Sentinel lymph node procedures in women with presumed early stage endometrial cancer

Get Complete Project Material File(s) Now! »

Chapter 2 Sentinel lymph node biopsy procedures in women with presumed early stage endometrial cancer

Introduction

Endometrial cancer is a common gynaecological cancer and in South African women it is the second most common cancer after cervical cancer. In countries with effective population based cervical cancer screening, endometrial cancer is the most common gynaecological cancer diagnosed and treated in gynaecologic oncology units [1]. The age standardised risk in South African women is estimated to be between 5.32 and 6.9/100 000 [2,3]. The estimated endometrial cancer lifetime risk of black South African females is 1:160, of Coloured women 1:125, while Asian and White women have lifetime risks of 1:76 and 1:130 respectively [4].
In developed countries roughly 70% of women with endometrial cancer are diagnosed with stage I disease, and the histological sub-type in more than 80% of cases are endometrioid adenocarcinoma [5]. Racial and population differences have been described [6,7]. Data from the University of Pretoria Gynaecologic Oncology Unit suggest local endometrial cancer figures are different from the developed world, with 62% of patients diagnosed with endometrioid histology and only 43% of women with FIGO stage I disease [8].
Surgery is the preferred treatment in women diagnosed with endometrial cancer. This involves hysterectomy, bilateral salpingo oophorectomy (BSO) and removal of pelvic and para-aortic lymph nodes. Most patients with high risk for recurrent disease, defined as endometrioid adenocarcinoma grade III and all other non-endometrioid histological sub-types, will undergo the abovementioned treatment. If para-aortic lymphadenectomy is performed it should be completed up to the level of the renal veins, as isolated positive para-aortic nodes between the aortic bifurcation and the inferior mesenteric artery is rare.

Lymphadenectomy in patients with presumed early stage endometrial carcinoma

In patients with presumed early stage low risk for recurrence disease, defined as endometrioid adenocarcinoma grade I and grade II, less than 50% myometrial invasion and tumour size less than 20 mm, lymphadenectomy can be omitted. The rationale is that if all three these parameters are present in the same patient, the risk for nodal metastases is less than 1% [9]. The caveat to consider is that the identification of this low risk group in the referenced study from the Mayo Clinic, was performed by means of very reliable intra-operative frozen section examination of the uterus, and lymphadenectomy was performed or omitted following information obtained from this frozen section examination. In most institutions frozen section examination is either not available or not as accurate and reliable as is the case at the Mayo Clinic who reported 89% accuracy of frozen section examination.
In the absence of intra-operative reliable and accurate frozen section examination, decisions on omission of lymphadenectomy is made on histological grade and pre-operative imaging with regard to tumour size and depth of myometrial invasion. This strategy however will result in missing the diagnosis of lymph node metastases in up to 11% of so-called low risk patients, and under staging of disease in up to 25% of cases [10-13]. In the absence of reliable intra-operative frozen section examination of the uterus, the diagnosis of low risk endometrial cancer becomes a retrospective diagnosis and theoretically means that all patients with endometrial cancer will require at least a full pelvic lymphadenectomy.
On the other hand, lymphadenectomy is associated with morbidity such as lymphocysts and lower limb oedema [14]. In addition, two prospective randomised trials and a Cochrane meta-analysis have failed to show survival benefit of lymphadenectomy in patients with endometrial cancer [15-17]. The low risk of lymph node metastases in FIGO stage I disease also implies that the vast majority of women will undergo this treatment unnecessary and without benefit if lymphadenectomy is performed as standard management in all patients with endometrial cancer.
The sentinel lymph node (SLN) algorithm has the potential to be a reasonable solution for the controversy regarding lymphadenectomy in women with possible low risk for recurrent endometrial cancer. It might also potentially provide a better alternative than complete omission of pelvic lymphadenectomy.

