Abstract
BACKGROUND: This study aims to investigate the causes of death in patients with esophageal cancer (EC) and report reliable and accurate estimates of adjusted conditional overall survival (COS). METHODS: We retrieved data on patients aged 18 years or older who were diagnosed with EC between 1975 and 2016 from the Surveillance Epidemiology End Results (SEER) registry. We estimated COS by using an inverse probability weighting method to adjust for the available covariates. Linear trends were analyzed via a weighted linear regression. RESULTS: A total of 40,142 confirmed patients were included in the final analysis. Of these, 20,971 were diagnosed with esophageal squamous cell carcinoma, and 19,171 were diagnosed with esophageal adenocarcinoma. EC was the leading cause of death, followed by heart disease (5.2%), stomach cancer (3.2%), and other diseases (2.9%). Five-year adjusted COS, given that patients had already survived 36 months was improved from 0.50 to 0.87 (P(Difference)<0.001) in comparison with the 5-year conventional overall survival. As the survival time increased from 0 to 48 months, the 5-year adjusted COS improved significantly. The adjusted conditional EC-specific survival also showed similar trends. CONCLUSIONS: For patients with EC who had survived over time, the 5-year adjusted COS and the conditional EC-specific survival improved dramatically. Other causes of death in EC patients should also be considered.