Sentinel lymph node biopsy in endometrial cancer: state of the art.

Endometrial cancer (EC) is the most common gynecological neoplasm in developed countries. In literature, there are discordant data regarding the therapeutic value of systematic lymphadenectomy whereas the importance of lymph node status for determining prognosis and the need for adjuvant treatment is undoubted. Given the low risk of lymph-node metastases in the apparent early-stage disease and the significant surgical and postoperative risks when performing a complete pelvic lymphadenectomy, the surgical approach in these patients is controversial, ranging from no nodal evaluation to comprehensive pelvic and aortic lymphadenectomy. The recent introduction of sentinel node mapping represents the mid-way between the execution and omission of node dissection in EC patients. Indeed, the sentinel node mapping has rapidly emerged as an alternative to complete lymphadenectomy to reduce morbidity. In the present review, we discuss the role of sentinel node mapping in the surgical management of EC evaluating all aspects of this procedure. 2020 Translational Cancer Research. All rights reserved.

Introduction

EC is a common gynecologic malignancy with an increasing incidence and an age of onset younger than in prior years (1,2).

EC typically has a well or moderately differentiated endometrioid histotype at early-stages with a good prognosis. The surgical management for women with clinically apparent early-stage EC consists of total hysterectomy and bilateral salpingo-oophorectomy with or without lymphadenectomy. The evaluation of lymph node status through pelvic and selective para-aortic node dissection is an invaluable prognostic factor for targeted adjuvant treatment (3).

Recently, it has been also discussed the role of intraoperative frozen section (FS) diagnosis during hysterectomy for early-stage EC to determine the depth of myometrial invasion and consequently to select the women to undergo lymphadenectomy. The Authors showed a concordance of FS and final pathology of 85.2%, with an under-diagnosis rate of 14%, over-diagnosis rate of 0.8%, and overall accuracy of FS in predicting myometrial invasion of 93.3% (4).

The role of systematic pelvic lymphadenectomy is a controversial topic in the surgical treatment of these patients and there is a disagreement among clinical practices, with surgical approaches varying from no nodal evaluation to full lymphadenectomy (5-7).

A non-standardized surgical approach is explained by discordant data regarding the therapeutic value of systematic lymphadenectomy.

A retrospective study documented the positive effects of pelvic and para-aortic lymphadenectomy on survival rates (8) whereas two randomized prospective European studies conflicted with these results (9,10) demonstrating no advantage in disease-free or overall survival. These trials were largely criticized in their study designs because included minimally invasive disease, limiting the ability to conclude.

Albeit the therapeutic role of lymphadenectomy is still questioned, its prognostic role is undoubted. Indeed prognosis of EC is correlated with lymph node status, which is itself related to the histologic grade, as well as the depth of myometrial invasion (11).

The patients with lymph node metastasis disease have significantly lower median survival rates in comparison to females with a tumor limited to the uterus: positive lymph node status restricts 5-year disease-free survival at stage 1 EC to 54% while it was 90% if the nodes were negative (12). In addition to the prognostic value, lymph node status assessment plays an essential role in tailoring adjuvant therapy.

The increasing awareness of some risks associated with lymphadenectomy, as intraoperative neurovascular injury and long-term complications such as leg lymphedema (13,14), called into question the benefit of the procedure in stage 1 where the incidence of lymph-node involvement is low. In the presence of G1 or G2 endometrioid EC and less than 50% myometrial invasion [Stage IA according to 2010 FIGO (International Federation of Gynecology and Obstetrics) staging)], the risk of lymph node metastasis is low. Conversely, in the case of G3 endometrioid EC or non-endometrioid adenocarcinoma (e.g., serous, clear cell carcinoma), independent of grade status, the risk of lymph node metastasis is high and an extended operative staging with lymphadenectomy is mandatory during surgery (4).

Chi et al. reported the incidence of pelvic nodal metastasis in patients classified as Stage I (2010 FIGO staging), concerning the grade of differentiation and depth of myometrial invasion. In this study, none of the 80 patients with grade 1 well-differentiated tumors had pelvic lymph node metastases irrespective of the depth of endometrial invasion. In patients with no myometrial invasion, only 2% (2/110) had nodal metastasis, which increased to 8% (13/64) for those with inner-half invasion and 20% (15/75) with outer half invasion (15).

Considered the low risk in early-stage disease of lymph-node metastases, performing a systematic lymphadenectomy may be more detrimental rather than helpful.

