Delayed Esophagectomy is Associated With Inferior Survival: A National Cancer Database Study

Abstract

Background: Locoregionally advanced esophageal cancer is typically treated with neoadjuvant chemoradiation followed by surgery 4 to 8 weeks later. Occasionally surgery is delayed >12 weeks; outcomes of this approach are not well studied. We hypothesized that delayed esophagectomy after chemoradiation would have inferior long-term overall survival relative to planned trimodality esophagectomy. Methods: Adult patients with locally advanced esophageal cancer (T2−4aN0M0, T0−4aN+M0) who received multi-agent chemotherapy, radiation, and esophagectomy were identified in the 2018 National Cancer Database. Esophagectomy performed within 90 days from end of chemoradiation were categorized as “trimodality” and those ≥90 days were categorized as “delayed.” Primary outcome was overall survival measured using Kaplan-Meier estimates and Cox proportional hazard models. Secondary outcomes included surgical margin status, hospital length of stay, and readmission. Results: Included were 19 698 patients, 3905 (19.8%) “delayed.” Median time to surgery for trimodality patients was 51 days (IQR 41-63) versus 110 days (IQR 98-131) for delayed patients. Delayed patients tended to be older, non-white, have non-private insurance, and have more comorbidities. Overall survival was shorter for delayed patients (34.8 months) versus trimodality patients (43.1 months, P ≤ .001). In multivariable analysis, delay was associated with inferior overall survival (HR 1.15, 95% CI 1.08-1.23). Length of stay and readmission rate were similar between cohorts, but delay was associated with a higher rate of positive surgical margins (6.7% vs 4.6%, P ≤ .001). Conclusions: In the National Cancer Database, delayed esophagectomy is associated with inferior long-term survival. Nonetheless, delayed esophagectomy may be appropriate for select patients; further research is needed to identify the optimal approach.