Endometrial carcinoma: Staging and surgical treatment

INTRODUCTION — Endometrial carcinoma (EC) is the most common gynecologic malignancy in the United States and is the fourth most common cancer in females. The standard cost effective, value-based approach to endometrial cancer is to (1) make the diagnosis based on histology on biopsy or dilation and curettage, (2) clinically assess the probability of metastatic disease based on histologic type and grade, (3) evaluate for metastatic disease if histologic type and grade justify evaluation (usually with computed tomography [CT] and cancer antigen 125 [CA 125]), and (4) perform hysterectomy, with lymph node assessment in patients at increased risk of metastasis.

Staging and surgical treatment of EC will be reviewed here. Related topics are discussed in detail separately, including:

●Overview (see "Overview of resectable endometrial carcinoma")

●Histopathology and pathogenesis (see "Endometrial cancer: Pathology and classification")

●Epidemiology and risk factors (see "Endometrial carcinoma: Epidemiology, risk factors, and prevention")

●Clinical features and diagnosis, and screening for high-risk patients (see "Endometrial carcinoma: Clinical features, diagnosis, prognosis, and screening")

●Adjuvant therapy (see "Overview of resectable endometrial carcinoma", section on 'Role of adjuvant therapy')

This is discussed separately. (See "Endometrial cancer: Pathology and classification" and "Overview of resectable endometrial carcinoma", section on 'Histopathology'.)

PREOPERATIVE EVALUATION — Prior to treatment, the patient should have a complete evaluation, including history, physical examination, and endometrial sampling, to establish the diagnosis of EC. Treatment planning is guided by the patient's ability to tolerate surgery and by the cancer histology and grade. Additional testing may include imaging or tumor marker studies to assess for metastatic disease if suspected.

History and physical examination

●A medical and surgical history is taken to evaluate for comorbidities that may impact surgical or adjuvant therapy planning.

●A thorough family cancer history should be taken as a means to screen for hereditary susceptibility to cancers (table 1). (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Cancer screening and management", section on 'Endometrial and ovarian cancer'.)

●If uterine size cannot be assessed with pelvic and abdominal examination (and if pelvic imaging was not done prior to evaluation), pelvic ultrasonography is usually the preferred modality to determine whether the patient is a candidate for minimally invasive hysterectomy.

Imaging studies — A chest radiograph should be performed as part of the initial assessment to exclude lung metastases.

Abdominal and pelvic imaging is rarely performed in patients with nonaggressive ECs but is often used to exclude metastasis from aggressive ECs. The goal of preoperative axial imaging in patients with aggressive cancers is to detect peritoneal disease and anticipate the need for laparotomy and cytoreduction rather than a minimally invasive approach. In such patients, additional chest imaging with CT is often performed to exclude metastatic disease even when chest radiography is negative.

Infrequently, imaging is performed for the purpose of clinical staging (although standard EC staging is surgical). This occurs when surgery is not planned because the patient cannot tolerate surgery or fertility preservation is desired. For clinical staging, contrast-enhanced magnetic resonance imaging (MRI) appears to be the best radiographic modality for detecting myometrial invasion, cervical involvement, or lymph node metastases when compared with nonenhanced MRI, ultrasound, or CT [1-6]. (See 'Fertility preservation' below and 'Inoperable patients' below.)

Tumor markers — The role of tumor markers is unclear, unvalidated, and of low value. The author does not routinely perform tumor marker testing prior to surgery for EC. However, serum cancer antigen 125 (CA 125) can be useful for surveillance of patients after initial treatment if the level was initially elevated. The author and section editor routinely order CA 125 in patients with type II endometrial cancer and in those with grade 3 endometrial cancers as there is evidence that there is an association with metastatic disease. If the histologic diagnosis from the biopsy changes after hysterectomy, the CA 125 can be drawn postoperatively when the final pathology is known.

STAGING SYSTEM — EC is surgically staged; the 2023 International Federation of Gynecology and Obstetrics (FIGO) staging system is shown in the table (table 2 and table 3) [7]. This topic will utilize the 2023 FIGO staging system except when describing studies in which older staging systems (eg, 2009 FIGO/2017 Tumor, Node, Metastasis [TNM] classification system (table 4)) are used.

The staging procedure includes total hysterectomy, bilateral salpingo-oophorectomy, and lymphadenectomy. Pelvic washings are not required for surgical staging; however, the presence of cancer cells in the peritoneal cavity is a poor prognostic factor [8]. (See "Overview of resectable endometrial carcinoma", section on 'Other factors'.)

APPARENTLY UTERINE-CONFINED DISEASE — Surgery with hysterectomy and salpingo-oophorectomy is the mainstay of management of EC.

The approach to surgery for EC has evolved significantly over time. Traditionally, patients underwent exploratory laparotomy through a midline incision. Today, the majority of patients with apparently early-stage disease undergo minimally invasive surgery (MIS). Additionally, changes in the approach to retroperitoneal lymph node assessment have varied across time and institutions.

We recommend MIS (laparoscopy or robotic surgery) for patients with disease apparently confined to the uterus (based on physical examination, with or without pelvic imaging) who are surgical candidates and in whom the procedure is expected to be able to be completed without conversion to laparotomy. This is based on consistent data from randomized trials that laparoscopic hysterectomy compared with open hysterectomy results in lower morbidity and in comparable oncologic outcomes [9-19].