The purpose of this review is to give an overview of the sentinel lymph node (SLN) procedure in the management of patients affected by early-stage EC.

Material and methods

The data research was conducted using the following databases MEDLINE, EMBASE, Web of Sciences, Scopus, ClinicalTrial.gov, OVID and Cochrane Library querying for all articles related to SLN biopsy in the EC from the inception of the database up to December 2019. The studies were identified with the use of a combination of the following text words: endometrial cancer, systematic pelvic lymphadenectomy, SLN biopsy, indocyanine green (ICG), ultrastaging. The selection criteria of our narrative review included randomized clinical trials, non-randomized controlled studies (observational prospective, retrospective cohort studies, case-control studies, case series) and review articles of the SLN biopsy in EC. A review of articles also included the abstracts of all references retrieved from the search. Article not in English language, conference papers and reviews, and studies with information overlapping another publication were excluded. In presence of overlapping studies, we took into account the most recent and/or most comprehensive manuscript.

Sentinel lymph-node mapping

The SLN mapping is a recent innovative method (16,17). This procedure permits the reduction of the morbidity associated with full lymphadenectomy without lacking prognostic information.

Besides, the minimally invasive laparoscopic hysterectomy approach with the combination of SLN biopsy constitutes an interesting alternative to contemporary surgical treatment particularly valuable in obese patients (16).

The use of SLN mapping is a well-known procedure in the treatment of melanoma (18) and breast cancer (19). It would seem to be useful in high-risk histologies, such as serous carcinoma, clear cell carcinoma, carcinosarcoma (20). The purpose of using this tool is to target the lymph nodes that are the primary tumor drainage area and thus the most likely to metastasize. The pathologic exam of the SLN may with accuracy inform about the status of the regional lymphatic basin, thereby minimizing the risk of morbidity associated with a complete lymphadenectomy (21). It is precisely in the cases of the surgical staging of apparent uterine-confined malignancy when no metastasis is demonstrated by imaging studies or no obvious extrauterine disease at exploration, that the SLN mapping can be considered. In a recent meta-analysis, Bodurtha Smith et al. (22) presented data from 55 studies and 4,915 women. The overall detection rate of SLN mapping was 81% (95% CI, 77–84%) with a 50% bilateral pelvic node detection rate (95% CI, 44–56%) and 17% paraaortic detection rate (95% CI, 11–23%). The detection rate and accuracy of fluorescence-guided SLN mapping have been evaluated even in patients undergoing robotic-assisted surgery, as demonstrated in the pilot study of Paley et al. [2016]. At least one sentinel node was detected in 119/123 (96.7%) patients. Ninety-nine patients (80%) had bilateral pelvic and/or periaortic sentinel nodes successfully mapped. The final hematoxylin and eosin (H&E) examination revealed the presence of lymph node metastasis in only 10.6% of the patients undergoing completion lymphadenectomy. In 44% of these cases, interestingly, the sentinel node was the only positive node founded (isolated metastasis in SLN). The procedure showed a sensitivity of 100%: the examination of non-sentinel nodes (upon completion lymphadenectomy) revealed the absence of metastatic disease among cases where H&E of the sentinel node was negative. As expected, the specificity was 100% because no false-positive results occurred. All women affected by Stage IIIC disease were identified using SLN mapping (23).

However, despite the potential benefits of SLN biopsy for EC, concerns remain regarding the false-negative rate mainly due to the bilaterality of lymphatic drainage of the uterus (24,25) as well as the detection and diagnostic accuracy for paraaortic nodes. Indeed, the uterus, as a midline organ, exhibits right and left pathways of lymphatic drainage. Furthermore, lesion located in the fundus may drain via the gonadal vessels to the high paraaortic area, normally difficult to detect trough lymph nodes sentinel mapping. Several studies have been performed to implement the relatively low bilateral detection rate as well as the low diagnostic accuracy rate for paraaortic nodes. To reduce the low bilateral detection rate, Barlin et al. [2012] first proposed an algorithm that takes into account a side specific pelvic lymph nodes dissection in case of non-mapping on a hemipelvis (). This algorithm entailed a significant decrease in the false-negative rate (26). The low diagnostic accuracy rate for paraaortic nodes has no substantial clinical impact. Whereas paraaortic metastases are a poor prognostic indicator, a survival benefit has been reported with paraaortic lymph node dissection (27). The likelihood incidence of isolated paraaortic metastases even among patients with high-risk characteristics in the bilateral absence of pelvic nodes positive is very low (28,29). Therefore, pelvic SLN mapping is likely sufficient in the majority of patients (22).

