Lymph node sampling and survival in endometrial cancer

Objective. To determine the impact of pelvic lymph node sampling on surviva l in women with FIGO stage I and II endometrial adenocarcinoma. Methods. We reviewed data from the National Cancer Institute's Surveillance , Epidemiology, and End Results program on 9185 women with stage I endometr ial cancer and 881 women with stage II endometrial cancer. Life table analy sis was used to compare survival rates. Results. Overall, lymph node sampling did not appear to convey survival ben efit. The 5-year relative survival for 6363 women with stage I endometrial cancer who did not undergo lymph node sampling was 0.98, compared to 0.96 f or 2831 women who did undergo lymph node sampling at the time of hysterecto my, a nonsignificant difference. Lymph node sampling (LNS) was associated w ith increased survival among those with stage I, grade 3 disease, but not g rade 1 or grade 2. Women with stage I, grade 3 disease who underwent LNS ha d a relative 5-year survival of 0.89, compared to 0.81 for those who did no t undergo LNS (P = 0.0110). Nonsignificantly improved survival associated w ith LNS for women with grade 3 disease was observed in International Federa tion of Gynecology and Obstetrics stages Ib, Ic, and II. Conclusions. The observed survival benefit associated with lymph node sampl ing may be due to identification of women with more advanced endometrial ca ncer. Accurate determination of grade and extent of tumor is necessary to d elineate which patients may benefit from lymph node sampling at hysterectom y. Effective cooperation between surgical pathology and gynecology services may be required to ensure adequate examination of the hysterectomy specime n. A surgeon with expertise in performing lymph node sampling should be ava ilable if operative findings render lymph node sampling appropriate, (C) 19 98 Academic Press.