Literature overview

Sentinel lymph node biopsy and the sentinel lymph node algorithm

The idea of SLNs in women treated for endometrial cancer was first described by Burke et al in 1996 [18]. Isosulfan blue dye was injected into the fundus and posterior and anterior midline in the uteri of 15 women who underwent surgery for endometrial cancer. At least one SLN was detected in 67% of cases by means of this procedure.
In many populations more than 70% of women diagnosed with endometrial cancer will present with FIGO stage I disease. Pelvic lymph nodes will have metastases in 6% of women with FIGO stage IA moderately differentiated and 10% with poorly differentiated endometrioid adenocarcinomas respectively (5). These figures indicate that many women with presumed early stage endometrial cancer undergo lymphadenectomy without benefitting from the procedure.
There is no consensus on the role and place of lymphadenectomy in women treated for endometrial cancer. In addition, pelvic lymphadenectomy in these women has not been well-defined or standardised, and practices in this regard varies considerably, ranging from complete omission of lymph adenectomy in some cases based on pre-operative histological and imaging findings, to various practices of node sampling and systematic pelvic lymphadenectomy with or without para-aortic lymphadenectomy on all women. The extent of systematic lymphadenectomy varies from pelvic alone to pelvic plus para-aortic below the superior mesenteric artery or para-aortic up to the renal vessels.
The effect of para-aortic lymphadenectomy on survival in women with endometrial cancer remains unclear, with some data suggesting improved survival [19]. Abu-Rustum et al published data on 1942 women treated for endometrial cancer and according to this prospectively maintained database only 1.6% of women with negative pelvic nodes will have positive para-aortic nodes. This low rate seems to be consistent for both low and high-grade histology-type tumours [20].
Ballester et al published the prospective multicentre SENTI-ENDO study in 2011 assessing SLNB in 113 women with FIGO stage I to II endometrial cancer [21]. Cervical injections with technetium the day before surgery and blue dye injected into the cervix pre-operatively after induction of anaesthesia were used and ultrastaging of negative SLNs was performed. Three patients had false negative SLNs giving a NPV of 97% (95% CI 91% – 99%) and a sensitivity of 84% (95% CI 62% – 95%). Using the hemi-pelvis as the unit of analysis, the NPV was 100% (95% CI 95% – 100%) and the sensitivity 100% (95% CI 63% – 100%). The long-term results of this trial were published in 2015 [22].
Levinson et al reviewed 19 studies on SLN dissection in women with endometrial cancer in 2013 [23]. Overall detection rates ranged from 62% to 100%, the false negative rate (FNR) ranged from 0% to 50%, and the NPV from 95% to 100%. No technique was superior with regard to the type of surgery modality, injection used, injection site, or pathology techniques.
Different injection sites for detection of SLNs in endometrial cancer have been investigated. These include the cervix, the fundus of the uterus as well as hysteroscopic peri-tumoural injections. Khoury-Collado et al published a literature review on lymphatic mapping in endometrial carcinoma in 2008 [24]. In seven studies using the subserosal myometrium of the uterine corpus as the injection site, SLN detection rates varied between 0% and 92%. Seven other studies employing the cervix as the injection site, reported detection rates varying between 84% and 100%, while in five studies reporting on the hysteroscopic peri-tumoural or endometrial injection, detection rates varied between 0% and 100%. The largest study in this review included 28 patients. In a prospective study of 80 women who had hysteroscopic injection of technetium followed by systematic pelvic and para-aortic lymph adenectomy, Solima et al observed a sensitivity of 90% (95% CI 55.5% – 95.5%) and an NPV of 98% (95% CI 89.4% – 100%) [25]. Other studies have also demonstrated less impressive figures at injection sites other than the cervix [26-28].
Barlin et al suggested an endometrial cancer SLN algorithm and determined its effectiveness in 498 patients with early stage endometrial cancer [29] (Figure 1). The algorithm is similar to the one used in cervical cancer, consisting of peritoneal washings, removal of all mapped SLNs with ultrastaging in nodes negative after H&E examination, removal of any suspicious enlarged nodes regardless of mapping, and side specific lymphadenectomy in the absence of SLN mapping on a hemi-pelvis. Para-aortic lymphadenectomy was performed at the discretion of the attending physician. Blue dye injected into the cervix was used as the only mapping method. At least one SLN was detected in 81% (n =of women of which 96% mapped only to the pelvis and 0.5% had SLNs only in the para-aortic region; The FNR was 14.9%. Applying the endometrial cancer algorithm resulted in a sensitivity of 98.1%, a FNR of 1.9%, and an NPV of 99.8%.
According to results of a study from Raimond et al, SLN biopsy is able to detect lymphatic disease in three times more women compared to conventional standard lymphadenectomy [31]. In this retrospective multi-centre study involving 304 women, the SLN procedure detected disease in 16.2% versus 5.1% detected by lymphadenectomy. The difference was attributable to ultrastaging of negative SLNs following H&E examination.
Cormier et al published a systematic review on the SLN procedure in endometrial cancer. The review included seventeen studies with each more than 30 patients, totalling 1 572 patients, and overall detection rates ranged from 60% to 100%. In studies with more than 100 patients the overall detection rate was more than 80%. When the algorithm was applied retrospectively, the sensitivity was 95%, the negative predictive value was 99% and the false negative rate was 5% [32]. Lin et al published a systematic review and meta-analysis on 44 studies. The pooled overall detection rate was 83% with bilateral detection of 56% and the pooled sensitivity was 91% but the false negative rate was not reported [33].
A systematic review and meta-analysis by Bodurtha Smith et al included 55 studies with 4 915 women and reported a detection rate of 81%, bilateral detection rate of 50% and sensitivity of 96%. The conclusion was that sentinel lymph node mapping may be considered an alternative standard of care in the staging of women with endometrial cancer [34]. Findings of the literature reviews and meta-analysis discussed here have also been confirmed by other reviews on this topic [35,35].