Figure 1

The algorithm of sentinel lymph node mapping for endometrial cancer. EC, endometrial cancer; SLN, sentinel lymph node; 99mTc, technetium-99m radiocolloid; ICG, indocyanine green; LND, lymph node dissection.

Lymphatic drainage and rationale of SLN mapping

The technique consists of the dye injection into the uterine cervix: there, its penetration towards the region of the uterine vessels and main uterine lymphatic trunks is excellent and leads to pelvic and occasionally paraaortic sentinel nodes because it condenses in the parametria and appears in the broad ligament.

The typical site of pelvic SLN is medial to the external iliac, ventral to the hypogastric, or above the obturator region since the obliterated umbilical artery is commonly crossed over by the uterine body lymphatic trunks.

The SLN can be also seen in the common iliac presacral region when the lymphatic trunks do not cross over the obliterated umbilical and move cephalad following the mesoureter: this location remains not so frequent (30).

Site of injection for sentinel nodes mapping

The adequacy of the injection, the site of injection and the substance injected are fundamental for the rate of detection of SLNs. A dedicated surgical team for a successful SNL mapping is required (31).

It seems to be a lack of consensus about the optimal injection site. Three are the injection sites for sentinel node mapping described: (I) subserosal/myometrial; (II) subendometrial peritumoral (hysteroscopic guided injection); (III) the cervix. The lymphatic drainage is ambiguous and complex and this makes the EC ideal for lymphatic mapping. Along the obturator, internal iliac, external iliac, common iliac, caval, and aortic vessels as well as in the parametrium and presacral space is possible found potential at-risk lymph node basins (31). Two major ways of uterus lymph-node drainage are described. The main one follows the uterine vessels through the parametrium (the one identified through cervical injection); the other follows the ovarian vessels to nodes located into the high para-aortic area. This path of lymphatic drainage is more effectively indicated by the uterine injection technique (32-36). The detection rates of the subserosal myometrial injection technique is around 45% (37). Altgassen et al. [2007] reported the highest detection rate (92%) described for subserosal injection with a new approach consisting of multiple subserosal injection sites (4 ventrally and 4 dorsally) (36). The positive results of the technique indicate that the detection rate increases with the number of injections. Hysteroscopic perilesional injections although more challenging and demanding guarantees a more accurate visualization of tumor lymphatic drainage of lesion located in the fundus (38,39). Martinelli et al. [2017] recently reported a technique consisting in the peri-tumoral injection with ICG; the rate of bilateral SNL detection was 89.5%. Interestingly, 47% mapped to the para-aortic nodes and 74.5% mapped to bilateral pelvic nodes (40). One concern is the risk of spreading malignant cells through the tubes when performing hysteroscopic injection in patients with EC (21). Hysteroscopic visualization of the endometrial cavity at low pressure is recommended to avoid tubal spillage. Gien et al. [2005] reported negative peritoneal washing for endometrial cells after the hysteroscopy (41). A cervical injection of dye into the stroma and submucosal at 3 and 9 o’clock, as previously reported by the pioneer of SLN in EC (42), has the advantage to be easy and less time-consuming. It is associated with an excellent detection rate compared with uterine injection (43) but exposes to the risk to ignore direct para-aortic drainage (44). In conclusion, cervical and corpus injections combined may help in detecting both pelvic and para-aortic basins (45-47).

Tracers for sentinel nodes mapping

Historically, blue dye and technetium-99m radiocolloid (99mTc) were used alone as tracers to map sentinel nodes with the same detection rate (48-50). The combination of the two tracers has been proposed (51). In two studies where a higher detection rate was achieved by the radioactive tracers compared with blue dye alone, no radioactivity was detected in metastatic nodes stained with blue dye (52,53). These results led to the consideration that SLN with combined 99mTc and the blue dye may be an accurate and reasonable surgical option (54,55).

