Background: While the Royal Marsden Hospital (RMH) prognostic model had been established in Western patients with advanced gastric cancer(AGC) (Chau, JCO, 2004), we identified four risk factors; performance status (PS) ≥ 1, number of metastatic sites ≥ 2, no prior gastrectomy, and serum alkaline phosphatase (ALP) > normal range, and proposed Japan Clinical Oncology Group (JCOG) prognostic index based on the number of these factors (good risk 0,1; moderate risk 2,3; poor risk 4) in the JCOG9912 trial (Takahari, Oncologist, 2014). Methods: To validate our JCOG prognostic index, it was applied to the combined cohort of other two Japanese phase III trials for AGC; SPIRITS trial comparing between S-1 and S-1 + Cisplatin (SP) (Koizumi, Lancet Oncol., 2008) and G-SOX trial comparing between SP and S-1 + Oxaliplatin (SOX) (Yamada, Ann. Oncol., 2015). Results: 936 (94.5%) out of the totally 990 patients randomized in these trials, whose complete data were available for multivariate analyses, were included in the present study (S-1 n=150, SP n=470, SOX n=316). The median survival time (MST) for all patients was 13.2 months, and each risk factor of the JCOG index remained a significant prognostic factor (Table). Three risk groups categorized by the JCOG prognostic index identified highly significant survival differences (compared with good risk group [n=338]; HR, [95%CI];1.71[1.46-2.01], p<0.001 in moderate risk group [n=537] and 3.32[2.47-4.46], p<0.001 in poor risk group [n=61]), associated with MST of 17.2, 12.0 and 7.8 months, respectively. Good and moderate risk groups categorized by RMH index showed no significant survival differences (compared with good [n=314]; HR, [95%CI];1.10[0.94-1.28], P =0.23 in moderate [n=598] and 1.98[1.27-3.10], P =0.003 in poor [n=24]), associated with MST of 14.9, 13.0 and 6.9 months, respectively. Conclusions: JCOG prognostic index was validated and can be used for patient stratification in the future clinical trials.