1. Primary Management
Surgery remains the primary treatment option for endometrial cancer (EC). It usually includes total hysterectomy (TH), where the uterus and the cervix are removed, with bilateral salpingo–oophorectomy (BSO), where both fallopian tubes and ovaries are also removed
[1][2][3]. Lymph node assessment is an incorporated part of the surgical management of EC and provides a strong predictor of survival, but the best method for this still raises controversy
[4]. Surgical staging is essential, as it strongly determines prognosis and guides the decision for adjuvant treatment
[1].
Generally, a meticulous abdominal and pelvic examination must be performed upon entering the peritoneal cavity. When possible, obvious extrauterine disease or suspicious lesions should be removed or biopsied
[1]. It is suggested, though, that cytology of the peritoneal fluid should not be routinely performed, as positive results could be attributed to uterus handling. For some high-grade histological types of non-endometrioid EC, such as clear cell or serous carcinoma, and carcinosarcoma, omentectomy should be included in the surgical procedure, as visual assessment of the omentum appears to be insufficient
[1][5][6]. When complete or optimal cytoreduction is attained, overall survival (OS) and progression-free survival (PFS) appear to be longer
[7].
Oophorectomy is included in the standard procedure to exclude any ovarian metastasis or primary ovarian tumors, especially in patients with Lynch syndrome. However, ovarian preservation can be considered in young, premenopausal patients who have endometrioid type I EC, with approximately 50% endometrial invasion and no sign of extrauterine disease in preoperative diagnostic procedures
[3]. Methods of fertility preservation, if desired, will be further discussed.
Radical hysterectomy, type II or III, is recommended in cases of clear cervical involvement
[1][2].
Vaginal hysterectomy is also an option that might be used if the patient is not a suitable candidate for systematic surgery
[2].
2. Routes of Surgery
The available routes of surgery are open surgery/laparotomy or minimally invasive surgery (MIS), which have emerged in recent years as the optimal methods and mainly include laparoscopic surgery (LS) and robot-assisted surgery (RS).
Combining the results of two of the largest randomized controlled trials, namely the Laparoscopy Compared with Laparotomy for Comprehensive Surgical Staging of Uterine Cancer: Gynecologic Oncology Group Study (LAP2) and the Laparoscopic Approach to Cancer of the Endometrium (LACE) trials, which compared laparotomy to laparoscopy, revealed multiple advantages of the latter
[8][9]. LS is associated with a shorter hospital stay, enhanced recovery, and equal-to-laparotomy detection rates for overall disease at advanced stages. Fewer postoperative complications, such as blood loss, need for blood transfusion, wound complication, or need for ICU admission were also noted
[7][10][11]. Furthermore, LS was shown to be more cost effective, especially in patients with higher BMI
[10]. Despite longer operation times, LS was not accompanied by an increase in intraoperative injuries
[11]. However, it is important to mention that the LAP2 trial enrolled patients with stage I to IIA EC, while the LACE trial included patients with stage I EC. These results concerning early-stage EC were confirmed in a recent Cochrane review
[12]. Another study in 2020, showed that MIS was equally safe for Stage IIIC, without impairing survival and complete resection of disease was achieved
[13]. More studies concerning high-grade ECs are needed.
An alternative effective type of MIS, RS has emerged and its use in EC is rising. Compared to both laparotomy and LS, RS resulted in even shorter hospital stays and fewer complications, namely, blood loss and blood transfusions
[10]. Although RS is more expensive
[1] and often associated with a longer duration of operation, a randomized trial found shorter operation times, and no conversions to open surgery
[14]. Generally, RS is associated with fewer conversions to laparotomy, which mostly happen due to inadequate exposure
[10]. The elderly, and patients with higher BMI, are also suitable candidates for RS
[10][15]. However, recent studies have questioned the long-term outcomes of RS. More specifically, Argenta et al.
[16] reported that in patients with stage I EC, RS is associated with poorer long-term outcomes, compared to the LS group, and led to poorer recurrence-free (Hazard Ratio-HR: 1.41; 95% CI: 1.12, 1.77), OS (HR: 1.39; 95% CI: 1.06, 1.83), and disease-specific survival (HR: 3.51; 95% CI: 2.19, 5.63). The study concludes that possibly, the significance of long-term effects has been underestimated due to positive short-term outcomes
[16], as also highlighted in the British Gynaecological Cancer Society (BGCS) guidelines
[3].
