Progression-free survival (PFS) and time to progression (TTP) have been reported to correlate with overall survival (OS) in several types of cancers. To our knowledge, however, their use in the evaluation of new agents for AGC has not been investigated. We evaluated the potential of PFS and TTP to act as surrogates of OS in clinical trial settings. Randomized trials of systemic chemotherapy for advanced gastric cancer were identified by comprehensive electronic and manual search. Correlations between PFS/TTP and OS were evaluated. Thirty-six trials with a total of 83 treatment arms and 10,484 patients were selected for analysis. The nonparametric Spearman rank correlation coefficient (ρ) between median PFS/TTP and OS was 0.70 (95% CI, 0.59 to 0.82) and the correlation coefficient between hazard ratios in PFS/TTP and OS was 0.80 (95% CI, 0.68 to 0.92). Correlation tended to be higher in trials reporting PFS (ρ = 0.85; 0.72-0.97) than in those reporting TTP (ρ = 0.60; 0.24-0.97), trials in Non-Asian countries (ρ = 0.80; 0.61-0.99) than Asia (ρ = 0.67; 0.39-0.94), trials in patients with measurable lesions only (ρ = 0.91; 0.77-1.00) than in those including non-measurable lesions (ρ = 0.71; 0.50-0.93), albeit that none of these differences was significant. Our results indicate that improvements in PFS/TTP in advanced gastric cancer strongly correlate with improvements in OS. Further research is needed to clarify the surrogacy of PFS/TTP for OS or the role of PFS as the true end point in future randomized clinical trials of chemotherapy for AGC.