reports on
POLEmut ECs showed that patients with confirmed pathogenic
POLE mutations have almost no recurrence or disease specific death events, even when their tumours appear to have unfavourable clinicopathological or molecular characteristics, and adjuvant treatment was not associated with outcomes (3). There are a growing number of molecular subtype-specific clinical trials and ALL EC patients should be assessed for trial eligibility. For patients not eligible for clinical trials, it is recommended to follow this molecular directed adjuvant therapy algorithm.
This algorithm should be applied to all surgically resected ECs of any histotype, including non-endometrioid carcinomas.
Carboplatin and paclitaxel 3 weekly schedule (GOENDCAT) is the recommended protocol) for 4- 6 cycles. If a patient has intolerance to GOENDCAT, consideration should be given to another taxane-platinum regimen. Carboplatin/liposomal doxorubicin (GOOVPLDC) can be considered if there is a contraindication to all taxanes.
Table 1. Adjuvant therapy guidelines for NSMP endometrioid carcinomas:
NB There is no level 1 evidence showing benefit from adjuvant chemotherapy used in addition to radiation in stage I-II NSMP ECs. We recommended referral for a discussion regarding adjuvant chemotherapy when indicated in table 1.
* For Stage IA or IB recommend pelvic radiotherapy if there is
extensive/substantial LVSI (≥3 foci)
**For Stage II, chemotherapy should be considered if there is at least one additional high-risk uterine factor (deep myometrial invasion, or grade 3 tumour, or extensive LVSI).
*** Hormonal therapy, can be considered as first-line therapy in carefully selected stage IVB grade 1-2 tumours if ER positive (see section 9. Advanced Endometrial Cancer).
Table 2. Adjuvant therapy guidelines for NSMP non-endometrioid carcinomas:
NB MMR and p53 immunohistochemistry are available in any pathology laboratory and is now considered standard of care for all ECs in BC.
POLE testing has been approved in BC since May 2022. Testing in the current funding scheme is being paid for, for patients in whom information on
POLE status would impact adjuvant treatment decisions (see
POLE testing section below).
Mismatch repair testing
Mismatch repair is a DNA repair pathway. Loss of function of mismatch repair enzymes can arise secondary to inherited susceptibilities (e.g Lynch Syndrome), can arise sporadically through somatic mutation or can be epigenetic where a key protein associated with mismatch repair (e.g. MLH1) is not present because of selective hypermethylation of the genome in a way that turns off translation of the gene. Mismatch repair is usually measured in BC by IHC testing for two proteins: PMS2 and MSH6. Testing is performed on tumour specimens—biopsy or hysterectomy (not needed on both) and loss of one or more of these proteins (with a positive internal control) is termed mismatch repair deficient. You may see comments on pathology reports that status of other mismatch repair proteins cannot be determined. This is because PMS2- MLH1 and MSH2- MSH6 form heterodimers and loss of one can cause weak or absent expression of the other. All you need to know as a clinician is 1) loss of expression of one or more MMR proteins means the case has been categorized as MMRd (sometimes also referred to as MMR loss or MMR abnormal), 2) HCP referral should be initiated if the carcinoma is MMRd UNLESS hypermethylation of MLH1 is identified. Note: methylation testing can take 4-8 weeks so referral to hereditary cancer program testing in the interim is recommended and this should be cancelled if hypermethylation is identified.
Mismatch repair deficiency may be identified by 4 protein IHC (loss of one or more defined mismatch repair deficient) or by DNA-based microsatellite instability (MSI) assay, but this is less common in BC.
p53 is a protein also commonly assessed by IHC in multiple gynecologic cancers, and is a key part of molecular classification of endometrial cancers. Criteria for categorization of p53 abnormalities by IHC differ slightly between tumour sites but in general, abnormal p53 expression (p53abn) encompasses three scenarios; i) complete loss of protein or ii) overexpression or iii) subclonal/zonal expression (distinct and regional abnormal staining). Normal p53 expression levels as measured by IHC (or
TP53 mutation status on a next generation sequencing panel), is termed 'wild type' or p53wt.
POLE is a gene also involved in DNA repair and when specific pathogenic mutations are identified in the
POLE gene this identifies a subset of EC patients at very low risk of recurrence and death from their disease. These patients may be candidates for de-escalated adjuvant therapy in clinical trials or as recommended by their physicians according to BC and international guidelines.
POLE testing has been approved in BC since May 2022. Testing in the current funding scheme is being paid for but ordered after hysterectomy for those patients in whom information on
POLE status would impact adjuvant treatment decisions. It may be ordered reflexively by the pathologist or, more commonly, is ordered by the treating oncologist when needed (see table below). Click
here for POLE requisition.
Correct interpretation of molecular results
Clarification of pathogenicity of
POLE mutation is imperative, with pathogenic list of mutations considered 'actionable' for
POLEmut assignment currently limited to 11 well characterized missense mutations (P286R, V411L, S297F, S459F, A456P, F367S, L424I, M295R, P436R, M444K, D368Y) (1).
Multiple-Classifier Positive Cases
How to manage the approximately 3-5% of ECs with more than one molecular feature (called multiple-classifiers) is also crucial.
TP53 mutations can occur as secondary events in ultramutated
POLEmut ECs and/or hypermutated MMRd ECs, with molecular features and clinical behaviour of
POLEmut and MMRd ECs, respectively (2). These findings suggest that these secondary
TP53 mutations are a later event acquired during tumour progression that do not affect the clinical outcome. These tumours should be managed according to the WHO and ProMisE order of segregation, with any EC harbouring a pathogenic
POLE mutation considered
POLEmut (even if p53abn and/or MMR abnormal IHC) and next any EC with loss of MMR proteins or MSI considered MMRd even if p53abn/TP53mut detected (3).