The benefits of MIS compared to open surgery have been established concerning both early and advanced EC stages as outlined in a retrospective cohort study in England published in 2020
[17]. The need for longer hospitalization periods with open surgery compared to MIS was confirmed (5.28 vs. 2.32 days), while the overall conversion rate for MIS was 6.6%. The only complication that was not significantly higher with open surgery was ureteric complications. Significantly higher overall 90-day mortality with open surgery (OR 0.34; 95% CI: 0.18–0.62;
p = 0.0002) was also noted
[17]. MIS has also been associated with reduced postoperative pain
[10].
Currently, the BGCS suggests MIS for suitable patients
[3], while European Society of Gynaecological Oncology (ESGO), European Society for Radiotherapy and Oncology (ESTRO), and European Society of Pathology (ESP) recommend MIS, even in patients with high-risk EC
[18].
3. Determining Myometrial Invasion
The depth of myometrial invasion is essential for both staging and prognosis and it can be determined either pre-or intra-operatively. Magnetic resonance imaging (MRI) and transvaginal sonography (TVS) are the usual preoperative methods used, while intraoperative methods include intraoperative gross examination (IGE) and intraoperative frozen section (IFS). A 2016 meta-analysis, which mostly included studies with high-risk EC cases, in which deep invasion of myometrium is more common, showed the superiority of IFS compared to IGE in determining myometrial invasion. Pooled sensitivity and specificity for IFS were 85% and 97%, respectively, while for IGE they were 71% and 91%, respectively
[19]. Another study compared the methods of MRI, TVS, IGE and IFS. It was observed that IFS had the highest sensitivity (90%), while IGE had the lowest sensitivity. For the preoperative methods, MRI had higher sensitivity than TVS, but both methods showed low positive predictive values
[20]. IFS may be more time consuming, more expensive and demand the presence of a pathologist, which is not feasible in every center, but it prevails on determining tumor grade and its results are congruent with the results of histologic examination
[21]. IGE, on the other hand, is quicker, cheaper, and simpler but vastly depends on the surgeon. It should be acknowledged that the tumor-invaded myometrium is not always macroscopically visible
[19][20].
The implementation of sentinel lymph node (SLN) mapping will probably lower the need for IFS, but IFS can still be useful, especially in cases where the preoperative histopathological information is unclear or vague
[21].
4. Fertility-Sparing Management
EC in younger patients might not be very common, but it is estimated that about 5% of cases involve patients under 40 years of age. In this group, it is important to consider the possible desire for fertility preservation. Fertility-sparing options, where the uterus and the ovaries are kept, can be considered if the following criteria are met: age younger than 40 years, endometrioid type EC, Stage I, with no evidence of myometrial invasion, and no evidence of metastatic disease, or lymph node involvement. Moreover, the expression of progesterone receptors on the endometrium is favorable and a strong predictor of remission, but it is not compulsory. Patients under 40 often meet the aforementioned criteria and they should be asked whether they desire to preserve their fertility so fertility-sparing options can be discussed. It should be clarified, though, that this is not the standard management of EC and existing risks and outcomes should be thoroughly discussed, accompanied by counseling
[7][22].
Before any intervention, regular blood work, including Ca125, urine exams and endometrial biopsy should be conducted in the context of clinical staging
[22]. Imaging, preferably with contrast–enhanced MRI, should also be performed to assess possible myometrial invasion and exclude any metastasis. Ovarian metastasis or synchronous ovarian cancer and the presence of Lynch syndrome should also be excluded. If the results are inconclusive or ambiguous, exploratory laparoscopy with peritoneal lavage, SLN biopsy, or biopsy of the ovaries could give more information. It is reported that in 5–30% of cases, these tests underestimate the tumor grade
[7].