The blue dye injection is a valid and not expensive alternative for SLN mapping. Allergic reactions to the blue dye were observed including anaphylactic shock. The radioactive option needs a preoperative injection of 99mTc and lymphoscintigraphy and an intraoperative gamma probe to detect the hot nodes. ICG is another tracer widely used in gynecological oncology, but the cost of the product is higher and there is the need for a dedicated near-infrared optical filter (54). In particular, the sensitivity and negative predictive value (NPV) of sentinel lymph-node mapping have been evaluated in the Fluorescence Imaging for Robotic Endometrial Sentinel lymph node biopsy (FIRES) trial where the robotic-assisted fluorescence imaging of the tracer ICG was used to detect the lymphatic metastases in women affected by EC, compared with the gold standard lymphadenectomy. The SLN technique showed a sensitivity of 97.2% in identifying nodal metastatic disease (95% CI: 85–100%; McNemar’s P=1). Two hundred fifty-seven out of 258 patients with negative SLN results turned out to have truly negative non-sentinel lymph nodes, resulting in an NPV of 99.6% (95% CI: 97.9–100%). In a post-hoc analysis, pathologically identified SLN specimens were significantly more likely to have metastatic cells than non-SLN specimens [58 (5%) of 1,098 vs. 63 (1%) of 5,416, P=0.0001]. This trial demonstrated how the sentinel-lymph-node biopsy is capable of detecting metastatic disease for EC with a sensitivity similar to that for melanoma, vulvar and breast cancers (56).

Several studies have suggested that the combination of blue dyes and 99mTc is equivalent to fluorescent ICG SLN in terms of accuracy (57-60). In the meta-analysis of Lin et al. [2017], that included 44 studies with a total of 2,236 cases, the overall detection rates and sensitivity of ICG were 93% and 91%, respectively, with an NPV of 96.4% (61). An increasing interest in the use of fluorescent dies has been recently registered for a claimed better detection rate in overweight patients (62). Additionally, the use of ICG has other advantages in comparison with radiocolloids that include less injection pain ease of use, an excellent toxicity profile, and quick visualization (58,63,64). For this reason, IGC should be regarded as the favorite tracer for SLN.

Ultrastaging

An evolving field in the use of SLN is the identification of low volume lymph node metastasis through a histopathological exam of a restricted number of lymph nodes that permits a more detailed examination of the node (ultrastaging). Low volume lymph node metastasis in the form of micrometastases (MMs) and isolated tumor cells (ITCs) can be missed with routine evaluation (65). Micrometastases are defined as a focus of tumor cells measuring more than 0.2 mm and less than 2 mm, whereas ITCs are single cells or microscopic clusters measuring ≤0.2 mm (66). Plante et al. [2017] evaluated the benefit of adjuvant treatment in terms of progression-free survival (PFS) and overall survival in patients with ITCs. The PFS at 36 months for ITCs patients was comparable to patients with negative node and MM (respectively 95.5%, 87.6%, 85.5%) while in patients with macrometastasis was statistically worse (58.6%). The overall survival rate at 36 months was reported to be respectively 92.4%, 87.9%, 87.5% and 75.1% for patients with negative nodes, ITCs, MMs and macrometastasis. The Authors concluded that the excellent outcome of patients with ITCs is independent of additionally adjuvant treatment (3). These low volume metastases could be detected by SLN mapping with ultrastaging, otherwise, they could go undetected with routine evaluation. According to Blakely et al. [2019], SLNs can be examined following Mount Sinai ultrastaging protocol, although the clinical value and the prognostic relevance of this approach is yet unknown. The lymph nodes are cut perpendicularly to the long axis into 3 mm sections, then four levels are cut from each section: levels 1, 3, and 4 are submitted for H&E stain while level 2 is submitted for immunohistochemistry ITCs and MMs, are described as positive. Non-SLN pelvic lymph nodes are cut perpendicularly to the long axis into 3 mm sections. One level is cut from each section and submitted for H&E staining (67).

Conclusions

It is still a subject for debate if there is a benefit (therapeutic value) of full lymphadenectomy in the treatment of EC. If we assume that lymphadenectomy has a pure diagnostic role, a complete lymphadenectomy should increase surgical complications without clear benefits in terms of survival and disease-free interval. The proposed aim of SLN mapping is to reduce the morbidity preserving the diagnostic relevance of the lymph nodes status assessment. To date, no randomized trials have been conducted on this technique in EC. Currently, a clinical trial (NCT03778255) evaluating the diagnostic value of different lymph node tracing methods in detecting SLN metastasis in EC is ongoing with an estimated completion date set for June 2020: the investigators intend to conduct a prospective and multicentric study to evaluate the effectiveness of different mapping methods to achieve a reliable lymph node assessment. Nevertheless, before this procedure can replace the standard lymphadenectomy in the staging of this malignancy further studies are needed clarifying how to achieve the proper selection of the patient to this procedure and identifying how to overcome the factors responsible for a still unsatisfactory diagnostic accuracy.