Ability to tolerate surgery is determined based upon preoperative medical evaluation of the patient's current condition and comorbidities. The likelihood of a successful minimally invasive procedure is based on several factors, including the patient's surgical history and the size of the uterus. (See "Hysterectomy (benign indications): Selection of surgical route", section on 'Route of hysterectomy for benign disease'.)

Choice of surgical approach — The goal of surgical management of EC is to remove the primary tumor and to identify prognostic factors to determine whether adjuvant therapy is required. This is done using the approach with the least possible surgical morbidity that does not compromise long-term survival, which in most cases is MIS (conventional laparoscopy or robotic surgery) [20,21]. Conventional total laparoscopic hysterectomy (TLH) with bilateral salpingo-oophorectomy (BSO) is an option for most patients, with use of robotic surgery if this makes MIS possible in patients who have a high risk of conversion to laparotomy (eg, patients with obesity) [22].

The type of surgical approach is planned based on the preoperative impression of stage, which is based on the pretreatment evaluation. (See 'Preoperative evaluation' above.)

Laparoscopy or robotic surgery — Randomized trials have demonstrated that minimally invasive approaches to EC staging, laparoscopic-assisted vaginal hysterectomy, or TLH, result in lower rates of peri- and postoperative complications than laparotomy, without negatively impacting oncologic outcomes:

●LAP2 – The Gynecologic Oncology Group LAP2 randomized trial assigned 1696 patients with clinical stage I to IIA uterine cancer to laparoscopy and 920 to laparotomy [18]. Patients in the laparoscopy group had similar rates of intraoperative complications and longer operative duration (median 204 versus 130 minutes), but fewer moderate to severe postoperative adverse events (14 versus 21 percent). Hospital stay of longer than two days was significantly lower in the laparoscopy group (52 versus 94 percent). Fewer patients in the laparoscopy compared with laparotomy group underwent pelvic and paraaortic lymphadenectomy (92 versus 96 percent), but among those who did there was no difference between groups in lymph node counts. There was no difference between groups in the rate of detection of advanced-stage disease (stage IIIA, IIIC, or IVB; 17 and 17 percent).

One concern regarding the LAP2 study was that 25.8 percent of patients assigned to laparoscopy were converted to laparotomy (only 4 percent were converted because of advanced disease). The conversion rate was correlated with body mass index (BMI), with 57 percent of patients with BMI ≥40 kg/m² converted to laparotomy. While the rate of conversion to laparotomy was high, it is important to note that, at the time of the LAP2 trial, many surgeons were relatively new to MIS for EC staging. Additional data suggest that patients with obesity who undergo MIS are not at any higher risk for disease-specific mortality compared with patients without obesity and EC (though patients with obesity and EC have a higher all-cause mortality rate).

Since the publication of LAP2, reported rates of conversion to laparotomy for planned laparoscopic staging procedures have declined substantially [9,10]. This is likely partly due to the lower rate of conversion with robotic surgery and partly to the use of selective complete or routine sentinel lymphadenectomy compared with protocol-directed lymphadenectomy in LAP2. In our practice, BMI does not influence a decision to proceed with MIS for EC.

Follow-up data from LAP2 showed similar survival for patients who underwent either laparoscopy or laparotomy (five-year overall survival: 89.8 percent in both groups) [11]. Estimated five-year recurrence rates were also similar (13.7 versus 11.6 percent). Quality of life (QOL) assessments were superior in the laparoscopy group for many measures, including Functional Assessment of Cancer Therapy-General (FACT-G) scores on physical functioning and body image, less pain and interference with QOL, and an earlier resumption of normal activities and return to work over the six-week recovery period [12].

●LACE – The Laparoscopic Approach to Cancer of the Endometrium (LACE) trial was a large, multinational randomized trial including 760 patients with stage I endometrioid EC who underwent staging with TLH or total abdominal hysterectomy (TAH) and found no difference in disease-free survival at 4.5 years (81.6 versus 81.3 percent) or overall survival (mortality: 7.4 versus 6.8 percent) [13]. This was the second largest trial to evaluate this question [14].

A subset of 361 participants were enrolled in a QOL study [15]. Patients in the TLH group reported significantly greater QOL compared with those who had TAH in the initial postoperative period. For some patients, the improved QOL persisted up to six months after surgery and continued to favor TLH. In a subsequent study, patients in the LACE trial were asked to complete a self-administered questionnaire (FACT-G); physical and functional well-being subscales favored patients in the TLH compared with TAH group [23].

A systematic review of 3644 patients with EC included data from eight randomized trials [14]. There was no difference in the risk of death or recurrence between patients who underwent laparoscopy compared with laparotomy (overall survival: hazard ratio [HR] 1.14, 95% CI 0.62-2.1; recurrence-free survival: HR 1.13, 95% CI 0.9-1.42). In two separate sub-meta-analyses (incorporating trials that met specific criteria), there was less blood loss with laparoscopy and fewer severe postoperative adverse events compared with laparotomy.

●Robotic surgery – There is evidence that robotic surgery is associated with similar short-term outcomes, shorter operating room time, and lower conversion rates to laparotomy than conventional laparoscopic surgery. However, in expert hands, we do not find conventional laparoscopic surgery to be inferior to robotic surgery. Given the higher costs of robotics, it remains unclear which cohort of patients derives the most benefit from a clinical and cost-effectiveness perspective [17]. For example, in patients with obesity, robotic hysterectomy may reduce conversions because of positional intolerance [22].

