Abstract
BACKGROUND: We evaluated the efficacy of adjuvant durvalumab after neoadjuvant concurrent chemoradiotherapy (CCRT) in patients with esophageal squamous cell carcinoma (ESCC). PATIENTS AND METHODS: This randomized, double-blind, phase II study included patients with ESCC who underwent curative surgery after neoadjuvant CCRT. Patients were randomized to receive either durvalumab (20 mg/kg/i.v. every 4 weeks for 12 months) or placebo in a 1:1 ratio and were stratified by age and pathologic tumor stage. The primary endpoint was disease-free survival (DFS). RESULTS: Between March 2016 and June 2018, 86 patients were randomized to the durvalumab (n = 45) or placebo (n = 41) arm. The median follow-up duration was 38.7 months. There was no difference in DFS [hazard ratio (HR) 1.18, 95% confidence interval (CI) 0.62-2.27, P = 0.61] or overall survival (HR 1.08, 95% CI 0.52-2.24, P = 0.85) between the two arms. Subgroup analysis was performed for patients for whom the post-CCRT programmed death-ligand 1 (PD-L1) expression profile could be assessed (n = 54). In the PD-L1-positive group, based on tumor proportion score ≥1%, durvalumab was associated with longer overall survival compared with the placebo (36-month survival rate: 94% versus 64%; HR 0.42, 95% CI 0.10-1.76), while in the PD-L1-negative group, it was associated with shorter overall survival (42% versus 55%; HR 1.53, 95% CI 0.48-4.83), showing the tendency of interaction between post-CCRT PD-L1 status and adjuvant durvalumab therapy for overall survival (interaction P = 0.18). CONCLUSIONS: We failed to demonstrate that adjuvant durvalumab improved survival after neoadjuvant CCRT in patients with ESCC. However, post-CCRT PD-L1 expression could predict the survival of patients who receive adjuvant durvalumab after neoadjuvant CCRT, which needs to be validated.