Hormonal methods are one of the options for fertility-sparing management. Oral progestin is used; and the most common regimens are medroxyprogesterone acetate and megestrol acetate. The response rates vary; in one study, 73% of cases responded to oral progestin, with a relapse rate of 36%
[23], while in another study on tumors expressing estrogen and progesterone receptors, the response rate ranged from 26–89% when receptors were present, and from 8–17% when receptors were absent
[22]. Alternatively, a levonorgestrel-releasing intrauterine device can be used. The use of gonadotropin-releasing hormone agonists (GnRHa) during chemotherapy in EC, to suppress ovarian function and limit ovarian damage is still under investigation
[24].
Cryopreservation techniques are also available and they include embryo, oocyte, or ovarian tissue cryopreservation. The first two options require ovarian stimulation, while the last option does not, since cortical ovarian tissue can be obtained through laparoscopy
[24]. Induction of ovulation seems not to be associated with higher relapse risk and in patients with estrogen-dependent tumors, there are strategies to maintain low estrogen levels during ovarian hyperstimulation
[22].
Furthermore, in patients for radiotherapy, uterine transposition is performed under laparoscopy to keep the uterus, cervix, and ovaries in the upper abdomen, away from the area that will be radiated
[25].
Once childbearing is complete, patients should undergo surgery with TH and bilateral salpingectomy
[3][7].
5. Lymphadenectomy
Historically, surgical staging of EC has included complete pelvic and para-aortic lymphadenectomy
[1], with the upper border at the left renal vein. As mentioned before, the nodal factor is essential in staging and provides important prognostic information
[26], thus determining the need for adjuvant therapy
[11], as well as the recurrence risk
[27].
However, the role of lymphadenectomy in EC has raised controversy. The decision to perform a lymphadenectomy could depend on disease progression and tumor grade, as determined by preoperative imaging and biopsies
[26].
It is known that the risk of lymph node involvement is correlated with the tumor grade, the depth of myometrial invasion and the high-risk histological types of EC. The results of early-stage EC from various studies tend to align. A recent Cochrane review, involving patients with Stage I EC who either underwent lymphadenectomy or did not, found no difference in overall and recurrence-free survival (RFS) between the two groups, concluding that routine use of lymphadenectomy is not recommended in early stages
[26]. Considering higher-grade EC, however, given the higher risk of nodal involvement
[4], many studies have shown a survival benefit when both pelvic and para-aortic lymphadenectomies are performed
[1]. The need for para-aortic lymphadenectomy is also underlined by the fact that in approximately 8% of high-risk EC cases, while the pelvic nodes are negative, the para-aortic nodes are positive
[7].
Nevertheless, full lymphadenectomy is associated with a number of adverse effects and high surgical morbidity
[11]. Lymphedema, formation of lymphocysts, injury of blood vessels and pain and numbness from the lower abdomen to the genitalia and the inner thigh due to genitofemoral nerve damage are among the most common adverse effects. The incidence of lymphedema and lymphocysts may be even higher than reported, since many studies tend to focus on short-term effects. Deep vein thrombosis and pulmonary embolism could also occur postoperatively. Moreover, in obese patients, who represent a considerable percentage of patients with EC, the procedure presents multiple difficulties, mostly of a technical nature
[26][27].
It is evident that randomized controlled clinical trials (RCTs) are needed to determine the role of lymphadenectomy in patients with higher-grade EC. In recent years, other less invasive methods have been in trial, such as the following:
- Selective lymph node sampling;
- Deciding whether to perform lymphadenectomy based on intrauterine risk factors mainly from IFS and;
- SLN mapping.
The last method seems to offer the best results
[11].
6. Adjuvant Radiotherapy
Radiotherapy is recognized among adjuvant modalities in EC standards of care and it includes pelvis external beam radiotherapy (PEBRT), whole pelvic radiotherapy, and vaginal brachytherapy (VBT). Risk stratification plays a key role in adjuvant radiotherapy (ART). Patients diagnosed with low-risk Stage I EC are not candidates for ART, as brachytherapy does not prevail over surgical management. While brachytherapy is recommended for intermediate-risk, high-risk and high-intermediate-risk EC, the presence of lymphovascular space invasion (LVSI) is of great importance regarding the latter EC group, in the presence of which external beam radiation therapy (EBRT) is also suggested. Several trials, such as the “A Study in the Treatment of Endometrial Cancer”/EN.5 (ASTEC/EN.5), Gynecologic Oncology Group (GOG)-99, Post Operative Radiation Therapy in Endometrial Carcinom (PORTEC)-1, and PORTEC-2, demonstrated no difference in OS, despite the effectiveness on pelvic and vaginal recurrences regarding early-stage intermediate-risk or high-risk EC. Although its effect on OS is limited, a pattern of concurrent chemoradiotherapy (CRT) is proposed for high-risk p53 + and high-risk Stages III C1 or C2 EC (PORTEC-3 protocol), where brachytherapy boost was indicated in cases of cervical involvement, LVSI or Grade 3 and Stage III EC
[28][29][30][31].
