Stop cancelling surgery for esophageal cancer: An analysis of treatment-related mortality in 60,000 patients.

Abstract

e16102 Background: The optimal management of squamous cell carcinoma (SCC) of the esophagus remains controversial. There is a growing perception that definitive chemoradiation (dCR) has equivalent outcomes as neoadjuvant chemoradiation followed by surgery (CRS). The esophagectomy is falsely regarded as adding no survival benefit, and only serves to expose patients to perioperative risk. We explore the role for surgery by evaluating treatment – related mortality risk with a focus on chemoradiation (dCR) compared to chemoradiation and surgery (CRS). We then performed sub-analyses by T-Stage, overall stage, surgical era (Before and After 2009), and patients who refused surgery to compare groups with similar clinical conditions. Methods: The National SEER database was queried for all records related to “Esophageal Cancer” (n = 74,918). Patients were excluded from this query if they were younger than 18 years old at diagnosis or had been diagnosed at autopsy or by death certificate. The sample was further restricted to only patients diagnosed after 2004. We evaluated patients for their hazard ratio (HR) for overall mortality by treatment type under multiple clinical conditions. Results: There were 59,754 patients with esophageal cancer; 19,806 patients had SCC and 34,484 patients had AC. The HR of mortality for patients undergoing dCR compared to CRS for SCC was 1.93. The overall hazard ratio of mortality for all patients was 2.15 and for AC was 2.39 (p < 0.0001). Patients with resectable T-stages (T1 – T3) also displayed a markedly higher risk of mortality for patients undergoing dCR compared to CRS at every T – Stage and for every cancer type. The lowest HR was 1.65 and the highest was 2.70 (p < 0.0001). For overall stage, there was again a markedly higher risk of mortality for patients undergoing dCR at every stage for every cancer type. The lowest increased hazard was 1.75 and the highest was 2.60 (p < 0.0001). After 2009, more patients underwent both dCR and CRS, but the rate of growth was higher for dCR. After 2009, for SCC patients, the HR for dCR increased from 1.82 to 2.15 and the HR for surgery alone decreased from 1.33 to 1.19. For all patients undergoing dCR, the HR increased from 2.07 to 2.47. The HR for surgery alone decreased from 0.86 to 0.72 (p < 0.0001). For AC patients, the HR increased from 2.37 to 2.78 and for surgery alone decreased from 0.72 to 0.58. Lastly, we identified a subgroup of 2,714 patients who were offered surgery, but did not undergo a procedure. The overall HR was 1.68, for SCC the HR was 1.54, and for AC the HR was 1.76. Conclusions: This study suggests that chemoradiation alone substantially increases the risk of death compared to chemoradiation with surgery in nearly all patients with resectable esophageal cancer. Research that supports non-surgical management should be critically evaluated, with particular attention to sample size, treatment protocols, and the impact of outdated operative mortality.