Declaration 
Ethics approval certificates
Acknowledgements 
Abstract 
List of Tables 
List of Figures 
List of abbreviations 
Chapter 1: Overview
1. Introduction
2. Cancer staging
2.1. Cervical cancer staging
2.2. Endometrial cancer staging
3. Information on lymph node status
4. Alternative assessment options of pelvic lymph nodes
4.1. Imaging
4.2. Sentinel lymph node biopsy
5. Research problem
6. Aims and objectives
6.1. Aims
6.2. Objectives
7. Materials and methods
7.1. Study population and setting
7.2. Inclusion criteria
7.3. Exclusion criteria
7.4. Data collection
7.5. Statistical analysis
7.5.1. Sentinel lymph node biopsy
7.5.2. Calculations
7.6. Sample size
7.7. Ethical considerations
7.8. Definitions
8. Results
8.1. Demographic data
8.2. HIV infection
8.3. Sentinel nodes
8.4. Nodal metastases
8.5. Sensitivity, specificity, and detection rates
9. Discussion
References
Chapter 2: Sentinel lymph node procedures in women with presumed early stage endometrial cancer
1. Introduction
1.1. Lymphadenectomy in patients with presumed early stage endometrial cancer
2. Literature overview
2.1. Sentinel lymph node biopsy and the sentinel lymph node, algorithm
2.2. Tracers and injection sites for sentinel lymph node mapping
2.3. Imaging of pelvic lymph nodes for detection of metastases
3. Aims and objectives
3.1. Aims
3.2. Objectives
4. Material and methods
4.1. Setting and study population
4.2. Inclusion criteria
4.3. Exclusion criteria
4.4. Pre-operative evaluation
4.5. Positron emission tomography/computed tomography
4.6. 99Technetium and lymphoscintigraphy
4.7. Intra-operative lymph node mapping
4.8. Surgical treatment
4.9. Histological examination
4.9.1. Frozen section examination
4.9.2. Routine haematoxylin and eosin examination
4.10. Ethical considerations
5. Results
5.1. Demographic data
5.2. Disease characteristics
5.3. Sentinel lymph node detection
5.4. Lymph node disease status
5.5. 18Fluoro-deoxy-glucose positron emission tomography / computed tomography scan
5.6. 99Technetium and lymphoscintigraphy
5.7. Sensitivity, specificity, positive and negative predictive value for the sentinel lymph node algorithm
6. Discussion
7. Conclusion
References
Chapter 3: Sentinel lymph nodes in women with cervical cancer
1. Introduction
2. Sentinel lymph node literature overview
3. Aims and objectives
4. Materials and methods
5. Results
6. Discussion
7. Conclusion
References
Chapter 4: The ability of 18Fluoro-deoxy-glucose positron emission tomography/computed tomography scan to accurately predict pelvic nodal status in women with uterine cancer
1. Introduction
2. Rationale for the study
3. Aim and objectives
4. Materials and methods
5. Results
6. Discussion
7. Conclusions
References
Chapter 5: High-risk human papilloma virus DNA in sentinel lymph nodes in women with cervical cancer
1. Introduction
2. Literature review
3. Rationale for the study
4. Aims and objectives
5. Materials and methods
6. Results
7. Discussion
8. Conclusion
References
Chapter 6: Analysis of women in whom sentinel lymph nodes were detected and the impact of different histological examinations of sentinel lymph nodes
1. Introduction
2. Histological evaluation of sentinel lymph nodes
3. Aims and objectives
4. Materials and methods
5. Results
6. Discussion
7. Conclusions
References
Chapter 7: Summary and recommendations
GET THE COMPLETE PROJECT