Keywords atypical endometrial hyperplasia - endometrial cancer - sentinel node biopsy - indocyanine green - surgery
Schlüsselwörter atypische Endometriumhyperplasie - Endometriumkarzinom - Sentinel-Lymphknoten-Biopsie - Indocyaningrün - chirurgischer Eingriff
Introduction
The publication of the Cancer Genome Atlas in 2013 marked a significant milestone in the understanding of cancer biology, particularly in the context of endometrial carcinoma [11 ]. This landmark study led to the establishment of novel molecular classifications, representing a paradigm shift from the traditional dualistic view of the disease [22 ]. Concurrently, there has been a notable trend toward reducing lymphadenectomy procedures, as evidenced in the German S3 Level guideline for endometrial cancer (EC) [33 ]
[44 ].
The introduction of sentinel lymph node biopsy (SNB) for intermediate-risk EC has emerged as a pivotal technique for providing a comprehensive means of lymph node staging while mitigating the morbidities associated with systematic lymphadenectomy. This innovative approach represents a substantial stride toward personalized, patient-centered care. The FIRES trial reported a negative predictive value of 99.6% (97.9–100%) for patients diagnosed with EC who underwent pelvic SNB followed by pelvic lymphadenectomy, with or without paraaortic lymphadenectomy [55 ].
In the realm of advanced endometrial carcinoma, the advent of molecular classification systems has catalyzed the development of personalized immunotherapeutic modalities. These treatments, such as dostarlimab and pembrolizumab, offer promising prospects for patients with advanced-stage EC [66 ]
[77 ]
[88 ].
However, the role for SNB in the context of early-stage endometrial carcinoma (International Federation of Gynecology and Obstetrics [FIGO] Grade Ia [G] 1–2) in the absence of p53 mutations and for atypical endometrial hyperplasia is still not finally determined [99 ]. The German S3 Level guideline for EC, which was recently updated, recommends that a sentinel node biopsy can be performed, and this is also in line with international recommendations in which SNB can be considered in patients with low-risk or intermediate-risk disease [1010 ]
[1111 ]. This is against the background of a 15% likelihood of lymph node metastases across all cancer disease stages — underscoring the clinical necessity for a more refined approach to lymph node evaluation [1212 ]. Moreover, the final histology in patients initially found to have AEH reveals EC in up to 60% of cases [1313 ]. Furthermore, 22% of patients initially classified as having FIGO I were found to have higher tumor stages postoperatively [1414 ]. This highlights discrepancies between preoperative and postoperative staging.
Indocyanine green (ICG) has gained recognition as a reliable SNB mapping modality, and it is now included in clinical practice guidelines [33 ]
[44 ]. Several studies have examined lymph node detection rates in patients with EC treated with ICG, reporting negative predictive values ranging from 99% to 99.5% [1515 ]
[1616 ]. The results of the SENTI-ENDO trial led to the analysis of early-stage EC patients in the trial, since EC patients with FIGO stages I and II were included. The authors found that SNB up-staged the final cancer stage in 10% of patients at low risk and in 15% of those at intermediate risk [1717 ].
The present study aimed to address three major topics. Firstly, the diagnostic FIGO stage, determined using vaginal sonography and gynecologic examination, was correlated with the final pathologic stage. The frequency of final tumor stage upgrades and how often the tumor grade changes influence final therapy decisions. Secondly, the feasibility and efficacy of laparoscopic SNB for AEH was to be examined; and thirdly, the feasibility and efficacy of laparoscopic SNB for early-stage endometrial carcinoma was to be assessed. In addition, clinical outcome parameters were evaluated.
Materials and Methods
Patient cohort
This retrospective, single-center cohort study included the period from January 2018 to December 2023. A total of 102 patients with AEH (n = 20) and early-stage EC (n = 82) according to diagnostic curettage were treated with laparoscopic pelvic sentinel node biopsy during the final operation. Patients with other malignant tumors, such as vaginal carcinoma or cervical cancer, and one duplicate dataset were excluded. In addition, only patients with early-stage FIGO Ia and Ib EC or AEH were included ([Fig. 1Fig. 1 ]).
Fig. 1
Fig. 1
CONSORT statement.