The majority of studies on robotic surgery for EC are single-institution or multi-institution cohorts that report that robotic surgical staging is feasible with good surgical and oncologic outcomes; however, robotic surgery is more costly:

•A randomized trial (n = 99) of robotic versus laparoscopic staging for EC found shorter operative duration for robotic group (139 minutes [range 86 to 197] versus 170 minutes [range 126 to 259]) [10]. There were no conversions to laparotomy in the robotic group compared with 5 of 49 cases in the laparoscopy group. There were no differences in the number of lymph nodes removed, bleeding, or the length of postoperative hospital stay. The rate of major postoperative complications showed no statistically significant difference in robotic surgery versus laparoscopy (10 versus 22 percent).

•A database study including 1027 patients with EC who underwent laparoscopic hysterectomy and 1437 who underwent robotic hysterectomy also found no difference in rate of complications with robotic compared with laparoscopic surgery [17].

•A population-based study in Washington state showed robotic surgery was associated with fewer early readmissions compared with laparoscopic surgery and resulted in an increased proportion of cases performed by MIS compared with the clinical experience prior to widespread adoption of robotic surgery [24].

•A single-institution retrospective review of patients who underwent laparoscopic hysterectomy (n = 461) or robotic hysterectomy (n = 566) and had at least five years of potential follow-up showed poorer disease-specific (HR 3.51, 95% CI 2.19-5.63) and overall (HR 1.39, 95% CI 1.06-1.83) survival for patients managed robotically [25].

In terms of future directions, investigation of single-site laparoscopic or robotic hysterectomy with lymph node evaluation have been gaining interest in an effort to continue to reduce the potential morbidity of EC surgery [26].

Laparotomy — For patients who are surgical candidates with disease apparently confined to the uterus but in whom MIS is not feasible (eg, due to excessive uterine size, inability to tolerate steep Trendelenburg position, or known adhesive disease), laparotomy through a midline incision is the appropriate surgical approach. In rare circumstances, hysterectomy for apparently early-stage EC through a transverse (Pfannenstiel, Maylard, or Cherney) incision may be considered [27].

Vaginal hysterectomy — In special circumstances (eg, patients who require regional anesthesia and in whom assessment of the peritoneal cavity and retroperitoneum is not necessary), vaginal hysterectomy may be considered for EC staging [28]. In these patients, retrospective series have reported that vaginal hysterectomy is associated with a low rate of perioperative complications and 5- to 10-year disease-specific survival rates of 80 percent or higher [29,30]. Additionally, the cost of vaginal hysterectomy (with laparoscopic lymphadenectomy as needed) was shown to be lower than with robotic hysterectomy (with robotic lymphadenectomy as needed) [31].

Hysterectomy and oophorectomy procedure — Standard primary surgical management of patients with EC apparently confined to the uterus is total hysterectomy (TH) and BSO. Techniques for hysterectomy and BSO are discussed in detail separately. The aspects relevant to EC staging are included below. (See "Hysterectomy: Abdominal (open) route" and "Hysterectomy: Laparoscopic" and "Oophorectomy and ovarian cystectomy".)

The basic components of the procedure are:

●The peritoneal cavity is entered via MIS or laparotomy, and the abdomen is explored to exclude peritoneal metastasis. In the presence of clinical evidence of metastasis, determination of whether cytoreduction is feasible is imperative. In the absence of metastasis, sampling of peritoneal cytology may be considered. While it has been eliminated from International Federation of Gynecology and Obstetrics staging of EC since 2009, some have suggested that peritoneal metastasis remains a prognostic factor for disease outcomes [32], while others have not recognized this association [33]. In our practice, routine collection of peritoneal cytology has been discontinued since 2009 given that the information does not influence decision making for adjuvant therapy. In rare circumstances (when cytology is required for clinical trial eligibility), it is collected upon entry into the peritoneal cavity. (See "Initial treatment of metastatic endometrial cancer", section on 'Surgical cytoreduction'.)

●Considerations during MIS hysterectomy include whether to use a uterine manipulator to facilitate exposure and division of the ectocervix from the vagina. The impact of uterine manipulator use on oncologic outcomes in patients with early-stage disease is uncertain [34-36]. In addition, if a uterus with endometrial neoplasia cannot be removed intact through the vagina, a power morcellator should not be used, since this may disseminate neoplastic cells. In such cases, use of a small laparotomy incision or placing the uterus in a bag with controlled vaginal morcellation are options [37]. (See "Uterine tissue extraction by morcellation: Techniques and clinical issues", section on 'Morcellation devices and techniques'.)

●In general, in the absence of clinical evidence of omental metastasis, omental biopsy is not performed given that the rate of microscopic disease is small [38]. Exceptions may be in the presence of nonendometrioid histologies, where the rate of microscopic omental metastasis may be as high as 5 percent [39].

●If patients will undergo lymphatic mapping and sentinel lymph node dissection (SLND), injection is performed prior to hysterectomy. The timing of the injection is variable based on surgeon preference and the goal of allowing at least 15 minutes of time between injection and mapping (to allow for adequate time to reach the sentinel node[s]) but less than one hour from injection to completion of SLND (to prevent secondary lymph nodes from mapping).

●TH and BSO are then performed.

