Can tests to identify the main draining lymph nodes in women with endometrial cancer accurately diagnose if the cancer has spread to the lymph nodes?

The issue
Women who have endometrial cancer (cancer arising from the lining of the womb) may have cancer cells spread to lymph nodes in the pelvis and/or the para-aortic area (main blood vessel in the upper abdominal cavity) and need additional radiotherapy treatment, with or without chemotherapy, following initial hysterectomy. Previous studies looked at the effect of removing as many of the pelvic lymph nodes as possible (lymphadenectomy), but did not show a benefit to survival. However, lymphadenectomy often causes problems with lymphoedema (swelling of the legs) and lymphocysts (collections of lymph fluid). Sentinel lymph node (SLN) biopsy (SLNB) involves identifying the first SLN draining the tumour, using a dye or radioactive tracer, or both. The SLNs are removed and examined under a microscope to check for cancer cells. Additional treatment decisions may change depending on these findings.

Why is this review important?
Several studies used dyes or traceable agents to identify SLNs in women with womb cancer. It is not clear whether all of these agents are sufficiently accurate to predict which women have cancerous spread to lymph nodes and whether it is better to inject dye to the cervix or the muscle of the womb. This review summarises the evidence and produces overall estimates of the relative accuracies of the available tests.

How was the review conducted?
We included studies that tested the accuracy of tracer agent/s to identify SLN against the standard method of removing all pelvic nodes, with or without para-aortic lymph nodes. We restricted the studies to those that examined the SLNs by taking multiple slices of the node and staining for cancer cells using antibody markers (immunohistochemistry (IHC)) before examining them under a microscope (ultrastaging).

What are the findings?
We included 33 studies (2237 women) that evaluated any techniques used to identify the SLN draining from the womb. These included: 11 for blue dye only; four for technetium-99m (a radioactive substance) only; nine for indocyanine green (ICG) dye that fluoresces under near infra-red light; 12 that used a combination of blue dye and technetium-99m and one that used a combination of ICG and technetium-99mand where tracers were injected into the cervix (neck of the womb) or directly into the muscle of the womb, or a combination. Overall, the methodological reporting was poor, which limited our ability to assess the quality of the studies.

Tests have two attributes.

1. The ability of the tests to find the SLN (detection rate) varied, with the blue dye test only detecting SLN in 77.8% of women, compared with 80.9% with technetium-99m alone, 86.3% for combined blue dye/technetium-99m, 92.4% for ICG alone, 96.7% for ICG/blue dye and 100% for ICG/technetium-99m. If SLN are not detected, they cannot be examined for cancer cells; therefore, these women may either need to undergo lymphadenectomy or treated based on the risk factors of the cancer within the womb.

2. If cancer cells were present in lymph nodes, did the SLNB identify a lymph node with cancer within it or did it miss a lymph node that contained cancer? This is called the sensitivity of the test and the ability of the test to avoid a false-negative result. If a SLN was found, all tests can identify cancer in the pelvic/para-aortic nodes with good accuracy (more than 90% of nodes with cancer will be accurately identified with any of the tests). In this setting a false-positive result cannot occur, as the histological examination of the SLN is unchanged by the results from any additional nodes removed at systematic lymphadenectomy.

What does this mean?
All of the different methods were able to identify a SLN, although techniques that used ICG, either alone or in combination with blue dyes or technetium-99m, may be more likely to find a node. If a node was found, this was likely to be the one that contained cancer cells; if the SLN did not contain any cancer cells (negative-SLN), the chance that other nodes contained cancer cells was less than 10%. However, these tests were limited to women who were likely to have early-stage disease and only small volume of cancer cells within lymph nodes. Where nodes or lymphatic channels contain many cancer cells, this can block lymphatic channels, affecting lymph drainage, and may have been a reason for failure to find SLN in some of the studies.

This review only looked at the accuracy of being able to identify a SLN and does not tell us whether this offers additional survival advantage to women with early-stage womb cancer. Other types of clinical trials are needed to show us whether performing a SLNB, to guide need for additional treatment, improves survival compared to basing treatment decisions on risk factors determined from examination of the womb alone. This would be important to know, as previous studies have not demonstrated a survival advantage to removing all of the pelvic lymph nodes.

Authors' conclusions: 

The diagnostic test accuracy for SLNB using either ICG alone or a combination of a dye (blue or ICG) and technetium-99m is probably good, with high sensitivity, where a SLN could be detected. Detection rates with ICG or a combination of dye (ICG or blue) and technetium-99m may be higher. The value of a SLNB approach in a treatment pathway, over adjuvant treatment decisions based on uterine factors and molecular profiling, requires examination in a high-quality intervention study.

Read the full abstract...