Vaginal ultrasonography and gynecological examination were used to define the FIGO stage for EC, and also for patients with AEH at the initial histology. Possible myometrial invasion is assessed using ultrasound. Hysteroscopy and curettage were performed in all patients to obtain pretherapeutic biopsies. All patients were treated in accordance with the German S3 Level guideline for EC, with hysterectomy and bilateral adnexectomy or salpingectomy if indicated in addition to SNB and, if indicated, with radical pelvic and paraaortic lymphadenectomy [33 ]
[44 ]. The SNB procedures were performed with ICG, following the algorithm set out in the German guideline, with intracervical injection of ICG at two to four points [33 ]
[44 ]. Systematic lymphadenectomy was not performed for patients with AEH if sentinel node mapping was not possible. This approach was reserved only for patients with EC who were found to have higher disease stages during surgery. The study was approved by the local ethics committee (ethical approval number: 252_20 B, first approval on June 29, 2020).
Collection of clinical and histopathological data
Clinical, histopathological, and follow-up data were prospectively documented as part of routine care at the University Gynecological Oncology Center in Erlangen. Data on tumor type, tumor grade, and molecular analysis, such as p53 mutation status, mismatch repair (MMR) protein status, and DNA sequencing for the POLE exonuclease domain, were obtained from the pathology files. The latest WHO classification for endometrial hyperplasia, published in 2014, was used [1818 ]. The new (revised 2020) FIGO/TNM classification of EC was used to define tumor stage [1919 ].
All surgical procedures were exclusively performed at the University Gynecological Oncology Center in Erlangen. SLB was conducted following the protocols described in the German S3 Level guideline for endometrial cancer [33 ]
[44 ].
Statistical analysis
Data management was facilitated using the Microsoft Access database. Descriptive analysis was conducted using IBM SPSS Statistics for Windows, Version 28.0 (IBM Corp., released 2021, Armonk, New York).
Results
The analysis included 102 patients with a mean age of 63 years who had been diagnosed with atypical endometrial hyperplasia or EC. Their mean body mass index (BMI) was 33. In the final pathologic staging, over 75% of the lesions were stage pT1a or pT1b, with 92.2% pN0. The predominant histological type was endometroid adenocarcinoma, at 88.2%. The final grades resulting from the hysterectomy were grade 1 in 37.2%, grade 2 in 46.1%, and only 7.8% in grade 3 ([Table 1Table 1 ]).
Table 1
Table 1
Demographic and clinical characteristics.
Characteristic
All patients
(n = 102)
Abbreviations: BMI = Body Mass Index; FIGO = The International Federation of Gynecology and Obstetrics; n = number
Age (Mean)
63.38
BMI (Mean)
33.0
Final Pathological Stage (%)
pT1a
54 (52.9)
pT1b
24 (23.5)
pT2
13 (12.7)
pT3a
2 (2.0)
Nodal status (%)
pN0
94 (92.2)
pN1
3 (2.9)
pNx
4 (3.9)
Histology
Atypical endometrial Hyperplasia
9 (8.8)
Endometrioid Adenocarcinoma
90 (88.2)
Uterine serous carcinoma
2 (2.0)
Adenosquamous endometrial cancer
1 (1.0)
Final Grading
Grade 1
38 (37.3)
Grade 2
47 (46.1)
Grade 3
8 (7.8)
Sentinel node biopsy for patients with atypical endometrial hyperplasia
As [Table 2Table 2 ] shows, no lymph node metastases were detected in the groups that were initially diagnosed with AEH. Bilateral SNL detection was also possible in all but one of the patients (95%).
Table 2
Table 2
Sentinel node detection rates and metastases.
Endometrial cancer
(n = 82)
Atypical endometrial hyperplasia
(n = 9)
Endometrial cancer (Atypical endometrial hyperplasia presurgery)
(n = 11)
SNL detection
n (%)
n (%)
n (%)
Abbreviations: n = number; PLD = Pelvic Lymphadenectomy; SNL = Sentinel lymph node
Unilateral SNL detection
1 (1.2)
0
1 (9.1)
Bilateral SNL detection
77 (94.9)
9 (100)
10 (90.9)
Total detection rate
78 (95.1)
9 (100)
11 (100)
No SNL detection
4 (4.9)
0
0
Surgical Lymph node assessment
SNL mapping only
76 (92.7)
9 (100%)
11 (100)
SNL mapping and PLD
6 (7.3)
0
0
Lymph node metastases
3
0
0
Sentinel node biopsy for patients with endometrial cancer
Among the 82 patients diagnosed with EC at the initial curettage, three (3.6%) were found to have lymph node metastases. Bilateral lymph node mapping was possible in 77 patients (95.1%); only unilateral SNL detection was possible in one patient (1.2%); and SNL detection was not successful in four patients ([Table 2Table 2 ]). Notably, these patients did not warrant systematic lymphadenectomy due to their presurgical FIGO stage. Among the three patients with lymph node metastases, the first and second underwent pelvic lymphadenectomy; the ascites cytology was positive, and adjuvant chemotherapy was therefore administered. In the third patient with a positive sentinel lymph node, a systematic pelvic and paraaortic lymphadenectomy was performed.