Treatment of recurrent endometrial cancer will depend on sites of recurrence and primary treatment.
In the setting of isolated disease in the absence of previous adjuvant therapy, cure may be possible with radiotherapy or surgery for isolated vaginal vault or pelvic disease. These patients should be referred to BC Cancer. Radiotherapy is generally reserved for those with localized recurrences, while chemotherapy is offered to those with distant disease. Pelvic exenteration may be considered for those who recur within a radiated field, when the recurrence is localized to the central pelvis and there is no evidence of distant disease based on PET/CT.
In the upfront treatment setting, combination chemotherapy with carboplatin and a taxane is the current standard of care. Treatment is usually for 6 cycles (GOENDCAT or GOENDCAD).
Second-Line Therapy - MMRd
For patients with MMRd endometrial cancer that has progressed after first-line chemotherapy, treatment with immune checkpoint inhibitors has been demonstrated to have a high response rate and long durations of treatment benefit. Access to immune checkpoint inhibitors is possible through patient access programs, and this can be determined by contacting the regional Drug Access Navigator (1-2).
Second-Line Therapy - MMRp
For patients with MMRp endometrial cancer that has progressed after first-line chemotherapy, a phase 3 clinical trial has demonstrated that the use of lenvatinib and pembrolizumab improves progression-free survival and overall survival (3). Access to this combination is currently possible through a patient access program. Contact your regional Drug Access Navigator for additional information.
Other treatment options
Grade 1 (and possibly grade 2) histology or significant comorbidities precluding chemotherapy- consider hormonal therapy if tumour is ER or PR positive (GOENDAI or GOENDH). Early initiation of palliative care/supportive care is recommended with or without chemotherapy. Radiation therapy may be considered for symptomatic disease (e.g. pain, bleeding).
1. Oaknin A, et al. Safety and antitumour activity of dostarlimab in patients with advanced or recurrent DNA mismatch repair deficient/microsatellite instability-high (dMMR/MSI-H) or proficient/stable (MMRp/MSS) endometrial cancer: interim results from GARNET-a phase I, single-arm study. J Immunother Cancer. 2022 Jan;10(1):e003777. doi: 10.1136/jitc-2021-003777. PMID: 35064011; PMCID: PMC8785197.
2. O'Malley DM, et al. Pembrolizumab in Patients With Microsatellite Instability-High Advanced Endometrial Cancer: Results From the KEYNOTE-158 Study. J Clin Oncol. 2022 Mar 1;40(7):752-761. doi: 10.1200/JCO.21.01874. Epub 2022 Jan 6. PMID: 34990208; PMCID: PMC8887941.
3. Makker V et al. Lenvatinib plus Pembrolizumab for Advanced Endometrial Cancer. N Engl J Med. 2022 Feb 3;386(5):437-448. doi: 10.1056/NEJMoa2108330. Epub 2022 Jan 19. PMID: 35045221.
Endometrial tumours that develop on estrogen replacement therapy tend to be well differentiated and minimally invasive, and therefore these women generally have a good prognosis. However, the known association between estrogen and EC causes concern when women with a history of EC request hormone replacement. Many of these patients are asymptomatic. There is a small proportion of patients (usually premenopausal) who suffer from extreme vasomotor symptoms. Unfortunately, there are limited non-hormonal alternatives to estrogen, such as clonidine or venlafaxine. For premenopausal women, thorough discussion of the risks and benefits of estrogen replacement is essential. The absence of hormone replacement therapy in young women who have surgery including bilateral salpingo-oophorectomy before age 50 is known to be associated with an increased risk of death secondary to coronary heart disease, osteoporosis, thromboembolic events, and cancers such as lung and colorectal cancer (1). There is no evidence that estrogen replacement therapy use alters the prognosis in women who have had treatment for endometrial cancer (2).
In all situations outlined below, history and physical exam, including pelvirectal examination, are recommended. Patients do not need routine bloodwork, pap smear, or imaging, unless indicated by symptoms or signs on examination 1,2.
These patients are at low risk of recurrence (<5%), which is most likely to occur within the first 2 years after primary treatment. The most likely site of recurrence is the vaginal vault, therefore these patients need to be counselled about vaginal bleeding. Their follow-up care can be provided by their referring physician.
Years 1 through 5: every 6 months
Year 5+: annually
These patients are still at low risk of recurrence (5-10%), which is most likely to occur within the first 2-3 years after primary treatment. They can recur in the pelvis/vault, but some will recur distantly. These patients need to be counselled about vaginal bleeding, pelvic pain, bloating, increased abdominal girth. Their follow-up care can be provided by their referring physician.
Years 1 2: every 3-6 months
Years 3-5: every 6 months
These patients are at high risk of recurrence, which may be distant or locoregional. Their recurrences are also most likely to occur within the first 2-3 years after primary treatment. These patients will typically be seen at least once following completion of radiotherapy by the radiation oncologist at BC Cancer to review/manage side-effects of treatment. Further follow-up care can then be provided by the referring physician.
Years 1-2: every 3-6 months
Year 5: annually
These patients should be followed more frequently as they have the potential for salvage treatment. Their follow-up care can be provided by their referring physician.
Years 1 and 2: every 3 - 4 months
Years 3 - 5: every 6 months
Year 5+: annually
BC Cancer. All Rights Reserved.