Molecular profiles and biomarkers tend to upend the established therapeutic scenario in terms of ART. Interestingly, molecular classification of EC combined with clinicopathological factors of EC patients may suggest a novel risk profile as a determinant of ART in Stage I–II high-intermediate-risk EC (PORTEC-4a randomized trial)
[28]. In a retrospective multicenter cohort study conducted by Reijnen et al.
[32], proactive molecular risk classifier for endometrial cancer (ProMisE) mismatch repair (MMR) status was proposed as a predictive biomarker concerning ART response and, therefore, the effect of such treatment options on survival. The majority of the included cases concerned Stage II high-risk endometrioid EC
[32]. Furthermore, Mohammadi et al.
[33] proposed the use of the Radiosensitivity Index (RSI), a genomic signature, as a prediction model in pelvic recurrence, ART decisions and treatment escalation, including radiosensitizing agents
[33].
7. Chemotherapy
Adjuvant chemotherapy remains the mainstay of high-risk EC treatment strategies. Carboplatin and paclitaxel are commonly used due to similar results in OS and PFS; they also appear to be less toxic. According to National Comprehensive Cancer Network (NCCN) guidelines, patients with clear cell or serous tumors of Stage IA characterized by myometrial invasion, Stage IB or Stage III could also be candidates for chemotherapy, with or without VBT. This could also be applied to Stages IB and II, Grade C EC. Stages III B and III C EC usually follow chemotherapy protocols in conjunction with radiotherapy, while adjuvant chemotherapy after cytoreductive surgery is suggested for resectable Stage IV disease.
Whether CRT is superior to adjuvant chemotherapy or radiotherapy alone remains controversial and current data lack prospective studies that compare adjuvant chemotherapy to adjuvant CRT. Radiation therapy (RT) could either be used before, after or in-between (sandwich fashion) chemotherapy.
The PORTEC-3 study showed that adjuvant chemotherapy (four cycles of carboplatin AwUC5 and paclitaxel 175 mg/m
2) preceded by concomitant CRT (two cycles of cisplatin 50 mg/m
2) improved five-year failure-free survival in patients with stage III EC, without increasing 5-year OS
[29]. However, (GOG-249) study, a Phase III trial that included intermediate-high-risk stage II EC and Stage I to II clear cell or serous tumors, concluded that VBT followed by chemotherapy (three cycles of carboplatin AwUC6 and paclitaxel 175 mg/m
2) (VBC/C) resulted in greater acute toxicity and nodal recurrences compared to the pelvic RT scheme. The differences between the two groups were minimal regarding late toxicity and 36-month OS and RFS
[34]. On the contrary, Matei et al.
[35] enrolled patients with Stage III or IVA (locally advanced) EC in a Phase III trial (GOG-258/NRG) and randomized them in a chemotherapy group and CRT (four cycles of carboplatin AwUC5-6 and paclitaxel 175 mg/m
2) group. RFS was superior in the latter group, which was associated with a lower 5-year incidence of vaginal recurrence and pelvic and para-aortic lymph node recurrence, but with more distant metastases
[35].
Results from the European Network of Gynaecological Oncological Trial-EN2- Danish Gynecological Cancer Group Phase III trial, in which postoperative chemotherapy (six cycles of paclitaxel-carboplatin) is compared with standard treatment (observation) in patients with node-negative Stage I and II intermediate-or high-risk EC, are expected. The primary outcome of the study is defined by OS
[36].
