Abstract
The study objectives were to identify variables associated with the use of adjuvant chemotherapy among patients with node-positive esophageal cancer who received induction therapy and to evaluate its relationship with overall survival.
Treatment data for patients with esophageal cancer receiving induction chemotherapy ± radiotherapy and esophagectomy were abstracted from the National Cancer Data Base. Pathologic node-positive patients were dichotomized by whether they received 2 or more cycles of adjuvant chemotherapy or none. Kaplan–Meier survival curves were generated, and a Cox proportional hazards model was done to identify factors associated with overall survival.
From 2006 to 2012, 3100 patients had pathologic positive nodes after induction therapy and esophagectomy. A total of 2625 patients (84.7%) did not receive adjuvant chemotherapy, and 475 patients (15.3%) did. N3 nodal stage was associated with an increased likelihood of receiving adjuvant chemotherapy (reference: N1, odds ratio, 1.82, 95% confidence interval, 1.15-2.97, P = .01), whereas increasing age (by year, odds ratio, 0.97, confidence interval, 0.96-0.98, P < .001), induction chemoradiation therapy (reference: induction chemotherapy, odds ratio, 0.39, confidence interval, 0.30-0.52, P < .001), and increasing inpatient length of stay after esophagectomy (per day: odds ratio, 0.98, confidence interval, 0.97-0.99, P = .007) were associated with a decreased likelihood. Patients receiving adjuvant chemotherapy had improved overall survival at each pathologic nodal stage: 31.6 months versus 22.7 months for N1 disease (P < .001), 32.4 months versus 19.2 months for N2 disease (P = .035), and 19.5 months versus 10.4 months for N3 disease (P < .001). Adjuvant therapy was independently associated with decreased mortality hazard (hazard ratio, 0.69, 95% confidence interval, 0.57-0.83, P < .001).
Patients receiving adjuvant chemotherapy after induction therapy and esophagectomy show a survival benefit at all positive nodal stages. Prospective studies may help further delineate this benefit.