Pelvic lymphadenectomy provides prognostic information for those diagnosed with endometrial (womb) cancer and provides information that may influence decisions regarding adjuvant treatment. However, studies have not shown a therapeutic benefit, and lymphadenectomy causes significant morbidity. The technique of sentinel lymph node biopsy (SLNB), allows the first draining node from a cancer to be identified and examined histologically for involvement with cancer cells. SLNB is commonly used in other cancers, including breast and vulval cancer. Different tracers, including colloid labelled with radioactive technetium-99, blue dyes, e.g. patent or methylene blue, and near infra-red fluorescent dyes, e.g. indocyanine green (ICG), have been used singly or in combination for detection of sentinel lymph nodes (SLN).


To assess the diagnostic accuracy of sentinel lymph node biopsy (SLNB) in the identification of pelvic lymph node involvement in women with endometrial cancer, presumed to be at an early stage prior to surgery, including consideration of the detection rate.

Search strategy: 

We searched MEDLINE (1946 to July 2019), Embase (1974 to July 2019) and the relevant Cochrane trial registers.

Selection criteria: 

We included studies that evaluated the diagnostic accuracy of tracers for SLN assessment (involving the identification of a SLN plus histological examination) against a reference standard of histological examination of removed pelvic +/- para-aortic lymph nodes following systematic pelvic +/- para-aortic lymphadenectomy (PLND/PPALND) in women with endometrial cancer, where there were sufficient data for the construction of two‐by‐two tables.

Data collection and analysis: 

Two review authors (a combination of HN, JM, NW, RG, and WH) independently screened titles and abstracts for relevance, classified studies for inclusion/exclusion and extracted data. We assessed the methodological quality of studies using the QUADAS‐2 tool. We calculated the detection rate as the arithmetic mean of the total number of SLNs detected out of the total number of women included in the included studies with the woman as the unit of analysis, used univariate meta‐analytical methods to estimate pooled sensitivity estimates, and summarised the results using GRADE.

Main results: 

The search revealed 6259 unique records after removal of duplicates. After screening 232 studies in full text, we found 73 potentially includable records (for 52 studies), although we were only able to extract 2x2 table data for 33 studies, including 2237 women (46 records) for inclusion in the review, despite writing to trial authors for additional information.

We found 11 studies that analysed results for blue dye alone, four studies for technetium-99m alone, 12 studies that used a combination of blue dye and technetium-99m, nine studies that used indocyanine green (ICG) and near infra-red immunofluorescence, and one study that used a combination of ICG and technetium-99m. Overall, the methodological reporting in most of the studies was poor, which resulted in a very large proportion of 'unclear risk of bias' ratings.

Overall, the mean SLN detection rate was 86.9% (95% CI 82.9% to 90.8%; 2237 women; 33 studies; moderate-certainty evidence). In studies that reported bilateral detection the mean rate was 65.4% (95% CI 57.8% to 73.0%) . When considered according to which tracer was used, the SLN detection rate ranged from 77.8% (95% CI 70.0% to 85.6%) for blue dye alone (559 women; 11 studies; low-certainty evidence) to 100% for ICG and technetium-99m (32 women; 1 study; very low-certainty evidence). The rates of positive lymph nodes ranged from 5.2% to 34.4% with a mean of 20.1% (95% CI 17.7% to 22.3%).

The pooled sensitivity of SLNB was 91.8% (95% CI 86.5% to 95.1%; total 2237 women, of whom 409 had SLN involvement; moderate-certainty evidence). The sensitivity for of SLNB for the different tracers were: blue dye alone 95.2% (95% CI 77.2% to 99.2%; 559 women; 11 studies; low-certainty evidence); Technetium-99m alone 90.5% (95% CI 67.7% to 97.7%; 257 women; 4 studies; low-certainty evidence); technetium-99m and blue dye 91.9% (95% CI 74.4% to 97.8%; 548 women; 12 studies; low-certainty evidence); ICG alone 92.5% (95% CI 81.8% to 97.1%; 953 women; 9 studies; moderate-certainty evidence); ICG and blue dye 90.5% (95% CI 63.2.% to 98.1%; 215 women; 2 studies; low-certainty evidence); and ICG and technetium-99m 100% (95% CI 63% to 100%; 32 women; 1 study; very low-certainty evidence). Meta-regression analyses found that the sensitivities did not differ between the different tracers used, between studies with a majority of women with FIGO stage 1A versus 1B or above; between studies assessing the pelvic lymph node basin alone versus the pelvic and para-aortic lymph node basin; or between studies that used subserosal alone versus subserosal and cervical injection. It should be noted that a false-positive result cannot occur, as the histological examination of the SLN is unchanged by the results from any additional nodes removed at systematic lymphadenectomy.