Effects of Facility Volume on Pathologic Complete Response and Overall Survival in Patients With Esophageal Adenocarcinoma Receiving Trimodality Therapy

Abstract

The standard of care for resectable locally advanced esophageal adenocarcinoma is tri-modality therapy. We hypothesize that centers with more expertise in delivering all aspects of neoadjuvant treatment consisting of chemoradiation followed by surgical resection at the same high volume facility may have improved outcomes, potentially through an increase in rates of pathologic complete response (pCR). The National Cancer Data Base was queried for patients undergoing tri-modality therapy at a single institution from 2004-2014. These patients received multiagent chemotherapy and radiation followed by surgery at the same Commission on Cancer facility. Facility volume was categorized into halves, as either a high (HVC) or low volume center (LVC), based on the treatment incidence and unique facility identifiers. Predictors of overall survival (OS) were analyzed with univariate and multivariate methods using Kaplan-Meier (compared via log-rank) and Cox proportional hazards analysis. We identified 2268 patients with esophageal adenocarcinoma in the NCDB. Median follow-up was 25.6 months. Patients who underwent tri-modality therapy at HVCs had an improvement in 3 year OS (53% vs. 46%) and estimated median survival (41 vs. 31 months, p=0.006), when compared to LVCs on univariate analysis (UVA). On multivariate analysis (MVA), HVC continued to predict for OS (HR 0.62, 95% CI 0.48-0.81, p<0.001). Other factor significantly affecting OS on MVA included female gender (HR 0.44, 95% CI 0.27-0.70, p<0.001). There was a trend towards improved survival for patients treated at Comprehensive Community Cancer Centers (HR 0.31, 95% CI 0.10-1.00, p=0.052) and Academic Comprehensive Cancer Centers (HR 0.35, 95% CI 0.11-1.13, p=0.080). pCR was significantly associated with survival on UV (HR 0.48, 95% CI 0.39-0.61, p<0.001) and MV analysis (HR 0.47, 95% CI 0.36-0.60, p<0.001). Patients with a pCR had improvement in their 3 year OS (60% vs. 36%) and estimated median survival (61 vs. 25 months) on UVA (p<0.001). There was no significant difference in percentage of patients with pCR based on high vs. low facility volume, 38.0% and 37.7% respectively (p=0.931). There was no significant difference in 30 or 90 day mortality after surgery between high and low volume centers (p=0.210 and p=0.080). This NCDB analysis is the first of its kind in esophageal adenocarcinoma to demonstrate significant effect on facility volume of tri-modality therapy on overall survival. High volume centers are associated with a significantly improved overall survival. This survival difference is not explained by a difference in pathologic complete response rates or 30 and 90 day post-surgery mortality.