The contemporary view of endometrial cancer suggests that most women, fortunately, have low-risk disease, with a low probability of recurrence following hysterectomy. Identifying the smaller subset of patients with disease spread at presentation or who may benefit from adjuvant therapy remains the challenge. Risk of disease recurrence and survival has been related to simple models evaluating uterine characteristics, nodal status, and use of adjuvant therapy.

Additional data based on patient age, tumor histology, and specific type of adjuvant therapy may fine-tune risk models even further. In 2011, the extent of disease spread, or stage, remains the single most important factor for defining use of adjuvant therapy and prognosis. Of the methods to assess disease spread, lymph node dissection is the best technique for identifying the approximately 9% of patients with otherwise unrecognized disease spread. Compared with patients with node negative disease, patients with node-positive disease carry different prognoses and largely receive different therapies. Compared with those patients with unresected/unrecognized positive nodes, those node positive patients treated with lymphadenectomy perhaps achieve better result with the same therapies once use.

The use of lymph node dissection as a routine practice or a selective one is the source of considerable debate and active research and largely relates the marginal benefit that may be achieved in a particular population.

In the late 1990s to the early 2000s, a paradigm based on increasing use of lymphadenectomy was discussed in the literature. Lymphadenectomy was suggested to be a therapeutic endeavor and could lead to a substitution of vaginal cuff brachytherapy for pelvic radiation therapy. Data also emerged suggesting that nodal status could be used as a fulcrum; node-negative patients would receive surveillance, node-positive patients would receive adjuvant therapy. Lymph node status as a single variable, however, probably produces an incomplete picture of disease behavior.

Understanding the importance of grade, depth of myometrial invasion (DOI), and lymphovascular space invasion (LVSI) in predicting recurrence and survival vs nodal status as a single variable has been difficult, and uterine characteristics may confound potential advantages of nodal dissection. Data from the Gynecologic Oncology Group study 99 demonstrated that in the setting of stage I-II disease with negative lymph nodes, information on patient age, LVSI, DOI, and tumor grade could define a population of patients with a 27% risk of recurrence treated by surgery alone, including 19% of patients recurring at a distant sites. In a population based study from Canada, Kwon et al suggested that high-risk uterine factors were more important determinants of survival than nodal status. Both studies have been criticized for the extent and quality of the nodal assessment, however.

The extent to which the quality of lymph node dissection alters outcomes is a source of controversy. The number of nodes removed (a surrogate of the completeness of dissection) and the extent of dissection (pelvic and/or paraaortic) are variables that may have an impact on outcomes but are seldom controlled. Both recent prospective randomized trials evaluating hysterectomy with or without routine pelvic lymphadenectomy were criticized for the absence of paraaortic nodal dissections. Both studies showed that pelvic lymphadenectomy did not alter survival.

A subsequent retrospective Japanese series suggested that a survival benefit may be achieved by removing paraaortic nodes in addition to pelvic nodes. That trial, however, did not control well for postoperative use of chemotherapy, another variable that may blur the outcomes associated with lymphadenectomy. The ASTEC (A Study in the Treatment of Endometrial Cancer) trial was criticized by the fact that 8% of patients on the nodal dissection arm had no nodal dissection at all, and 35% had less than 10 lymph nodes removed.

Sharma and colleagues have evaluated a large population database and suggested that the knowledge of lymph node status changes use of adjuvant therapy in select patient populations. In the Surveillance, Epidemiology, and End Results (SEER) analysis, a higher lymph node number (≥10) was associated with less use of pelvic radiation, perhaps suggesting greater confidence in avoiding pelvic radiation in patients with thorough staging. What factors contribute to a less extensive resection (patient and physician) and how these may alter outcomes are, again, unknown. The SEER data do not report the extent of pelvic and/or paraaortic dissections.

It is interesting to note that in the SEER analysis by Sharma and colleagues, lymphadenectomy did not alter use of adjuvant pelvic radiation in low- or high-risk patients but was associated with a change in the type of radiation therapy used in intermediate-risk patients (more common use of vaginal brachytherapy). Perhaps striking, however, was the finding that regardless of age group (<60 or ≥60 years old), patients who underwent a lymph node dissection with stage IB, Gr2-3 disease received any radiation (pelvic or vaginal) more commonly compared with those who did not undergo lymph node dissection (25% vs 20%). This interesting finding from the data suggests that we may be overtreating patients or that lymphadenectomy is not altering our decision making in terms of using any radiation or not.

Is the study by Sharma and colleagues then another indictment of lymphadenectomy? It would seem the authors’ discussion that “optimal adjuvant treatment for most women with endometrial cancer is unknown” is the crux of the lymphadenectomy debate. If patients are at low risk for recurrence or do not benefit from adjuvant therapy, then there is probably a limited role for lymphadenectomy in this population as well. In patients at high risk for recurrence, regardless of nodal status, does the effect of adjuvant therapy blunt benefit from lymphadenectomy?

Today lymphadenectomy is the discriminating test for risk, but tomorrow tissue or serum biomarkers may be better tools. Demonstrating the value and effect of lymph node dissection continues to be difficult because of the multiple variables that are related to use of adjuvant therapy, recurrence, and survival. Many of us think that lymphadenectomy can help at least some subset of our patients and believe that the information from the procedure fine-tunes our use of therapy. The data Sharma and colleagues support the belief that a lymph node dissection alters the use of postoperative radiation but perhaps not as much as we would like to think. Given the variety of variables that may confound the single question of what does a nodal dissection offer, trials that control for the extent of nodal dissection and the use of postoperative therapy are essential.