Objectives We investigated factors associated with cause-specific survival (CSS) after isolated lymphatic recurrence (ILR) in endometrial cancer (EC).

Methods We identified patients who developed ILR among 4,216 EC patients surgically treated at the Mayo Clinic between 1984 and 2017. ILR was defined as the first and unique evidence of recurrence in lymph node-bearing areas (with or without (±) vaginal recurrence). Univariate and multivariable Cox regression analysis was used to evaluate factors associated with CSS after ILR.

Results We observed 70 cases of ILR: 12 pelvic, 15 paraaortic, 14 pelvic and paraaortic, and 29 distant (± pelvic and/or paraaortic). Most women (90.0%) underwent pelvic and/or paraaortic lymphadenectomy during primary surgery, and 68.3% had positive nodes. Among 70 patients, 50 died of disease with median survival after ILR of 1.4 years. Patients who did not die of EC had a median follow-up after ILR of 6.6 (IQR 4.8–10.0) years. By univariate analysis, histologic grade, lymphovascular space invasion, ILR site, concomitant vaginal recurrence, and ILR treatment were significantly associated with CSS after ILR. CSS after ILR was not associated with primary lymphadenectomy, stage, or adjuvant therapy. Results of the multivariable analysis are reported in the Table.

Conclusions Histologic grade 2 or 3 of the primary tumor and concomitant recurrence in the pelvic and paraaortic lymph node basins or at the vaginal cuff were independent predictors of poor CSS after ILR. The choice to surgically treat ILR in some patients was associated with improved CSS.