Lymphadenectomy in endometrial cancer

The Lancet, Volume 373, Issue 9670, Pages 1169 - 1170, 4 April 2009 <Previous Article|Next Article>doi:10.1016/S0140-6736(09)60676-0Cite or Link Using DOI

Frédéric Amant a, Patrick Neven a, Ignace Vergote a

The ASTEC study group1 conclude that a systematic lymphadenectomy in endometrial cancer cannot be recommended as a routine procedure because of lack of benefit in terms of recurrence-free and overall survival. However, there are several reasons why the ASTEC trial did not show improved overall survival with routine lymphadenectomy.

First, the number of lymph nodes resected was insufficient in many patients. Although the median number resected overall was 12, 35% of patients in the lymphadenectomy group had nine or fewer lymph nodes removed. Cragun and colleagues2 showed that removal of more than 11 pelvic nodes had an effect on overall survival. Chan and colleagues3 showed that, in intermediate-risk and high-risk endometrial cancer, patients with more than 10 nodes harvested have an improved outcome.

Second, many patients with low-risk endometrial carcinoma, and hence a low risk of lymph-node involvement, were included (eg, only 41% had stage IC—IIB disease, and only 22% presented with poorly differentiated tumours). The high rate of inclusion of low-risk patients and the low number of lymph nodes removed are the reasons for the low rate of involved lymph nodes seen in the lymphadenectomy group (9%).

Third, the study group did not assess the para-aortic nodes. However, up to 67% of patients with lymph-node metastases have involved para-aortic nodes.4 Fourth, the ASTEC trial was too small to detect an overall survival difference because the expected proportion of isolated pelvic lymph-node recurrences is as low as 2—3% in early endometrial carcinoma.

In conclusion, we believe that there is still an indication to do a comprehensive lymphadenectomy to select patients at high risk of pelvic side wall recurrence. The selection of patients for a lymphadenectomy should be based on myometrial invasion, grade, and diameter of the tumour.5

We declare that we have no conflict of interest.