●Retroperitoneal lymphadenectomy is performed if indicated. If selective lymphadenectomy will be performed based on intraoperative assessment of uterine factors, the uterus is bivalved from the ectocervix, through the lateral uterus bilaterally and to the fundus (picture 1). Clinical or pathologic assessment of the tumor size and depth of invasion is then performed based on surgeon and institutional practice. If an intraoperative decision as to whether a lymph node dissection is to be performed, then either gross or pathologic evaluation of the endometrium (for depth of invasion, tumor diameter, and tumor grade) is performed at this time. (See 'Lymph node evaluation' below.)

●Pathology evaluation should include testing of the specimen for microsatellite instability in all patients with EC to detect those who need further testing for Lynch syndrome, since these patients are also at increased risk of ovarian, colon, and other cancers (table 5) [40]. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis", section on 'Tumor MSI/IHC testing'.)

Surgical management of cervical involvement — For patients with clinically apparent extension of EC to the cervix, management may include (1) radical hysterectomy with postoperative radiation therapy based on pathologic factors, (2) extrafascial hysterectomy with postoperative radiation, or (3) primary radiation therapy followed by extrafascial hysterectomy [41].

Retrospective studies have generally found that radical hysterectomy does not significantly improve survival in stage 2 EC [42,43]. Given the absence of a prospective comparison of these treatment approaches, the management of patients with disease extension to the cervix must be individualized based on patient, pathologic, and provider factors. For patients with a normal clinical examination (with neither expansion of the cervix into a "barrel shape" nor a visible lesion on the ectocervix) but with endocervical curettage (ECC) positive for EC, extrafascial hysterectomy followed by volume-directed radiation (based on final pathologic features) is generally used. This is based on the finding that less than 50 percent of patients with clinical stage II EC because of a positive ECC alone have pathologic extension of EC to the cervical stroma (pathologic stage II) [44].

LYMPH NODE EVALUATION — The approach to lymph node evaluation in patients with EC is a subject of debate. Practice varies across different institutions or surgeons. Data informing this choice are limited, since all options have not been directly compared in well-designed studies and the available randomized trials are regarded as high quality by some experts and as limited by design flaws by others.

In general, the options for management of retroperitoneal lymph nodes (in the absence of grossly metastatic disease) include:

●No lymph node dissection (LND)

●Systematic LND only if the risk of lymph node metastasis exceeds a certain threshold

●Routine sentinel LND following lymphatic mapping

●Systematic LND in all patients

Since the decision to recommend adjuvant therapy after surgery for EC is strongly influenced by surgical stage (and thus the histologic status of the retroperitoneal lymph nodes), different lymph node management strategies will be associated with different rates (and modalities) of the use of adjuvant therapy [45].

Significant controversy exists regarding the management of retroperitoneal lymph nodes in EC. From the initial cohort studies that demonstrated the rate of lymph node metastasis is increased with increasing tumor grade and depth of myometrial invasion [46], the dissection of lymph nodes has been used for staging, prognostic information, and to determine the need for adjuvant therapy.

The surgical management of lymph nodes in EC has changed dramatically. Prior to the mid-1980s, a systematic retroperitoneal (pelvic and paraaortic) LND was not performed. Then, in the 1980s, the factors that increase the risk of lymph node metastasis became better defined. In 1987, Gynecologic Oncology Group 33 study described the increased risk for pelvic lymph node metastasis in ECs that were more deeply invasive into the myometrium and higher grade [46]. At that time, complete LND was considered to be a reasonable approach to the management, with suggestion of both a therapeutic and prognostic benefit to removal of the pelvic and aortic lymph nodes. A disadvantage of this approach was the added the morbidity of lower extremity lymphedema.

Since the majority of patients with EC are without metastasis, investigators then set out to determine whether there was a therapeutic value of lymphadenectomy. The paradigm of routine complete pelvic and paraaortic LND was challenged with the publication of two randomized trials. These trials demonstrated that systemic LND did not offer a survival advantage in patients with EC [47,48]:

●One trial described 514 patients with clinical stage I EC (with frozen-section documentation of myometrial invasion) assigned to systematic pelvic LND versus removal only if enlarged lymph nodes were present [47]. A median of 30 lymph nodes were removed and adjuvant therapy was provided at the discretion of the treating clinician, but fewer patients received adjuvant therapy in the LND group than the control group (17 versus 25 percent). After 49 months median follow-up, systematic pelvic LND compared with removal only for enlarged nodes had comparable multivariable hazard ratios (HRs) for relapse (12.9 versus 13.2 percent, HR 1.2, 95% CI 0.75-1.91) and death (overall survival 86 versus 90 percent, HR for death 1.16, 95% CI 0.67-2.02).

●The A Study in the Treatment of Endometrial Cancer (ASTEC) trial included 1408 patients with clinical stage I or II EC assigned to systematic pelvic LND (with removal of suspicious aortic lymph nodes at the surgeon's discretion) versus removal of lymph nodes only if suspicious and at the discretion of the surgeon. A median of 12 lymph nodes were removed, and a second randomization to radiation versus observation was performed in a subset of patients (intermediate- and high-risk early-stage patients) regardless of the status of the lymph nodes. After 37 months of median follow-up, removal of suspicious lymph nodes compared with systematic LND showed no difference in overall survival (81 versus 80 percent, HR 1.04, 95% CI 0.74-1.75) and recurrence-free survival (79 versus 73 percent, HR 1.25, 95% CI 0.93-1.66). Likewise, a stratified subgroup analysis (low, intermediate, high risk, and advanced) showed no difference in overall or recurrence-free survival.

