Novel Calculator to Estimate Overall Survival Benefit from Neoadjuvant Chemoradiation in Patients with Esophageal Adenocarcinoma.

BACKGROUND

Our group reported that patients with clinically node-negative esophageal adenocarcinoma do not derive overall survival (OS) benefit from neoadjuvant chemoradiation (nCRT) compared with clinically node-positive patients. The aim of this study was to develop a calculator that could more easily identify which patients derive OS benefit from nCRT.

STUDY DESIGN

Using the National Cancer Data Base (2006 to 2012), patients with clinical status T1b to T4a, N-/+, M0 adenocarcinoma of the esophagus who underwent resection were selected. Of this cohort, 80% were randomly selected to develop and test the prediction model using Cox regression. The remaining 20% were used to internally validate the model, and performance was evaluated using receiver operating characteristic curves and area under the curves.

RESULTS

A total of 8,974 patients met study criteria. Using the model testing cohort (7,179 patients), variables that were independently associated with OS in multivariable analysis were included in the model. These variables included Charlson-Deyo comorbidity score, tumor grade, clinical T and N status, and nCRT before surgery. Factors associated with increased risk of death were higher grade and higher T or N status. Receipt of nCRT was associated with improved OS. After validation, model performance showed an area under the curve of 0.630 and 0.682 for 1-year and 3-year OS, respectively.

CONCLUSIONS

A novel OS calculator was developed for esophageal adenocarcinoma that reasonably predicts which patients are expected to derive OS benefit from nCRT. This tool can be helpful in determining OS benefit from nCRT to assist with treatment decision making.