To determine which patients with stage III endometrial cancer receiving adjuvant chemotherapy derive benefit from the addition of vaginal brachytherapy, as stage III is a highly heterogeneous population with substantial variations in practice.

Patients with FIGO stage III endometrial carcinoma diagnosed 2004-2016 who underwent at least total hysterectomy and adjuvant multiagent chemotherapy were identified in the National Cancer Database. The primary outcome was overall survival according to receipt of brachytherapy, stratified by histologic type, pathological features, and status of pelvic external beam radiotherapy (EBRT), and analyzed using the Kaplan-Meier method and Cox multivariable regression.

In total, 9369 patients were identified (24% stage IIIA, 5% stage IIIB, 71% stage IIIC; 61% endometrioid, 39% nonendometrioid histology), and 28% received brachytherapy. In the endometrioid cohort, brachytherapy was associated with a 5% absolute increase in 3-year overall survival (87% vs. 82%, p < 0.0001), which persisted in multivariable analysis (adjusted hazard ratio 0.74, 95% confidence interval 0.64-0.84, p < 0.0001). The benefit of brachytherapy was greater in patients not also receiving EBRT, and in patients with vaginal/parametrial extension, grade 3 disease, lymphovascular invasion, and/or deep myometrial invasion. In the nonendometrioid cohort, brachytherapy was associated with a significant survival benefit in univariable but not multivariable analysis, regardless of EBRT status or pathological features.

Factors predictive of brachytherapy benefit were endometrioid histology and pathological risk factors for local recurrence. Additionally, brachytherapy appeared more beneficial in patients not already receiving pelvic EBRT. Further research is warranted to determine which stage III patients may be best served by brachytherapy, EBRT, or both.