Challenges to the design of these studies and the quality control of the surgery has led many surgeons to discount these data and wait for clinical trials designed to address these limitations. However, well-designed studies to clarify the findings of the previous trials have not been done and it is unlikely that they will in the future. Thus, there will likely remain gaps in the definitive data regarding the role of lymphadenectomy in EC.

In addition to the lack of evidence of a therapeutic benefit of LND, less invasive techniques have become standard management of EC. The technique for lymphatic mapping and sentinel lymph node dissection (SLND) in EC has been refined. Prospective studies have demonstrated the feasibility of this technique and found a low false-negative rate for detection of positive lymph nodes [49], leading many surgeons to use SLND rather than complete lymphadenectomy. This is also based on the perception of lower rates of complications, such as lymphedema. SLND has been the subject of enthusiasm from many gynecologic oncologists and is supported by organizations such as the National Comprehensive Cancer Network (NCCN) and the Society of Gynecologic Oncology (SGO) [50].

There are limited data about use of SLND in place of LND. There are no randomized trials comparing SLND with other strategies (complete LND or no LND) of managing the lymph nodes in EC. No studies have been published that address the comparative morbidity of SLND with alternative strategies (which is important since low-volume disease discovered in an SLND may be treated, whereas conventional LND may not have detected the disease at all and thus not received adjuvant therapy).

Despite the lack of high-quality evidence, for many surgeons, sentinel lymphadenectomy following lymphatic mapping has replaced complete lymphadenectomy or no lymphadenectomy in patients with apparently uterine-confined EC. In the absence of mapping of a hemipelvis, uterine factors are used to assess the risk of retroperitoneal lymph node metastasis and in patients at increased risk. Experts advise complete lymphadenectomy on the unmapped side of the pelvis following hysterectomy if the risk of lymph node metastasis exceeds a certain (individual) threshold based on the Mayo criteria [51]. (See 'Selective lymphadenectomy' below.)

There is no consensus regarding the best approach to lymph node evaluation in patients with EC that is apparently confined to the uterus. In current practice, factors that are used to determine the management of an individual patient may include estimates of:

●The a priori probability of lymph node metastasis

●The feasibility of surgically accessing the lymph nodes

●The risk for and acceptance of the risk of lymphedema

●The acceptable rate of the use of adjuvant therapy (especially radiation therapy)

Patients should be counseled about the options and engage with the surgeon in shared decision making regarding the approach to lymphadenectomy. In practice, some patients have read about lymphedema and want to avoid it at all cost (thus, no LND), while others are adverse to adjuvant therapy (thus, LND will diminish the rate of adjuvant therapy because we will have information to inform decision making rather than guessing).

Complete pelvic lymphadenectomy — Routine complete bilateral pelvic and aortic LND provides the most information about the extent of malignancy for purposes of planning adjuvant therapy, but the primary concern regarding its universal use is the morbidity of lower extremity lymphedema and associated cellulitis.

The reported risk of lymphedema varies widely (5 to 38 percent) [52,53]. Results from a prospective evaluation in patients with EC demonstrated that the rate of lymphedema (defined as a change in lower leg volume of >10 percent) was 34 percent, a risk that peaked at four to six weeks and extended through the evaluation at 24 months [54]. Factors associated with lymphedema included advancing age and having more than eight lymph nodes removed. In another series of patients with EC, the risk of lower extremity lymphedema attributable to LND was 23 percent using a validated survey [55]. In addition, lymphedema was associated with clinically significant reductions in nearly every quality of life domain tested. Risk of lymphedema appears to increase with obesity and the use of radiation therapy. Management of lymphedema and cellulitis following pelvic LND is discussed in detail separately. (See "Clinical staging and conservative management of peripheral lymphedema".)

For patients with histologic confirmation of lymph node metastasis, use of adjuvant chemotherapy (with or without volume-directed radiation) is generally preferred. However, when the lymph nodes are negative, adjuvant therapy is less commonly recommended for patients with high-intermediate risk disease than for similar patients in whom the lymph nodes were not dissected. (See "Adjuvant treatment of intermediate-risk endometrial cancer".)

Aortic lymph node dissection — The decision to dissect the aortic lymph nodes in EC is based on data suggesting that the risk for aortic metastasis increases in the presence of pelvic lymph node metastasis [56,57], that isolated aortic lymph node recurrences occur [57], and that survival is improved when aortic LND is included with pelvic LND [58]. Identification of patients at lower risk for aortic metastasis has led to many surgeons omitting aortic LND when the risk of metastasis is predicted to be lower than the procedure to remove the lymph nodes. In practice, when performing a lymphadenectomy, most surgeons will perform a pelvic dissection and a selective aortic dissection (based on the feasibility of completing this procedure safely, balancing this with the risk of aortic metastasis based on preoperative factors (ie, grade, histology, imaging findings) or uterine factors identified intraoperatively).

A prospective series reported that 77 percent of patients with paraaortic nodal metastases have disease above the inferior mesenteric artery [59]. Some experts extend the paraaortic LND superiorly to the level of the renal veins. To reach this level with minimally invasive surgery requires advanced skills and, possibly, special techniques. One potential solution is the use of extraperitoneal laparoscopic staging. In the extraperitoneal approach, small bowel is less likely to compromise visualization and successful removal of the paraaortic lymph nodes up to the level of the renal vein has been reported in over 90 percent of unselected patients with a body mass index up to 51 kg/m². Although infrarenal LND has not been proven to impact outcome, if it is performed, the extraperitoneal approach offers the most reliable method to consistently perform a thorough dissection. (See "Pelvic and paraaortic lymphadenectomy in gynecologic cancers", section on 'Extraperitoneal laparoscopic access'.)