Chemotherapy is also an effective tool against recurrent or metastatic EC, which usually does not respond to hormonal therapy. The GOG-209 trial compared the paclitaxel-doxorubicin-cisplatin regimen with carboplatin/paclitaxel (TC) in Stage III or IV EC. Both OS and PFS were similar between the two groups, while health-related quality of life was superior in the TC group
[37]. GOG-286B, an ongoing Phase II/III study of Paclitaxel/Carboplatin/Metformin versus Paclitaxel/Carboplatin/Placebo may shed light on the beneficial effect of metformin in EC management concerning advanced and recurrent disease
[38].
8. Immunotherapy
Recently, molecular classification has allowed targeted therapies to be developed. The angiogenesis pathway, the phosphatidylinositol 3-kinase (PI3K)/ serine/threonine kinase (AKT) /mammalian target of rapamycin (mTOR) pathway and glucose metabolism are under thorough investigation in studies, but currently there is no approved targeted therapy for this cancer beyond hormonal therapy
[39].
A number of immune checkpoints and biomarkers are expressed in EC and immune system cells. Thus, their use in diagnostic tests, targeted therapeutic management, and predictive value could be promising. PD-1 and PD-L1 are recognized among these biomarkers and checkpoints, with the latter being mostly expressed in the catalytic subunit (POLE) and microsatellite instability (MSI) EC microenvironment. An algorithm presented by BGCS in 2022, suggests that MMR, p53 and Estrogen receptor (ER) immunochemistry should be performed on all EC, while POLE next generation sequencing should only be performed in cases with abnormal MMR and/or p53, stage I/II non endometrioid, grade 3 endometrioid, stage IA with no/focal LVSI, or endometrioid with either estrogen receptor (ER)-negative, or stage IA with substantial LVSI, or stage IB/II
[40]. Monoclonal antibody-based therapies, known as immune checkpoint blockade, have recently shown robust evidence in EC immunotherapy
[41]. The “A Clinical Trial of Pembrolizumab (MK-3475) Evaluating Predictive Biomarkers in Subjects with Advanced Solid Tumors”, KEYNOTE-158, multicohort phase II study demonstrated the efficacy of pembrolizumab monotherapy, an anti-PD-1 agent, in MSI-H/dMMR advanced EC by denoting the antitumor activity and presenting a safe toxicity profile accompanied by a median PFS of 13.1 months
[42]. Immunotherapy with a PD-1 inhibitor, combined with an antiangiogenic agent, when managing pretreated recurrent EC has been shown to be quite beneficial in patients’ survival. According to NCT03367741, a translational Phase II trial, cabozantinib-nivolumab combined therapy significantly improved PFS (5.3 vs. 1.9 months)
[43]. Finally, a Phase III randomized placebo controlled trial aims to assess the addition of atezolizumab, an IgG1 PD-L1 inhibitor, to standard chemotherapy in advanced or recurrent EC; the results are expected in 2023
[44][45].
9. Sentinel Lymph Node Biopsy
SLN biopsy (SLNB) in EC is gaining ground in the staging of the disease and could be used as an alternative to selective lymphadenectomy (SLAD) and IFS. It is performed using tracer dyes (TD) and is oriented as the detection of at least one SLN in either or each hemipelvis. Blue dyes such as isosulfan blue, methylene blue and patent blue detected by colorimetry, indocyanine green (ICG) detected by near-infrared method, and Tc-99 m detected by radionuclide scanning are the most representative examples of TD-detector pairs. On the importance of preference, the near-infrared ICG method is usually chosen, as it is associated with higher detection rates and provides not only quick transcutaneous real-time visualization, but also low toxicity and cost. Its administration follows the pattern of “large volume and low concentration”. Detection rate is affected not only by the type of tracer, but also by the site of injection, LVSI, clinically enlarged lymph nodes, BMI, surgeon’s experience and RT history.
In uterine lymphatic draining systems, lymph nodes in the upper paracervical lymphovascular tissue are in favor of metastatic disease. The site of the TD injection could be cervical, hysteroscopic myometrial/peritumoral and transabdominal subserosal/myometrial. Despite the low risk of isolated para-aortic metastasis, a higher detection rate of para-aortic SLN mapping is offered by hysteroscopic injection. Nevertheless, cervical injection is considered the safest and easiest-performed method.