FIGO stage and tumor grade changes
A significant finding was that 11 of the 20 patients (55%) who had initially been diagnosed with AEH were found to have EC at the final pathology ([Table 3Table 3 ]).
Table 3
Table 3
Change of presurgical FIGO classification and postsurgical pathological classification for patients with endometrial cancer in final pathology.
Number of patients
(n = 93)
Abbreviations: FIGO = The International Federation of Gynecology and Obstetrics; G = grade; n = number
No FIGO stage available
1
FIGO IA
53
FIGO IA to pT1a
43
Change FIGO IA to pT1b
6
Change FIGO IA to pT2 or higher
4
FIGO IB
39
FIGO IB to pT1b
18
Change FIGO IB to pT1a
10
Change FIGO IB to pT2 or higher
11
Change of histological type (%)
Atypical endometrial hyperplasia
20
No change
9 (45)
Change to endometrial cancer
11 (55)
Change of histological grade (%)
(n = 82)
No grading available in curettage
2
No change for G1
22 (58.9)
G1 to G2
14 (36.8)
G1 to G3
2 (5.2)
No change for G2
31 (77.5)
G2 to G1
6 (15)
G2 to G3
3 (7.5)
No change for G3
2
A change in the FIGO stage occurred in 31 patients (33.7%), aggregated over all stages. The diagnostic FIGO stage and final pathologic tumor stage differed for patients with FIGO stage Ia (n = 53): six patients (11.3%) were upgraded to pT1b, and four patients (7.5%) were upgraded to pT2 or higher. In patients with FIGO stage Ib disease (n = 39), the postsurgical tumor stage was downgraded to pT1a in 10 patients (25.6%) and upgraded to pT2 or higher in 11 patients (28.2%) ([Table 3Table 3 ]).
Another important factor for therapy decisions is tumor grade. The grading was determined for the curettage and final pathology findings, and the results differed in 33.3% of the patients. In the group with grade 1 tumors, 12 of 38 patients (31.6%) had the tumor grade increased to grade 2, and two tumors (5.3%) were found to be grade 3 at the final histology. Most grade 2 tumors were confirmed as grade 2, in 31 patients (77.5%); they were downgraded to grade 1 in six patients (15%), and upgraded to grade 3 in three patients (7.5%) ([Table 3Table 3 ]). Those of clinical significance were mostly changes from low-grade to high-grade, which occurred in five patients (6.2%).
Surgical outcomes
The duration of the operations ranged from 38 to 268 minutes, with a mean of 97.5 minutes. The mean duration of laparoscopic operations alone was shorter, at 92.3 minutes. The patients’ mean hospital stay was shorter with laparoscopic procedures (3.6 days) than for the overall procedure (4.4 days). Postoperative outcomes were documented, including the absence of lymphedema and typical ranges for drain fluid volume, pain scores, and antibiotic use ([Table 4Table 4 ]). Eight patients underwent conversion from laparoscopic surgery to laparotomy. Conversion to laparotomy took place because the FIGO stage was higher than expected, the uterus was larger, vaginal constriction occurred, or the patient was overweight and had a positive SNB.
Table 4
Table 4
Postsurgical outcome.
All patients
(n = 102)
Abbreviations: max = maximum; min = minimum; n = number; NRS = numeric rating scale
* Surgery with conversion to open laparotomies for systematic lymphadenectomy is included
Time of surgery (minutes; min to max)
97.58 (38 to 268*)
Hospitalization (days, min to max)
4.38 (2 to 35)
Documented Lymphedema (n)
0
Mean total amount of drain fluid (ml)
312.63
Postsurgical antibiotic therapy (n)
7
Conversion to laparotomy (n)
8
Mean Pain Score (NRS 1 to 10)
2.52
Discussion
The findings of this study contribute to understanding the management of early-stage endometrial carcinoma and atypical endometrial hyperplasia (AEH) in the context of pelvic sentinel node biopsy (SNB). The data show the efficacy and feasibility of this approach, particularly in relation to clinical outcome parameters such as surgical duration, drainage fluid, and hospitalization period. The conversion rate from AEH to EC in 55% of the patients is in line with findings in the literature of between 42.6% and 60% — emphasizing the clinical importance of close monitoring and aggressive management in AEH patients [1313 ]
[2020 ].