Given the importance of lymph node involvement to staging and treatment decisions, when performed, lymph node assessment is best performed by experienced surgeons, such as gynecologic oncologists [47,53,60].

Selective lymphadenectomy — Use of selective LND has emerged as an alternative to complete pelvic and paraaortic LND for patients in whom LND could be safely omitted. Given that the majority of decisions regarding adjuvant therapy in patients in whom complete LND is performed are not impacted by the results from the LND [61], selecting the population likely to derive the greatest impact from the procedure is a means to balance the risks and potential benefits of LND.

With this approach, LND is performed only in patients with a higher risk for lymph node metastasis. Although no consensus exists as to the definition of "high-risk" patients, the factors that appear to be associated with an increased probability of retroperitoneal lymph node metastasis include:

●Aggressive neoplasms, including grade 3 endometrioid ECs, carcinosarcoma, serous, clear cell, mixed cell, undifferentiated, mesonephric-like, and gastrointestinal type mucinous carcinomas.

●Deeply invasive (invading one half or more of the myometrium) tumors on preoperative imaging or intraoperative assessment.

●Larger (>2 cm) tumors on preoperative imaging or intraoperative assessment.

Both preoperative endometrial sampling results and intraoperative gross inspection and/or frozen section are used to make decisions regarding performance or omission of LND after hysterectomy.

The above criteria are based on a study of 328 patients that is the source of the "Mayo criteria" that found that patients with EC with grade 1 to 2 endometrioid histology, tumor size of 2 cm or less, and 50 percent or less myometrial invasion (clinically or on frozen section) have a 5 percent or lower chance of nodal metastasis, and, thus, it is reasonable to omit LND in this population [51].

These factors have also been validated as predictive markers for lymph node metastasis in large prospective datasets [59] and population-based cohorts [62]. These studies have shown that in the absence of these factors, the risk of retroperitoneal lymph node metastasis is approximately 1 percent [63]. A nomogram that predicts the risk for lymph node metastasis based on these factors has also been developed [64].

Similarly, if a pelvic LND is planned, a concurrent aortic LND may be omitted in over three-quarters of patients (those with low-grade disease, with <50 percent myometrial invasion, and without clinical evidence of pelvic lymph node metastases) [65].

A limitation of selective LND is the reliance on intraoperative frozen section. Although diagnostic effectiveness of frozen section is high in many centers [66], others have found it to be unreliable compared with permanent section, particularly in determining tumor grade [67]. A prospective blinded evaluation of the accuracy of frozen section for the surgical management of EC found that frozen section for histologic grade and depth of myometrial invasion in EC correlates poorly with final pathology [67]. However, the use of grade and histology from preoperative endometrial biopsy together with tumor diameter assessed at the time of hysterectomy has been shown to be reliable, obviating the need for expert frozen section in low-resource settings [64].

Sentinel lymph node evaluation — SLND following lymphatic mapping has become a standard option for the management of the retroperitoneal lymph nodes in EC. In a hospital-based study including over 45,000 patients in the United States, SLN mapping increased from 1.8 percent in 2012 to 25.3 percent in 2018 [68]. Increasing utilization of SLND is not associated with an increase in cancer-specific mortality compared with complete LND [69].

Data supporting the use of SLND include prospective and retrospective studies showing increased detection of lymph node metastasis and a low false-negative rate compared with systematic LND [49,70,71], including in people with high-grade endometrial cancer. In a meta-analysis of nine prospective studies including 429 patients with high-grade endometrial carcinoma, SLND correctly identified 80 of the 87 patients with positive lymph nodes (pooled sensitivity: 92 percent, 95% CI 84-96 percent) [72]. The false-negative rate was 8 percent, which is similar to that observed in low-grade endometrial cancer.

The NCCN [73] and the SGO [50] both support the role of sentinel lymphadenectomy in the management of patients with EC. Both organizations note, however, the absence of randomized studies comparing sentinel lymphadenectomy with alternative strategies in evaluating the retroperitoneal lymph nodes.

Technique of lymphatic mapping — Lymphatic mapping with technetium-99m (99mTc) is the most common radiolabeled colloid injected. Blue-colored dyes (including 1% isosulfan blue and 1% methylene blue) are also used for direct visualization of lymphatic channels and sentinel lymph nodes. However, if a near-infrared camera is available for detection, use of indocyanine green (ICG) as a single agent has become the routine approach for detection of sentinel nodes [74] and is the agent used in our practice. Retrospective data suggest that ICG alone or in combination with other tracers is superior to that of blue dye alone (either isosulfan blue or methylene blue) for detection of sentinel nodes [75,76]. In a meta-analysis including 33 studies evaluating the diagnostic accuracy of various tracers for SLN assessment in patients with early-stage EC, ICG used alone or in combination with blue dye or 99mTC appeared to have a higher detection rate compared with non-ICG tracers (mean SLN detection rate: 92.4 to 100 percent versus 77.8 to 86.9 percent, respectively) [77]. However, sensitivities (ie, the ability to identify malignancy in a detected SNL) were all high (>90 percent) and similar between groups.

Whichever product is used for lymphatic mapping, 1 mL (0.5 mg) is injected using a 27-gauge needle into the superficial (2 mm) and deep (1 cm) cervical stroma at both the 3 and 9 o'clock positions prior to hysterectomy (achieving a total dose of 2 mL [1 mg]) [49].