Positive SLN mapping includes macrometastases, micormetastases and isolated tumor cells. Hematoxylin-eosin stain is used for SLN pathological analysis. In negative results, ultrastaging is recommended, which consists of deep serial sections and cytokeratin immunohistochemical stain
[11][46].
The most commonly used surgical algorithm for SLN mapping was described by Persson et al. in 2017. After the cervical injection, ICG display is bilaterally evaluated in the upper and lower paracervical pathway. If all four pathways are visible, the presacral avascular plane is opened in order to identify and remove SLNs along the lower paracervical pathway bilaterally. The next step is to dissect the paravesical and pararectal planes to identify and remove SLNs along the upper paracervical pathway. Lastly, the upper parametrium is removed and, along with the SLNs, is checked with ultrastaging and immunochemistry. If all pathways are not visible thoroughly, exploration and reinjection are the only available options before full lymph node dissection is required
[47].
SLNB was considered experimental at the ESMO/ESGO/ESTRO consensus conference on EC in 2016
[48]. However, in 2021, the ESGO/ESTRO/ESP guidelines recommended SLNB for surgical staging in patients with low-risk or intermediate-risk EC, while systematic lymphadenectomy was not recommended in this group. In patients with high-intermediate-risk or high-risk disease, SLNB is considered an acceptable alternative to systematic lymphadenectomy
[18].
According to Stewart et al.
[46], surgical staging of both low-and high-risk EC through the SLN algorithm reduced operative time and the use of IFS. However, the SLN algorithm was not superior in terms of hospital charges or intraoperative and postoperative complications
[46].
The prospective “A comparison of sentinel lymph node biopsy to lymphadenectomy for endometrial cancer staging” (FIRES) trial demonstrated SLN mapping using ICG as a technique with high sensitivity and negative predictive value in the detection of nodal metastases, enough to succeed staging through lymphadenectomy. The FIRES trial interpreted that SLN mapping with ICG has high diagnostic accuracy in detecting EC metastases and can therefore be a safe alternative to lymphadenectomy, with the benefit of exposing fewer patients to the morbidity of a complete lymphadenectomy. Specifically, the SLN technique had a sensitivity of 97.2% (95% CI: 85.0–100) and a negative predictive value of 99.6% (95% CI: 97.9–100), as 257 of the 258 patients had truly negative non-sentinel lymph nodes. Furthermore, SLN mapping proved to be superior to traditional complete lymphadenectomy, as pathologically identified SLNs were significantly more likely to contain metastatic disease compared to non-SLN, thus, pathologists were required to ultra-stage less, but more crucial, nodes
[27].
Moreover, the distribution and typical position of SLNs in high-risk EC were investigated in the FIRES trial with SLN-ICG identifying SLNs in the obturator area, left and right, in approximately 60% of patients, in the external iliac area, bilaterally, in 80% of patients, in the presacral area in half the patients, and only 13% of patients had an SLN identified in the common iliac area. Metastatic SLNs were found mostly in the obturator area (25% of node-positive patients left/36% right) and in the external iliac area (41% left/25% right)
[47].
A meta-analysis by Bogani et al. in 2019
[49] showed that SLN mapping was not only superior to lymphadenectomy, but also, in combination with pathological ultrastaging, it achieves a higher detection rate of nodal disease in comparison to lymphadenectomy and an accurate detection of positive nodes, even in high-risk EC. Low-risk patients who underwent SLN mapping had a higher detection rate of positive pelvic nodes (OR: 3.12 (95% CI: 1.32–7.39) and a comparable detection rate of paraortic nodes (OR: 1.38 (95% CI: 0.39–4.83)). This outcome was also observed in the intermediate and high-risk EC groups, with a higher detection rate in pelvic nodes (OR: 2.04 (95% CI: 1.19–3.48), and a similar detection rate between groups for positive para-aortic nodes. These data additionally support the effectiveness of SLN mapping in terms of oncologic outcome, as no statistical difference in RFS was observed between groups (OR: 0.90 (95% CI: 0.58–1.38)
[49].