Efficacy of Chemoradiation and Patterns of Practice in the Treatment of Locally Advanced Endometrial Stromal Sarcoma: An NCDB Population-Based Study

Abstract

Endometrial stromal sarcoma (ESS) is a rare neoplasm of the mesenchymal elements of the uterus. Radical surgery is the primary therapy for non-metastatic disease. The role of adjuvant therapy after surgery remains controversial and is mostly consistent with systemic therapy. Given the rarity of this disease, it was hypothesized that analysis of current patterns of practice and their efficacy, as identified in a large hospital-based registry dataset, could inform more uniform and effective approaches to management. The National Cancer Database (NCDB) was queried for patients with Stage III-IVA ESS age 18 to 90 years diagnosed from 2001-2015 and treated with surgery +/- RT +/- chemotherapy. Patients were excluded if vital status was unavailable, if they received chemotherapy or RT alone, or if they died within 90 days of surgery. Fisher’s exact test and multivariate logistic regression were performed to analyze factors associated with delivery of a particular treatment modality, with corrections made to control the family-wise error rate. A survival analysis was performed using Kaplan-Meier estimates, a Cox proportional hazards regression model, and log-rank testing. A total of 214 patients met search criteria with a median age of 50 (19-87) years; 89 (41.5%) patients received surgery alone, 35 (16.4%) received surgery + RT, 68 (31.8%) received surgery + chemotherapy, and 22 (10.3%) received tri-modality therapy. Median follow-up was ∼27 months. Overall survival for all patients was 62% at 3 years, and 56% at 5 years. Age <50 years, low grade, and administration of RT were associated with improved overall survival (p < 0.001). For patients with high grade disease, the addition of RT to surgery and chemotherapy was associated with significant survival benefit (p < 0.004). Number of nodes examined in high-grade patients was associated with improved survival. Mean time from diagnosis to the start of RT was 107 (18-399) days, and the mean time between surgery and the start of RT was 90.5 (8-379) days. These were not significant predictors of survival. Integrated Cancer Network facilities had higher rates of RT use at 40% vs. ∼20-25% at other facilities. East North Central and East South Central census regions had higher rates of RT use at 40-47% vs. 11-29% in other regions. The addition of RT to surgery for patients with locally advanced endometrial stromal sarcoma appears to provide a survival benefit, particularly in patients with high-grade histology receiving surgery and chemotherapy. The rate of RT utilization was low overall, and there was significant variability in dose delivered and timing of treatment. More robust multi-institutional studies are required to better elucidate appropriate management in this cohort.