Other lymphatic mapping techniques, such as fundal injection or hysteroscopic injection, do not have the convenience or the sensitivity for detection of sentinel lymph nodes, though they may be associated with an increased detection of aortic sentinel lymph nodes [78-81].

Technique of sentinel lymph node dissection — Following injection of the mapping product into the cervix, optimal detection of the dye or ICG will occur between 15 and 60 minutes [82]. Successful mapping of a hemipelvis is defined by observing a channel leading from the cervix directly to at least one candidate lymph node. Common iliac or aortic sentinel lymph nodes are also dissected if present. Identified sentinel lymph node(s) are then retrieved and sent for pathologic evaluation. If either hemipelvis does not map, individualization of the management of the patient is necessary.

Utilizing a standardized strategy when intending to perform an SLND in patients with EC has been shown to improve the SLND detection rate and decrease the rate of complete pelvic LND [83,84]. When such an algorithm is applied, bilateral mapping was seen in 81 percent, unilateral in 12 percent, and no mapping of either hemipelvis in 6 percent of patients. If a selective LND strategy is utilized to determine the management of a nonmapping hemipelvis, fewer than 10 percent of patients will require a complete pelvic LND [84] without compromising the ability to detect metastatic disease in the lymph nodes.

Importantly, standardization of the pathologic assessment of removed SLND is critical to the correct use of this technique. Sentinel lymph nodes are generally cut at 3 mm intervals, in a bread-loaf fashion, or bivalved if less than 1.5 cm in any dimension. Two paraffin-embedded slides are created from each section, each 50 microns apart. One slide is generally stained for hematoxylin and eosin (H&E) and the other is reserved for immunohistochemistry staining. If no metastatic disease is identified on the first H&E slide, the reserved slide is generally stained for cytokeratin AE1 and AE3 [85].

Further studies are needed to confirm that utilization of sentinel node biopsy reduces lower extremity lymphedema, particularly given the number of patients ultimately requiring full LND.

Additionally, the populations in which lymphatic mapping and SLND are appropriate are being evaluated. While initially utilized in patients at low risk for lymph node metastasis (with a complete LND in those at higher risk), many centers have moved to utilizing SLND in all patients, with selective complete LND in specific circumstances. It does appear that SLND is feasible in patients with nonendometrioid histology [86,87], with lymph node metastasis identified in approximately 20 percent of patients regardless of whether the procedure was performed using an SLND or systematic approach [86].

ADVANCED-STAGE DISEASE — In patients who have clinical or radiographic evidence of metastatic disease, surgery is typically performed to establish a definitive diagnosis, for palliation of symptoms and signs of disease, and/or to improve oncologic outcomes [88]. In general, these procedures are performed via laparotomy through a midline incision. Cytoreduction is performed in some patients. Indications for cytoreduction and outcomes are discussed elsewhere. (See "Initial treatment of metastatic endometrial cancer", section on 'Overview of treatment'.)

Similar to the treatment paradigms in ovarian cancer, a small study investigating the role of neoadjuvant chemotherapy and interval cytoreduction (rather than primary cytoreduction) demonstrated a higher rate of resection of all visible disease (70 versus 32 percent) and a shorter operating time with neoadjuvant chemotherapy [89]. Surgically, the approach to primary cytoreduction of EC is identical to that for patients with advanced ovarian cancer. (See "Cancer of the ovary, fallopian tube, and peritoneum: Surgical cytoreduction".)

POSTOPERATIVE TREATMENT AND SURVEILLANCE — Postoperative treatment, surveillance, complications, and side effects, including postmenopausal hormone therapy and sexual health, are discussed separately. (See "Overview of approach to endometrial cancer survivors".)

SPECIAL CLINICAL SITUATIONS

Fertility preservation — Reproductive-age patients with low-risk EC should be asked about whether they wish to preserve fertility. For patients who are candidates for fertility preservation, the most common approach is progestin therapy and deferral of surgical staging (including hysterectomy and bilateral salpingo-oophorectomy) until after completion of childbearing.

Use of a fertility-preserving treatment approach is limited to appropriate candidates at low risk for advanced or recurrent disease. Patients should be counseled that the risk of recurrent or persistent disease is likely higher than with hysterectomy.

Prior to initiating conservative therapy, a thorough evaluation is necessary to assess cancer grade and depth of myometrial invasion. This consists of medical history, physical examination for uterine size and mobility and evidence of metastatic disease, endometrial sampling, and pelvic and abdominal imaging.

Additionally, patients should be assessed for their potential risk for having developed EC as a result of Lynch syndrome, since the safety of fertility preservation in the setting of a genetic mutation is unknown (especially given the increased risk for the development of ovarian cancer in a woman with Lynch syndrome). (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Cancer screening and management", section on 'Endometrial and ovarian cancer'.)

Fertility-sparing therapy for EC is discussed in detail separately. (See "Fertility preservation in patients with endometrial carcinoma".)

Inoperable patients — For patients with clinical stage I EC who are unfit or unwilling to undergo surgery, primary radiation therapy may be acceptable treatment. Clinical staging should be performed for these patients according to the International Federation of Gynecology and Obstetrics system adopted in 1971 [90]. A clinical staging procedure includes examination under anesthesia, sounding of the uterus, endocervical curettage, hysteroscopy, cystoscopy, proctoscopy, and imaging studies.

