Although multimodality therapy has been shown to improve outcomes for patients with high-risk endometrial carcinoma, little consensus has been reached on the type and timing of adjuvant therapies. We sought to determine whether sequence of adjuvant therapy is associated with overall survival in high-risk endometrial cancer. Patients with stage I-IVA endometrial carcinoma diagnosed from 2004 to 2015, and treated with surgery, chemotherapy, and radiation were identified in the National Cancer Database. Adjuvant treatment was categorized as sequential radiation followed by chemotherapy (RT-CT), concurrent chemoradiation (CCRT), or sequential chemotherapy followed by radiation (CT-RT). RT-CT was defined as CT start >7 days after RT start, CCRT was defined as CT start within 7 days of RT start, and CT-RT was defined as RT start >7 days after CT start. Analyses were repeated using 14- and 21-day cutoffs, and yielded similar results. Survival analyses were performed using Kaplan-Meier estimates, log-rank test, and Cox proportional hazards regression. A total of 17,070 patients were identified, including 12,402 (72.7%) treated with RT-CT, 2,153 (12.6%) treated with CCRT, and 2,515 (14.7%) treated with CT-RT. Median follow-up for the cohort was 44.3 months. Five-year overall survival (OS) by adjuvant treatment regimen was 77.3% (95% CI 76.4%-78.2%), 74.3% (95% CI 72.0%-76.3%), and 74.4% (95% CI 72.5%-76.3%), respectively (p < 0.001). On multivariate Cox regression, sequence of adjuvant therapy (HR 1.09, p < 0.001), higher grade, (HR 1.06, p < 0.001), higher stage (HR 1.18, p < 0.001), government insurance (HR 1.10, p < 0.001), and non-white race (HR 1.07, p = 0.007) were independently associated with worse overall survival. Higher income (HR 0.81, p < 0.001) and >9 regional nodes resected (HR 0.84, p < 0.001) were independently associated with improved overall survival. With the exception of race, these covariates remained significantly associated with OS after correcting for multiple comparisons. Of note, treatment facility type and lymphovascular invasion were not associated with OS. When stratified by stage, the benefit of adjuvant RT-CT was limited to patients with stage III disease. Patients with stage III disease who received adjuvant RT-CT had 5-year overall survival of 73.9% versus 69.6% and 69.9% with CCRT and CT-RT, respectively (p < 0.001). In contrast, there was no difference in 5-year OS (49.7% for RT-CT, 48.3% for CCRT, and 50.3% for CT-RT) among stage IVA patients (p = 0.806). Our findings suggest a survival benefit with adjuvant RT-CT compared to CT-RT or CCRT in patients undergoing trimodality therapy for endometrial cancer. This survival benefit was exclusively seen in patients with stage III disease.
Read full abstract