Predictors of Pelvic Recurrence in Stage IB Grade 2-3 Endometrioid Endometrial Adenocarcinoma Treated with Post-Operative Vaginal Brachytherapy

Abstract

Optimal adjuvant management of surgically staged FIGO stage IB grade 2 (G2) and grade 3 (G3) endometrial endometrioid adenocarcinoma (EAC) is controversial. The objective of this study is to report outcomes utilizing adjuvant vaginal brachytherapy (VB) without pelvic radiotherapy (RT) for patients with surgically staged IB G2 and G3 EAC and to analyze rates and predictors of pelvic recurrence. Retrospective review of an institutional database identified patients diagnosed between 2005-2015 with 2009 FIGO stage IB G2 and G3 EAC. All patients underwent comprehensive surgical staging and VB as the sole adjuvant RT as per institutional guidelines during the study period. Disease free survival (DFS) was calculated from the date of primary surgery to date of last follow up or recurrence, and a minimum of 6 months follow-up was required for inclusion. Kaplan-Meier methods were used to estimate DFS. Pearson chi-squared test was used to determine association of pathologic variables and pelvic recurrence. A total of 78 patients were identified for inclusion. The median age at diagnosis was 67 (range, 42-93). Surgical staging included hysterectomy, bilateral salpingo-oopherectomy, pelvic washings, pelvic lymph node dissection (98.7%), and paraaortic node sampling (78.2%). Fifty-five patients (71%) had G2 EAC and 23 (29%) had G3 EAC. Lymphovascular invasion (LVI) was observed in 37 (47%), lower uterine segment involvement (LUSI) in 38 (49%), and outer 1/3rd myometrial invasion (MMI) (defined as ≥ 67% MMI) in 39 (50%). Median number of pelvic lymph nodes removed was 18. The majority (88.5%) received VB alone and 11.5% received VB in combination with chemotherapy. Median follow up was 45.5 months (range 6.3-151.8) and 3-year estimated DFS was 84.7% for the entire cohort. There were a total of 11 recurrences, of which 3 (27.3%) were distant only. The remainder (72.3%) had a pelvic component including 5 which were pelvic and distant and 3 which were isolated to the pelvis. Median time to recurrence was 27.8 months (range, 11.5 - 46.9). On univariate analysis, G3 versus G2 (p = 0.600), LVI present versus absent (p = 0.388), and ≥20 versus <20 pelvic nodes dissected (p = 0.051) were not associated with higher pelvic recurrence risk. Presence of LUSI (p = 0.03) and outer 1/3rd MMI (p = 0.03) were associated with increased risk of pelvic failure. Stage IB G2 and G3 patients treated with adjuvant VB following surgical staging overall fared well, and isolated pelvic recurrences were few in number. While grade and LVI were not associated with increased pelvic failure in this cohort, the association of LUSI and outer 1/3rd MMI is intriguing. Further investigation into risk factors for pelvic recurrence in stage I EAC would be valuable to identify subgroups which benefit the most from pelvic RT.