Event-Free Survival as a Surrogate for Overall Survival in Gastric and Gastroesophageal Junction Adenocarcinoma: A Meta-analysis in the Neoadjuvant ? Adjuvant Setting

Purpose: This study assessed the trial-level association between event-free survival (EFS) and overall survival (OS) in gastric or gastroesophageal junction (GEJ) adenocarcinoma in the neoadju-vant +/- adjuvant settings.Experimental Design: A systematic literature review was con-ducted to identify randomized controlled trials (RCT) that evalu-ated neoadjuvant therapies with or without adjuvant therapies for gastric or GEJ adenocarcinoma. A meta-analysis was performed using weighted linear regressions of the treatment effect of OS on the treatment effect of EFS. The coefficient of determination (R2) and associated 95% confidence interval (CI) were used to evaluate the association between treatment effects of EFS and OS. The threshold used for defining good trial-level surrogacy was a correlation coefficient (R) of 0.8 or R2 of 0.65, based on prior literature. Sensitivity analyses were performed to assess the robustness of the association with divergent study designs, including study popula-tion, inclusion of adjuvant therapy, and definitions of EFS and OS.Results: The main analysis included 16 comparisons from 15 RCTs. The log(HR) of EFS was a significant predictor of log(HR) of OS, with an estimated coefficient of 0.72 (P < 0.001) and R2 = 0.75 (95% CI, 0.49-0.95), indicating that EFS was a good surrogate outcome for OS. The results of the sensitivity analyses were con-sistent with the primary results, with R2 ranging from 0.76 to 0.89.Conclusions: This study suggests that EFS is a good surrogate for OS in gastric or GEJ adenocarcinoma in the neoadjuvant +/- adjuvant setting.