One critical aspect of the study’s findings is the discrepancy observed between the preoperative FIGO stage and the final pathological stage. The upgrades from FIGO stage Ia to pT1b and higher in a notable percentage of patients reflect the complexities of EC staging, as noted by Creasman et al. [1414 ].
The 94.1% success rate in bilateral sentinel node mapping in the present study and the low complication rate are consistent with recent literature reports advocating the accuracy and safety of SNB in treating EC [1515 ]
[2121 ]. These results further endorse the effectiveness of SNB in accurately staging early-stage EC, reducing the need for more invasive procedures and associated morbidities.
The integration of molecular profiling into EC management offers a promising avenue for enhancing the precision of staging and personalizing treatment strategies [11 ]. This approach is especially relevant in a setting in which conventional staging may underestimate the severity of disease.
The emerging therapeutic options for treating EC — pembrolizumab with or without the combination of lenvatinib, or dostarlimab as monotherapy, or both immunotherapies in combination with standard chemotherapy — present promising treatment options, especially in patients with advanced-stage EC [66 ]
[77 ]
[2222 ]
[2323 ]. These advances in targeted therapies and immunotherapies are indicative of a shift toward personalized medicine in oncology; precise lymph node staging is therefore crucial.
Nevertheless, none of the patients initially diagnosed with AEH had any lymph node metastases, a finding that is in line with recent literature reports. In the 2021 study by Sullivan and coworkers, for example, 141 patients with preinvasive endometrial lesions were diagnosed. Fifty-one patients (36%) had a final diagnosis of cancer, a rate that is slightly lower than the cancer rate of 56% in the present study; the majority of the patients (96%) had stage IA grade 1 endometrioid carcinomas. Among these patients, seven (5%) met the Mayo criteria of 2000 used for SNB and did not have any lymph node metastases [2424 ]
[2525 ]. In the present cohort, none of the 20 patients with AEH met the criteria for SNB in accordance with recent guidelines, as all of the carcinomas were classified as pT1a grade 1, with one exception being pT1a grade 2 [33 ]
[44 ]. Another study described multiple models estimating the risk for lymph node metastasis in patients with AEH and reported that the estimated risk of lymph node spread was 1.6–2.1% for all women with a preoperative diagnosis of AEH and 4.4–6.8% for the 55 women with EC included [2626 ]. A recent meta-analysis of 1044 patients with AEH showed comparable unilateral and bilateral detection rates (89% and 79%) and a lymph node involvement rate of less than 2% — again underlining the view that SNB needs to be carefully indicated in patients with AEH [2727 ].
In contrast, for low-risk EC, Convery and coworkers found lymph node metastases in 1.6% of patients with grade 1 or 2 endometrial carcinoma [2828 ]. A post hoc analysis of the GOG LAP2 study also reported a 1% rate of lymph node metastases in 389 EC patients [2929 ]. In both studies, low risk was assessed using the Mayo criteria: < 50% myometrial invasion, tumor size < 2 cm, and grade 1 or 2 [2525 ]
[2929 ]. In the cohort described, 41 patients with pT1a grade 1 or 2 EC in the final pathology report did not have any lymph node metastases, but the indication for at least SNB depended on the molecular pathology in one patient who had a grade 3 tumor according to the final pathology findings.
Limitations
The retrospective nature and single-center design of the present study limit the extent to which its findings can be generalized. The absence of systematic lymphadenectomy in patients with unsuccessful SLN mapping might affect staging accuracy. In addition, only iliac sentinel nodes were detected and resected, and no sacral or paraaortic nodes, which represents a selection bias in comparison with literature. Potential selection bias might be present due to omission of AEHs that were not included in the study because a sentinel node biopsy was not performed. Furthermore, the number of patients with AEH in this trial was only 20, so that only limited conclusions can be drawn. Future studies, particularly prospective and multicenter trials, are essential for further validation of the role of SNB in early EC.
Conclusion
Sentinel node biopsy (SNB) can be safely performed in patients with precursor lesions and early-stage EC without notably extending surgical times or increasing postoperative morbidity. It can be discussed as a safe method with patients diagnosed with AEH, but it appears to have a fairly small impact on these patients, and it seems to be safe for them not to undergo SNB. In addition, variations in presurgical grading and FIGO staging, in both directions, suggest the potential need for tailored treatment strategies.
Funding: This research received no external funding.
Informed Consent Statement: Informed consent was obtained from all the patients involved in the study.
Institutional Review Board Statement: The ethics committee of Erlangen University Hospital provided approval for the study (ref. number 252_20_B).
Data availability statement: The datasets are available from the corresponding author upon reasonable request.