The majority of studies report that enhanced magnetic resonance imaging has a sensitivity of approximately 80 to 90 percent (range 57 to 100 percent) to detect myometrial invasion. Findings of sensitivity for cervical invasion were inconsistent, ranging from 56 to 100 percent. Thus, a negative study does not provide assurance that local invasion does not exist [91].

Adjuvant therapy for patients with EC who cannot tolerate surgery is discussed in detail separately. (See "Treatment of low-risk endometrial cancer", section on 'Limited role of adjuvant therapy'.)

Synchronous ovarian and endometrial cancer — Synchronous primary cancers of the endometrium and ovary are found in 5 percent of patients with EC and 10 percent of patients with ovarian cancer [92,93]. For patients with EC, the risk appears to be higher in premenopausal patients, in whom 5 to 29 percent have a synchronous ovarian malignancy [94].

For patients with ovarian involvement, the following criteria are used to distinguish those with disease limited to low-grade endometrioid carcinoma involving the endometrium and ovary from extensive spread of the endometrial carcinoma to the ovary(s) (table 2) [7]:

●Invasion of <50 percent of myometrium, and

●No extensive/substantial lymphovascular space invasion (LVSI), and

●No additional metastases, and

●Ovarian tumor is unilateral, limited to the ovary, without capsule invasion and/or rupture

Histology is often the same for both groups, making it unclear whether there are two separate primary tumors, metastasis from the endometrium to the ovary, or (less frequently) from the ovary to the endometrium. However, prognosis and management differ. Patients with low-grade endometrioid carcinoma involving the endometrium and ovary have a good prognosis and are not typically treated with adjuvant therapy [7]. By contrast, those with extensive spread of the endometrial carcinoma to the ovary have more severe disease and are typically treated with adjuvant therapy.

Given that young patients with EC may be candidates for ovarian preservation, recognition of the risk for synchronous ovarian cancer in this population will facilitate shared decision making [95]. Additionally, young patients should be assessed for their potential risk for having developed their endometrial and/or ovarian cancers as a result of Lynch syndrome, since the safety of fertility preservation in the setting of a genetic mutation is unknown. (See "Treatment of low-risk endometrial cancer", section on 'Fertility preservation as alternative in selected patients' and "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Cancer screening and management", section on 'Endometrial and ovarian cancer'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Uterine cancer".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Uterine cancer (The Basics)")

●Beyond the Basics topic (see "Patient education: Endometrial cancer diagnosis, staging, and surgical treatment (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Preoperative evaluation

•Prior to treatment, patients with endometrial carcinoma (EC) should have a complete evaluation, including history (including screening for hereditary susceptibility to cancers (table 1)), physical examination, and endometrial sampling, to establish the diagnosis of EC. (See 'History and physical examination' above.)

•A chest radiograph should be performed as part of the initial assessment to exclude lung metastases. Abdominal and pelvic imaging is rarely performed in patients with nonaggressive ECs but is often used to exclude metastasis from aggressive ECs. (See 'Imaging studies' above.)

●Surgical staging – EC is surgically staged (table 2 and table 3). The staging procedure includes total hysterectomy, bilateral salpingo-oophorectomy (BSO), and lymphadenectomy. (See 'Staging system' above.)

•Apparently uterine-confined disease – For EC staging in patients with disease apparently confined to the uterus (based on physical examination, with or without pelvic imaging) who are surgical candidates and in whom the procedure is expected to be able to be completed without conversion to laparotomy, we recommend minimally invasive surgery (MIS; laparoscopy or robotic surgery) rather than laparotomy (Grade 1A). Conventional total laparoscopic hysterectomy with BSO is the preferred option for most patients, with use of robotic surgery if this makes MIS possible in patients who have a high risk of conversion to laparotomy (eg, patients with obesity). (See 'Apparently uterine-confined disease' above.)

•Apparent extension to the cervix – For patients with clinically apparent extension of EC to the cervix, management may include (1) radical hysterectomy with postoperative radiation therapy based on pathologic factors, (2) extrafascial hysterectomy with postoperative radiation, or (3) primary radiation therapy followed by extrafascial hysterectomy. (See 'Surgical management of cervical involvement' above.)

•Evidence of metastatic disease – In patients who have clinical or radiographic evidence of metastatic disease, surgery is typically performed to establish a definitive diagnosis, for palliation of symptoms and signs of disease, and/or to improve oncologic outcomes. In general, these procedures are performed via laparotomy through a midline incision. Cytoreduction is performed in some patients. (See 'Advanced-stage disease' above.)

●Approach to lymph node evaluation – The approach to lymph node evaluation in patients with EC is a subject of debate and no option has emerged as superior based on the available data. Options for management of retroperitoneal lymph nodes (in the absence of grossly metastatic disease) include no lymph node dissection (LND), systematic LND only if the risk of lymph node metastasis exceeds a certain threshold, routine sentinel lymph node dissection (SLND) following lymphatic mapping, or systematic LND in all patients. Patients should be counseled about the options and engage with the surgeon in shared decision making regarding the approach to lymphadenectomy. (See 'Lymph node evaluation' above.)

●Clinical staging – Clinical staging may be an appropriate option for patients who cannot tolerate surgery or who desire fertility preservation. Evaluation in these patients includes medical history, physical examination for uterine size and mobility and evidence of metastatic disease, endometrial sampling, and pelvic and abdominal imaging. (See 'Special clinical situations' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Steven C Plaxe, MD, who contributed to an earlier version of this topic review.