Clinical Understaging, Treatment Response, and Survival Among Esophageal Adenocarcinoma Patients

Selecting appropriate management for patients with esophageal adenocarcinoma (EA) is predicated on accurate clinical staging information. Inaccurate information could lead to inappropriate treatment and suboptimal survival. We investigated the relationship between staging accuracy, treatment, and survival. This was a national cohort study of EA patients in the National Cancer Data Base (2006-2015) treated with upfront resection or neoadjuvant therapy (NAT). Clinical and pathological staging information was used to ascertain staging concordance for each patient. For NAT patients, Bayesian analysis was used to account for potential downstaging. We evaluated the association between staging concordance, receipt of NAT, and survival through hierarchical logistic regression and multivariable Cox regression. Among 7635 EA patients treated at 877 hospitals, 3038 had upfront resection and 4597 NAT followed by surgery. Relative to accurately staged patients, understaging was associated with a lower likelihood (odds ratio [OR] 0.04 95% confidence interval [CI] 0.02-0.05) while overstaging was associated with a greater likelihood of receiving NAT (OR 1.98 [1.53-2.56]). Relative to upfront surgery, treatment of cT1N0 patients with NAT was associated with a higher risk of death (HR 3.08 [2.36-4.02]). For accurately or overstaged cT3-T4 patients, NAT was associated with a lower risk of death whether downstaging occurred (ypN0 disease-HR 0.67 [0.49-0.92]; N+ disease-HR 0.55 [0.45-0.66]) or not (ypN + disease-HR 0.78 [95% CI 0.65-0.93]). Clinical understaging is associated with receipt of NAT which in turn may have a stage-specific impact on patients' survival regardless of treatment response. Guidelines should account for the possibility of inaccurate clinical staging.